category
stringclasses 1
value | source
stringclasses 1
value | url
stringlengths 74
77
| text
stringlengths 16
2.15M
|
|---|---|---|---|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12710317&blobtype=pdf
|
# Comparison of urine and semen samples for HPV detection: a cross-sectional study in fertility clinic patients
Dayanara Delgado López, Andrea Cabrera-Andrade, Roque Rivas-Párraga, Pedro Gonzalez, Bernardo Crespo, Vivian Neira
## Abstract
Human papillomavirus (HPV) is a prevalent sexually transmitted infection linked to genital warts and various cancers. While effective screening methods exist for women, detecting HPV in men remains challenging due to asymptomatic infections and the lack of standardized, non-invasive diagnostic tools. This study aimed to assess the correlation between HPV DNA detection in urine and semen samples from men attending a fertility clinic in Ecuador. This cross-sectional study included 106 sexually active male patients referred for semen analysis at the Fertility Unit of Hospital del Río, Cuenca, Ecuador, between June 2024 and March 2025. Each participant provided paired semen and urine samples, which were analyzed for HPV genotyping. Sociodemographic and behavioral data were also collected. HPV DNA was detected in 15% of semen samples and 9.4% of urine samples. Both low-and high-risk genotypes were identified in semen and urine. In semen samples, high-risk HPV was detected in 9.4% and low-risk HPV in 5.7%. In urine, high-risk HPV was detected in 8.5% and low-risk HPV in 0.9% of samples. Among high-risk genotypes, HPV 58 was the most frequent in urine (20%), while HPV 81 was the most frequent in semen (19%). Genotypes such as HPV 16, 18, 39, 52, and 66 were detected exclusively in semen. Compared with semen sampling, urine sampling showed a sensitivity of 37.5% and a specificity of 95%. Agreement between the two sample types was fair (kappa = 0.39). Although urine sampling offers a non-invasive option, its limited sensitivity restricts its utility as an HPV detection method in asympto matic men. Semen analysis provides higher detection rates and may be more reliable for HPV screening in male fertility populations. IMPORTANCE This study addresses a critical knowledge gap regarding human papillomavirus (HPV) detection in men, a topic insufficiently covered in current control and prevention strategies. While screening and vaccination efforts in women have advanced considerably, the absence of standardized, well-accepted, and non-invasive methods for HPV detection in men limits the recognition of their roles as reservoirs and transmitters in the HPV transmission pathway. Evaluating alternative solutions for HPV testing, such as urine and semen sampling, not only contributes valuable scientific knowledge to improve detection in this population but also establishes the foundation for public health policies on male HPV testing. Ultimately, this work supports the goal of reducing global HPV prevalence and its associated diseases. KEYWORDS male self-sampling, non-invasive male screening, HPV genotyping H uman papillomavirus (HPV) is one of the most common sexually transmitted infections worldwide, associated with a spectrum of anogenital and oropharyngeal diseases, including genital warts and several cancers (1, 2). While HPV screening and prevention strategies have been well developed for women, particularly through cervical cytology and HPV DNA testing, HPV detection in men remains less standardized. Male
HPV infections often go undiagnosed due to the asymptomatic nature of the infection and the absence of a universally accepted, non-invasive screening method (3,4).
In line with current recommendations, HPV vaccination is primarily offered to girls according to the WHO (5). However, female-only vaccination is insufficient to eradicate HPV infection or to achieve the 90-70-90 target (6,7). According to Garolla et al., women have more than 80% probability of acquiring HPV infection, while men have a risk exceeding 90% (8). Currently, three vaccines approved by the US Food and Drug Administration (FDA) are available, all of which cover HPV types 16 and 18, among others (9). Vaccination of boys should be strongly encouraged, as the ultimate goal of HPV immunization is the eradication of cervical cancer and other HPV-related diseases through a gender-neutral vaccination strategy (10). In Ecuador, the Ministry of Public Health (MSP) has traditionally vaccinated only girls aged 9 to 11 years; however, since May 2024, vaccination has been extended to both sexes (11,12).
Traditional methods for HPV detection in men rely on clinician-collected swabs from the penile shaft, urethra, or anal region. However, these approaches are invasive, may cause discomfort, and have variable sensitivity depending on the anatomical site and sampling technique (13). Importantly, no universally validated or widely accepted method for HPV detection in the male population currently exists (14). In response to these limitations, urine sampling has emerged as a non-invasive, self-collectable alternative. Several studies have demonstrated that urine samples, particularly first-void urine, can detect HPV DNA in males with moderate sensitivity (15)(16)(17). For example, Golijow et al. detected high-risk HPV in 18.7% of urine samples from men in a high-risk population in Argentina, and Jin et al. found a prevalence of 35.1% among asymptomatic male partners of women with sexually transmitted infections (17,18).
Nonetheless, the clinical performance of urine-based HPV testing in men remains suboptimal. Aung et al. found that urine sample positivity varied significantly by circumcision status and the anatomical location of genital warts, with lower detection in circumcised men and those with penile shaft lesions (19). Similarly, a study performed by Koene et al. compared urine and penile swabs and reported higher detection in swabs (61.4%) than in urine (35.1%) (20). These findings are echoed in the review by Enerly et al. who emphasized the need for standardized protocols and noted that variability in sample processing and assay choice significantly affects detection rates (21).
Although urine has been widely studied, semen remains an underexplored but biologically plausible sample type for HPV detection in men. HPV DNA has been detected in semen in multiple studies, and a recent systematic review and meta-analysis by Laprise et al. reported prevalence rates ranging from 2% to 31% across populations and methodologies (22). The presence of HPV in semen has also been associated with impaired sperm quality, suggesting a potential role in fertility and viral shedding (23)(24)(25). Despite these implications, semen is rarely used in routine HPV screening or research protocols.
A small number of studies have begun to address this gap. Gupta et al. (26) included semen among several sample types for HPV genotyping and noted its potential utility in improving the accuracy of male risk profiling, while Melchers et al. demonstrated that HPV DNA could be successfully detected in semen using PCR techniques (16,26).
Given the limitations of urine sampling and the biological relevance of semen, there is a clear need to compare these two non-invasive sample types for their performance in HPV detection among men. The primary objective of this study was to evaluate the correlation between urine and semen samples for the detection of HPV DNA in male patients attending a fertility clinic. The aim was to determine whether urine could serve as a reliable, non-invasive alternative to semen for HPV genotyping in this population. To the best of our knowledge, this is the first study of its kind conducted in Ecuador.
## MATERIALS AND METHODS
## Study design and setting
This was an observational, cross-sectional, and analytical study conducted at the Fertility Unit of Hospital del Río in Cuenca, Ecuador. The study period spanned from June 2024 to March 2025. The sociodemographic characteristics and sexual history of the population can be observed in Table 1.
## Study population
Eligible participants were male patients referred to the Fertility Unit for semen analysis. All individuals were invited to participate by the attending fertility specialist. To be included in the study, participants had to meet the following criteria: be ≥18 years of age, be sexually active, have abstained from sexual activity for at least 72 h prior to sample collection, and have been referred for semen analysis (semen parameters are presented in Table 2 according to HPV status). Patients diagnosed with varicocele or with a history of vasectomy were excluded. A total of 106 patients were included, each of whom provided both a semen and a urine sample.
## Sample collection
Following informed consent, participants completed a standardized data collection form that included sociodemographic information and details regarding their sexual and reproductive history. Each participant received two sterile containers, one for urine and one for semen collection. Participants were instructed to first collect the urine sample without prior cleansing of the genital area, capturing the first-void urine until a volume of 50 mL was reached. Subsequently, a semen sample was obtained via masturbation. All specimens were anonymized using an alphanumeric code and promptly transported to the Molecular Biology Laboratory of the University of Cuenca. Urine samples were centrifuged within a maximum of 6 h post-collection to prevent degrada tion of biological material.
## HPV genotyping
Genomic DNA was extracted from both urine and semen samples for HPV genotyping. Urine samples were centrifuged at 3,000 rpm for 15 min to obtain a cellular pellet used for DNA extraction. For semen samples, 300 µL of ejaculate was processed. DNA extraction was performed using the Invitrogen commercial kit (Jetflex Genomic DNA Purification kit), following the manufacturer's protocol for both types of specimens.
HPV genotyping was performed using the 24 HPV Typing Kit (Jiangsu Mole Bio science), which enables the identification of HPV genotypes through real-time PCR. The identified genotypes were categorized as follows: low-risk types are (6,11,42,43, 44, and 81), and high-risk types (16,18,26,31,33,35,39,45,51,52,53,56,58,59, 66, 68, 73, and 82). All samples were processed with the inclusion of an internal amplification control, as well as positive and negative controls. Amplification was carried out using the Exicycler 96 thermal cycler (Bioneer). The amplification protocol followed the manufac turer's recommendations.
## Data analysis
All data from the data collection form were entered into an Excel spreadsheet for analysis. Descriptive statistics were used to characterize the study population, employing the R software package for statistical analysis. To assess the level of agreement between urine and semen samples for HPV detection, the kappa statistic was calculated. A kappa value of 0 indicated no agreement better than chance, while a value of 1 represented perfect agreement. Intermediate kappa values were interpreted as follows: 0.00-0.20 (poor agreement), 0.21-0.40 (fair agreement), 0.41-0.60 (moderate agreement), 0.61-0.80 (good agreement), and >0.81 (excellent agreement).
Additionally, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of urine analysis were calculated, using semen analysis as the reference method. These measures were computed to evaluate the diagnostic accuracy of urine samples for HPV detection in comparison to semen samples.
## RESULTS
The mean age of the participants was 36 years, with 61% between 30 and 39 years old. Regarding educational level, 73% had attended university. Additionally, 68% were married, and 76% had not had biological children. A total of 77% reported no history of abortion, and 72% had their first sexual intercourse between the ages of 15 and 19 years. Furthermore, 96% had no prior HPV diagnosis, and 37% reported having had 5 to 9 sexual partners. A total of 62% stated that they used protection during sexual intercourse, and 77% reported not having sexual activity with high-risk partners. Moreover, 92% had never undergone HPV testing. Regarding ejaculatory abstinence, 68% of men reported 2 to 4 days. Finally, 93% indicated that the reason for semen analysis was fertility assessment.
In Table 2, seminal parameters are presented. Overall, 15.1% of individuals were HPV positive. In this group, progressive motility was lower (34.5 [20.2-46]) compared with the HPV-negative group. Most participants exhibited sperm morphology values at or below the WHO lower reference limit (≤4%), with this alteration observed in 84% of the total study population, 81.1% of HPV-negative individuals, and 100% of HPV-positive individuals. No differences in pH were observed between the groups. The overall prevalence of HPV in semen samples was 15.1%, compared with 9.4% in urine samples (Table 3). Specifically, high-risk HPV was detected in 9.4% of semen samples and 8.5% of urine samples. Low-risk HPV was identified in 5.7% of semen samples and 0.9% of urine samples. Notably, urine samples also demonstrated the capacity to detect high-risk genotypes.
The distribution of HPV genotypes between semen and urine samples revealed notable differences in detection patterns (Table 4). Several high-risk genotypes-such as HPV types 16, 18, 39, 52, and 66-were exclusively observed in semen samples. HPV types 6,31,42,44,45,53,56,58,59, and 81 were detected in both semen and urine samples. Among the high-risk genotypes, HPV type 58 was the most frequently detected in urine samples (20%) compared with semen samples (9.5%). Regarding low-risk HPV, type 81 was the most frequently detected in semen samples (19%), while type 44 was the most common in urine samples (13.3%). Semen samples were considered the gold standard, as they demonstrated a higher overall detection rate. Although both semen and urine samples detected high-risk genotypes, urine samples failed to detect certain low-risk genotypes, which were identified exclusively in semen samples.
The correlation between urine and semen sampling for HPV detection showed that urine sampling had a sensitivity of 37.5%, indicating that only one-third of the positive cases detected by semen were also detected by urine. In contrast, the specificity was 95%, suggesting that urine sampling accurately identified the majority of HPV-negative cases. The PPV was 60%, meaning that 60% of the positive results were true positives. The NPV was 89.6%, indicating that approximately 90% of the negative results were true negatives. The kappa coefficient was 0.39, reflecting an equal agreement between semen and urine sampling. Although urine sampling demonstrated high specificity, its limited sensitivity suggests that it may not be a suitable alternative for HPV detection in the male population (Table 5).
## DISCUSSION
The present study evaluated the diagnostic performance of urine sampling in compari son to semen sampling for the detection of HPV in men. Regarding sociodemographic characteristics, more than 60% of participants were between 30 and 39 years of age, and the mean age at first sexual intercourse was 17 years, which is consistent with the findings of Malhotra et al. (27). Participants reported a mean of eight lifetime sexual partners, aligning with the data presented by Buttman et al. (28). Concerning protective sexual practices, 62.3% of participants reported consistent condom use during sexual activity, which is comparable to the findings of Borges et al. (29). However, this differs from Malhotra et al. (27), where over half of the study population reported not using protection. Additionally, 77.4% of participants in our study reported no history of sexual activity with high-risk partners, similar to Buttman et al. (28) and further supported by their later study (30).
Our findings show an overall HPV prevalence of 15.1% in semen samples compared with 9.4% in urine samples. These results are consistent with those reported by Lyu et al. High-risk HPV types, including HPV 16, 18, 39, 52, and 66, were detected exclusively in semen samples and not in urine. This finding is in agreement with the study con ducted by Astori et al. (32), which demonstrated a higher detection rate of HPV in semen compared to urine. This finding is particularly relevant, as HPV types 16 and 18 are responsible for more than 70% of precancerous cervical lesions and cervical cancers worldwide (33)(34)(35)(36). These results suggest that semen sampling may represent an appropriate and minimally invasive diagnostic method for HPV detection in men.
Both low-and high-risk genotypes were identified in semen samples. The most frequent was HPV 81 (19%), followed by HPV types 42, 44, and 58, each with a preva lence of 9.5%. Clinically, the detection of high-risk types, such as HPV 16, 18, and 31, is highly relevant given their well-established association with intraepithelial lesions and anogenital cancers (9,13,14,23,37). These results highlight the primary role of men as transmitters and reservoirs in the HPV transmission pathway (18,38). In addition, the detection of low-risk genotypes, typically associated with genital warts, underscores the importance of considering semen not only as a vector for viral transmission but also as a biological fluid in which a broad spectrum of HPV genotypes can be identified.
In terms of diagnostic accuracy, urine sampling showed high specificity (95.6%) but low sensitivity (37.5%), suggesting that it may not be an appropriate diagnostic tool for HPV detection in asymptomatic men. These values are compared with those reported by Koene et al. (20), who found a specificity of 95% and a sensitivity of 41%. Söderlund-Strand et al. (39) reported even lower sensitivity, at only 7.9% for urine samples. Although urine collection is convenient, non-invasive, and well-tolerated (17), its diagnostic utility in asymptomatic men is limited. Sensitivity improves in symptomatic individuals or when the sexual partner is infected (17).
While urine sampling remains appealing due to its feasibility and acceptability, its low sensitivity restricts its clinical applicability (20,39). Conversely, although semen sampling may be perceived as slightly more invasive, it remains a well-tolerated, practical, and demonstrated a higher detection rate in our study, including identification of high-risk HPV types. These findings are consistent with Astori et al. (32), who also support semen as a viable diagnostic specimen for HPV detection.
In the context of fertility, previous studies have shown an association between HPV presence in semen and impaired sperm quality. Weinberg et al. (25), Moreno-Sepulveda and Rajmil (40), Foresta et al. (41), and Capra et al. (42) have reported a significant relationship between HPV infection and decreased sperm concentration, morphology, and motility. These findings suggest the need for further research to determine whether the higher prevalence of HPV observed in semen contributes to male infertility.
A limitation of this study is the absence of a general urine analysis, as samples were processed only by centrifugation to obtain cells for DNA extraction. This may have affected HPV detection and potentially contributed to lower sensitivity.
## Conclusion
In summary, while urine sampling offers clear advantages in terms of acceptability, feasibility, and non-invasiveness, its limited sensitivity restricts its utility for HPV diagnosis in asymptomatic men. Urine sampling may be more appropriate in symptomatic individuals or when a partner is known to be HPV-positive. Semen sampling, on the other hand, demonstrated a higher detection rate, including high-risk HPV types, and should be considered a convenient, minimally invasive, rapid, and reliable method for accurate HPV diagnosis.
## References
1. Sung, Ferlay, Siegel et al. (2021) "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries" *CA Cancer J Clin*
2. Mcbride (2024) "Human malignancies associated with persistent HPV infection" *Oncologist*
3. Giuliano, Nyitray, Kreimer et al. (2015) "EUROGIN 2014 roadmap: differences in human papillomavirus infection natural history"
4. "transmission and human papillomavirus-related cancer incidence by gender and anatomic site of infection" *Int J Cancer*
5. Dunne, Nielson, Stone et al. (2006) "Prevalence of HPV infection among men: a systematic review of the literature" *J Infect Dis*
6. (2017) "Human papillomavirus vaccines: WHO position paper" *Vaccine (Auckl)*
7. Gezimu, Bekele, Bekana et al. (2024) "Males' access to human papillomavirus vaccination in resource-limited settings" *Immunotargets Ther*
8. Harder, Wichmann, Klug et al. (2018) "Efficacy, effectiveness and safety of vaccination against human papillomavirus in males: a systematic review" *BMC Med*
9. Garolla, Graziani, Grande et al. (2024) "HPV-related diseases in male patients: an underestimated conundrum" *J Endocrinol Invest*
10. Sucato, Buttà, Bosco et al. (2023) "Human papillomavirus and male infertility: what do we know?" *Int J Mol Sci*
11. Bosco, Serra, Fasciana et al. (2021) "Potential impact of a nonavalent anti HPV vaccine in Italian men with and without clinical manifestations" *Sci Rep*
12. Quiroz (2025) "La Vacuna Del HPV En Ecuador Se Enfoca En Niños y Niñas de 9 Años ¿Qué Pasa Con El Resto? Available from"
13. (2014) "Vacuna Contra El Virus Del Papiloma Humano Previene Cáncer Uterino En El Ecuador -Ministerio de Salud Pública Available Online"
14. Giuliano, Nielson, Flores et al. (2007) "The optimal anatomic sites for sampling heterosexual men for human papillomavirus (HPV) detection: the HPV detection in men study" *J Infect Dis*
15. Foresta, Pizzol, Moretti et al. (2010) "Clinical and prognostic significance of human papillomavirus DNA in the sperm or exfoliated cells of infertile patients and subjects with risk factors" *Fertil Steril*
16. Bianchi, Frati, Panatto et al. (2013) "Detection and genotyping of human papillomavirus in urine samples from unvaccinated male and female adolescents in Italy" *PLoS One*
17. Melchers, Schift, Stolz et al. (1989) "Human papillomavirus detection in urine samples from male patients by the polymerase chain reaction" *J Clin Microbiol*
18. Jin, Kim, Lee (2021) "Human papillomavirus prevalence in urine samples of asymptomatic male sexual partners of women with sexually transmitted diseases" *Int J Environ Res Public Health*
19. Golijow, Pérez, Smith et al. (2005) "Human papillomavirus DNA detection and typing in male urine samples from a high-risk population from Argentina" *J Virol Methods*
20. Aung, Fairley, Tabrizi et al. (2018) "Detection of human papillomavirus in urine among heterosexual men in relation to location of genital warts and circumcision status" *Sex Transm Infect*
21. Koene, Wolffs, Brink et al. (2016) "Comparison of urine samples and penile swabs for detection of human papillomavirus in HIV-negative Dutch men" *Sex Transm Infect*
22. Enerly, Olofsson, Nygård (2013) "Monitoring human papillomavirus prevalence in urine samples: a review" *Clin Epidemiol*
23. Laprise, Trottier, Monnier et al. (2014) "Prevalence of human papillomaviruses in semen: a systematic review and meta-analysis" *Hum Reprod*
24. Zhaffal, Salame (2023) "Semen human papillomavirus (HPV) shedding in males: frequency, clinical significance, and reproductive outcomes-literature review" *Middle East Fertil Soc J*
25. Foresta, Noventa, Toni et al. (2015) "HPV-DNA sperm infection and infertility: from a systematic literature review to a possible clinical management proposal" *Andrology (Los Angel)*
26. Weinberg, Nahshon, Feferkorn et al. (2020) "Evaluation of human papilloma virus in semen as a risk factor for low sperm quality and poor in vitro fertilization outcomes: a systematic review and meta-analysis" *Fertil Steril*
27. Gupta, Sherpa, Lucksom et al. (2024) "Evaluating different samples & techniques for hr-HPV DNA genotyping to improve the efficiency of risk profiling for oral & cervical cancers in Sikkim, India" *Indian J Med Res*
28. Malhotra, Kant, Ahamed et al. (2019) "Health behaviors, outcomes and their relationships among young men aged 18-24 years in a rural area of north India: a cross-sectional study" *PLoS One*
29. Buttmann, Nielsen, Munk et al. (2011) "Sexual risk taking behaviour: prevalence and associated factors. a population-based study of 22,000 Danish men" *BMC Public Health*
30. Borges, Duarte, Cabral et al. (2021) "Uso de preservativo masculino e dupla proteção por homens adolescentes no Brasil" *Rev saúde pública*
31. Buttmann, Nielsen, Munk et al. (2014) "Young age at first intercourse and subsequent risk-taking behaviour: an epidemiological study of more than 20,000 Danish men from the general population" *Scand J Public Health*
32. Lyu, Feng, Li et al. (2017) "Human papillomavirus in semen and the risk for male infertility: a systematic review and meta-analysis" *BMC Infect Dis*
33. Astori, Pipan, Muffato et al. (1995) "Detection of HPV-DNA in semen, urine and urethral samples by dot blot and PCR" *New Microbiol*
34. Bakir, Alacam, Karabulut et al. (2021) "Evaluation of human papillomavirus genotype distribution in cervical samples" *J Cytol*
35. Ye, Jones, Wang et al. (2024) "Comprehensive overview of genotype distribution and prevalence of human papilloma virus in cervical lesions" *Gynecol Obstet Clin Med*
36. De Sanjose, Alemany, Geraets et al. (2010) "Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study" *Lancet Oncol*
37. (2021) "World Health Organization WHO guidelines for screening and treatment of precancerous lesions for cervical cancer prevention"
38. Moghimi, Zabihi-Mahmoodabadi, Kheirkhah-Vakilabad et al. (2019) "Significant correlation between high-risk HPV DNA in semen and impairment of sperm quality in infertile men" *Int J Fertil Steril*
39. Hauwers, Tjalma (2009) "Screening for human papillomavirus: is urine useful?" *Indian J Cancer*
40. (2025) *Full-Length Text Journal of Clinical Microbiology*
41. Söderlund-Strand, Wikström, Dillner (2015) "Evaluation of human papillomavirus DNA detection in samples obtained for routine Chlamydia trachomatis screening" *J Clin Virol*
43. Moreno-Sepulveda, Rajmil (2021) "Seminal human papillomavirus infection and reproduction: a systematic review and meta-analysis" *Andrology (Los Angel)*
44. Foresta, Garolla, Zuccarello et al. (2010) "Human papillomavirus found in sperm head of young adult males affects the progressive motility" *Fertil Steril*
45. (1016)
46. Capra, Notari, Buttà et al. (2022) "Human papillomavirus (HPV) infection and its impact on male infertility" *Life (Basel)*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12688151&blobtype=pdf
|
# Seasonal dynamics and genetic diversity of human adenoviruses in patients with acute respiratory infection in Thailand, 2024
Jiratchaya Puenpa, Lita Tantipraphat, Ratchadawan Aeemjinda, Preeyaporn Vichaiwattana, Sumeth Korkong, Yong Poovorawan
## Abstract
Human adenoviruses (HAdVs) are a significant cause of acute respiratory infections (ARIs), particularly in pediatric populations. Continuous molecular surveillance is essential to understand their epidemiology, genetic diversity, and seasonal dynamics. In 2024, a surveillance study was conducted in Bangkok, Thailand, involving 8,130 nasopharyngeal swabs collected from ARI patients. HAdV was detected in 5.3% (429/8,130) of ARI cases, with peak weekly positivity reaching 17.7% during weeks 10-12 (March). These HAdV-positive samples were subsequently genotyped through partial hexon gene sequencing and phylogenetic analysis using the maximum likelihood method. HAdV-B3 was the predominant genotype (70.7%), followed by HAdV-C2 (11.6%) and HAdV-C1 (10.4%). Genotype diversity increased toward the end of the year, with the emergence of HAdV-B7, C5, and C6. The majority of cases occurred in children aged 0-9 years, with HAdV-B3 dominating across all pediatric groups. Phylogenetic analysis revealed close genetic relationships between Thai strains and reference strains from China, Japan, the USA, and Europe, indicating both local circulation and international linkages. Despite the detection of HAdV-B7, no severe outcomes were reported in this cohort. This study also reports potential relevant sites under episodic positive selection in the hexon gene, suggesting adaptive evolution at specific codon positions. This study provides updated insight into the molecular epidemiology of HAdV in Thailand. The findings highlight seasonal and age-specific patterns in genotype distribution and underscore the importance of continued genomic surveillance to detect emerging variants and guide public health responses.GenBank database at https://www.ncbi.nlm.nih. gov/nucleotide/.
## Introduction
Human adenoviruses (HAdVs), classified within the Mastadenovirus genus of the Adenoviridae family, possess a linear, non-segmented, double-stranded DNA genome approximately 28-38 kilobases in length, flanked by inverted terminal repeats [1]. The Adenoviridae family is taxonomically divided into five genera, with Mastadenovirus comprising isolates from both human and nonhuman primate hosts [2]. HAdVs are non-enveloped, icosahedral viruses measuring 70-90 nm in diameter, composed of 252 capsomeres. Twelve penton capsomeres occupy the vertices and feature protruding filamentous glycoproteins with terminal knobs, while the remaining 240 hexon capsomeres constitute the structural faces of the capsid [2].
To date, more than 110 HAdV types have been identified and classified into seven species (A-G), initially based on serological neutralization profiles and now distinguished by their genomic features [3,4]. Recombination plays a pivotal role in HAdV evolution, particularly in cases of severe or persistent infection, and contributes to the emergence of novel genotypes with potential clinical significance [5][6][7].
HAdVs initiate infection by binding to specific cellular receptors on the airway or nasal epithelium. For most respiratory HAdVs, the coxsackievirus and adenovirus receptor (CAR) and/or CD46 serve as primary attachment sites, with subsequent engagement of integrins facilitating internalization via clathrin-mediated endocytosis [8,9]. Following entry, viral particles are trafficked to the nucleus where viral DNA is released, initiating early gene transcription and modulating host cell responses [10]. Infection triggers host innate immune signaling pathways, including activation of NF-κB and type I interferon responses, which coordinate antiviral defenses and influence the severity of inflammation in the respiratory epithelium [11,12]. In addition, host cell autophagy has been shown to modulate early stages of HAdV infection in airway epithelial cells, affecting viral entry, replication efficiency, and the magnitude of antiviral immune responses [13]. Variations in receptor usage, downstream signaling, and autophagic responses may collectively contribute to differences in tissue tropism, pathogenicity, and clinical outcomes among distinct HAdV types.
These molecular interactions at the cellular level underpin the broad tissue tropism exhibited by HAdVs, enabling them to infect epithelial mucosal cells across the respiratory, gastrointestinal, genitourinary, and ocular systems [14][15][16]. Respiratory tract infections are particularly common and can range from mild upper respiratory illnesses, such as pharyngitis and pharyngoconjunctival fever, to more severe lower respiratory conditions including bronchitis, bronchiolitis, and pneumonia, particularly among young children and immunocompromised individuals [17,18].
In Thailand, where acute respiratory infections remain a leading cause of pediatric morbidity, epidemiological surveillance of HAdVs has been limited. A study conducted between 2009 and 2012 reported a 1.0% positivity rate for HAdV among patients with respiratory tract infections, with HAdV-B3, HAdV-C1, and HAdV-C2 identified as the most frequently detected genotypes [19]. Children under the age of three accounted for the highest proportion of HAdV-positive cases, comprising 63.29% of all infections [19]. These findings highlight the pressing need for continued molecular surveillance and virological studies to clarify the burden and circulating genotypes of HAdVs in Thailand, particularly in young children.
Despite the significant public health burden posed by HAdVs, therapeutic options remain limited, with no specific antiviral agents currently approved for routine clinical use. The absence of targeted therapies underscores the urgent need for the development of effective antivirals and vaccines. Bridging this gap requires a deeper understanding of HAdV molecular biology, genetic diversity, and epidemiological dynamics to support evidence-based prevention and treatment strategies.
The specific objectives of this study were to determine the prevalence of HAdV among acute respiratory infection (ARI) cases in Thailand, identify circulating genotypes through partial hexon gene sequencing, and describe the temporal and seasonal patterns of infection. In addition, phylogenetic analysis was conducted to explore the genetic relationships between local HAdV strains and those reported globally, providing insights into molecular evolution and potential implications for public health interventions.
## Materials and methods
## Ethical statement
This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University, Thailand (Approval Number: IRB0977/67). To protect patient confidentiality, all data were fully anonymized prior to analysis. Due to the retrospective nature of the study, the requirement for informed consent was waived by the Institutional Review Board. Nasopharyngeal swab specimens were accessed for research purposes on 04/03/2025. Patient demographic information, including age and sex, was obtained retrospectively from the records accompanying the nasopharyngeal swab specimens. All data were fully anonymized prior to analysis.
## Real-time PCR-based detection of HAdV in clinical samples
In 2024, a total of 8,130 nasopharyngeal swab specimens were collected from patients hospitalized with ARIs in two tertiary-care hospitals in Bangkok that have been long-term collaborators with our study team. All collected specimens were submitted to our center for comprehensive virological analysis and tested for multiple respiratory viruses, including HAdV, influenza A and B, SARS-CoV-2, respiratory syncytial virus (RSV), parainfluenza viruses, human metapneumovirus, seasonal coronaviruses, and rhinovirus, using specific real-time PCR assays for each virus. For the purpose of this study, we specifically focused on analyzing the prevalence, genotype distribution, and epidemiological characteristics of HAdV. Viral nucleic acids were extracted from 200 μL of the sample supernatant using the magLEAD 12gC automated extraction system (Precision System Science, Chiba, Japan), in accordance with the manufacturer's instructions. This high-throughput platform ensured consistent and efficient extraction suitable for large-scale surveillance studies.
Each extracted nucleic acid sample was then subjected to real-time PCR to screen for HAdV. The assay utilized primers and probes specifically targeting the conserved region of the hexon gene, a key structural component of the adenoviral capsid, following previously validated protocols [20].
## HAdV genotyping
To determine the genotypes of HAdV, 429 HAdV-positive samples were randomly selected for further analysis. A conventional PCR assay was employed to amplify a portion of the hexon gene, producing an amplicon approximately 956 base pairs in length, using primer sequences previously published in the literature [19]. Among these 429 samples, 345 yielded amplicons of sufficient quality for successful sequencing, while the remaining samples failed to amplify or produced low-quality products, as summarized in the flow diagram (S1 Fig).
The specific primers used for amplification were ADV_F2 (5′-TTY CCC ATG GCN CAC AAC AC-3′) and ADV_R2 (5′-GYY TCR ATG AYG CCG CGG TG-3′). PCR reactions were prepared in a final volume of 25 μL, comprising 2-3 μL of extracted DNA (concentration range: 100 ng to 1 µg), 10 mM of each primer, 1X Perfect Taq MasterMix (5 PRIME, Darmstadt, Germany), and nuclease-free water. The thermocycling protocol included an initial denaturation at 94°C for 3 minutes, followed by 40 cycles of denaturation at 94°C for 30 seconds, annealing at 50°C for 30 seconds, and extension at 72°C for 1 minute and 45 seconds. A final extension step was performed at 72°C for 10 minutes. Amplicons were subsequently sent for sequencing and analysis, which were carried out by First BASE Laboratories Sdn Bhd, located in Selangor Darul Ehsan, Malaysia.
## Phylogenetic analysis
Sequence alignment was performed using the CLUSTAL W algorithm available through the European Bioinformatics Institute (EBI) web platform [21]. Phylogenetic trees were constructed using MEGA software version 10 [22], applying the maximum likelihood (ML) method. Statistical support for the tree topology was evaluated through 1,000 bootstrap replicates to ensure robustness.
To determine the most suitable nucleotide substitution model for each dataset, a model selection procedure incorporating correction factors was implemented. For the hexon gene analysis, the General Time Reversible (GTR) model with a proportion of invariant sites (I) and a gamma distribution (Γ) to account for rate variation among sites was selected as the best-fitting model.
## Statistical analysis
The relationship between gender and infection status, both categorical variables, was evaluated using Pearson's chisquare test of independence. To compare infection rates across different age groups, pairwise proportion tests were applied, with the Bonferroni method used to correct for multiple comparisons. A p-value threshold of less than 0.05 was considered indicative of statistical significance. All statistical procedures were carried out using R software (version 4.4.2) [23].
## Estimation of site-specific selection pressure
To explore the evolutionary forces acting on the hexon gene of HAdV, we applied a comprehensive set of codon-level analytical tools that capture various modes of selection, including purifying, directional, and episodic diversifying selection. The analysis incorporated four established models: Single-Likelihood Ancestor Counting (SLAC), Fixed Effects Likelihood (FEL), Mixed Effects Model of Evolution (MEME), and the Fast Unconstrained Bayesian AppRoximation (FUBAR) [24][25][26]. Each method estimates site-specific ratios of non-synonymous (dN) to synonymous (dS) substitutions using either maximum likelihood (SLAC, FEL, MEME) or a Bayesian statistical framework (FUBAR), based on aligned nucleotide sequences and a corresponding phylogenetic tree. All analyses were performed via the Datamonkey web server (http://www.datamonkey.org) [27]. Positive selection was considered when identified by at least 1 method, with codons having a P threshold <0.1 or a posterior probability >0.90.
## Nucleotide sequence accession numbers
The novel sequence data generated in this study are available in the GenBank database under accession numbers PX020986-PX021330, and are publicly accessible through the NCBI GenBank database at https://www.ncbi.nlm.nih.gov/ nucleotide/. Detailed accession numbers for all sequenced samples are provided in S1 Table.
## Results
## Prevalence of HAdV detection in Thailand
In 2024, a surveillance study was conducted at a collaborating hospital in Bangkok to investigate the prevalence and temporal distribution of HAdV among patients presenting with ARI. A total of 8,130 nasopharyngeal swab specimens were collected and analyzed using real-time PCR. Of these, 429 samples were confirmed positive for HAdV, corresponding to an overall positivity rate of 5.3% (429/8,130), as illustrated in Fig 1a . Weekly trends in HAdV detection revealed marked temporal variability throughout the year. The weekly positivity rate ranged from 0% to a maximum of 17.7%, while the number of ARI samples tested per week varied between 45 and 214. A pronounced increase in HAdV activity was observed during the summer months, particularly between epidemiological weeks 9 and 17 (March), peaking at approximately 18% in week 12. This surge coincided with a substantial rise in ARI case numbers. Following week 41 (October to December), which corresponds to the winter season, both the number of ARI cases and HAdV-positive detections declined, although sporadic minor peaks were noted in weeks 46 2a). However, this difference was not statistically significant (χ² = 3.23, p = 0.072). In this study, age groups were classified as follows: Infants (0-2 years), Pre-school children (3-5 years), Primary school children (6-12 years), Secondary school adolescents (13-18 years), Young adults (19-30 years), Middle-aged adults (31-60 years), and Older adults (over 60 years). In contrast to gender, a significant variation in HAdV infection rates was observed across age groups (Fig 2b). Specifically, Pre-school children (14.7%) and Primary school children (11.4%) had significantly higher infection rates compared to Secondary school adolescents and all adult age groups (p < 0.001, Bonferroni-adjusted pairwise comparisons). No statistically significant differences in infection proportions were observed among individuals aged 13 years and older.
## Molecular characterization of the hexon Gene in circulating HAdV
To investigate the genetic diversity of circulating human adenoviruses (HAdVs), 345 HAdV-positive clinical samples with high-quality PCR amplicons were selected for genotypic analysis. Targeting the hypervariable region of the hexon gene, partial sequences were amplified using conventional PCR. Phylogenetic analysis was performed to assess genetic relationships between the circulating strains and reference sequences obtained from the GenBank database. Among the successfully genotyped samples (n = 345; Fig 3), HAdV-B3 was the predominant genotype, representing 70.7% of cases. Additional genotypes included HAdV-C2 (11.3%), HAdV-C1 (10.4%), HAdV-C5 (5.8%), HAdV-C6 (1.2%), and HAdV-B7 (0.6%).
Phylogenetic analysis revealed that the HAdV-B3 strains formed a closely related cluster with reference strains previously reported from China (2015), the USA (2011), Japan (2023), and Germany (2023), suggesting global dissemination and genetic conservation. The two HAdV-B7 strains identified in this study (B52150 and B55900) were closely related to strains reported from Finland in 2024. Most HAdV-C1 Thai isolates clustered with strains from the USA (2012), China (2021), and Japan (2018-2019). Notably, a distinct subcluster composed of three Thai strains (B50277, B51350, and B50258) was observed alongside a genetically similar strain from Japan (2018), indicating possible regional lineage divergence.
HAdV-C5 strains segregated into two distinct phylogenetic groups: one composed exclusively of 11 Thai strains from this study, and another including nine Thai strains clustered with reference sequences from the USA, Russia, China, and Japan. Most HAdV-C2 strains grouped with reference strains from Singapore (2016), Russia (2019), and the USA (2011); however, three isolates (B55066, B49004, and B49474) formed a separate lineage. All four Thai HAdV-C6 strains clustered tightly and showed close phylogenetic relationships with strains from Argentina (2002), Japan (2019), the USA (2015), and China (2015), reflecting a potentially conserved but globally dispersed lineage.
## Genotypic trends by month and age
The monthly distribution of HAdV genotypes throughout 2024 is illustrated in Fig 4 . HAdV-B3 remained the most consistently detected genotype across the year, comprising 60-80% of circulating strains from January through September. Other genotypes, including C1, C2, C5, and C6, were detected at lower frequencies and showed intermittent circulation. A notable shift in genotype composition emerged in the final quarter of the year. In October, B3 declined in relative prevalence, coinciding with an increase in C5 and C6. This trend intensified in November and December, where genotype diversity increased markedly. During these months, B3 accounted for less than 50% of cases, while B7 reemerged prominently in November, and C5 and C6 demonstrated elevated proportions through the end of the year.
Among the 345 HAdV-positive cases, genotype B3 was the most prevalent across all age groups, particularly in children (Fig 5a and5b). It accounted for the majority of infections in Primary school children (87.4%), Pre-school children (68.1%), and Infants (39.3%). Notably, B3 was the sole genotype detected in Secondary school adolescents (100%) and Older adults (100%), and was dominant in Young adults (85.7%) and Middle-aged adults (83.3%).
Other genotypes, including C1, C2, C5, and C6, were detected at lower frequencies and primarily among younger age groups. For instance, genotype C2 was found in 33.9% of Infants and 9.9% of Pre-school children, while genotype C1 was most common in Pre-school children (14.9%) and Infants (12.5%). Genotypes C5 and C6 were rarely detected and only in children, with proportions below 11%. Genotype B7 was infrequently detected overall, identified in only three cases: one each in Infants, Pre-school children, and Middle-aged adults. These findings indicate a clear age-related distribution of HAdV genotypes, with greater genotype diversity observed among Infants and Pre-school children, and B3 predominating in older age groups.
## Site-specific selective pressure acting on the hexon gene
To investigate the adaptive molecular evolution of HAdV, codon-specific selective pressures acting on the hexon gene were examined by estimating the ratio of non-synonymous (dN) to synonymous (dS) substitution rates across the phylogeny. A suite of complementary codon-based models implemented in multiple selection detection algorithms (SLAC, FEL, MEME, and FUBAR) was employed.
Using SLAC (Fig 6a), the overall dN/dS ratio across the hexon region was estimated at 0.095, with 45 codons showing significantly lower dN than dS (dN/dS < 1, p ≤ 0.1). FEL identified 157 codons with significant evidence of purifying selection (p ≤ 0.1), while FUBAR confirmed pervasive purifying selection at 153 codons (posterior probability ≥ 0.9), with no evidence of pervasive positive selection.
In contrast, MEME detected three codon positions (19, 39, and 174) that were evolving under episodic diversifying selection (p ≤ 0.1) (Fig 6b).
## Discussion
To better understand the epidemiology of human adenovirus (HAdV) infections, this study analyzed the genetic profiles of HAdV strains from ARI patients in Bangkok in 2024. We assessed seasonal patterns, genotype distribution, and phylogenetic relationships to clarify local transmission dynamics. Weekly surveillance data and genotype analysis revealed insights into strain prevalence and evolution. Comparison with global reference sequences highlighted the genetic links between Thai and international HAdV strains, suggesting both local circulation and global connectivity of emerging variants.
The detection of HAdV in 5.3% of ARI cases in this study underscores its continued relevance as a respiratory pathogen in the Thai population. Although this prevalence is lower than many reports from other regions, it still reflects a measurable burden. For instance, studies in India have documented positivity rates ranging from 6.8% to as high as 18.6% [28,29], while findings from China report slightly higher rates than ours, between 6.9% and 10.8% [30][31][32]. Significantly elevated rates are often associated with closed or outbreak-prone settings; for example, the Korean military reported a 36.0% positivity rate during outbreaks [33], and in northern Vietnam, a large pediatric outbreak at the end of 2022 showed HAdV in 54.5% of cases [34]. Additionally, surveillance data from the United States National Adenovirus Type Reporting System (NATRS) indicated a positivity rate of up to 77.0% among submitted specimens from 2017 to 2023, reflecting continued virus circulation despite reduced testing volume post-COVID-19 [35]. Such disparities may result from differences in surveillance strategies, testing policies, seasonal trends, and population characteristics across regions.
The HAdV outbreak in Bangkok peaked between epidemiological weeks 10 and 12 of 2024, with the highest detection rate of 20.3% reported in March. This seasonal pattern aligns with regional variations observed globally. In southern China, peak HAdV activity was documented earlier, in January 2024, while in Jining City, China, the peak occurred later, in July 2024 [32,36]. Similarly, in the Republic of Korea, two distinct peaks were observed, occurring in August-September 2023 and April-May 2024, respectively, suggesting multiple waves of transmission [37]. In the United Kingdom, HAdV cases peaked in April 2024 [38], while in Wales, the peak was reported in January of the same year, highlighting intra-country variability. In contrast, Australia reported sporadic HAdV detections throughout the year, without a clear seasonal peak [39]. These temporal variations likely reflect climatic and population factors, suggesting a seasonal pattern of HAdV circulation in Bangkok and emphasizing the need for continued surveillance.
Co-infections with multiple respiratory viruses were detected in over half of the HAdV-positive cases in this study. Such mixed infections are frequently reported in acute respiratory illness and may influence viral interactions, transmissibility, and disease outcomes such as pneumonia, intensive care unit (ICU) admission, prolonged hospital stay, and death [40]. Although clinical data were not collected to evaluate the clinical severity associated with co-infection, the high frequency observed here suggests that HAdV often circulates concurrently with other respiratory pathogens. This finding highlights the complexity of attributing acute respiratory symptoms to a single viral agent and emphasizes the need for integrated molecular and clinical surveillance in future studies.
In this study, the highest proportion of HAdV cases was observed in the 5-9 years age group, followed by children aged 0-4 years. This age distribution aligns with findings from previous studies that highlight the vulnerability of young children to HAdV infections. For example, a recent study from China (2023-2024) reported that 59.5% of cases occurred in children aged 0-6 years, with species B adenoviruses showing a higher infection rate among school-aged children [41]. Additionally, a severe outbreak in a nursery in Dakar in April 2024 involved four infants and was associated with subgroup B1 strains, underscoring the potential for significant HAdV transmission and disease severity in younger age groups [42].
The predominant genotype identified in this study was HAdV-B3, followed by HAdV-C2 and HAdV-C1. HAdV-B3 is well-documented as a major cause of respiratory illness in children and has been frequently implicated in outbreaks across various regions [34]. In contrast, HAdV-B7, though less prevalent, is often associated with more severe clinical outcomes. Interestingly, our data also indicate a seasonal pattern in genotype distribution, with greater heterogeneity and the emergence of less common genotypes during the colder months. Notably, a 2013-2014 outbreak in the United States reported HAdV-B7 infections with intensive care unit (ICU) admission rates reaching 46% [43], and a 2024 outbreak in Finland caused unusually severe respiratory illness among military conscripts, resulting in six deaths [44]. While many prior studies have reported severe illness associated with HAdV-B7 [43,44], particularly in very young children or military recruits, community-wide outbreaks involving a broader pediatric population have been less frequently documented. In this study, only two HAdV-B7 strains were identified among more than three hundred genotyped specimens, and most participants were aged 3-12 years. The absence of reported severe cases may therefore reflect both the limited number of B7 detections and the study's design, in which detailed clinical data were not systematically collected. Additionally, strain-specific genetic variations could influence viral virulence and clinical outcomes. Future investigations combining comprehensive clinical information with full-genome characterization of circulating B7 strains are warranted to elucidate these relationships.
Comparison with recent systematic reviews highlights regional heterogeneity in adenovirus circulation. A meta-analysis from China reported winter peaks in northern regions, dual peaks in southern regions, and HAdV-7/B55 as predominant types among adults [45]. In contrast, our Bangkok data show lower overall positivity, predominance of HAdV-B3, and higher infection rates among preschool and primary school children. These differences likely reflect climatic variation, host immunity, and population characteristics, emphasizing the need for region-specific surveillance.
In the current study, a predominant signal of purifying selection was observed across the hexon region, as indicated by the low overall dN/dS ratio (0.095) estimated by SLAC and the large number of codons under significant purifying selection identified by FEL (157 codons) and confirmed by FUBAR (153 codons). This widespread negative selection likely reflects the functional constraints on the hexon protein, which is essential for viral structure and immune recognition. Notably, despite the dominance of purifying selection, MEME detected episodic diversifying selection at three codon positions (19, 39, and 174). Codons 19 and 39 in our amplified sequences correspond to positions 650 and 670 in the reference hexon protein, located within the viral jellyroll connector (VC) domain, which is important for hexon trimer assembly and capsid stability [46]. Codon 174 corresponds to position 805 in the FG2 loop, a surface-exposed region contributing to antigenicity [46]. Variation at these sites may therefore influence capsid stability or immune recognition, suggesting potential adaptive significance. These findings partially contrast with those of Saha et al., who reported no evidence of positive selection in either the hexon or fiber genes, though purifying selection was also observed [28]. Conversely, a study from Taiwan on HAdV-3 and HAdV-7 reported positive selection sites within the hexon protein, consistent with our findings [47]. Such intertype and regional variations, together with the identified positively selected sites, highlight the complex evolutionary dynamics and potential adaptive changes within the adenovirus hexon gene, warranting further functional investigation.
This study has several limitations that should be considered. First, only a partial region of the hexon gene was amplified using previously published primers, which may not equally detect all genetic variants. This limitation might have influenced the detection of certain HAdV strains, the observed genotype distribution, and the identification of recombinant forms. Comprehensive genomic characterization, including sequencing of the penton base and fiber genes, would be required to more accurately assess recombination events and viral evolution. Second, clinical data were not collected alongside the molecular findings, which restricts our ability to correlate specific genotypes with clinical severity, such as pneumonia or other severe outcomes. Third, this study focused exclusively on HAdV detected in acute respiratory illness. Including samples from other clinical syndromes, such as gastroenteritis or conjunctivitis, would provide a more complete picture of genotype distribution across different disease presentations and help understand the broader impact of HAdV in the population. Fourth, the study was conducted over a single one-year period, which may limit the generalizability of the findings to other years or seasonal variations.
In conclusion, the present study enhances our understanding of the genomic epidemiology and evolution of human adenoviruses (HAdVs) in Bangkok, Thailand. The dominant circulating genotype identified was HAdV-B3, with a marked increase in activity beginning in March 2024. Phylogenetic analysis revealed multiple distinct clusters within each genotype, suggesting several independent introductions and ongoing local transmission. Our findings demonstrate the value of molecular epidemiology in tracking HAdV diversity and identifying emerging genotypes, which may inform public health strategies and help mitigate the impact of future outbreaks.
## References
1. Benkő, Aoki, Arnberg et al. (2022) "ICTV Virus Taxonomy Profile: Adenoviridae 2022" *J Gen Virol*
2. Wachtman, Mansfield (2012) "Viral diseases of nonhuman primates"
3. Robinson, Singh, Lee et al. (2013) "Molecular evolution of human adenoviruses" *Sci Rep*
4. Gonzalez, Hayes, Balansay et al. (2023) "Two novel recombinant human mastadenovirus D genotypes associated with acute respiratory illness" *J Med Virol*
5. Akello, Kamgang, Barbani et al. (2021) "Genomic analyses of human adenoviruses unravel novel recombinant genotypes associated with severe infections in pediatric patients" *Sci Rep*
6. Dhingra, Hage, Ganzenmueller et al. (2019) *Molecular Evolution of Human Adenovirus (HAdV) Species C. Sci Rep*
7. Zhou, Chen, Zhang et al. (2025) "Identification and characterization of a novel human adenovirus type HAdV-D116" *Front Microbiol*
8. Stasiak, Stehle (2020) "Human adenovirus binding to host cell receptors: a structural view" *Med Microbiol Immunol*
9. Wickham, Mathias, Cheresh et al. (1993) "Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment" *Cell*
10. Greber, Suomalainen (2022) "Adenovirus entry: Stability, uncoating, and nuclear import" *Mol Microbiol*
11. Chen, Wang, Zhong et al. (2024) "Activation of the RIG-I/MAVS signaling pathway during human adenovirus type 3 infection impairs the pro-inflammatory response induced by secondary infection with Staphylococcus aureus" *Int J Mol Sci*
12. Li, Fan, Zhou et al. (2023) "Human adenovirus infection induces pulmonary inflammatory damage by triggering noncanonical inflammasomes activation and macrophage pyroptosis" *Front Immunol*
13. Zeng, Carlin (2013) "Host cell autophagy modulates early stages of adenovirus infections in airway epithelial cells" *J Virol*
14. Thompson, De Vries, Paulson (2019) "Virus recognition of glycan receptors" *Curr Opin Virol*
15. Menon, Zhou, Spurr-Michaud et al. (2016) "Epidemic Keratoconjunctivitis-Causing Adenoviruses Induce MUC16 Ectodomain Release To Infect Ocular" *Surface Epithelial Cells. mSphere*
16. Shieh (2022) "Human adenovirus infections in pediatric population -An update on clinico-pathologic correlation" *Biomed J*
17. Kunz, Ottolini (2010) "The role of adenovirus in respiratory tract infections" *Curr Infect Dis Rep*
18. Chau, Lee, Peiris et al. (2014) "Adenovirus respiratory infection in hospitalized children in Hong Kong: serotype-clinical syndrome association and risk factors for lower respiratory tract infection" *Eur J Pediatr*
19. Sriwanna, Chieochansin, Vuthitanachot et al. (2009) "Molecular characterization of human adenovirus infection in Thailand" *Virol J*
20. Jiang, Huang, Xie et al. (2022) "Development of a diagnostic assay by three-tube multiplex real-time PCR for simultaneous detection of nine microorganisms causing acute respiratory infections" *Sci Rep*
21. Madeira, Park, Lee et al. (2019) "The EMBL-EBI search and sequence analysis tools APIs in 2019" *Nucleic Acids Res*
22. Kumar, Stecher, Li et al. (2018) "MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms" *Mol Biol Evol*
23. Core (2014) "R: A Language and Environment for Statistical Computing"
24. Pond, Frost (2005) "Not so different after all: a comparison of methods for detecting amino acid sites under selection" *Mol Biol Evol*
25. Murrell, Wertheim, Moola et al. (2012) "Detecting individual sites subject to episodic diversifying selection" *PLoS Genet*
26. Murrell, Moola, Mabona et al. (2013) "FUBAR: a fast, unconstrained bayesian approximation for inferring selection" *Mol Biol Evol*
27. Weaver, Shank, Spielman et al. (2018) "Datamonkey 2.0: A Modern Web Application for Characterizing Selective and Other Evolutionary Processes" *Mol Biol Evol*
28. Saha, Majumdar, Chaudhuri et al. (2023) "Molecular epidemiology of circulating human adenoviruses among acute respiratory infection patients seeking healthcare facilities in West Bengal, India" *Virology*
29. Nath, Choudhury, Gogoi et al. (2024) "Molecular characterization of human adenovirus associated with pediatric severe acute respiratory infections in a tertiary care hospital in North East India" *Front. Virol*
30. Wang, Liu, Mi et al. (2021) "Clinical features and epidemiological analysis of respiratory human adenovirus infection in hospitalized children: a cross-sectional study in Zhejiang" *Virol J*
31. Xu, Chen, Wu (2022) "Molecular typing and epidemiology profiles of human adenovirus infection among hospitalized patients with severe acute respiratory infection in Huzhou, China" *PLoS One*
32. Wang, Yang, Wu et al. (2024) "An Outbreak of Human Adenovirus Infection Among Children Post COVID-19 Pandemic in Southern China" *J Med Virol*
33. Kim, Lee, Eom et al. (2024) "Prevalence and Burden of Human Adenovirus-Associated Acute Respiratory Illness in the Republic of Korea Military, 2013 to 2022" *J Korean Med Sci*
34. Nguyen, Phung, Tran et al. (2023) "Molecular subtypes of Adenovirus-associated acute respiratory infection outbreak in children in Northern Vietnam and risk factors of more severe cases" *PLoS Negl Trop Dis*
35. Abdirizak, Winn, Parikh et al. (2024) "Surveillance of human adenovirus types and the impact of the COVID-19 pandemic on reporting -United States, 2017-2023" *MMWR Morb Mortal Wkly Rep*
36. Dou, Chen, Song et al. (2025) "Epidemiological characteristics and genomic analysis of respiratory adenovirus in Jining City from February 2023 to July 2024" *BMC Genomics*
37. Lee, Woo, Lee et al. (2023) "Analysis of adenovirus detection trends in the Republic of Korea" *Public Health Weekly Report*
38. (2025) "Surveillance of influenza and other seasonal respiratory viruses in the UK, winter 2023 to 2024"
39. Government (2025) "Annual Australian Respiratory Surveillance Report -2024"
40. Niu, Gao, Zhang et al. (2025) "Systematic Review and Meta-Analysis of the Association Between Clinical Severity and Co-Infection of Human Adenovirus With Other Respiratory Pathogens in Children" *J Med Virol*
41. Jin, Qin, Li et al. (2025) "Increased circulation of adenovirus in China during 2023-2024: Association with an increased prevalence of species B and school-associated transmission" *J Infect*
42. Jallow, Sall, Diagne et al. (2024) "Outbreak of severe acute respiratory infections caused by recombinant human adenovirus type B 7/3 in hospitalized infants from a nursery in Dakar" *IJID Reg*
43. Scott, Chommanard, Lu et al. (2016) "Human Adenovirus Associated with Severe Respiratory Infection" *Emerg Infect Dis*
44. Heinonen, Erra, Lundell et al. (2024) "Adenovirus type 7d outbreak associated with severe clinical presentation" *Euro Surveill*
45. Liu, Xu, Li et al. (2023) "Prevalence of human infection with respiratory adenovirus in China: A systematic review and meta-analysis" *PLoS Negl Trop Dis*
46. Rux, Kuser, Burnett (2003) "Structural and phylogenetic analysis of adenovirus hexons by use of high-resolution x-ray crystallographic, molecular modeling, and sequence-based methods" *J Virol*
47. Lin, Lu, Lin et al. (2015) "Molecular Epidemiology and Phylogenetic Analysis of Human Adenovirus Caused an Outbreak in Taiwan during 2011" *PLoS One*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12710318&blobtype=pdf
|
# Comparison of manual and automated ultrasensitive assays for residual HIV-1 in plasma from individuals on suppressive antiretroviral therapy
Sonia Coco, Mars Stone, Xutao Deng, Yunfei Wang, Wes Rountree, Salvatore Scianna, Leilani Montalvo, Melanie Dimapasoc, Martin Stengelin, George Sigal, Guoxin Wu, Bonnie Howell, Sarah Palmer, Cheryl Jennings, Douglas Richman, Robert Gorelick, Gregory Laird, Albine Martin, Jana Jacobs, John Mellors, Steven Deeks, Michael Busch
## Abstract
Several nucleic acid and antigen assays have been developed to detect HIV in plasma at levels below the detection limit of standard clinical assays; however, comparative assessment of performance of these ultrasensitive assays on identical panels has been limited. Here, we report the relative performance characteristics of three manual (ultracentrifugation concentration and laboratory-developed PCR) and two automated commercial assays for single-copy detection of HIV RNA, as well as two ultrasensitive HIV p24 assays, using blinded panels of plasma samples containing HIV at low concentrations. Independent sample sets were studied in two phases: in the first phase, qualification panels consisted of analytic standards in serial dilution and clinical plasma samples from virologically suppressed people with HIV (PWH). HIV detection in clinical samples was infrequent using the ultrasensitive p24 assays (mean 11%). In contrast, a higher proportion of the same samples were detected using single-copy RNA assays (mean 61%). In the second phase, evaluation panels of clinical plasma samples (n = 80) from virally suppressed PWH and low-copy analytic standards assessed performance of one automated and two manual single-copy RNA assays with the highest qualification phase detection frequencies. The automated assay performed comparably in three separate laboratories and consistently detected HIV RNA in over half (mean 57%) of samples from virally suppressed PWH, whereas the manual assays detected HIV in ≤40% of the same samples. An automated single-copy assay provides a scalable method for measuring residual HIV viremia that may be broadly useful in pathogenesis and cure-directed studies.
IMPORTANCE Most people with HIV (PWH) on antiretroviral therapy have viral loads below the detection limit of clinical assays, yet virus is often present and detectable at very low levels using ultrasensitive research assays. Clinical trials evaluating curative interventions and interpreting outcomes of analytical treatment interruptions depend on reliable assays to assess and quantify changes in HIV persistence often at very low levels. We conducted a two-stage head-to-head, blinded comparison of multiple ultrasensitive HIV RNA and p24 assays, first using 50 low viral load plasma samples, then further evaluating the top-performing assays on a 144-member blinded panel composed of duplicate contrived and clinical specimens from well-suppressed PWH. Single-copy RNA methods performed better than p24 assays, and a fully automated, 9-replicate commercial RNA assay demonstrated high sensitivity, reproducibility across laboratories, and practical scalability that can be applied to measure the impact of interventions in HIV cure trials.
## KEYWORDS antiretroviral therapy, HIV persistence, residual HIV viremia, HIV
A ntiretroviral therapy (ART) suppresses HIV viremia to levels that are undetectable by clinical viral load assays and (generally) requires lifelong adherence to ART to sustain viral suppression. However, single-copy assays can detect HIV RNA well below the limit of detection of clinical assays (1,2). Low levels of HIV in plasma reflect viral persistence, largely due to cells that were infected near the time of ART introduction, subsequently resulting in clonal or oligoclonal proviruses that, following discontinuation of ART, become highly transcriptionally active and continue to release viral particles (3)(4)(5). In some people, viremia can persist at a level detectable by clinical viral load assays, despite optimal ART adherence and absence of HIV drug resistance. This non-suppressi ble viremia (NSV) can arise from proviruses with small deletions and mutations that affect the major splice donor site in the 5′ leader region (6,7).
In the quest for a cure for HIV, many interventional clinical trials are underway utilizing various strategies that target the latent reservoir in ART-suppressed participants (8). The HIV reservoir can be quantified by assessing the levels and replication competence of proviruses in peripheral blood mononuclear cells (PBMC), such as total HIV DNA and intact proviral DNA, as well as by expression of viral RNA, protein, and replication competence based on viral outgrowth assays (9). One strategy used to assess the effect of a curative intervention on the replication-competent reservoir is the use of analytical treatment interruption (ATI), with measurement of time to and/or magnitude of plasma viral rebound following ATI. Timing of HIV viral rebound has been correlated with timing of ART initiation, with the strongest predictor of faster time to rebound being higher residual plasma viremia before treatment interruption in early-treated individuals and higher levels of intact proviral DNA in individuals treated during chronic infection (10). Theoretically, circulating virus-which is measured by plasma RNA tests-reflects the total body burden of virus-producing cells. Thus, plasma RNA tests that sample and quantify circulating virus reflective of the clinically relevant HIV reservoir are increasingly appreciated as a key measure of the impact of curative interventions.
The Reservoir Assay Validation and Evaluation Network (RAVEN) program was developed to establish performance characteristics and limitations of currently available and novel assays for detection and quantitation of the HIV reservoir in blood of ARTsuppressed individuals, with the goal of informing their application to selection and monitoring of ART-suppressed HIV-infected populations of interest for cure interven tions, including those incorporating ATI. Given that different anatomical and cellular compartments can harbor persistent virus, and that currently available and novel assays measure various aspects of the HIV provirus and its functionality, the RAVEN program included evaluations of the quantitative viral outgrowth assay (QVOA) and other viral outgrowth assays for measuring the replication-competent HIV reservoir in PBMC (11)(12)(13)(14), assays for measuring levels of intact or inducible viral nucleic acid or protein in PBMC (15), and ultrasensitive assays for residual viremia in plasma (16). The latter study focused on the performance of a fully automated replicate testing assay for HIV RNA, demonstrating that persistent low-level viremia can be quantified in individuals on ART and that viral levels decline with length of time on suppressive ART.
Here, addressing the RAVEN program objective to evaluate ultrasensitive assays for residual viremia in plasma, we describe a head-to-head comparison of multiple ultrasensitive assays for detection of HIV RNA or p24 antigen in plasma using blinded panels of serially diluted analytic standards, clinical samples from PWH on ART with viremia undetectable by standard clinical viral load assays, and samples from people without HIV. The study consisted of two phases, beginning with a qualification phase that evaluated five HIV RNA and two p24 assays with a smaller panel, followed by an evaluation phase on the most sensitive assays with a larger panel.
## MATERIALS AND METHODS
## Sample selection and panel construction
RAVEN qualification phase plasma panels were built separately for HIV subtypes B and C, each including 50 samples consisting of (i) low viral load plasma samples from 20 PWH characterized as HIV RNA-negative antibody (Ab)-positive, (ii) five-step threefold serial dilutions in duplicate of two well-characterized HIV RNA-positive/Ab-negative plasma samples diluted in HIV-negative human serum, and (iii) plasma samples from 10 people without HIV (Table 1). The HIV-positive samples were obtained from plasma units acquired through regular blood donations and identified by pooled (for subtype B) or individual (for subtype C) nucleic acid testing, as well as by serology testing as HIV RNA-positive/Ab-negative or RNA-negative/Ab-positive. Although plasma samples were not tested to determine HIV subtype, it is highly likely that they consisted of subtype B and subtype C infections based on prevalence within the geographic location of blood collection sites in the USA and South Africa, respectively (17). Blood donations were collected with ethical review and approval of the collection protocol and informed consent as previously established for the donor organizations' collection activities. For HIV-positive samples identified as RNA-negative/Ab-positive through blood donor screening, additional testing was performed using (i) 10 to 20 replicates of a nucleic acid testing assay used in the blood bank setting to determine specimen reactivity for HIV and (ii) nine replicates of the Aptima HIV-1 Quant Dx assay. A total of 20 samples (each for subtypes B and C) were selected, representing a range of low viral loads, based on reactivity in at least one replicate during the additional testing. RNA-positive/Ab-neg ative samples were quantified using the Abbott m2000 RealTime HIV-1 Viral Load assay at an intermediate dilution 1 log 10 above the highest concentration to be included in the panel, prior to preparing the final serial dilutions to nominal concentrations of 27 to 0.3 copies/mL in defibrinated human plasma (Gemini Biosciences). All samples were frozen into 5 mL aliquots at -80°C and assembled into blinded panels.
For the RAVEN evaluation phase, samples were collected via apheresis (with specific consent for collections and testing, as approved by the UCSF Committee on Human Research), processed, and stored from a well-characterized UCSF cohort of PWH with HIV-1 subtype B infection who had started ART ≤6 months (early) or ≥1 year (late) after infection and had been well-suppressed for ≥1 year, with an average of two apheresis collections approximately one year apart (16). The cohort also included elite controllers (PWH in whom viral suppression is maintained for several years in the absence of ART), non-suppressed PWH, and people without HIV. These samples were used to construct 144-member blinded subtype B plasma evaluation panels, consisting of duplicate aliquots of (i) low viral load clinical samples from early-treated (n = 10) and late-treated (n = 10) PWH (each at two visits), (ii) clinical samples from untreated elite controllers (n = 2) and non-suppressed participants (n = 3), (iii) analytic standards of five-step half-log 10 serially diluted HIV RNA-positive/Ab-negative plasma and of HIV viremic plasma, and (iv) HIV-negative plasma (Table 1). Two sets of analytic standards were included. Standards 1 and 2 were derived from serially diluted HIV RNA-positive/Abnegative plasma. Standard 1 was identical to that used in the subtype B qualification phase panel (nominal concentrations of 27 to 0.3 copies/mL), and Standard 2 had concentrations of 45 to 0.45 copies/mL. Standards 3 and 4 were serially diluted HIV+ non-suppressed plasma to concentrations of 45 to 0.45 copies/mL in 2 diluents; Standard 3 in commercial HIV-negative defibrinated human plasma and Standard 4 in pooled aviremic plasma from recalled fully-suppressed participants with undetectable HIV RNA (based on 45-replicate Hologic Aptima HIV-1 Quant Dx assay). All samples were frozen into 5 mL aliquots at -80°C and assembled into blinded panels.
Single-copy assay controls obtained from the Virology Quality Assurance (VQA) program (contract # HHS75N93019C00015) at Duke University were quantified by the Roche cobas HIV-1 test, then spiked and diluted into HIV-1 negative plasma at 20, 5, and 0 copies/mL. For each concentration, 18 aliquots (1.8 mL each) were randomly selected across the entire production of each control level for testing using the three-replicate Hologic Aptima HIV-1 Quant Dx assay.
## Single-copy HIV RNA assays
The gSCA, HMMC gag, iSCA, and iSCA v2 assays were performed as previously descri bed (2,(18)(19)(20). The gSCA assay is performed using a primer set targeting the 5′ end of gag, whereas the iSCA (v1 and v2) assay targets the 3′ region of pol. The iSCA primer and probe binding sites were selected in a highly conserved sequence in the integrase region; sequence conservation agreement was excellent across subtypes, except for subtype C, where a one-nucleotide substitution was present in the iSCA forward primer-binding site. Therefore, the subtype C panel was not tested using the iSCA assay. For the HMMC gag assay, samples were centrifuged for 30 min, 10°C at 43,100 × g (Beckman OPTIMA XE-90IVD with Ty70.1Ti rotor), and nucleic acids were isolated and analyzed as described previously (2, 21) using the gag primer set designed for the gSCA assay, as well as additional primers and probe described in Somsouk et al. (18), in which the forward primer spans the major splice donor site in the 5′ leader region. For the ultrasensitive Roche CAP/CTM assay, samples provided were centrifuged for 1 h at 41,860 × g (Sorvall RC6 + with rotor model SM-24) at 4°C. From the 4.9 mL total volume, 3.7 mL of plasma supernatant was removed from each sample, leaving 1.2 mL behind (Table 2). Then, the remaining pellet volume was tested using 1.1 mL of sample on the COBAS AmpliPrep/COBAS TaqMan HIV-1 Test, v2.0, per the package insert, targeting two regions, gag and the long terminal repeat (LTR). The concentration of each sample was corrected by a factor of 4.08 (4.9 mL/1.2 mL), wherein the raw result was divided by 4.08 to correct for the total volume originally used in the assay. For the Automated 18×, 9×, and 3× assays, replicates (18, 9, or 3 replicates, respectively) of the Aptima HIV-1 Quant Dx assay on the Panther platform were performed according to the manufacturer instructions, targeting two regions: pol and 5′ LTR. The Panther instrument was programmed to generate up to nine independent replicate tests for each specimen aliquot tube containing up to 5 mL, with each replicate test processing 0.5 mL input. Quantification was based on Poisson modeling of the number of negative (at least one) out of total replicates tested. For samples where no negative replicates were observed, viral load for each replicate was imputed using the assay's normal calibration combined with offline analysis, and then averaged across all replicates (16,22). For testing the VQA single-copy assay controls using the Automated 3× assay, the algorithm for estimating viral load from replicate measurements is provided in the Supplemental Materials.
## Ultrasensitive p24 antigen assays
Testing using the Quanterix Simoa HIV p24 assay was performed as previously described (23). The Meso Scale Discovery (MSD) S-PLEX ® HIV p24 assay was performed as described by Stengelin et al. (24). In the current study, samples underwent acid dissociation of immune complexes and 40-fold concentration prior to detection. Four milliliters of each blinded plasma sample from the RAVEN qualification phase panel, as well as positive and negative plasma control samples (prepared by MSD), were treated with 2 mL of HCl, followed by neutralization with 2 mL of HEPES-buffered NaOH. The volume was then reduced to 100 µL using capture agents that have reversible binding (Table 2). Concentrated samples were frozen at -80°C and later tested in batches in a randomized and blinded fashion. An eight-point calibration curve, run in duplicate, was included on each plate, and the data were fitted with a weighted four-parameter logistic curve fit. Concentrations of each sample were calculated from the calibrator curves, taking into account sample pretreatment. The mean of two measurements was derived for each sample.
## Statistical analysis
For the RAVEN evaluation phase, probit modeling was used to estimate the concentra tion, where the probability of detection of HIV in analytic standards is 50% (LOD 50 ) for each assay. The probit models were fitted using the R drc package. The estimated means and 95% confidence intervals of the LOD were calculated. Assay sensitivity and precision were compared relative to the reference HMMC gag assay. This assay was selected as the benchmark due to its well-established application in many clinical trials. Sensitivity was defined as the probability that an assay returns a positive result when there is a single HIV RNA copy present in the sample assayed. Specifically, we estimated the ratios of the probability of detecting a positive result for an assay to that of a reference assay (HMMC gag) by fitting a Poisson regression model. The detailed method for assessing relative sensitivity and precision is provided in the Supplemental Materials. Separately, to assess reproducibility of the Automated 9× assay across the three separate testing laboratories, we first calculated pairwise (Labs 1 and 2; Labs 1 and 3) interlab variances and means, and then derived pairwise interlab CVs using equation (3) from the same section. Finally, we estimated the relative pairwise interlab precision by taking the logarithm of the ratio of the two pairwise interlab CVs, with P-values for the null hypothesis of equal interlab variation.
## RESULTS
## Performance of ultrasensitive HIV RNA assays using the RAVEN qualification phase plasma panels
The HMMC gag and Automated 18× assays detected the lowest nominal concentration (0.3 cp/mL) in at least one of the duplicate aliquots for both subtype-specific analytic standards, whereas the ultrasensitive CAP/CTM assay detected the lowest concentration in one of the duplicates for both subtype C analytic standards, but not in either of the subtype B analytic standards. The gSCA and iSCA assays were unable to detect the 0.3 cp/mL concentration (Fig. 1). At 1 cp/mL, the HMMC gag, iSCA, and Automated 18× assays detected at least one of the duplicates for all analytic standards tested. The ultrasensitive CAP/CTM assay did not detect the 1 cp/mL concentration in one of the two subtype C analytic standards, while the gSCA assay did not detect this concentration in one each of the two subtype-specific analytic standards. At nominal concentrations ≥3 cp/mL, all five single-copy assays detected HIV in the analytic standards. The reported viral loads were higher at higher nominal HIV RNA concentrations.
Specificity was evaluated using plasma samples from 10 people without HIV per panel. For the HMMC gag, iSCA, ultrasensitive CAP/CTM, and Automated 18× assays, 100% specificity was observed. For the gSCA assay, false positives were observed for 3 of 10 negative controls included in the subtype C panel (at measured HIV RNA concentra tions of 1, 1, and 13 cp/mL), whereas no false positives were observed for the same 10 negative controls included in the subtype B panel.
Detection of HIV in clinical samples was compared using 20 subtype-specific plasma samples from HIV-positive individuals with undetectable HIV RNA by blood donation testing. The Automated 18× assay had the highest clinical sample detection frequency, with 95% and 85% detection of subtype B and subtype C clinical samples, respectively (Fig. 2). Detection frequencies for the subtype B clinical samples were 65%, 60%, 55%, and 50% for the HMMC gag, ultrasensitive CAP/CTM, iSCA, and gSCA assays, respectively, whereas for the subtype C clinical samples, they were 72%, 40%, and 25% for the HMMC gag, gSCA, and ultrasensitive CAP/CTM assays, respectively.
## Performance of ultrasensitive HIV p24 assays using the RAVEN qualification phase plasma panels
The Simoa ultrasensitive p24 assay detected HIV antigen above the cutpoint at all concentrations of the analytic standards; however, the detected p24 antigen levels and the nominal HIV RNA concentrations were not correlated (Fig. S1). The S-PLEX ultrasensi tive p24 assay did not detect any of the analytic standards, except for one of the two 1 cp/mL replicates for one of the subtype C analytic standards.
No detection of HIV in plasma samples from HIV-uninfected individuals was observed using the S-PLEX p24 assay, whereas HIV was detected (at 5.6 fg/mL) in 1 of 10 negative controls using the Simoa p24 assay. Low detection frequencies were observed for the clinical samples, with 10% of subtype B and 20% of subtype C samples detected above the cutpoint for the Simoa p24 assay, while the S-PLEX p24 assay detected 5% of subtype B and 10% of subtype C clinical samples.
## Ultrasensitive HIV RNA assay evaluation
Based on assay performance, with the highest detection frequencies in the RAVEN qualification phase plasma panels, the HMMC gag, iSCA, and Automated 18× assays were chosen for further testing in the RAVEN evaluation phase plasma panels. In this phase, the iSCA v2 assay, with higher sensitivity and simpler sample processing (20), was performed in lieu of the iSCA v1 assay. In addition, the Automated 18× assay was replaced with the Automated 9× assay to match the sample volume (5 mL rather than 10 mL) used for the other two single-copy assays, allowing for a more direct compari son. Whereas the HMMC gag and iSCA v2 assays rely on manual methods involving multiple steps, the Automated 9× assay is based on replicate testing using a commercial automated platform. Therefore, this assay was performed in three different laboratories to assess reproducibility of the results.
At the lowest nominal concentration (0.3 cp/mL) for analytic Standard 1 (that was included in both the subtype B qualification phase panel and the evaluation phase panel), detection was observed for one of the duplicate aliquots using the iSCA v2 assay and the Automated 9× assay in two of the three testing labs (Fig. 3), whereas FIG 1 Nominal and measured viral load (VL) using single-copy assays in qualification phase plasma panels. For each HIV subtype (B and C), plasma from two different individuals, identified as HIV RNA-positive and Ab-negative through routine blood donor testing, was serially diluted and blinded, and each concentration was included in duplicate in the panel for measurement of VL using single-copy assays. Open symbols indicate that HIV was not detected in both replicates using the single-copy assay. LOQ, limit of quantification.
## FIG 2
Measured VL in plasma samples from individuals identified as HIV RNA-negative and Ab-positive through routine blood donor testing. For each HIV subtype (B and C), plasma samples were blinded and included in singlicate in qualification phase panels for testing using single-copy assays. Open symbols indicate that HIV was not detected. Symbols were left blank if the measurement was invalid for a sample. Participant ID samples are ordered from highest to lowest measured VL using the Automated 18× assay. In the bar graph of proportion of samples detectable by single-copy assay, the number of samples with valid measurements is shown inside the bars. no detection was observed using the HMMC gag assay and the Automated 9× assay in one of the testing labs (where no detection was observed for the 1 cp/mL nominal concentration as well). All assays detected HIV in at least one of the two duplicate aliquots in analytic Standards 2 and 3 at the lowest nominal concentration (0.45 cp/mL), except for the iSCA v2 assay in Standard 3.
To determine whether fibrin or other insoluble complexes in plasma impair nucleic acid isolation for HIV recovery by the different single-copy assays, we assessed HIV detection sensitivity in standards diluted in commercial defibrinated plasma and in aviremic plasma, with no difference in detectability observed for any of the assays (Fig. 3). In the aviremic HIV+ plasma diluent, detection was not observed (by HMMC gag and Automated 9× in two of the three testing labs) or was observed for one of the duplicate aliquots at the assay limit of detection of 0.30 to 0.38 cp/mL for iSCA v2 and Automated 9×.
The LOD 50 for detection sensitivity on the analytic standards was calculated for each assay, resulting in similar estimates (Table 3). For the two manual assays, HMMC gag and iSCA v2, the LOD 50 was 0.4 cp/mL (95% CI, 0.34 to 0.45 cp/mL) and 0.41 cp/mL (95% CI, 0.02 to 0.8 cp/mL), respectively. For the Automated 9× assay, the LOD 50 was 0.25 cp/mL (95% CI, 0 to 0.51 cp/mL), 0.94 cp/mL (95% CI, 0.17 to 1.7 cp/mL), and 0.27 cp/mL (95% CI, 0.01 to 0.52 cp/mL) in Lab 1, Lab 2, and Lab 3, respectively. Specificity was evaluated in five HIV-negative clinical samples and commercial HIV-negative defibrinated human plasma, each in duplicate, with 100% specificity observed for the iSCA v2 and Automated 9× assays in two of the three testing labs. False positives were observed for one of two replicates of one HIV-negative clinical sample on the HMMC gag assay (at the assay limit of detection of 0.26 cp/mL) and the Automated 9× assay in one of the testing labs (at the assay limit of detection of 0.38 cp/mL). Plasma from an HIV-positive RAVEN participant with high VL was serially diluted in two different diluents (with or without anti-HIV antibodies), and each concentration was included in duplicate in the panel. Samples diluted in HIV-positive aviremic plasma are indicated in gray compared to samples diluted in HIV-negative defibrinated plasma. Open symbols indicate that HIV was not detected in both replicates using the single-copy assay. Detection of clinical samples was evaluated on 40 plasma samples from 20 HIV+ ART-suppressed RAVEN participants, each at two visits approximately one year apart. The Automated 9× assay had the highest clinical sample detection frequency, with 90% (52.5% in both replicates), 82.5% (62.5% in both replicates), and 72.5% (55% in both replicates) of samples detected in at least one of the two replicates for the three different testing labs (Fig. 4). The HMMC gag assay detected 72.5% of the samples (37.5% in both replicates), whereas the iSCA v2 assay detected 60% of the samples (40% in both replicates).
Plasma from two elite controller RAVEN participants was detectable on all assays, except for one Automated 9× assay testing lab which failed to detect one sample (Fig. S2). All assays demonstrated good reproducibility of viral load on both replicates of three non-suppressed RAVEN participant samples and were within approximately fivefold across the assays, except for the iSCA v2 assay which quantified viral load two orders of magnitude lower than that for the other assays on one non-suppressed participant sample.
Assay sensitivity was compared using statistical modeling of the results from the diluted analytical standards of the clinical samples as well as the low viral load clinical samples from treated PWH and the clinical samples from untreated elite controllers. The HMMC gag assay, which has been applied to a large number of clinical studies, was used as the benchmark for comparison. Sensitivity of the iSCA v2 assay did not differ significantly relative to that of the HMMC gag assay, whereas the sensitivity of the Automated 9× assay was higher relative to that of the HMMC gag assay. For Lab 1 and Lab 3, sensitivity was 1.6-fold (95% CI, 1.1-fold to 2.3-fold) and 1.8-fold (95% CI, 1.2-fold to 2.5-fold) higher, respectively. For Lab 2, the difference was not statistically significant (Fig. S3).
Precision was compared by estimating the ratios of the assays' CVs using HMMC gag as the reference assay. Thus, the CV of the iSCA v2 assay did not differ significantly from that of the HMMC gag assay, whereas the CV of the Automated 9× assay was lower relative to that of the HMMC gag assay. For Labs 1, 2, and 3, CV was 1.3-fold (95% CI, 1.1-fold to 1.6-fold), 1.3-fold (95% CI, 1.1-fold to 1.5-fold), and 1.3-fold (95% CI, 1.1-fold to 1.5-fold) lower, respectively.
Reproducibility of the Automated 9× assay across the three labs was assessed by estimating pairwise interlab CVs, with no significant difference found in reproducibility across the three labs.
## Characterization of Virology Quality Assurance single-copy assay standards
The Virology Quality Assurance (VQA) program prepares and distributes standards consisting of laboratory-cultured HIV virions diluted in plasma to low concentrations, for use in assessing assays that measure low-level HIV viremia in PWH on ART in HIV cure trials. To verify low-copy-number concentrations (0, 5, and 20 cp/mL) and homogeneity across aliquots of a VQA production batch of single-copy assay controls, 18 aliquots of each concentration were tested. Each 1.8 mL aliquot was tested in triplicate (Automated 3× assay). All 54 replicates of the 0 cp/mL control were not detected. Testing of the 5 cp/mL control resulted in two of three reps detected for 10 aliquots, and one of three reps detected for six aliquots, with the remaining two aliquots having no detection or all three reps detected (Fig. 5). The majority (16 aliquots) of the 20 cp/mL control showed detection in all three reps, while two aliquots showed detection in two of three reps. The distribution of positive replicates across aliquots at each concentration suggests acceptable homogeneity. The concentration of the 5 cp/mL control was inferred at 2.46 cp/mL (95% CI 1.55, to 3.38 cp/mL) based on detection in a total of 29 of 54 reps (Supplemental Materials). The 20 cp/mL control concentration was calculated at 17.98 cp/mL (95% CI, 14.04 to 21.91 cp/mL).
## DISCUSSION
Here, we evaluated the performance characteristics of ultrasensitive assays for HIV RNA and p24 antigen in plasma using blinded panels containing contrived specimens prepared from acutely infected blood donor plasma, clinical samples with low levels of HIV RNA from unsuppressed and virally suppressed PWH on ART, and samples from HIV uninfected individuals. These ultrasensitive research assays are essential tools in HIV curative interventions in monitoring for viral rebound after treatment interrup tion, providing a reliable baseline against which post-intervention outcomes can be compared and identifying subtle differences in rebound kinetics among individuals (25-27), as well as for confirming HIV infection and quantifying low-level viremia during follow-up in cases of breakthrough HIV infection in people on PrEP (28,29).
Most of the assays employed in this study had previously been described in publications detailing their development and validation. The original single-copy assay, gSCA (2), detected concentrations down to 0.78 cp/mL using HIV virions diluted in human plasma and measured concentrations ranging from 1 to 32 cp/mL in plasma samples (7 mL volume) from 15 of 15 HIV-infected ART-suppressed individuals. However, a comparison of the gSCA assay to a commercial HIV viral load assay for plasma samples from 20 untreated individuals with viral load >1,000 cp/mL showed that 15% of samples were markedly under-quantified by the gSCA assay, suggesting that genetic variation in HIV gag affected amplification from clinical samples using the single-copy assay. Nonetheless, the gSCA assay has been successfully applied to measure residual viremia in several seminal studies of HIV persistence and clinical interventions (30)(31)(32)(33).
The HMMC gag assay was designed based on the gSCA assay with the addition of primer sets targeting an additional gag region (18). The amplification reactions are set up in 12 replicates, with quantification based on standard curve interpolation (where all replicates were positive) or application of Poisson distribution analysis (where fewer than 12 replicates were positive). Using this assay on plasma samples (9 mL volume) from 33 ART-suppressed participants in a clinical trial, 91% of participants had detectable viremia (median 0.6 and 2.1 cp/mL at baseline in the two arms of the trial [18]). The HMMC gag assay has been applied to numerous clinical trials and studies of HIV persistence and cure interventions on the HIV reservoir (18,27,(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45)(46)(47)(48).
Another assay, iSCA (19), targeting the conserved integrase region in the pol gene, rather than gag, was developed to decrease the potential for mismatches of primer sets with sequences from clinical samples. In addition, nucleic acid recovery was improved by modifying the sample processing after ultracentrifugation through addition of a chaotropic agent during protein digestion and other changes to subsequent steps. Comparison of iSCA and gSCA assays on plasma samples (3 mL volume) from 25 ART-suppressed individuals showed that 36% and 16% had concordant detectable and undetectable viremia, respectively. The remaining were detectable only by iSCA (32%) or gSCA (16%). Similar to other single-copy assays, the iSCA assay has been applied to numerous studies assessing the impact of HIV curative interventions or association of host characteristics with residual viremia (49)(50)(51).
Subsequently, iSCA v2 (20) was developed as a simpler and more sensitive assay based on iSCA v1 using microcentrifugation rather than ultracentrifugation and assaying a larger fraction of the total nucleic acid extract. A rigorous assessment of the iSCA v2 95% limit of detection estimated it to be 0.994 cp/mL (95% CI, 0.62 to 2.55 cp/mL) for a 5 mL volume of low-copy-number standards of diluted plasma from a viremic HIV-positive individual. Furthermore, testing of plasma samples (1 to 5 mL volume) from 60 ART-suppressed individuals demonstrated that 45% had detectable viremia by both iSCA v1 and v2, 23% had undetectable viremia by both assays, 28% had viremia detectable only by iSCA v2, and 3% had viremia initially detectable only by iSCA v1, although these samples were detectable by iSCA v2 upon subsequent testing. The iSCA v2 assay has been employed in investigations of the HIV reservoir in a clinical case of ART-free remission following transplantation with CCR5Δ32/Δ32 stem cells (52), as well as rare cases of HIV infection identified in the treatment arm of PrEP clinical trials (53).
The gSCA, HMMC gag, and iSCA (both v1 and v2) assays all rely on manual centrifu gation and nucleic acid extraction steps with low throughput. Consequently, modified assays using automated HIV viral load assay platforms have been utilized to allow higher throughput. The modifications include sample concentration through centrifugation on a density cushion (54) or ultracentrifugation (55) combined with extraction and amplification using commercial tests, such as the Abbott RealTime HIV-1 assay, Roche CAP/CTM HIV-1 test (as described in this study), and Hologic Aptima HIV-1 Quant Dx assay (22). Alternatively, a multi-replicate strategy, rather than sample concentration, has been reported for the Aptima HIV assay (16,22,56,57). Bakkour et al. developed a Poisson modeling method to calculate HIV RNA concentration based on the combina tion of positive and negative replicates. Using the Automated 45× replicate assay, HIV RNA was detected (median 0.54 cp/mL) in plasma from each of 50 individuals within the RAVEN cohort (16). Furthermore, in that study, no detection was observed in 447 replicates with valid results from plasma samples from 10 people without HIV. Jacobs et al. performed a comparison of the iSCA v2, Automated 1× concentrated, and Automated 9× replicate assays, demonstrating that the iSCA v2 method had the highest variation and lowest 95% LOD. The Automated 1× concentrated method had reduced analytic sensitivity and was not further evaluated for clinical sensitivity. The latter was evaluated using plasma samples from 50 ART-suppressed individuals, showing higher detection frequency with Automated 9× Aptima assay (82%) than iSCA v2 (62%). Thus, based on prior studies comparing single-copy HIV RNA assays, clinical sensitivity was higher for iSCA compared to gSCA, for iSCA v2 compared to iSCA v1, and for Automated 9× compared to iSCA v2. To our knowledge, this study is the first head-tohead comparison of multiple ultrasensitive assays for detection of residual HIV in plasma. In addition to assays detecting HIV RNA, we included ultrasensitive HIV p24 assays due to their small volume requirements, high throughput, and lower cost (58)(59)(60)(61). The ultrasensitive p24 assays have been used to measure HIV gag p24 protein in plasma from acutely infected HIV-positive individuals. However, their application to samples from ART-suppressed individuals has focused on CD4 + T cells (14,23,(62)(63)(64) rather than plasma, due to the formation of immune complexes after seroconversion during chronic infection.
Based on the results of this study, the ultrasensitive p24 assays, as performed, were not suitable for detection in plasma samples from HIV Ab-positive individuals with residual low-level viremia and may require enhancements to the methodology to improve detection and reduce background. Such improvements have been developed to enrich p24 through immunocapture and elute in assay-compatible format, allowing sensitive and specific detection in lysates of cells from ART-suppressed HIV-positive individuals (65-67).
We showed herein that the gSCA assay had lower clinical sensitivity on subtype B plasma samples than the iSCA, HMMC gag, and ultrasensitive assays using commercial HIV tests, confirming and extending previously reported findings. We also found that the clinical sensitivity of the iSCA v2 assay was lower than that of the HMMC gag and Automated 9× assays. For subtype C plasma samples, the gSCA assay had higher clinical sensitivity than the ultrasensitive CAP/CTM assay, but lower than the HMMC gag and Automated 18× assays. It should be noted that the CAP/CTM system has been replaced with higher throughput systems (such as the cobas 6800/8800 platform) in many laboratories. The newer system has been reported to have higher sensitivity in detecting residual viremia in ART-suppressed individuals for both B and non-B subtypes (68). In the current study, the use of the high-throughput Automated 9× assay in three different laboratories showed reproducible results. The wide availability of such automated systems has led to their use in current global clinical trials networks, such as the AIDS Clinical Trials Group (10), and in other multi-center studies, such as those performed by the Martin Delaney Collaboratories for HIV Cure Research (69).
In this study, clinical sensitivity was evaluated using plasma samples from PWH with undetectable viremia using clinical assays. As noted earlier, non-suppressible viremia has been shown to arise from proviruses containing defects in the 5′ leader sequence. It is possible that such defective RNA contributes to residual viremia in people receiving suppressive ART. The single-copy HIV RNA assays employed in the current study target regions outside of the 5′ leader defective sequences and/or have dual target regions. Therefore, it is unlikely that such mutations in the 5′ leader sequence would affect the performance of the single-copy assays on the clinical samples included in the panels.
Further evaluation of assay performance on HIV subtype C and other non-B subtypes was not performed in this study and warrants further investigation, given the geo graphic distribution of these subtypes in regions where clinical trials are being under taken worldwide. Of note, the iSCA v2 assay requires an alternative primer set for adequate quantification of subtype C isolates (22). We found that both the HMMC gag and the Automated 18× assays had good clinical sensitivity (72% and 85%, respectively) using subtype C samples from blood donors with undetectable HIV RNA using routine screening assays. Another limitation of our study consists of a lack of further comparison between commercial assays with pre-processing by centrifugation and manual or replicate single-copy assays. Although ultracentrifugation followed by Automated 1× assay had been investigated in a prior study (22) and found to have lower analytic sensitivity than the other single-copy assays tested, we found that the ultrasensitive CAP/CTM assay had reasonable clinical sensitivity (60%) on subtype B samples. Generating additional comparative data on the widely used commercial viral load platforms would be informative, using sample concentration prior to testing relative to replicate testing of samples without pre-processing, in order to gain insight into the potential effect of interfering substances that could be concentrated along with the virus and affect target recovery.
In conclusion, automated replicate testing using the Aptima HIV Quant assay provides a scalable alternative to manual single-copy assays for detecting residual viremia in HIV-infected, ART-suppressed individuals, offering high clinical sensitivity, precision, and reproducibility. With many ongoing cure-directed clinical trials (9,70,71), the wide dynamic range and high throughput of the replicate testing-based single-copy assay will be advantageous for measuring the in vivo burden of HIV reservoirs.
## References
1. Dornadula, Zhang, Vanuitert et al. (1999) "Residual HIV-1 RNA in blood plasma of patients taking suppressive highly active antiretroviral therapy" *JAMA*
2. Palmer, Wiegand, Maldarelli et al. (2003) "New real-time reverse transcriptase-initiated PCR assay with single-copy sensitivity for human immunodeficiency virus type 1 RNA in plasma" *J Clin Microbiol*
3. Bailey, Sedaghat, Kieffer et al. (2006) "Residual human immunodeficiency virus type 1 viremia in some patients on antiretroviral therapy is dominated by a small number of invariant clones rarely found in circulating CD4+ T cells" *J Virol*
4. Anderson, Archin, Ince et al. (2011) "Clonal sequences recovered from plasma from patients with residual HIV-1 viremia and on intensified antiretroviral therapy are identical to replicating viral RNAs recovered from circulating resting CD4+ T cells" *J Virol*
5. Simonetti, Sobolewski, Fyne et al. (2016) "Clonally expanded CD4+ T cells can produce infectious HIV-1 in vivo" *Proc Natl Acad Sci*
6. White, Wu, Yasin et al. (2023) "Clonally expanded HIV-1 proviruses with 5'-leader defects can give rise to nonsuppressible residual viremia" *J Clin Invest*
7. Mohammadi, Etemad, Zhang et al. (2023) "Viral and host mediators of non-suppressible HIV-1 viremia" *Nat Med*
8. Deeks, Archin, Cannon et al. "International AIDS Society (IAS) Global Scientific Strategy working group. 2021. Research priorities for an HIV cure: international AIDS society global scientific strategy 2021"
9. Abdel-Mohsen, Richman, Siliciano et al. (2020) "Recommendations for measuring HIV reservoir size in cure-directed clinical trials" *Nat Med*
10. Li, Melberg, Kittilson et al. (2024) "Predictors of HIV rebound differ by timing of antiretroviral therapy initiation" *JCI Insight*
11. Rosenbloom, Bacchetti, Stone et al. (2019) "Assessing intra-lab precision and inter-lab repeatability of outgrowth assays of HIV-1 latent reservoir size" *PLoS Comput Biol*
12. Wonderlich, Subramanian, Cox et al. (2019) "Effector memory differentiation increases detection of replication-competent HIV-l in resting CD4+ T cells from virally suppressed individuals" *PLoS Pathog*
13. Stone, Rosenbloom, Bacchetti et al. (2021) "Assessing the suitability of next-generation viral outgrowth assays to measure human immunodeficiency virus 1 latent reservoir size" *J Infect Dis*
14. Kuzmichev, Lackman-Smith, Bakkour et al. (2023) "Application of ultrasensitive digital ELISA for p24 enables improved evaluation of HIV-1 reservoir diversity and growth kinetics in viral outgrowth assays" *Sci Rep*
15. Levy, Hughes, Roychoudhury et al. (2021) "A highly multiplexed droplet digital PCR assay to measure the intact HIV-1 proviral reservoir" *Cell Rep Med*
16. Bakkour, Deng, Bacchetti et al. (2020) "Replicate aptima assay for quantifying residual plasma viremia in individuals on antiretroviral therapy" *J Clin Microbiol*
17. Nair, Gettins, Fuller et al. (2024) "Global and regional genetic diversity of HIV-1 in 2010-21: systematic review and analysis of prevalence" *Lancet Microbe*
18. Somsouk, Dunham, Cohen et al. (2014) "The immunologic effects of mesalamine in treated HIV-infected individuals with incomplete CD4+ T cell recovery: a randomized crossover trial" *PLoS One*
19. Cillo, Vagratian, Bedison et al. (2014) "Improved single-copy assays for quantification of persistent HIV-1 viremia in patients on suppressive antiretroviral therapy" *J Clin Microbiol*
20. Tosiano, Jacobs, Shutt et al. (2019) "A simpler and more sensitive single-copy HIV-1 RNA assay for quantification of persistent HIV-1 viremia in individuals on suppressive antiretroviral therapy" *J Clin Microbiol*
21. Cline, Bess, Piatak et al. (2005) "Highly sensitive SIV plasma viral load assay: practical considerations, realistic performance expectations, and application to reverse engineering of vaccines for AIDS" *J of Medical Primatology*
22. Jacobs, Tosiano, Koontz et al. (2020) "Automated multireplicate quantification of persistent HIV-1 viremia in individuals on antiretroviral therapy" *J Clin Microbiol*
23. Wu, Swanson, Talla et al. (2017) "HDAC inhibition induces HIV-1 protein and enables immune-based clearance following latency reversal" *JCI Insight*
24. Stengelin, Roy, Aghvanyan et al. (2015) "HIV p24 immunoassay with the sensitivity of PCR methods. Abstr American association for clinical chemistry (AACC) annual meeting and clinical lab Expo"
25. Gandhi, Zheng, Bosch et al. "AIDS Clinical Trials Group A5244 team. 2010. The effect of raltegravir intensification on low-level residual viremia in HIV-infected patients on antiretroviral therapy: a randomized controlled trial" *PLoS Med*
26. Rasmussen, Rajdev, Rhodes et al. (2021) "Impact of anti-PD-1 and anti-CTLA-4 on the human immunodeficiency virus (HIV) reservoir in people living with HIV with cancer on antiretroviral therapy: the AIDS malignancy consortium 095 study" *Clin Infect Dis*
27. Uldrick, Adams, Fromentin et al. (2025) *Full-Length Text Journal of Clinical Microbiology*
28. Yarchoan, Maldarelli, Cheever et al. (2022) "Pembrolizumab induces HIV latency reversal in people living with HIV and cancer on antiretroviral therapy" *Sci Transl Med*
29. Marzinke, Grinsztejn, Fogel et al. (2021) "Characteri zation of human immunodeficiency virus (HIV) infection in cisgender men and transgender women who have sex with men receiving injectable cabotegravir for HIV prevention: HPTN 083" *J Infect Dis*
30. Koss, Gandhi, Halvas et al. (2024) "First case of HIV seroconversion with integrase resistance mutations on long-acting cabotegravir for prevention in routine care" *Open Forum Infect Dis*
31. Maldarelli, Palmer, King et al. (2007) "ART suppresses plasma HIV-1 RNA to a stable set point predicted by pretherapy viremia" *PLoS Pathog*
32. Palmer, Maldarelli, Wiegand et al. (2008) "Low-level viremia persists for at least 7 years in patients on suppressive antiretroviral therapy" *Proc Natl Acad Sci*
33. Dinoso, Kim, Wiegand et al. (2009) "Treatment intensification does not reduce residual HIV-1 viremia in patients on highly active antiretroviral therapy" *Proc Natl Acad Sci*
34. Mcmahon, Jones, Wiegand et al. (2010) "Short-course raltegravir intensification does not reduce persistent low-level viremia in patients with HIV-1 suppression during receipt of combination antiretroviral therapy" *Clin Infect Dis*
35. Swanstrom, Gorelick, Welker et al. (2023) "Long-acting lenacapavir protects macaques against intravenous challenge with simian-tropic HIV" *EBioMedicine*
36. Mystakelis, Wilson, Laidlaw et al. (2023) "An open label randomized controlled trial of atorvastatin versus aspirin in elite controllers and antiretroviral-treated people with HIV" *AIDS*
37. Gay, James, Tuyishime et al. (2022) "Stable latent HIV infection and low-level viremia despite treatment with the broadly neutralizing antibody VRC07-523LS and the latency reversal agent vorinostat" *J Infect Dis*
38. Gatechompol, Zheng, Bao et al. (2021) "Prevalence and risk of residual viremia after ART in low-and middle-income countries: a cross-sectional study" *Medicine (Baltimore)*
39. Kroon, Ananworanich, Pagliuzza et al. (2020) "A randomized trial of vorinostat with treatment interruption after initiating antiretroviral therapy during acute HIV-1 infection" *J Virus Erad*
40. Gay, Kuruc, Falcinelli et al. (2020) "Assessing the impact of AGS-004, a dendritic cell-based immunotherapy, and vorinostat on persistent HIV-1 Infection" *Sci Rep*
41. Reid, Suazo, Lensing et al. (2020) "Pilot trial AMC-063: safety and efficacy of bortezomib in AIDS-associated kaposi sarcoma" *Clin Cancer Res*
42. Boulougoura, Gabriel, Laidlaw et al. (2019) "A phase I, randomized, controlled clinical study of CC-11050 in people living with HIV with suppressed plasma viremia on antiretroviral therapy (APHRODITE)" *Open Forum Infect Dis*
43. Crowell, Colby, Pinyakorn et al. (2019) "Safety and efficacy of VRC01 broadly neutralising antibodies in adults with acutely treated HIV (RV397): a phase 2, randomised, double-blind, placebo-controlled trial" *Lancet HIV*
44. Scully, Gandhi, Johnston et al. (2019) "Sex-based differences in human immuno deficiency virus type 1 reservoir activity and residual immune activation" *J Infect Dis*
45. Lee, Elliott, Mcmahon et al. (2019) "Population pharmacokinetics and pharmacodynamics of disulfiram on inducing latent HIV-1 transcription in a phase IIb trial" *Clin Pharmacol Ther*
46. Colby, Trautmann, Pinyakorn et al. (2018) "Rapid HIV RNA rebound after antiretroviral treatment interruption in persons durably suppressed in Fiebig I acute HIV infection" *Nat Med*
47. Lynch, Boritz, Coates et al. (2015) "Virologic effects of broadly neutralizing antibody VRC01 administration during chronic HIV-1 infection" *Sci Transl Med*
48. Elliott, Mcmahon, Chang et al. (2015) "Short-term administration of disulfiram for reversal of latent HIV infection: a phase 2 dose-escalation study" *Lancet HIV*
49. Caskey, Klein, Lorenzi et al. (2015) "Viraemia suppressed in HIV-1infected humans by broadly neutralizing antibody 3BNC117" *Nature*
50. Riddler, Zheng, Durand et al. (2018) "Randomized clinical trial to assess the impact of the broadly neutralizing HIV-1 monoclonal antibody VRC01 on HIV-1 persistence in individuals on effective ART" *Open Forum Infect Dis*
51. Cyktor, Bosch, Mar et al. (2021) "Association of male sex and obesity with residual plasma human immunodeficiency virus 1 viremia in persons on long-term antiretroviral therapy" *J Infect Dis*
52. Henrich, Bosch, Godfrey et al. (2024) "Sirolimus reduces T cell cycling, immune checkpoint marker expression, and HIV-1 DNA in people with HIV" *Cell Rep Med*
53. Hsu, Van Besien, Glesby et al. (2025) "International Maternal Pediatric Adolescent AIDS Clinical Trials Network (IMPAACT) P1107 Team. 2023. HIV-1 remission and possible Full-Length Text Journal of Clinical Microbiology December"
54. "cure in a woman after haplo-cord blood transplant" *Cell*
55. Eshleman, Fogel, Piwowar-Manning et al. (2022) "Characterization of human immunodeficiency virus (HIV) infections in women who received injectable cabotegravir or tenofovir disoproxil fumarate/emtricitabine for HIV prevention: HPTN 084" *J Infect Dis*
56. Yukl, Li, Fujimoto et al. (2011) "Modification of the Abbott RealTime assay for detection of HIV-1 plasma RNA viral loads less than one copy per milliliter" *J Virol Methods*
57. Gupta, Mccoy, Mok et al. (2019) "HIV-1 remission following CCR5Δ32/Δ32 haematopoietic stem-cell transplantation" *Nature*
58. Hatano, Yukl, Ferre et al. (2013) "Prospective antiretroviral treatment of asymptomatic, HIV-1 infected controllers" *PLoS Pathog*
59. Landay, Golub, Desai et al. (2014) "HIV RNA levels in plasma and cervicalvaginal lavage fluid in elite controllers and HAART recipients" *AIDS*
60. Chang, Song, Fournier et al. (2013) "Simple diffusionconstrained immunoassay for p24 protein with the sensitivity of nucleic acid amplification for detecting acute HIV infection" *J Virol Methods*
61. Cabrera, Chang, Stone et al. (2015) "Rapid, fully automated digital immunoassay for p24 protein with the sensitivity of nucleic acid amplification for detecting acute HIV Infection" *Clin Chem*
62. Wilson, Rissin, Kan et al. (2016) "The Simoa HD-1 analyzer: a novel fully automated digital immunoassay analyzer with single-molecule sensitivity and multiplex ing" *SLAS Technol*
63. Stone, Bainbridge, Sanchez et al. (2018) "Comparison of detection limits of fourthand fifth-generation combination HIV antigen-antibody, p24 antigen, and viral load assays on diverse HIV isolates" *J Clin Microbiol*
64. Passaes, Bruel, Decalf et al. (2017) "Ultrasensitive HIV-1 p24 assay detects single infected cells and differences in reservoir induction by latency reversal agents" *J Virol*
65. Stuelke, James, Kirchherr et al. (1971) "Measuring the inducible, replicationcompetent HIV reservoir using an ultra-sensitive p24 readout, the digital ELISA viral outgrowth assay" *Front Immunol*
66. Levinger, Howard, Cheng et al. (2021) "An ultrasensitive planar array p24 Gag ELISA to detect HIV-1 in diverse biological matrixes" *Sci Rep*
67. Wu, Cheney, Huang et al. (2021) "Improved detection of HIV gag p24 protein using a combined immunoprecipita tion and digital ELISA method" *Front Microbiol*
68. Falcinelli, Peterson, Turner et al. (2022) "Combined noncanonical NF-κB agonism and targeted BET bromodomain inhibition reverse HIV latency ex vivo" *J Clin Invest*
69. Wietgrefe, Anderson, Duan et al. (2023) "Initial productive and latent HIV infections originate in vivo by infection of resting T cells" *J Clin Invest*
70. Wirden, Palich, Abdi et al. (2022) "More HIV-1 RNA detected and quantified with the Cobas 6800 system in patients on antiretroviral therapy" *J Antimicrob Chemother*
71. Henrich, Schreiner, Cameron et al. (2021) "Everolimus, an mTORC1/2 inhibitor, in ART-suppressed individuals who received solid organ transplantation: a prospective study" *Am J Transplant*
72. Ta, Malik, Anderson et al. (2022) "Insights into persistent HIV-1 infection and functional cure: novel capabilities and strategies" *Front Microbiol*
73. Armani-Tourret, Bone, Tan et al. (2024) "Immune targeting of HIV-1 reservoir cells: a path to elimination strategies and cure" *Nat Rev Microbiol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12802318&blobtype=pdf
|
# EBNA1 SUMOylation by PIAS1 suppresses EBV lytic replication and enhances episome maintenance
Gunawan Febri, Sugiokto, Kun Zhang, Yunash Maharjan, Renfeng Li
## Abstract
Epstein-Barr virus nuclear antigen 1 (EBNA1) is essential for the replication and stable maintenance of the viral episome in infected cells. Here, we identify the SUMO E3 ligase PIAS1 as a key regulator of EBNA1 through site-specific SUMOylation. Our chromatin immunoprecipitation sequencing analysis revealed that PIAS1 is enriched at the viral origin of plasmid replication (oriP), where it physically associates with EBNA1 and catalyzes its SUMOylation. Using mutational analysis, we identified three lysine residues on EBNA1 (K17, K75, and K241) as major SUMOylation sites. Disruption of these sites compromises EBNA1's ability to restrict Epstein-Barr virus (EBV) lytic replication. In addition, both PIAS1 depletion and the disruption of EBNA1 SUMOylation lead to reduced retention of EBNA1-OriP-based EBV mini-replicon, indicating the importance of EBNA1 SUMOylation in viral episome maintenance. Together, these results uncover a conserved post-translational mechanism by which PIAS1-mediated SUMOylation modulates EBNA1 function and EBV episome maintenance, suggesting a broader role for SUMOylation in viral latency, lytic replication, and persistence. IMPORTANCE Epstein-Barr virus (EBV) persists in infected cells by maintaining its episome through the viral protein EBNA1. We discovered that PIAS1 SUMOylates EBNA1 at specific sites, a process essential for EBNA1 to retain the viral episome and suppress reactivation. When SUMOylation is disrupted, the EBV-based replicon becomes less stable, and EBV is more likely to reactivate. These findings reveal a new layer of host control of EBV latency and reactivation and highlight PIAS1-mediated EBNA1 SUMOyla tion as a key mechanism regulating viral persistence.
In addition to EBV, other DNA viruses such as Kaposi's sarcoma-associated herpesvi rus (KSHV) and human papillomavirus (HPV) also maintain their genomes as episomes during latent or persistent infection (9). These viruses also encode episome maintenance proteins, namely LANA in KSHV and E2 in HPV, that perform analogous functions critical for viral genome persistence. LANA tethers the KSHV episome to host chromatin through interactions with histones and chromosomal proteins such as BRD2/4 while also recruiting cellular replication machinery to ensure genome maintenance (10)(11)(12)(13). Similarly, the E2 protein of HPV binds to specific sequences in the viral origin and interacts with host mitotic chromosomes, via Brd4 and TopBP1, to facilitate episome segregation during cell division (14,15). Like EBNA1, both LANA and E2 are also involved in regulating viral transcription and replication. This functional conservation highlights a shared evolutionary strategy among diverse viruses to ensure long-term persistence in host cells through episome tethering, DNA replication, and transcriptional control (9).
SUMOylation is a highly regulated post-translational modification (PTM) process involving the covalent attachment of small ubiquitin-like modifier (SUMO) proteins-SUMO1, SUMO2, and SUMO3 in humans-to lysine residues on target proteins. This reversible modification is crucial for various cellular processes, including transcriptional regulation, DNA repair, nucleocytoplasmic transport, signal transduction, and protein stability. The SUMOylation pathway comprises a well-coordinated enzymatic cascade, initiated by the activation of SUMO precursors through the E1 activating enzyme complex (SAE1 and SAE2, also known as UBA2). Subsequently, SUMO is transferred to the E2 conjugating enzyme UBC9, which uniquely serves as the sole E2 enzyme responsible for SUMO transfer in mammals (16).
While UBC9 can mediate SUMO conjugation independently, E3 ligases significantly enhance substrate specificity and conjugation efficiency. The Protein Inhibitor of Activated STAT (PIAS) family proteins, comprising PIAS1, PIAS2 (also known as PIASx), PIAS3, and PIAS4 (also known as PIASy), play an important role in substrate selection as E3 SUMO ligases. PIAS proteins facilitate the transfer of SUMO from UBC9 to sub strate proteins, which could influence the function and the subcellular localization of SUMOylated targets. These E3 ligases have been implicated in various biological processes, such as immune response, genome stability, and the control of oncogenic signaling pathways (17).
Our previous research has established PIAS1 as an EBV restriction factor and an E3 SUMO ligase that synergizes with SAMHD1 and YTHDF2 to regulate EBV replication through SUMOylation (18)(19)(20)(21)(22). In this study, we demonstrated that PIAS1 is localized in EBV oriP, where it interacts with EBNA1. We found that PIAS1 enhances EBNA1 SUMOyla tion at three specific lysine residues: K17, K75, and K241. The interaction between EBNA1 and PIAS1, along with the PIAS1-mediated SUMOylation process, plays an important role in regulating viral replication and genome maintenance.
## RESULTS
## PIAS1 is enriched at EBV oriP in EBV+ lymphoma cells
In our previous study, we demonstrated that PIAS1 plays a crucial role in restricting EBV lytic replication. Specifically, PIAS1 inhibits lytic viral gene transcription by binding to viral promoters (20). To gain further insight into PIAS1 binding across the entire EBV genome, we conducted a chromatin immunoprecipitation sequencing (ChIP-seq) experiment using Akata (EBV+) Burkitt lymphoma cells as our model system. Unexpect edly, we observed that PIAS1 peaks were significantly enriched at the EBV oriP (FR and DS) region. We also compared our PIAS1 ChIP-seq data with published EBNA1 ChIP-seq data set and found that PIAS1 peaks overlapped with Raji EBNA1 ChIP-seq peaks at oriP (7,23) (Fig. 1A andB).
To validate these findings, we performed ChIP-qPCR, which confirmed the enrichment of PIAS1 at the EBV oriP with anti-PIAS1 antibody compared to the IgG control (Fig. 1C). These results suggest that PIAS1 may regulate EBV replication and genome maintenance, potentially through interaction with the viral oriP or EBNA1.
## PIAS1 interacts with EBNA1
The EBV oriP region is known to be occupied by EBNA1, which is crucial for maintaining the EBV genome and facilitating replication (6,7). Based on PIAS1 ChIP-seq results (Fig. 1), we hypothesized that PIAS1 interacts with EBNA1 at the site of oriP. To test this hypothesis, we employed multiple experimental approaches in various cell lines. First, we transfected HEK-293T cells with plasmids expressing EBNA1 and PIAS1. Co-immunopreci pitation (co-IP) using anti-V5 antibody-conjugated magnetic beads, followed by western blotting (WB) analysis with anti-PIAS1 antibody, demonstrated that PIAS1 co-immuno precipitated (co-IPed) with EBNA1 (Fig. 2A). To validate these findings in the context of the EBV genome, we repeated the experiment in HEK-293 (EBV+) cells and obtained similar results (Fig. 2B).
To further confirm the interaction between PIAS1 and EBNA1 under physiological conditions, we conducted a proximity ligation assay (PLA) in Akata (EBV+) B cells. The cells were subjected to PLA with or without anti-PIAS1 and anti-EBNA1 antibodies. No PLA signals were detected in the absence of antibodies, whereas dot-like signals were observed in cells treated with both antibodies, primarily localized in the nucleus (Fig. 2C). To extend these findings to EBV+ epithelial cells, we repeated the PLA experiment in SNU-719, an EBV+ gastric cancer cell line. Consistent with those results in Akata (EBV+) cells, we observed strong PLA signals in SNU-719 cells using anti-PIAS1 and anti-EBNA1 antibodies (Fig. 2D). Collectively, these findings provide strong evidence for the interaction between PIAS1 and EBNA1 across multiple cell types, including B cells and epithelial cells.
To elucidate the specific regions of PIAS1 responsible for this interaction, we conducted co-IP experiments using HEK-293T cells transfected with plasmids expressing full-length HA-EBNA1 and either full-length or individual fragments of V5 tagged PIAS1 (Fig. 3A). The results revealed that full-length PIAS1 and all PIAS1 fragments except the C-terminal region (aa 409-651) are co-IPed with HA-EBNA1 (Fig. 3B, lanes 2, 3, and 5). Probes were then added for ligation and amplification. Cell nuclei were visualized using Nikon AXR after staining with 4′,6-diamidino-2-phenylindole (DAPI). The interaction between EBNA1 and PIAS1 in situ was indicated by red dots representing proximity ligation assay (PLA) signals.
These observations suggest that the SAP and RING domains of PIAS1 are crucial for its interaction with EBNA1.
To identify the region(s) of EBNA1 involved in PIAS1 binding, we co-transfected HEK-293T cells with HA-EBNA1 fragments and V5-PIAS1 (Fig. 3C). Subsequent immu noprecipitation (IP) of PIAS1 using anti-V5 antibody-conjugated beads revealed that the C-terminal region (aa 216-405) of EBNA1, which contains the DNA binding and dimerization domain, is essential for PIAS1 interaction (Fig. 3D, lane 6 vs lane 4). Collectively, our findings indicate that the SAP and RING domains of PIAS1 interact specifically with the DNA-binding domain (DBD) of EBNA1 (Fig. 3E).
## PIAS1 enhances EBNA1 SUMOylation in both in vivo and in vitro
Our previous studies have demonstrated that PIAS1 enhances the anti-viral activity of SAMHD1 and YTHDF2 by promoting their SUMOylation on multiple lysine residues (18,19). The E3 ligase responsible for EBNA1 SUMOylation remained unidentified, despite a previous report confirming EBNA1 as a SUMOylated protein (24). We hypothesized that PIAS1 functions as the E3 ligase facilitating EBNA1 SUMOylation. To test this hypothesis, we transfected HEK-293T cells with plasmids expressing EBNA1, PIAS1, and SUMO2 (Fig. 4A). Our results showed that while individual transfection of PIAS1 slightly increased total SUMOylation levels, co-transfection of SUMO2 and PIAS1 significantly enhanced SUMOylation (Fig. 4A, lane 4 vs lane 3).
To further investigate PIAS1-mediated SUMOylation of EBNA1, we performed IP using anti-V5 antibody-conjugated magnetic beads. WB analysis with anti-SUMO2/SUMO3 antibody revealed a strong SUMOylated EBNA1 band when SUMO2 and PIAS1 were co-expressed, suggesting that EBNA1 is indeed targeted for SUMOylation by PIAS1 (Fig. 4A, lane 8).
Given the robust interaction between PIAS1 and EBNA1, coupled with the strong SUMOylated EBNA1 signal when PIAS1 is present, we reasoned that PIAS1 directly SUMOylates EBNA1. To examine this further, we conducted in vitro SUMOylation experiments using purified V5-EBNA1 and purified PIAS1 in combination with E1, E2 (UBC9), and SUMO2. Our findings showed that PIAS1 also enhances SUMOylation of EBNA1 in vitro (Fig. 4B, lane 3 vs lanes 1 and 2), suggesting that EBNA1 is a direct substrate for PIAS1.
To validate EBNA1 SUMOylation by PIAS1 in the relevant EBV-positive B cells, we used Akata (EBV+) cells as a model. Because endogenous SUMOylation is difficult to detect by traditional IP and WB analysis, we utilized PLA for EBNA1 and SUMO2/3 as a surrogate for EBNA1 SUMOylation. We showed that control cells display strong PLA signals, while PIAS1-depleted cells have significantly reduced signals (Fig. 4C andD). These results suggested that EBNA1 is SUMOylated by PIAS1 in Akata (EBV+) cells.
## PIAS1 mediates SUMOylation of EBNA1 at three lysine residues
SUMOylation typically occurs on lysine residues within the consensus motif ΨKxE/D or the inverted motif E/DxKΨ (Ψ represents a hydrophobic amino acid, and x can be any amino acid). However, SUMOylation can also occur outside these consensus sequences (25). To identify SUMOylation sites on EBNA1, we first used GPS-SUMO 2.0 webserver (https://sumo.biocuckoo.cn/advanced.php) to predict putative SUMOylation sites.
The program identified three high-scoring SUMOylation sites: K241, K17, and K75. K241 (PKFE) and K17 (QKED) are located within the KxE/D motif, but the motif in K17 lacks a hydrophobic amino acid, while K75 (QKRP) belongs to a non-consensus motif (Fig. 5A). To verify PIAS1-mediated SUMOylation at these sites, we created individual lysine-to-arginine mutants and performed in vitro SUMOylation assays, including pCEP4 as an EBV replicon containing oriP to mimic EBNA1 binding conditions.
Although K241 was previously reported as the major SUMOylation site, our in vitro assays showed substantial SUMOylation signals for each individual mutant (Fig. S1, lanes 4, 6, and 8 vs lane 2). We then generated a triple mutant, K17R/K75R/K241R (RRR), which greatly reduced the SUMOylation signal compared to the wild-type (WT) protein in the in vitro assay (Fig. 5B, lane 8 vs lane 4). The addition of oriP plasmid stimulated EBNA1 SUMOylation by PIAS1, suggesting oriP binding enhances the interaction between PIAS1 and EBNA1 (Fig. 5B, lane 3 vs lane 4). These results demonstrate that K17 (N-terminal), K75 (N-terminal), and K241 (C-terminal) are the three major SUMOylation sites on EBNA1 mediated by PIAS1, with individual sites capable of compensating for the loss of others (Fig. 5C).
According to the AlphaFold3-predicted three-dimensional structure of EBNA1 ( 27), lysine residues K17 and K75 are situated within an intrinsically disordered region, whereas K241 resides within an α-helix region. The localization of lysines within or adjacent to these flexible regions should enhance structural accessibility, thereby facilitating SUMOylation (Fig. 5D).
To investigate the conservation of EBNA1 SUMOylation sites across various EBV strains and the closely related species, we aligned the amino acid sequences of six EBV EBNA1 with two cyno-EBV EBNA1 sequences from viruses that infect cynomolgus macaques (28). Notably, we found that the amino acids corresponding to K75/K241 of Akata EBNA1 are conserved among all examined species. However, K17 of Akata EBNA1 is conserved only in EBV strains but absent in cyno-EBV EBNA1 (K-R). This observation suggests that EBNA1 SUMOylation by PIAS1 is likely conserved in different EBV and cyno-EBV strains (Fig. 5E).
To investigate whether SUMOylation site mutations affect EBNA1 dimerization, we co-transfected WT HA-EBNA1 and V5-EBNA1 (WT or SUMO-deficient mutants) into HEK-293 (EBV+) cells and performed co-IP using anti-V5 magnetic beads. WB analysis of HA-EBNA1 showed no significant difference in dimerization between WT and SUMOy lation-deficient EBNA1 (Fig. 6A). We also assessed DNA binding affinity of WT EBNA1 and the RRR mutant using electrophoretic mobility shift assays (EMSA) with increasing protein concentrations (0-500 nM) and a 2× FR probe (10 nM). The EMSA results revealed no detectable difference in DNA-binding capacity between WT EBNA1 and RRR mutant (Fig. 6B). To examine whether PIAS1 binds to oriP DNA directly, we also performed EMSA using PIAS1 protein and a 2× FR probe. We found that there are weak EMSA signals only at higher concentrations of PIAS1 (Fig. 6C).
To evaluate whether PIAS1 affects EBNA1's DNA binding capability, we then performed EMSA with a fixed concentration of EBNA1 and increasing amounts of PIAS1. We found that with increasing PIAS1 concentrations, there is a gradual decrease of free probes and a trace increase of EMSA signals (Fig. 6D), suggesting PIAS1 and EBNA1 may coordinate to bind to oriP DNA.
To determine whether PIAS1 affects EBNA1 binding to oriP, we performed ChIP analysis for EBNA1 in WT and PIAS1-knockout (PIAS1-KO) Akata (EBV+) cells (Fig. 7A). Interestingly, we found that PIAS1-KO significantly reduces EBNA1 binding to oriP (Fig. 7B). SUMOylation of substrate also could, in turn, affect its interaction with PIAS1, as PIAS1 has multiple SUMO-interacting motifs (29). To examine whether SUMOylation-deficient EBNA1 affects its binding with PIAS1, we performed a co-IP experiment comparing PIAS1 binding to WT and RRR mutant EBNA1. We noticed that PIAS1 binding to EBNA1- RRR mutant is significantly reduced compared to WT EBNA1 (Fig. 7C, lane 3 vs lane 2), suggesting that SUMOylation of EBNA1 promotes its interaction with PIAS1.
## PIAS1 synergizes with EBNA1 to repress EBV lytic replication
As previously reported, EBNA1 expression inhibits spontaneous EBV reactivation (30). To investigate the role of EBNA1 SUMOylation in EBV replication, we compared WT EBNA1 with various EBNA1 SUMOylation site mutants using HEK-293 (EBV+) cells as a model. We found that the SUMOylation-deficient RRR mutant was impaired to restrict EBV replication compared to WT EBNA1, individual K-R mutants, and K17R/K241R double mutant (Fig. 8A, lane 7 vs lane 2-6).
To examine whether PIAS1 synergizes with EBNA1 to restrict EBV replication, we co-transfected HEK-293 (EBV+) cells with PIAS1, WT EBNA1, and the RRR mutant. Without PIAS1, the RRR mutant showed increased EBV lytic replication compared to WT EBNA1 (Fig. 8B, lane 3 vs lane 2). In the presence of PIAS1, both WT EBNA1 and RRR mutant further inhibited EBV replication, with WT EBNA1 having the strongest effect (Fig. 8B, FR probe (10 nM). The shifted and free probes were resolved on 1.4% agarose gel in 1× TBE buffer. (C) EMSA was performed using increasing concentrations of purified PIAS1 and a 2× FR probe (10 nM). (D) EMSA was performed using a fixed concentration of EBNA1, increasing concentrations of PIAS1 and a 2× FR probe (10 nM). lanes 4 and 5 vs lanes 2 and 3). These results together suggested that both EBNA1 SUMOylation and its interaction with PIAS1 contribute to reduced EBV lytic replication.
## PIAS1 promotes the maintenance of oriP-based plasmid by EBNA1 through SUMOylation
EBNA1 is known to function in EBV genome maintenance. To investigate the role of SUMOylation in this process, we utilized pCEP4 plasmid as an EBV oriP-based replicon that expresses EBNA1 and contains oriP for plasmid maintenance. We established a pCEP4 stable cell line in HEK-293T cells carrying non-targeting control sgRNA (NC) or PIAS1 targeting sgRNA (sg-PIAS1) cells under hygromycin B selection (Fig. 9A andB).
To study plasmid maintenance, we conducted serial passages every 3 days without hygromycin B to monitor plasmid maintenance rates. We observed faster plasmid loss in PIAS1-depleted cells compared to control cells over 9 days (Fig. 9C), suggesting that PIAS1 contributes to pCEP4 plasmid retention.
To investigate the direct impact of EBNA1 SUMOylation on plasmid maintenance, we then mutated EBNA1 within pCEP4 to generate SUMOylation-deficient EBNA1 (RRR, K17R/K75R/K289R). The K289 residue corresponds to K241 in our EBNA1 dGAr due to additional GA repeats in pCEP4.
We then transfected HEK-293T cells with both pCEP4 EBNA1-WT and the pCEP4 EBNA1-RRR plasmids, followed by hygromycin B selection (Fig. 10A). After establishing stable cell lines (Fig. 10B), we conducted plasmid retention assays in the absence of hygromycin B. We found that there is an increased loss of pCEP4 EBNA1-RRR plasmid compared to pCEP4 EBNA1-WT plasmid over 9 days (Fig. 10C), indicating that EBNA1 SUMOylation promotes oriP-based plasmid retention. To determine whether SUMOyla tion affects EBNA1's association with oriP, we performed EBNA1 ChIP using cells carrying pCEP4 EBNA1-WT and EBNA1-RRR plasmids. Our results showed that EBNA1-RRR has significantly reduced binding to oriP compared with EBNA1-WT (Fig. 10D). These results suggest that EBNA1 SUMOylation promotes its binding to oriP.
Our previous study showed that PIAS1 is cleaved by caspases during lytic reactivation (20). We hypothesized that EBNA1 SUMOylation level decreases during EBV reactiva tion. To test this hypothesis, we induced EBV reactivation in Akata (EBV+) cells using anti-human IgG and monitored EBNA1-SUMO2/3 PLA signals over the course of EBV reactivation. We observed that EBNA1-SUMO2/3 PLA signals decrease upon reactivation for 24 h and 48 h (Fig. S2).
Furthermore, from 24 h to 48 h post-reactivation, we noticed that a gradual increase of EBNA1-SUMO2/3 PLA signals localized in the cytoplasm (Fig. S2), suggesting that SUMOylated EBNA1 is progressively translocated from the nucleus to the cytoplasm. This cytoplasmic translocation of SUMOylated EBNA1 likely reflects a loss of its nuclear function essential for maintaining EBV latency.
## DISCUSSION
EBNA1 is an essential EBV protein that plays a key role in viral genome replica tion, episome maintenance, and transcriptional regulation. It is the only viral protein consistently expressed across all types of EBV latency and in EBV-associated malignan cies. Structurally, EBNA1 contains a GAr region, DNA-binding and dimerization domains, and nuclear localization signals. Its primary functions include tethering the viral episome to host chromosomes during cell division and modulating the expression of both viral and host genes (3,5,6,8).
It was reported that EBNA1 is regulated by various PTMs, such as phosphorylation, arginine methylation, lysine hydroxylation, and SUMOylation. Phosphorylation is a key modification for EBNA1, with 10 phosphorylated residues identified by mass spectrome try. These phosphorylation sites are located in the N-terminal GAr, glycine/arginine-rich domains (GR1 and GR2), and near the nuclear localization sequence. Phosphorylationdeficient mutants show reduced oriP-dependent transcription and episome mainte nance, while retaining normal half-life and nuclear localization, thereby highlighting the importance of this modification (31)(32)(33).
In addition, phosphorylation of EBNA1 at Ser393 by viral and cellular kinases may influence its antigenicity, modulate antibody response, and promote cross-reactivity with GlialCAM, a phenomenon observed in clonally expanded B cells in multiple sclerosis (34)(35)(36).
Arginine methylation, catalyzed by protein arginine methyltransferases, is another crucial PTM affecting EBNA1 stability, protein interactions, transcription activation, and episome maintenance. GAr region is particularly important for segregation and transcriptional activation functions (32,37). EBNA1 stability and DNA replication activity are regulated by PLOD1-mediated lysine hydroxylation, as well as ubiquitin-proteasomedependent degradation (38,39).
SUMOylation, the covalent attachment of SUMO proteins, has been implicated in the regulation of EBNA1's functions. Previous studies suggested that loss of EBNA1 SUMOylation at K477 impairs viral DNA persistence and enhances spontaneous EBV reactivation (24). It was reported that SUMO2 is covalently attached to K1140 of the KSHV-LANA, promoting viral genome maintenance and repressing lytic reactivation through inhibition of replication and transcription activator (RTA) expression (40). In the case of HPV16, another study identified K292 of the E2 protein as a SUMOylation site, although its functional consequences remain unexplored (41).
In this study, we demonstrated that PIAS1 is specifically enriched at EBV oriP (Fig. 1), where it binds to and colocalizes with EBNA1 (Fig. 2). We further demonstrated that these interactions are mediated by the N-terminal and central regions of PIAS1 and the C-terminal DBD of EBNA1 (Fig. 3).
Importantly, we discovered that PIAS1 serves as an E3 SUMO ligase for EBNA1 (Fig. 4). In addition to the previously characterized K477 (corresponding to K241 in GAr deleted EBNA1), we identified two novel SUMOylation sites in EBNA1, namely K17 and K75 (Fig. 5). These two sites reside in non-consensus SUMOylation motifs. Mutation of the individual lysine residue to arginine did not affect EBNA1 SUMOylation. However, the mutation of three sites abolished EBNA1 SUMOylation signals. These findings underscore the functional relevance and prevalence of non-consensus SUMOylation sites in EBNA1 regulation within cells (Fig. 5).
The binding of PIAS1 to EBNA1 was also implicated in EBNA1's DNA binding, where PIAS1 slightly promotes EBNA1 binding to oriP DNA (Fig. 6). The loss of PIAS1 also diminished EBNA1 binding to oriP in Akata (EBV+) cells (Fig. 7), suggesting that PIAS1 plays a critical role in EBNA1's chromatin binding activity at oriP.
As previously reported, both PIAS1 and EBNA1 have been implicated in EBV lytic reactivation (20,30). Intriguingly, we found that SUMOylation-deficient EBNA1 is compromised in limiting EBV lytic replication, and PIAS1 synergizes with EBNA1 to further restrict viral replication (Fig. 8). These findings expand the function of PIAS1mediated SUMOylation from our previously identified targets, SAMHD1 and YTHDF2 (18,19), to include EBV EBNA1.
EBNA1 was recently reported to coordinate with H2A.Z for epigenetic reprogramming of EBV episomes (42). Interestingly, we demonstrated that both PIAS1 binding and SUMOylation of EBNA1 contribute to the retention of EBV oriP-based replicon (Fig. 9 and 10), suggesting an important regulatory mechanism for EBV episome maintenance.
Our study paves the way for exploring whether KSHV LANA (40) and HPV E2 (41) are regulated by PIAS1. In co-transfection systems, we observed that PIAS1 interacts with KSHV LANA and HPV-16 E2, promoting their SUMOylation in vitro (Fig. S3). The functional significance of these interactions warrants further investigation, particularly in light of previous proteomic screens that identified PIAS1 as a potential E2-interacting partner (43). In summary, we identify PIAS1 as a key E3 SUMO ligase for EBNA1, enriched at EBV oriP, where it binds and colocalizes with EBNA1. We uncover two novel SUMOylation sites (K17 and K75) along with K241, which plays important roles in regulating EBV lytic replication and episome retention (Fig. 11). These findings reveal a novel role for PIAS1 in controlling EBV latency through EBNA1 SUMOylation and open new avenues to investigate whether similar mechanisms regulate KSHV LANA and HPV E2.
## MATERIALS AND METHODS
## Cell lines and cultures
Akata (EBV+) cells were cultured in Roswell Park Memorial Institute medium (RPMI 1640) supplemented with 10% fetal bovine serum (FBS) (Cat. # 26140079, Thermo Fisher Scientific) in 5% CO 2 at 37°C (20,(44)(45)(46)(47). HEK-293 (EBV+) cells with B95.8 EBV genome were maintained in 150 µg/mL Hygromycin B (Cat. # J60681MC, Thermo Fisher Scientific). HEK-293 (EBV+) and HEK-293T cells were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% FBS in 5% CO 2 at 37°C (48,49). See also Table 1 for cell line sources.
## Chromatin immunoprecipitation sequencing
ChIP assays were performed using the SimpleChIP Enzymatic Chromatin IP Kit (Magnetic Bead; Cat. #9003S, Cell Signaling Technology) according to the manufacturer's proto col. Briefly, 4-5 × 10 6 Akata (EBV+) cells were cross-linked with 1% formaldehyde and subsequently digested with micrococcal nuclease to yield chromatin fragments ranging from 150 to 900 bp. Two percent of the digested chromatin was reserved as an input control. For IP, 6 µg of chromatin was incubated with 1 µg of anti-PIAS1 antibody (Cat. #ab77231, Abcam). Following IP, cross-links were reversed, and DNA was purified for downstream analysis.
For ChIP-seq, 0.1-20 ng of ChIP-derived DNA was further sheared to an average fragment size of ~200 bp using a Covaris system (2 min sonication in 15 µL microTUBE-15 AFA Beads Screw-Cap tubes, Cat. #520145, Covaris). Sequencing libraries were prepared using the NEBNext Ultra II DNA Library Prep Kit for Illumina (Cat. #E7645L, New Eng land Biolabs), which features a low-input optimized adapter ligation system to reduce bias. Final libraries were quantified using both Qubit fluorometric quantitation and
## ChIP-seq analysis
ChIP-seq FASTQ files were processed and analyzed using the GALAXY web server (53).
Initially, raw reads were assessed for quality using FastQC, and low-quality sequences were trimmed accordingly. High-quality reads were then aligned to the human reference genome and the EBV Akata strain (GenBank: KC207813) using Bowtie2. Peak calling was performed with MACS2, using input DNA as a control to identify significant enrichment regions (54). For data visualization, aligned reads and peaks were loaded into the Integrative Genomics Viewer (55).
The Raji EBNA1 ChIP-seq data set was retrieved from GEO (accession number GSM982656) and aligned to the B95.8 EBV reference genome (NCBI accession NC_007605) (7,23).
## Chromatin immunoprecipitation qPCR
ChIP process and qPCR were previously described (19). DNA-protein complexes were immunoprecipitated with anti-PIAS1 antibody (Cat. #ab77231, Abcam) and rabbit IgG control (Cat. #2729, Cell Signaling Technology). ChIP was performed using an Enzymatic Chromatin IP kit (Cell Signaling Technology, SimpleChIP Enzymatic Chromatin IP kit) as described previously (20). IPed DNA was quantified by qPCR using oriP-specific primers (Table S1).
## Plasmid construction
Halo-PIAS1, Halo-V5-PIAS1 (full length and aa 1-100), and V5-PIAS1 (full length, aa 1-415, aa 409-651, and aa 101-433) plasmids were previously described (20).
EBNA1 dGAr insert was amplified from p3xFlag_CMVm-EBNA1 dGAr_Akata, a gift from Kathy Shair (51) by PCR with Q5 high-fidelity DNA polymerase (Q5-PCR) and cloned into the pHTN-CMV-Neo vector with an N-terminal V5 tag via Gibson Assembly. Halo-V5-EBNA1 was used as a template to create K17R, K75R, and K241R mutants using site-directed mutagenesis kit with Pfu Ultra II Fusion HotStart DNA Polymerase (Cat. #600672, Agilent) according to the manufacturer's instructions on Quickchange II system (Cat. #200523, Agilent). The SUMOylation-deficient EBNA1 in pCEP4 was mutated using the same approach. Halo-HA-EBNA1 (aa 1-215) was generated by adding a stop codon after aa 215. For Halo-HA-EBNA1 (aa 216-405), the corresponding DNA was amplified by Q5-PCR and then digested using EcoRI and NotI-HF and cloned into pHTN-CMV-Neo vector.
HPV16-E2 insert was amplified from pCDNA-HPV16-E2 (a gift from Iain Morgan) by Q5-PCR and cloned into pHTN-CMV-Neo vector with an N-terminal V5-tag via Gibson Assembly. See also Table 1 for construct sources. All primer sequences are listed in Table S1.
## In situ PLA
PLA was modified as previously described (18). Briefly, cells were blocked with 3% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) at room temperature for 1 h, then incubated with PBS control or a mixture of mouse anti-EBNA1 (Cat. #sc-81581, Santa Cruz) and rabbit anti-PIAS1 (Cat. #ab77231, Abcam) or rabbit anti-SUMO2/3 (Cat. # 11251-1-AP, Proteintech) antibodies (1:50 dilution in 3% BSA) at 4°C overnight. Then the probes were incubated at 37°C for 1 h, followed by ligation and amplification (NaveniFlex Cell Red #NC.MR.100, Navinci). Cell nuclei were stained using Duolink in situ mounting media with DAPI and visualized by Nikon AXR confocal microscope.
## Plasmid retention assay
HEK-293T (non-targeting control and sg-PIAS1) cells were established previously (18). Two micrograms of pCEP4 (WT EBNA1 or SUMOylation-deficient mutant) was transfected into the cells using PEI max for 24 h, the culture medium was changed, and the cells were selected under 150 µg/mL hygromycin B until stable cell lines were established around 14 days post-transfection (35). The cells were split into 3 × 10 5 cells/mL in 10 cm plate with 10 mL of DMEM + 10% FBS and incubated in 5% CO 2 at 37°C without hygromycin B. One portion of the cells was harvested for WB analysis. Every 3 days, the cells were passaged and reseeded at the same density (3 × 10 5 cells/mL) in fresh 10 cm plates with 10 mL of DMEM + 10% FBS. This process was repeated for a total of 9 days. At each passage, 6 × 10 5 cells were harvested for pCEP4 DNA detection. Total genomic DNA was extracted using the Genomic DNA Purification Kit (Cat. #A1120, Promega). Relative plasmid copy numbers were similarly measured by qPCR using previously described primers (35) and normal ized with β-actin gene (Table S1).
## Cell lysis, immunoblotting, and IP
Cell lysis, IP, and immunoblotting (WB) were performed as previously described (18), with minor modifications. Cells were harvested, lysed in 2× SDS-PAGE sample buffer, and boiled for 5 min. Proteins were resolved on 4%-20% TGX gels (Cat. #4561096; Bio-Rad), transferred to polyvinylidene fluoride (PVDF) membranes, and probed with the indicated primary antibodies followed by horseradish peroxidase-conjugated secondary antibodies. See also Table 1 for antibody sources.
For IP, cells were lysed in buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.1% NP-40, and protease inhibitor cocktail (Cat. #4693116001; Sigma-Aldrich) on ice for 30 min. Lysates were sonicated (10 s on/10 s off, three cycles, 35% amplitude) and clarified by centrifugation at 14,600 × g for 15 min at 4°C. Ten percent of the supernatant was reserved as input, and the remainder was incubated with the indicated magnetic beads. Input and immunoprecipitated proteins were analyzed by immunoblotting using the indicated antibodies. Anti-HPV16-E2 antibody is a gift from Iain Morgan.
## Protein expression and purification
Halo-tagged PIAS1, EBNA1, and HPV-16 E2 proteins were expressed and purified as previously described (18), with minor modifications. Briefly, HEK-293T cells were transfected with 18 µg plasmid DNA and 54 µg PEI Max and harvested 48 h later. Cells were lysed in Halo purification buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.005% NP-40, and 1 mM DTT), sonicated (10 s on/10 s off, three cycles, 35% amplitude), and clarified by centrifugation at 14,600 × g for 15 min at 4°C. The supernatant was incubated with 200 µL pre-washed Halo resin at 4°C overnight. Beads were washed three times with Halo purification buffer and subsequently incubated with Halo purification buffer containing TEV protease at 4°C overnight and then the cleaved proteins were eluted with Halo purification buffer.
## In vitro SUMOylation assay
In vitro SUMOylation assay was performed using the SUMO2 conjugation kit as previously described (18,19), with minor modifications. Reactions were carried out in a buffer containing 40 mM Tris (pH 7.1), 40 mM NaCl, 1 mM β-mercaptoethanol, and 5 mM MgCl₂. The substrates (EBNA1, LANA, or HPV-16 E2) were incubated with 100 nM SAE1/ SAE2 (E1), 2 µM His₆-Ube2I/UBC9 (E2), 50 µM SUMO2, and 4 mM ATP, with PIAS1 as the E3 ligase in the presence or absence of 50 ng/µL pCEP4 or pOri16M (52) plasmid. Reactions were incubated at 37°C for 3 h, and SUMOylation was assessed by immunoblotting.
For in vitro SUMOylation of LANA, HEK-293T cells were transfected with 10 µg of p3×FLAG-CMV-LANA, plasmid from Diane Hayward's Lab collection, and harvested 48 h post-transfection. Cells were lysed in lysis buffer, and LANA was immunoprecipita ted using Anti-FLAG M2 magnetic beads (Cat. #M8823; Millipore Sigma). SUMOylation reactions were performed directly on the bead-bound protein with gentle agitation.
## Lytic induction and EBV copy number detection
For lytic induction of EBV in HEK-293 (EBV+) cells, the cells were transfected with EBV ZTA plus other plasmids as indicated using PEI max for 48 h as described previously (18). Meanwhile, for lytic induction, Akata (EBV+) cells were seeded at a density of 1 × 10 6 cells/mL in six-well plates. After 3 h, anti-human IgG (50 µg/mL; Cat. #0855087, MP Biomedicals) was added to the cells, which were then harvested at the indicated time points as previously described (20).
Extracellular viral DNA was extracted and quantified following established protocols (18,22). Briefly, EBV-containing media were treated with RQ1 RNase-free DNase (Cat. #M6101; Promega) to remove naked DNA, and the reaction was terminated with the supplied stop buffer. Proteinase K (Cat. #BIO-37084; Meridian Bioscience) and SDS were then added to digest viral proteins and to release virion-associated DNA. EBV DNA was purified by phenol-chloroform extraction and precipitated with isopropanol, sodium acetate, and glycogen at -80°C overnight. DNA pellets were washed with 70% ethanol, air dried, and resuspended in Tris-EDTA buffer (10 mM Tris and 1 mM EDTA, pH 8.0). EBV DNA was detected by PCR using BALF5-specific primers (18).
## Structure prediction by AlphaFold3
AlphaFold3 algorithm (27) was employed to predict the three-dimensional structure of EBNA1 (dGAr) dimer with 1× FR DNA sequence (5′-GGATAGCATATACTACCCGGATATAGATT A-3′). Molecular graphics of EBNA1 were performed with UCSF ChimeraX (56), developed by the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco, with support from National Institutes of Health R01-GM129325 and the Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases. Model 1 of the prediction was used to display EBNA1 structure.
## Electrophoretic mobility shift assay
Purified V5-EBNA1 WT and EBNA1 RRR proteins were serially diluted (0-500 nM) and incubated with a 2× FR DNA probe labeled with IRDye 700 (10 nM) (5) in Halo purification buffer for 3 h at 30°C. Reactions were stopped by the addition of 2× nucleic acid loading buffer (50 mM Tris-HCl pH 8.0, 20% glycerol, 2 mM EDTA, and 0.1% Bromphenol Blue) and resolved on a 1.4% agarose gel at 80 V in 1× TBE running buffer. DNA-protein complexes were visualized using a LI-COR Odyssey Fc imaging system under the 700 nm channel.
## Quantification and statistical analysis
Statistical analyses were performed using a two-tailed Student t-test with Microsoft Excel software. A P value less than 0.05 was considered statistically significant. The values are presented as means and standard deviations for biological replicate experiments as specified in the figure legends. Figure 11 was created using BioRender.
## References
1. Epstein, Achong, Barr (1964) "Virus particles in cultured lymphoblasts from Burkitt's lymphoma" *Lancet*
2. Damania, Kenney, Raab-Traub (2022) "Epstein-Barr virus: biology and clinical disease" *Cell*
3. Sugiokto, Li (2025) "Targeting EBV episome for anti-cancer therapy: emerging strategies and challenges" *Viruses*
4. Lista, Martins, Billant et al. (2017) "Nucleolin directly mediates Epstein-Barr virus immune evasion through binding to G-quadruplexes of EBNA1 mRNA" *Nat Commun*
5. Dheekollu, Wiedmer, Ayyanathan et al. (2021) "Cell-cycle-dependent EBNA1-DNA crosslinking promotes replication termination at oriP and viral episome mainte nance" *Cell*
6. Frappier, Donnell (1991) "Epstein-Barr nuclear antigen 1 mediates a DNA loop within the latent replication origin of Epstein-Barr virus" *Proc Natl Acad Sci*
7. Lu, Wikramasinghe, Norseen et al. (2010) "Genome-wide analysis of host-chromosome binding sites for Epstein-Barr virus nuclear antigen 1 (EBNA1)" *Virol J*
8. Mei, Messick, Dheekollu et al. (2022) "Cryo-EM structure and functional studies of EBNA1 binding to the family of repeats and dyad symmetry elements of Epstein-Barr virus oriP" *J Virol*
9. Leo, Calderon, Lieberman (2020) "Control of viral latency by episome maintenance proteins" *Trends Microbiol*
10. Hellert, Weidner-Glunde, Krausze et al. (2013) "A structural basis for BRD2/4-mediated host chromatin interaction and oligomer assembly of Kaposi sarcoma-associated herpesvirus and murine gammaherpesvi rus LANA proteins" *PLoS Pathog*
11. Toptan, Fonseca, Kwun et al. (2013) "Complex alternative cytoplasmic protein isoforms of the Kaposi's sarcomaassociated herpesvirus latency-associated nuclear antigen 1 generated through noncanonical translation initiation" *J Virol*
12. Grundhoff, Ganem (2003) "The latency-associated nuclear antigen of Kaposi's sarcoma-associated herpesvirus permits replication of terminal repeat-containing plasmids" *J Virol*
13. Hu, Garber, Renne (2002) "The latency-associated nuclear antigen of Kaposi's sarcoma-associated herpesvirus supports latent DNA replication in dividing cells" *J Virol*
14. Mckinney, Kim, Chen et al. (2016) "Brd4 activates early viral transcription upon human papillomavirus 18 infection of primary keratinocytes" *mBio*
15. Prabhakar, James, Fontan et al. (2023) "Direct interaction with the BRD4 carboxyl-terminal motif (CTM) and TopBP1 is required for human papillomavirus 16 E2 association with mitotic chromatin and plasmid segregation function" *J Virol*
16. Gareau, Lima (2010) "The SUMO pathway: emerging mechanisms that shape specificity, conjugation and recognition" *Nat Rev Mol Cell Biol*
17. Shuai, Liu (2005) "Regulation of gene-activation pathways by PIAS proteins in the immune system" *Nat Rev Immunol*
18. Sugiokto, Saiada, Zhang et al. (2024) "SUMOylation of the m6A reader YTHDF2 by PIAS1 promotes viral RNA decay to restrict EBV replication" *mBio*
19. Saiada, Zhang, Li (2021) "PIAS1 potentiates the anti-EBV activity of SAMHD1 through SUMOylation" *Cell Biosci*
20. Zhang, Lv, Li (2017) "B cell receptor activation and chemical induction trigger caspase-mediated cleavage of PIAS1 to facilitate Epstein-Barr virus reactivation" *Cell Rep*
21. Zhang, Lv, Li (2019) "Conserved herpesvirus protein kinases target SAMHD1 to facilitate virus replication" *Cell Rep*
22. Zhang, Zhang, Maharjan et al. (2021) "Caspases switch off the m 6 A RNA modification pathway to foster the replication of a ubiquitous human tumor virus"
23. Arvey, Tempera, Tsai et al. (2012) "An atlas of the Epstein-Barr virus transcriptome and epigenome reveals host-virus regulatory interactions" *Cell Host Microbe*
24. Wang, Du, Zhu et al. (2020) "STUB1 is targeted by the SUMOinteracting motif of EBNA1 to maintain Epstein-Barr Virus latency" *PLoS Pathog*
25. Tammsalu, Matic, Jaffray et al. (2014) "Proteome-wide identification of SUMO2 modification sites" *Sci Signal*
26. Sievers, Wilm, Dineen et al. (2011) "Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega" *Mol Syst Biol*
27. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
28. Ohara, Hayashi, Teramoto et al. (2000) "Sequence analysis and variation of EBNA-1 in Epstein-Barr virus-related herpesvirus of cynomolgus monkey" *Intervirology*
29. Lussier-Price, Mascle, Cappadocia et al. (2020) "Characterization of a C-terminal SUMO-interacting motif present in select PIAS-family proteins" *Structure*
30. Sivachandran, Wang, Frappier (2012) "Functions of the Epstein-Barr virus EBNA1 protein in viral reactivation and lytic infection" *J Virol*
31. Duellman, Thompson, Coon et al. (2009) "Phosphorylation sites of Epstein-Barr virus EBNA1 regulate its function" *J Gen Virol*
32. Shire, Kapoor, Jiang et al. (2006) "Regulation of the EBNA1 Epstein-Barr virus protein by serine phosphorylation and arginine methylation" *J Virol*
33. Noh, Park, Joo et al. (2016) "ERK2 phosphor ylation of EBNA1 serine 383 residue is important for EBNA1-dependent transactivation" *Oncotarget*
34. Lanz, Brewer, Ho et al. (2022) "Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM" *Nature*
35. Zhu, Liao, Shan et al. (2009) "Protein array identification of substrates of the Epstein-Barr virus protein kinase BGLF4" *J Virol*
36. Kang, Lee, Soni et al. (2011) "Roscovitine inhibits EBNA1 serine 393 phosphorylation, nuclear localization, transcription, and episome maintenance" *J Virol*
37. Holowaty, Zeghouf, Wu et al. (2003) "Protein profiling with Epstein-Barr nuclear antigen-1 reveals an interaction with the herpesvirus-associated ubiquitin-specific protease HAUSP/USP7" *J Biol Chem*
38. Dheekollu, Wiedmer, Soldan et al. (2023) "Regulation of EBNA1 protein stability and DNA replication activity by PLOD1 lysine hydroxylase" *PLoS Pathog*
39. Chen, Addepalli, Soldan et al. (2025) "USP7 inhibitors destabilize EBNA1 and suppress Epstein-Barr virus tumorigenesis" *J Med Virol*
40. Cai, Cai, Zhu et al. (2013) "A unique SUMO-2-interacting motif within LANA is essential for KSHV latency" *PLoS Pathog*
41. Wu, Roark, Bian et al. (2008) "Modification of papillomavi rus E2 proteins by the small ubiquitin-like modifier family members (SUMOs)" *Virology (Auckl)*
42. Castro-Muñoz, Maestri, Yoon et al. (2025) "Histone variant H2A.Z cooperates with EBNA1 to maintain Epstein-Barr virus latent epigenome" *mBio*
43. Muller, Jacob, Jones et al. (2012) "Large scale genotype comparison of human papillomavirus E2-host interaction networks provides new insights for E2 molecular functions" *PLoS Pathog*
44. Li, Zhu, Xie et al. (2011) "Conserved herpesvirus kinases target the DNA damage response pathway and TIP60 histone acetyltransferase to promote virus replication" *Cell Host Microbe*
45. Li, Wang, Liao et al. (2012) "SUMO binding by the Epstein-Barr virus protein kinase BGLF4 is crucial for BGLF4 function" *J Virol*
46. Li, Liao, Nirujogi et al. (2015) "Phosphoproteomic profiling reveals Epstein-Barr virus protein kinase integration of DNA damage response and mitotic signaling" *PLoS Pathog*
47. Lv, Zhang, Li (2018) "Interferon regulatory factor 8 regulates caspase-1 expression to facilitate Epstein-Barr virus reactivation in response to B cell receptor stimulation and chemical induction" *PLoS Pathog*
48. Feederle, Mehl-Lautscham, Bannert et al. (2009) "The Epstein-Barr virus protein kinase BGLF4 and the exonuclease BGLF5 have opposite effects on the regulation of viral protein production" *J Virol*
49. Li, Walsh, Lam et al. (2019) "A single phosphoacceptor residue in BGLF3 is essential for transcription of Epstein-Barr virus late genes" *PLoS Pathog*
50. Prabhakar, James, Youssef et al. (2024) "A human papillomavirus 16 E2-TopBP1 dependent SIRT1-p300 acetylation switch regulates mitotic viral and human protein levels and activates the DNA damage response" *mBio*
51. Warner, Patel, Wang et al. (2024) "The Epstein-Barr virus nuclear antigen 1 variant associated with nasopharyngeal carcinoma defines the sequence criteria for serologic risk prediction" *Clin Cancer Res*
52. Taylor, Morgan (2003) "A novel technique with enhanced detection and quantitation of HPV-16 E1-and E2-mediated DNA replication" *Virology (Auckl)*
53. Galaxy (2024) "The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update" *Nucleic Acids Res*
54. Feng, Liu, Qin et al. (2012) "Identifying ChIP-seq enrichment using MACS" *Nat Protoc*
55. Robinson, Thorvaldsdóttir, Winckler et al. (2011) "Integrative genomics viewer"
56. Pettersen, Goddard, Huang et al. (2021) "UCSF ChimeraX: structure visualization for researchers, educators, and developers" *Protein Sci*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12613979&blobtype=pdf
|
# Complete genome sequences of Plantago lanceolata latent virus isolated from asymptomatic ribwort plantain plants in France, Italy, and Spain
Denis Filloux, Serge Galzi, Philippe Roumagnac
## Abstract
Plantago lanceolata latent virus (PlLV) that infects ribwort plantain, initially discovered in Finland, is also present in Western Europe. Four complete genome sequences of PlLV from France, Italy, and Spain were obtained that shared 94.17%-97.85% genome-wide pairwise identity with complete genomes of PlLV previously obtained from Finland. KEYWORDS geminivirus, ribwort plantain P lantago lanceolata latent virus (PlLV) that belongs to the genus Capulavirus (Geminiviridae family) infects ribwort plantain (Plantago lanceolata L.) (1), a perennial herb that is native from Eurasia and is now widespread all over the world (2). This virus was discovered in 2015 from asymptomatic, uncultivated ribwort plantain plants collected in the Åland archipelago of southwestern Finland (1). We here investigated to what extent the virus was present outside of Finland, particularly in Western Europe. One hundred and thirty-one leaves from asymptomatic ribwort plantain plants were collected in 2018 from France (94 samples), Italy (17 samples), and Spain (20 sam ples). Total DNA of ribwort plantain plants was extracted using the DNeasy Plant Mini Kit (Qiagen, Germany). PCR-mediated detection of PlLV from the 131 plant samples was performed using PCR primers F-PlLV_729 (5′-AAGGGAAAGGCTGGTTATGG-3′) and R-PlLV_1013 (5′-GAATCTCTTCTCTGAATCGTGGTC-3′). Amplification conditions consisted of 95°C for 5 min; 30 cycles at 95°C for 45 s, 50°C for 45 s, 72°C for 30 s; and 72°C for 5 min. The PCR amplicons corresponding to visible agarose gel bands were Sanger sequenced by Azenta (Germany) using the ABI3730xl sequencer (Applied Biosystems). PCR analysis revealed that PlLV was present in 2/94 samples in France, 1/17 samples in Italy, and 3/20 samples in Spain. PCR amplification of the complete genome of 4/6 positive samples from France, Italy, and Spain (one sample was selected from each region: Brittany, Galicia, Languedoc, and Emilia-Romagna; Table 1) was performed using a pair of abutting primers, Pla_pstIF (5′-CTGCAGATCATTGTATAAATACTGTCCCAAATACG -3′) and Pla_pstIR (5′-CTGCAGTATCTGTGATATTTGTATACAAATTTCTGAC-3′), as previously described (1). Amplicon products of approximately 2.8 Kbp were excised, gel-purified, and cloned into the plasmid pGEM-T Easy and further Sanger sequenced by primer walking (Azenta). Four PlLV complete genome sequences were obtained, ranging in size from 2,832 nt to 2,834 nt (%GC content of 41.8%). Pairwise identity analyses of the full genome nucleotide sequences were further carried out using SDT v1.2 using default settings (3), and indicated that the four PlLV complete genome sequences obtained in this study and both complete genomes obtained previously from Finland (1) shared 94.17-97.85% genome-wide pairwise identity (Fig. 1A). The observed degree of similarity exceeds the species demarcation threshold recommended for the Capulavirus genus (4), indicating that the four isolates from France, Italy, and Spain could be classified
as PlLV variants. The evolutionary relationships of the six PlLV isolates (four from this study and two from the study conducted in Finland) and representative members of capulaviruses were reconstructed using the complete nucleotide genome sequences, the replication-associated protein (REP), and the coat protein (CP) amino acid sequen ces. Alignment was carried out using MAFFT (5). Block mapping and gathering with entropy (6) were used, and a Maximum-Likelihood tree was inferred by FastTree with 1,000 bootstrap iterations (7). The three phylogenetic analyses based on the complete nucleotide genome sequences (Fig. 1B), the CP protein (Fig. 1C), and the REP protein (Fig. 1D) sequences all indicate that the four PlLV genomes from France, Italy, and Spain cluster with the other two Finnish isolates previously described. This study demonstrates that PlLV is not confined to a few islands in the Baltic Sea but is widespread in Western Europe.
## References
1. Susi, Laine, Filloux et al. (2017) "Genome sequences of a capulavirus infecting Plantago lanceolata in the Åland archipelago of Finland" *Arch Virol*
2. Upadhyay, Bhandari, Sharma et al. (2024) "Plantago lanceolata L"
3. Muhire, Varsani, Martin (2014) "SDT: a virus classification tool based on pairwise sequence alignment and identity calculation" *PLoS One*
4. Varsani, Roumagnac, Fuchs et al. (2017) "Capulavirus and Grablovirus: two new genera in the family Geminiviridae" *Arch Virol*
5. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
6. Criscuolo, Gribaldo (2010) "BMGE (Block Mapping and Gathering with Entropy): a new software for selection of phylogenetic informative regions from multiple sequence alignments" *BMC Evol Biol*
7. Price, Dehal, Arkin (2010) "FastTree 2--approximately maximumlikelihood trees for large alignments" *PLoS One*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12509549&blobtype=pdf
|
# Complete genome sequence of Baby Chick Ranikhet Disease Vaccine (BCRDV) strain in Bangladesh
Nasreen Parveen, Md Karim, Shushoma Happy, Tahmina Begum, Anowar Hossen, Asm Uddin, Tania Ahmed, Sultana Jahan, Md Kamal, Mohammed Samad
## Abstract
Newcastle disease (ND), a highly contagious viral disease of poultry, is commonly controlled using live-attenuated vaccines. The BCRDV vaccine in Bangladesh was developed from the lentogenic strain of ND virus through serial passaging in embryonated chicken eggs. Here, we present the complete genome sequence of the BCRDV vaccine strain used in Bangladesh.KEYWORDS complete genome sequence, Newcastle disease virus, BCRDV vaccine N ewcastle disease (ND) is an acute and highly contagious viral disease of poul try caused by Newcastle disease virus (NDV) or avian paramyxovirus type 1 (APMV-1), a negative-sense, single-stranded RNA virus of the genus Orthoavulavirus of the family Paramyxoviridae (1). NDV exhibits neurological, respiratory, gastrointestinal, and reproductive signs, especially in unvaccinated birds (2). The genome encodes six structural proteins, with the fusion (F) protein, which determines virulency (3). NDV strains are categorized into lentogenic or mild, mesogenic or moderate, and velogenic or very virulent pathotypes. Biosecurity and vaccination, especially using live and inactiva ted or genotype-matched vaccines, are essential for prevention and control of ND (4).Here, we report the complete genome sequence of the BCRDV vaccine strain (GenBank accession no. PV809488). The complete genome sequence was determined using next-generation sequencing (NGS) on the Illumina platform, by aligning it with the complete genome sequence of NDV reference sequence (NC_075404.1). Total RNA was extracted from the culture supernatant of the master seed of BCRDV vaccine strain in 9-10-day-old embryonated chicken eggs using the Quick-RNA Viral Kit (Cat. No. R1035, Zymo Research), following the manufacturer's protocol. A sequencing library was prepared using the Illumina Standard Total RNA Prep, Ligation with Ribo-Zero Plus kit, resulting in an average fragment size of 400 bp. Sequencing was performed on the Illumina NextSeq 2000 platform with 2 × 150 bp paired-end chemistry, yielding approximately 1.62 million read pairs. Raw read quality was assessed using FastQC v0.11.3 (5). Adapter trimming and quality filtering were conducted with Trimmomatic v0.39.2 (6), applying the following parameters: minimum quality score >30, minimum read length >35 bp, and 5′ clip of 15 bases for both forward (R1) and reverse (R2) reads. Reads specific to BCRDV were identified using Kraken2 v2.1.3 (7) in conjunction with the viral reference database (version: k2_viral_20210517). The extracted BCRDV reads were assembled de novo using SPAdes v3.14.0 with k-mer sizes of 77 and 99 (8). The resulting assembly was manually curated to confirm the final viral genome sequence with median depth coverage, 1,971×, which was determined to be 15,186 bp in length, containing six open reading frames. The presence of six open reading frames was identified using Prokka annotation. Completeness of the genome was confirmed by alignment with the NDV reference sequence (NC_075404.1) in GenBank, which showed full coverage of all coding sequences (CDS) as well as the non-coding 5′ and 3′ UTRs, indicating that
sequencing extended to both genome ends. The average GC content of the genome was 46%. Genome annotation was initially performed using Prokka v1.14.6 (9), and annotations were manually refined before submission to NCBI (10).
Sequence analysis revealed that the complete genome of the vaccine strain, designated BCRDV, is 15,186 nucleotides (nt) in length, comprising six transcriptional units in the order 3′-NP-P-M-F-HN-L-5′, with respective gene lengths of 1,470 nt (NP), 1,188 nt (P), 1,095 nt (M), 1,662 nt (F), 1,719 nt (HN), and 6,615 nt (L). Multiple sequence alignment of the deduced amino acid sequences was conducted to assess genetic and antigenic features of our study sequence in comparison with reference NDV strains retrieved from GenBank, including LaSota (JF950510), Hitchner-B1 (AF309418), Komarov (KT445901), and Herts/33 (AY741404). Analysis of the fusion (F) protein cleavage site demonstrated the presence of the motif 112 G-R-Q-G-R↓L 117 , which is characteristic of lentogenic NDV strains, confirming the low-virulence nature of the BLRI-LRI-BCRDV vaccine strain. WebLogo analysis of the deduced amino acid sequences further supported the multiple sequence alignment findings. The monobasic amino acid at the C terminus of the cleavage site and leucine residue at 117 suggest low pathogenicity. In BLASTn analysis, the nucleotide of BLRI-LRI-BCRDV sequence shows 99.69% identity with the NDV strain from Croatia (KJ670427.1), 99.66% identity with the lentogenic NDV strain F from Bangladesh (ON713865.1), and 99.62% identity with the ndv59/F strain from India (KM056355.1) in the NCBI Core nucleotide database (core_nt). A phylogenetic tree was constructed using the maximum likelihood method coupled with the Kimura two-parameter model with bootstrap analysis of 100 replicates in MEGA, version X (11). Phylogenetic analysis showed that the BCRDV vaccine strain clusters with other representative lentogenic pathotypes (Fig. 1). The VaxiJen server (12) also indicated that the BCRDV strain possesses higher antigenic potential compared to its corresponding standard pathotype strains, with an overall protective antigen prediction score of 0.5620.
## References
1. Amarasinghe, Ayllón, Bào et al. (2019) "Taxonomy of the order Mononegavirales: update" *Arch Virol*
2. Miller, Koch, Swayne et al. (2013) "Newcastle disease"
3. Rima, Balkema-Buschmann, Dundon et al. (2019) "ICTV virus taxonomy profile: Paramyxoviridae" *J Gen Virol*
4. Nedeljković, Mazija, Cvetić et al. (2022) "Comparison of chicken immune responses to immunization with vaccine La Sota or ZG1999HDS strain of Newcastle disease virus" *Life (Basel)*
5. Andrews (2010) "FastQC: a quality control tool for high throughput sequence data"
6. Bolger, Lohse, Usadel (2014) "Trimmomatic: a flexible trimmer for Illumina sequence data" *Bioinformatics*
7. Wood, Lu, Langmead (2019) "Improved metagenomic analysis with Kraken 2" *Genome Biol*
8. Bankevich, Nurk, Antipov et al. (2012) "SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing" *J Comput Biol*
9. Seemann (2014) "Prokka: rapid prokaryotic genome annotation" *Bioinformatics*
10. Samad, Hossen, Karim et al. (2025) "Complete genome sequence of a Bangladesh-developed live-attenuated lumpy skin disease (LSD) vaccine strain" *Microbiol Resour Announc*
11. Kumar, Stecher, Li et al. (2018) "MEGA X: molecular evolutionary genetics analysis across computing platforms" *Mol Biol Evol*
12. Doytchinova, Flower (2007) "VaxiJen: a server for prediction of protective antigens, tumour antigens and subunit vaccines" *BMC Bioinformatics*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792841&blobtype=pdf
|
# P-1801. Clinical Evaluation of a Novel, Extraction-free Isothermal Nucleic Acid Amplification Test for the Detection and Differentiation of HSV-1 and HSV-2 from Various Pediatric Sample Sources
Benjamin Liu
## Abstract
Background. HSV is a DNA virus that causes various infections in adults and children. While HSV-1 and HSV-2 share a preference for epithelial and neuronal tissues, HSV-1 usually causes cold sores and HSV-2 commonly leads to genital herpes or lesions. It is essential to rapidly and accurately detect and differentiate HSV-1 and -2 from various sample sources, especially from pediatric populations. However, there are significant diagnostic gaps in FDA-approved PCR tests for detection and differentiation of HSV-1 and -2, for example, limited approved sources, false positives and inter-type cross-reactivity. Herein we aimed to perform clinical evaluation of a novel, extraction-free isothermal nucleic acid amplification test (iAMP HSV 1/2, Atila BioSystems) for the detection and differentiation of HSV-1 and -2 from various pediatric sample sources.Methods. Swab samples collected from oral, skin, genital and eye sites of pediatric patients at Children's National Hospital were detected using iAMP HSV 1/2 Detection Kit and standard-of-care Simplexa HSV 1 & 2 (Diasorin). Positive, negative and overall percentage agreement of iAMP test was determined against Simplexa test. Accuracy of type differentiation of iAMP HSV 1/2 Detection Kit was determined. Discrepant analysis by patient chart review was performed for any samples with discrepant results between both tests.Results. 178 swab samples were collected from pediatric patients aged ranging from 1 month to 19 years. Among 104 Simplexa negative samples, 100.0% (104/104) were tested negative by iAMP test. Among 74 Simplexa positive samples, 96.0% (71/74) were tested positive by iAMP test, with 3 discrepant samples (Simplexa positive but iAMP negative). The accuracy of type differentiation of iAMP test was 100% (71/71) compared to Simplexa test. After discrepancy analysis, final positive, negative and overall percentage agreement of iAMP test were 98.6% (71/72), 100% (106/106), and 99.4% (177/178) compared to Simplexa test.Conclusion. This study demonstrated excellent clinical performance characteristics of iAMP HSV 1/2 Kit. It is a useful addition to a suite of molecular assays for detection and differentiation of HSV-1 and -2 from various pediatric sample sources and warrants further validation.Disclosures.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793418&blobtype=pdf
|
# P-2187. Integrated Metabolomic and Single-Cell Transcriptomic Analysis Reveals Neutrophil Heterogeneity and Metabolic Pathways in Severe Hand, Foot, and Mouth Disease Caused by Coxsackievirus A6 Infection Yaping Li, Doctor's degree 1 ; Meng Zhang, Doctor's degree 2 ; Chenrui Liu, Doctor's degree 1 ; huiling Deng, n/a 3 ; Muqi Wang, Doctor's degree 2 ; Yufeng Zhang, Doctor's degree 3 ; Yuan Chen, Doctor's degree 3 ; Mei Li, Doctor's Degree 2 ; Shuangsuo Dang, PhD 4
Background. Hand, foot, and mouth disease (HFMD) is a significant infectious disease that can lead to neurological damage in children, with Coxsackievirus A6 (CA6) emerging as the predominant pathogen in recent years. We aim to explore the metabolic and transcriptomic profiles of CA6-induced severe HFMD in children.
Methods. Peripheral blood mononuclear cells (PBMCs) and plasma samples were collected from children during the acute and recovery phases of CA6-associated HFMD. Single-cell sequencing and high-throughput targeted metabolomic analysis were conducted, followed by an integrated analysis of the interaction networks.
Results. Single-cell sequencing identified seven distinct cell types in PBMCs. Significant shifts were observed in the proportions of CD4+ naive T cells, neutrophils, CD16-monocytes and naive B cells. Metabolomic analysis of 25 common amino acids revealed significant changes in the glycine, serine, and threonine metabolic pathways. Neutrophils were found to have the highest metabolic scores for glycine, serine, and threonine pathways by scMetabolism. Furthermore, CellChat analysis of the PBMC microenvironment revealed potential molecular interactions between neutrophils and other cell types, including ADGRE5-CD55, ANXA1-FPR1, and CD22-PTPRC. The interaction between neutrophils and these metabolic pathways suggested that FCGR3B and AOC3 may play key roles in mediating these effects.
Conclusion. Alterations in glycine, serine, and threonine metabolic pathways in neutrophils may contribute significantly to the pathogenesis of severe CA6 HFMD.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548449&blobtype=pdf
|
# Evolution of BA.2.86 to JN.1 reveals that functional changes in non-structural viral proteins are required for fitness of SARS-CoV-2
Shuhei Tsujino, Masumi Tsuda, Naganori Nao, Kaho Okumura, Lei Wang, Yoshitaka Oda, Yume Mimura, Jingshu Li, Rina Hashimoto, Yasufumi Matsumura, Rigel Suzuki, Saori Suzuki, Kumiko Yoshimatsu, Miki Nagao, Jumpei Ito, Kazuo Takayama, Kei Sato
## Abstract
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of coronavirus disease 2019 , is still circulating among humans, leading to the continuous evolution. SARS-CoV-2 Omicron JN.1 evolved from a distinct SARS-CoV-2 lineage, BA.2.86, and spread rapidly worldwide. It is unclear why BA.2.86 did not become dominant and was quickly replaced by JN.1, which possesses one amino acid substitution in the spike protein (S:L455S) and two in the non-spike proteins NSP6 and ORF7b (NSP6:R252K and ORF7b:F19L) compared to BA.2.86. Here, we utilized recombinant viruses to elucidate the impact of these mutations on the virological characteristics of JN.1. We found that the mutation in the spike attenu ated viral replication, while the non-spike mutations acted synergistically to enhance replication. This suggests that the mutations in the non-spike proteins compensate for the one in the spike, improving viral fitness, as the mutations in the spike contribute to further immune evasion. Our findings suggest that functional changes in both the spike and non-spike proteins are necessary for the evolution of SARS-CoV-2, enabling evasion of adaptive immunity within the human population while sustaining replication. IMPORTANCE Because the spike protein is strongly associated with certain virological properties of SARS-CoV-2, such as immune evasion and infectivity, most previous studies on SARS-CoV-2 variants have focused on spike protein mutations. However, the nonspike proteins also contribute to infectivity, as observed throughout the evolution of Omicron subvariants. In this study, we demonstrate a "trade-off" strategy in SARS-CoV-2 Omicron JN.1 in which the reduced infectivity caused by spike mutation is compensated by non-spike mutations. Our results provide insight into the evolutionary scenario of the emerging virus in the human population.
and the second was the evolution from Omicron BA.2 to BA.2.86 (2,10). Between Delta and BA.1, 38 mutations were identified in the viral spike (S) protein and 49 in other viral genes. Similarly, BA.2.86 exhibited 32 mutations in S and 14 in other genes compared to BA.2 (Nextstrain; https://nextstrain.org/ncov/gisaid/global/6m) (2,10). Then, just as BA.2 emerged from BA.1, the descendant JN.1 emerged in the United States in September 2023 and outcompeted BA.2.86 to become the dominant variant (11). As of December 2024, direct descendants of JN.1, including KP.3 and KP.3.1.1, have become predominant globally (12,13).
Because the JN.1 lineage surged and rapidly outcompeted previously dominant variants in early 2024, the effective reproduction number (R e ) and immune-evasive properties of the JN.1 variant have been of great interest to researchers, including ourselves (11,14,15). JN.1 showed even greater immune evasion than BA.2.86 but exhibited reduced binding affinity for the SARS-CoV-2 receptor angiotensin-convert ing enzyme 2 (ACE2). Cryo-EM observations revealed that the mutation L455S in the receptor-binding domain (RBD) of the JN.1 S protein disrupted the interaction between RBD and human ACE2 (16). These findings suggest that the lower affinity of the JN.1 vs BA.2.86 S protein for ACE2 impairs viral entry and viral adaptation against host immune defenses. Thus, despite the reduced ACE2 binding affinity of the JN.1 S, the mechanisms underlying its selective advantage and rapid replacement of BA.2.86 remain incompletely understood. In the evolution from BA.2.86 to JN.1, two additional amino acid substitutions-NSP6:R252K and ORF7b:F19L-were acquired in non-spike proteins (https://jbloomlab.github.io/SARS2-mut-fitness/). NSP6 is a multi-spanning transmem brane protein essential for the formation of replication organelles (17). Recent studies have shown that NSP6 can restrict autophagosome expansion and impair lysosomeautophagosome fusion, potentially enhancing viral replication (18). Notably, the R252K mutation in NSP6 has been implicated in enhanced viral RNA replication in single-round infection assay (19). ORF7b is a small transmembrane protein implicated in modulating host responses. It has been shown to disrupt epithelial barrier integrity, induce cell death (20,21), and localize to the Golgi and endoplasmic reticulum, possibly interfering with intracellular trafficking and innate immune signaling (22). These observations raise the possibility that the acquisition of non-spike mutations may have compensated for the fitness cost associated with reduced receptor binding, thereby contributing to the successful spread of JN.1.
Reverse genetics systems have played a central role in studying viral replication, pathogenicity, and the impact of specific mutations. Since a rapid reverse genetics system for SARS-CoV-2 has been developed (23,24), we have investigated the viral characteristics of Delta, BA.2, XBB.1.5, EG.5.1, and XEC, elucidating the roles of specific mutations (1,8,9,25,26). In this study, we used recombinant viruses to examine the differences in viral characteristics between JN.1 and its direct ancestor, BA.2.86. To this end, we analyzed the mutation frequency and identified the mutation(s) that character ize JN.1. Furthermore, we generated recombinant viruses with these mutations and investigated their replication efficiency and intrinsic pathogenicity.
## RESULTS AND DISCUSSION
To investigate the replication efficiency and intrinsic pathogenicity, we inoculated VeroE6 cells expressing transmembrane serine protease 2 (TMPRSS2) (27) with clinical isolates of JN.1 (GISAID ID: EPI_ISL_18771637) or BA.2.86 (GISAID ID: EPI_ISL_18233521) (10). In quantifying the infectious viral titers and viral RNA load of the supernatants, we found that the replication properties of BA.2.86 and JN.1 were comparable (Fig. 1A, left and right). Next, we intranasally inoculated hamsters-our established animal model for COVID-19 (1-10)-with either BA.2.86 or JN.1 under anesthesia. The body weights of the hamsters were comparable for the two viruses and, as expected, lower than those of uninfected hamsters (Fig. 1B). These findings suggest that under our experimental conditions, the replication efficiency and intrinsic pathogenicity of JN.1 and BA.2.86 are similar.
To gain insight into the evolutionary transition from BA.2.86 to JN.1, we examined the frequency of key mutations across multiple Omicron subvariants (Fig. 2A). In BA.2.86, the NSP6:R252K and ORF7b:F19L mutations were present at low frequencies but were subsequently acquired as convergent mutations in JN.1, together with the S:L455S substitution. The JN.1 variant was first detected in September 2023 and became globally dominant by January 2024 (Nextstrain, clade 24A; https://nextstrain.org/ncov/gisaid/ global/6m) (11). It then gave rise to several descendant lineages, including KP.2, KP.3, and KP.3.1.1 (Pango nomenclature; https://github.com/cov-lineages/pango-designation) (12,13,28). Of note, the three signature mutations originally observed in JN.1-NSP6:R252K, ORF7b:F19L, and S:L455S-have been retained in currently circulating variants, such as KP.3.1.1 (Fig. 2A).
Previous studies, including ours, showed that the S:L455S mutation in JN.1 has a negative effect on RBD-ACE2 binding affinity (11,16). However, our in vitro analysis demonstrated similar growth kinetics for BA.2.86 and JN.1 (Fig. 1A). To further evaluate the impact of spike and non-spike mutations on viral growth in cell culture, we gener ated recombinant viruses: rBA.2.86; rJN.1; rBA.2.86/NSP6:R252K+ORF7b:F19L (a mutant carrying the two substitutions in NSP6 and ORF7b described above that initially emerged in humans); and rBA.2.86/S:L455S (a mutant carrying the S:L455S mutation).
VeroE6/TMPRSS2 and Calu-3 cells were inoculated with each of the four recombinant viruses. In both cell types, the viral titers and RNA load of rBA.2.86 and rJN.1 were almost identical (Fig. 2B andC), consistent with our results using the clinical isolates. In the supernatants of VeroE6/TMPRSS2 cells, the titers and RNA levels of rBA.2.86/ NSP6:R252K+ORF7b:F19L were higher than those of rBA.2.86 (Fig. 2B, left and right). In Calu-3 cells, the titer of rBA.2.86/S:L455S was lower than those of rBA.2.86. On the other hand, the titer of rBA.2.86/NSP6:R252K+ORF7b:F19L was significantly higher than those of rBA.2.86 (Fig. 2C,left). Viral RNA loads also showed a similar trend as well (Fig. 2C, right). These results suggest that the S:L455S mutation decreases replication efficiency, while the NSP6:R252K and ORF7b:F19L mutations increase replication efficiency in vitro. Of note, a modest reduction in rBA.2.86/S:L455S replication was observed in Calu-3 cells, but not in VeroE6/TMPRSS2 cells. This difference may reflect the presence of functional antiviral signaling in Calu-3 cells, which is absent in interferon-deficient VeroE6 cells (29,30). These findings raise the possibility that the replication phenotype associated with the L455S substitution is modulated by cell-intrinsic immune responses. These mutations in these non-spike proteins in JN.1 may affect replication efficiency. To further investigate this possibility, two additional recombinant JN.1 viruses containing the mutations to revert to the BA.2.86 sequence were generated (rJN.1/NSP6:K252R and rJN.1/ORF7b:L19F). rJN.1, containing an S mutation to revert back to the BA.2.86 sequence, was also evaluated. While designated in these experiments as rJN.1/S:S455L, this virus is the same sequence as rBA.2.86/NSP6:R252K+ORF7b:F19L.
In VeroE6/TMPRSS2 cells, the growth of rJN.1/S:S455L was significantly higher than those of rBA.2.86, consistent with the greater cytopathic effect observed (Fig. 3A andC). In Calu-3 cells, the titer of rJN.1/S:S455L was significantly higher than those of rBA.2.86 and rJN.1 (Fig. 3B,left). Viral RNA in the supernatants also showed a similar trend (Fig. 3B, right). In both cell lines, the non-spike protein mutants (rJN.1/NSP6:K252R and rJN.1/ ORF7b:L19F) did not significantly change growth rate compared to the parental rJN.1. Altogether, these results shown in Fig. 2 suggest that in evolving from BA.2.86 to JN.1, the S:L455S mutation attenuates replication efficiency in vitro, while mutations in NSP6 and ORF7b contribute to higher replication efficiency. To investigate the in vivo dynamics and pathogenicity of these viruses, Syrian hamsters were intranasally inoculated with rBA.2.86, rJN.1, and the different rJN.1 mutants. Consistent with the in vitro findings for the clinical isolates, changes in weight were comparable between hamsters infected with rJN.1 and rBA.2.86. Of the two rJN.1 viruses carrying the single mutation, only rJN.1/S:S455L infection led to significant weight loss compared with the parental rJN.1 infection, which showed the greatest change among all viruses evaluated (Fig. 4A,left). On the other hand, the weight loss of hamsters infected with rJN.1/NSP6:K252R was significantly lower than that of hamsters infected with rJN.1. Hamsters infected with rJN.1/ORF7b:L19F only showed slightly less weight loss compared to hamsters infected with rJN.1. SARS-CoV-2 infection causes a decline in pulmonary function (31), and the degree of deterioration can be used as an index of viral pathogenicity (2). Thus, we assessed the pulmonary function of infected hamsters by measuring enhanced pause (Penh) values. rJN.1/S:S455L infection tended to result in higher Penh values than rJN.1 infection (Fig. 4A, right).
Moreover, to evaluate viral spread in respiratory tissues, we collected the lungs of infected hamsters at 2 and 5 days post-infection (d.p.i.), separating the tissues into the hilum and peripheral regions. In the hilum, the viral RNA load of rJN.1/S:S455L-infected hamsters was comparable to that of rJN.1-infected hamsters (Fig. 4B,middle). In contrast, in the lung periphery region, the viral titer and RNA load of the rJN.1/S:S455L-infected hamsters were significantly higher than those of the rJN.1-and rJN.1/ORF7b:L19Finfected hamsters (Fig. 4B, left and right). At 5 d.p.i., viral titers and RNA load were generally lower across all groups, as expected, and no statistically significant differences were observed among the viruses (Fig. S1). These results suggest that the efficacy of viral spread in the lung is greater with rJN.1/S:S455L than with rJN.1 or rJN.1/ORF7b:L19F. We also performed immunohistochemistry (IHC) to evaluate the presence of viral N protein in the respiratory tissues of infected hamsters (Fig. 4C; Fig. S2A). In the lung 2 d.p.i., N-positive cells were more strongly detected in the bronchial/bronchio lar epithelia of rJN.1/S:S455L-infected hamsters than in those infected with rBA.2.86, rJN.1, rJN.1/NSP6:K252R, or rJN.1/ORF7b:L19F (Fig. 4C). Then, to evaluate the severity of inflammation upon infection with the mutant viruses, histopathological analyses were performed on the lung tissue (Fig. 4D; Fig. S2B andC). At 2 d.p.i., alveolar damage around the bronchi was prominent in rJN.1/S:S455L-infected hamsters (Fig. 4D). On the other hand, inflammation in bronchi/bronchioles tended to be more limited in rJN.1/ NSP6:K252R-and rJN.1/ORF7b:L19F-infected hamsters than in rBA.2.86 and rJN.1. At 5 d.p.i., the alveolar architecture appeared more severely destroyed by alveolar damage and the expansion of type II pneumocytes in rJN.1/S:S455L-infected hamsters (Fig. 4D). No significant differences were found between rBA.2.86-, rJN.1-, rJN.1/NSP6:K252R-, and rJN.1/ORF7b:L19F-infected hamsters. Taken together, these findings suggest that the S:L455S mutation, acquired in the evolution of BA.2.86 to JN.1, attenuates viral growth and pathogenicity.
This study aimed to investigate the virological characteristics of SARS-CoV-2 JN.1 and understand the evolution of this variant from BA.2.86. Our findings suggest that the balance between properties of the spike and non-spike proteins is impor tant for viral fitness and continued circulation of SARS-CoV-2 in humans. Compared with its direct ancestor BA.2.86, JN.1 has three mutations: S:L455S, NSP6:R252K, and ORF7b:F19L (Fig. 2A). Using recombinant mutant viruses generated in both the BA.2.86 and JN.1 backbones, we demonstrated that the S:L455S mutation attenuates replica tion efficiency and pathogenicity, while the NSP6:R252K and ORF7b:F19L mutations appear to compensate for this attenuation. BA.2.86 did not become predominant in human populations and was quickly replaced by JN.1 (32). This is likely because BA.2.86 exhibited weaker immune evasion than previously dominant variants (33)(34)(35). Acquisition of the S:L455S mutation may have helped enhance immune evasion but at the cost of impaired viral replication. Thus, additional mutations in non-spike proteins (NSP6:R252K and ORF7b:F19L), which had already been observed in a minor population of BA.2.86, increased the viral fitness of JN.1 and enabled more efficient circulation in humans. Interestingly, while each single revertant in the JN.1 background (NSP6:K252R or ORF7b:L19F) had little effect on viral replication compared to JN.1 (Fig. 3A andB), the double revertant (rBA.2.86/NSP6:R252K+ORF7b:F19L) in the BA.2.86 background showed increased replication (Fig. 2B andC). These findings indicate that the combined functional effects of the two non-spike mutations acted synergistically, resulting in a phenotypic change that became evident only when both mutations were present. Although direct physical interaction between these proteins is unlikely, combinations of non-spike mutations are known to exert epistatic effects on viral fitness and adaptation (36)(37)(38). Such synergy may have contributed to restoring overall viral fitness in JN.1, despite the fitness cost associated with the S:L455S mutation.
Throughout the evolution of Omicron subvariants, SARS-CoV-2 has demonstrated improved immune evasion while maintaining infectivity (39). Acquisition of mutations in the spike protein imposes a weakness on viral fitness. To overcome this, muta tions in non-spike proteins could enhance viral fitness by modulating certain virolog ical properties. This "trade-off" strategy has been consistently observed during the circulation of Omicron subvariants in humans. For instance, in BA.1, mutations in both spike and NSP6 were reported to contribute to attenuated pathogenicity (40). In BA.2, we reported that the spike mutation S:L371F enhanced fusogenicity and pathogenicity, while multiple non-spike mutations attenuated replication efficiency and pathogenicity (25). Furthermore, the impairment of major histocompatibility complex suppression due to dysfunctional ORF8 in XBB.1.5 was shown to influence viral pathogenicity (8). Taken together, this knowledge demonstrates that investigating the impact of mutations not only in the spike protein but also in the non-spike proteins is crucial for understanding SARS-CoV-2 evolution.
As we demonstrated for a variety of SARS-CoV-2 Omicron subvariants in the past (2-13, 25, 28), elucidating the virological features of newly emerging SARS-CoV-2 variants is important to determine their potential risk to human society and to understand the evolution of this virus in humans. Accumulating knowledge of the evolutionary traits of newly emerging pathogenic viruses in the human population will be beneficial in preparing for future outbreaks and pandemics.
## MATERIALS AND METHODS
## Cell culture
VeroE6/TMPRSS2 cells (VeroE6 cells stably expressing human TMPRSS2; JCRB Cell Bank, JCRB1819) (20) were maintained in DMEM (low glucose) (Cat#041-29775; FUJIFILM WAKO, Osaka, Japan) containing 10% FBS and 1 mg/mL G418 (Cat#09380-44; Nacalai Tesque, Kyoto, Japan). Calu-3 cells (ATCC, HTB-55) were maintained in Eagle's minimum essential medium (EMEM) (Cat#056-0838; Sigma-Aldrich, MO, USA) containing 10% FBS and 1% penicillin-streptomycin (Cat#09367-34; Nacalai Tesque).
## Epidemic dynamics analysis and mutation frequency calculations
In this study, we analyzed the viral genomic surveillance data stored in the GISAID database (https://www.gisaid.org; downloaded on 19 May 2025) (41). We used the data collected for SARS-CoV-2 from 1 January 2022 to 1 January 2025 for this analysis. We excluded any data that (i) did not have a collection date and Pango lineage information; (ii) were retrieved from non-human animals; and (iii) were sampled during quarantine. As BA.2.86 has diverged into multiple sublineages, BA.
## Plasmid construction
The nine pmW118 plasmids containing the partial genomes of SARS-CoV-2 BA.2.86 were previously generated (42). To generate the recombinant JN.1 viruses, mutations were introduced by inverse fusion PCR cloning into the plasmids encoding the corresponding BA.2.86 genes. Sequences of all the plasmids used in this study were confirmed by a SeqStudio Genetic Analyzer (Thermo Fisher Scientific, MA, USA) and an outsourced service (Fasmac, Kanagawa, Japan). Primer and plasmid information can be provided upon request.
## SARS-CoV-2 preparation and titration
The working stocks of SARS-CoV-2 virus were prepared and titrated as previously described (8). In this study, stocks were prepared using clinical isolates of BA.2.86 (strain TKYnat15020; GISAID ID: EPI_ISL_18233521) (10) and JN.1 (strain LG0688; GISAID ID: EPI_ISL_18771637).
Recombinant viruses were generated by a circular polymerase extension reaction (CPER) (23). The resultant CPER products were transfected into VeroE6/TMPRSS2 cells as described previously (8). All the viruses were stored at -80°C until use, and viral genome sequences were confirmed by Sanger sequencing (see "Plasmid construction, " above).
## Titration and growth kinetics
The infectious titers of supernatants from infected cell cultures were determined by quantifying the 50% tissue culture infectious dose (TCID 50 ) (43). For growth kinetics, VeroE6/TMPRSS2 cells or Calu-3 cells were inoculated with the virus in 12-well plates at a multiplicity of infection (MOI) of 0.01 or 0.1, respectively. The infectious titers of supernatants collected at the indicated time points were then determined.
## References
1. Saito, Irie, Suzuki et al. (2022) "Enhanced fusogenicity and pathogenicity of SARS-CoV-2 Delta P681R mutation" *Nature*
2. Suzuki, Yamasoba, Kimura et al. (2022) "Attenuated fusogenicity and pathogenicity of SARS-CoV-2 Omicron variant" *Nature*
3. Kimura, Yamasoba, Tamura et al. (2022) "Virological characteristics of the SARS-CoV-2 Omicron BA.2 subvariants, including BA.4 and BA" *Cell*
4. Tamura, Yamasoba, Oda et al. (2023) "Comparative pathogenicity of SARS-CoV-2 Omicron subvariants including BA.1, BA.2, and BA" *Commun Biol*
5. Saito, Tamura, Zahradnik et al. (2022) "Virological characteristics of the SARS-CoV-2 Omicron BA.2.75 variant" *Cell Host Microbe*
6. Ito, Suzuki, Uriu et al. (2023) "Convergent evolution of SARS-CoV-2 Omicron subvariants leading to the emergence of BQ.1.1 variant" *Nat Commun*
7. Tamura, Ito, Uriu et al. (2023) "Virological characteristics of the SARS-CoV-2 XBB variant derived from recombination of two Omicron subvariants" *Nat Commun*
8. Tamura, Irie, Deguchi et al. (2024) "Virological characteristics of the SARS-CoV-2 Omicron XBB.1.5 variant" *Nat Commun*
9. Tsujino, Deguchi, Nomai et al. (2024) "Virological characteris tics of the SARS-CoV-2 Omicron EG.5.1 variant" *Microbiol Immunol*
10. (2025) *Full-Length Text Journal of Virology*
11. Tamura, Mizuma, Nasser et al. (2024) "Virological characteris tics of the SARS-CoV-2 BA.2.86 variant" *Cell Host Microbe*
12. Kaku, Okumura, Padilla-Blanco et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium. 2024. Virological characteristics of the SARS-CoV-2 JN.1 variant" *Lancet Infect Dis*
13. Kaku, Yo, Tolentino et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium. 2024. Virological characteris tics of the SARS-CoV-2 KP.3, LB.1, and KP.2.3 variants" *Lancet Infect Dis*
14. Kaku, Uriu, Okumura et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium. 2024. Virological characteristics of the SARS-CoV-2 KP.3.1.1 variant" *Lancet Infect Dis*
15. Yang, Yu, Xu et al. (2024) "Fast evolution of SARS-CoV-2 BA.2.86 to JN.1 under heavy immune pressure" *Lancet Infect Dis*
16. Planas, Staropoli, Michel et al. (2024) "Distinct evolution of SARS-CoV-2 Omicron XBB and BA.2.86/JN.1 lineages combining increased fitness and antibody evasion" *Nat Commun*
17. Yang, Guo, Wang et al. (2024) "Structural basis for the evolution and antibody evasion of SARS-CoV-2 BA.2.86 and JN.1 subvariants" *Nat Commun*
18. Ricciardi, Guarino, Giaquinto et al. (2022) "The role of NSP6 in the biogenesis of the SARS-CoV-2 replication organelle" *Nature*
19. Zhang, Jiang, Liu et al. (2024) "SARS-CoV-2 NSP6 reduces autophagosome size and affects viral replication via sigma-1 receptor" *J Virol*
20. Taha, Ezzatpour, Hayashi et al. (2025) "Enhanced RNA replication and pathogenesis in recent SARS-CoV-2 variants harboring the L260F mutation in NSP6" *PLoS Pathog*
21. Nguyen, Palfy, Fogeron et al. (2024) "Analysis of the structure and interactions of the SARS-CoV-2 ORF7b accessory protein" *Proc Natl Acad Sci*
22. Deshpande, Li, Li et al. (2024) "SARS-CoV-2 accessory protein Orf7b induces lung injury via c-myc mediated apoptosis and rerroptosis" *Int J Mol Sci*
23. Xiao, Fu, You et al. (2024) "Inhibition of the RLR signaling pathway by SARS-CoV-2 ORF7b is mediated by MAVS and abrogated by ORF7bhomologous interfering peptide" *J Virol*
24. Torii, Ono, Suzuki et al. (2021) "Establishment of a reverse genetics system for SARS-CoV-2 using circular polymerase extension reaction" *Cell Rep*
25. Amarilla, Sng, Parry et al. (2021) "A versatile reverse genetics platform for SARS-CoV-2 and other positive-strand RNA viruses" *Nat Commun*
26. Kimura, Yamasoba, Nasser et al. (2023) "Multiple mutations of SARS-CoV"
27. Omicron "2 variant orchestrate its virological characteristics" *J Virol*
28. Tsujino, Tsuda, Ito et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium. 2025. A non-spike nucleocapsid R204P mutation in SARS-CoV-2 Omicron XEC enhances inflammation and pathogenicity"
29. Matsuyama, Nao, Shirato et al. (2020) "Enhanced isolation of SARS-CoV-2 by TMPRSS2-expressing cells" *Proc Natl Acad Sci*
30. Kaku, Uriu, Kosugi et al. (2024) "Virological characteristics of the SARS-CoV-2 KP.2 variant" *Lancet Infect Dis*
31. Desmyter, Melnick, Rawls (1968) "Defectiveness of interferon production and of rubella virus interference in a line of African green monkey kidney cells (Vero)" *J Virol*
32. Osada, Kohara, Yamaji et al. (2014) "The genome landscape of the african green monkey kidney-derived vero cell line" *DNA Res*
33. Huang, Tan, Wu et al. (2020) "Impact of coronavi rus disease 2019 on pulmonary function in early convalescence phase" *Respir Res*
34. Wang, Lu, Jiang (2024) "SARS-CoV-2 evolution from the BA.2.86 to JN.1 variants: unexpected consequences" *Trends Immunol*
35. Qu, Xu, Faraone et al. (2024) "Immune evasion, infectivity, and fusogenicity of SARS-CoV-2 BA.2.86 and FLip variants" *Cell*
36. Zhang, Kempf, Nehlmeier et al. (2024) "SARS-CoV-2 BA.2.86 enters lung cells and evades neutralizing antibodies with high efficiency" *Cell*
37. Wang, Guo, Liu et al. (2023) "Antigenicity and receptor affinity of SARS-CoV-2 BA.2.86 spike" *Nature*
38. Hossain, Akter, Rashid et al. (2022) "Unique mutations in SARS-CoV-2 Omicron subvariants' non-spike proteins: Potential impacts on viral pathogenesis and host immune evasion" *Microb Pathog*
39. Zeng, Dichio, Horta et al. (2020) "Global analysis of more than 50,000 SARS-CoV-2 genomes reveals epistasis between eight viral genes" *Proc Natl Acad Sci*
40. Innocenti, Obara, Costa et al. (2024) "Real-time identification of epistatic interactions in SARS-CoV-2 from large genome collections" *Genome Biol*
41. Parsons, Acharya (2023) "Evolution of the SARS-CoV-2 Omicron spike" *Cell Rep*
42. Chen, Chin, Kenney et al. (2023) "Spike and nsp6 are key determinants of SARS-CoV-2 Omicron BA.1 attenua tion" *Nature*
43. Khare, Gurry, Freitas et al. (2021) "GISAID's role in Full-Length Text Journal of Virology October"
44. *China CDC Wkly*
45. Kawashiro, Suzuki, Nogimori et al. (2024) "Neutralizing antibody responses and cellular responses against SARS-CoV-2 Omicron subvariants after mRNA SARS-CoV-2 vaccination in kidney transplant recipients" *Sci Rep*
46. Reed, Muench (1938) "A simple method of estimating fifty per cent endpoints12" *Am J of Epidemiol*
47. Motozono, Toyoda, Zahradnik et al. (2021) "SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity" *Cell Host Microbe*
48. Tamura, Ito, Torii et al. (2024) "Akaluc biolumines cence offers superior sensitivity to track in vivo dynamics of SARS-CoV-2 infection"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793629&blobtype=pdf
|
Margaret Lind, Emily Goldmann, Gregory Lalonde
Background. Influenza vaccination coverage in the United States remains below 50%. Historical racial and ethnic disparities in coverage have been previously documented, but recent trends suggest shifts these disparities. NH-Black individuals, a historically under-vaccinated group, have experienced consistent increases in coverage over the past few seasons while data suggests that historically better vaccinated groups, such as NH-, have experienced declining coverage rates. In this study, we examine racial and ethnic disparities in influenza vaccine coverage since the onset of the COVID-19 pandemic, assessing whether these disparities persisted after stratification by likely drivers of coverage including age, healthcare worker status, underlying health conditions, and access to care. Conclusion. Despites notable gains among people who identified as NH-Black, disparities persisted and, as of the 2022-2023 season, coverage among this population remained 25.1 percentage points below the Healthy People 2030 goal of 70%. While all other racial and ethnic groups saw increases in coverage, people who identified as NH-White experienced declines, particularly young children and insured individuals with financial barriers. These findings suggest that vaccine uptake is nuanced and that, addressing financial barriers and targeted messaging may help improve uptake.
Disclosures. All Authors: No reported disclosures
Poster Abstracts • OFID 2026:13 (Suppl 1) • S1329
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792344&blobtype=pdf
|
Caroline Rosario, Stephanie Welch, Kiran Gajurel, Rupal Jaffa, Roper St, Francis
Background. Ganciclovir (GCV) exhibits high interindividual variability in pharmacokinetic models. A combination of the inherent complexity of patients with cytomegalovirus (CMV) and the adverse effect profile of GCV has led to variable dosing practices. The purpose of this study was to assess GCV dosing practices for treatment of CMV and associated clinical outcomes across a large health system.
## Figure 1 Definitions
## Figure 2 Baseline Characteristics
Methods. This was a retrospective, multicenter cohort study of adults with CMV treated with ≥72 hours of GCV between May 2022 and August 2024. The primary outcome was the initial GCV regimen prescribed (Figure 1). Patients were categorized as primarily standard dose (PSD) or primarily high dose (PHD), defined by the regimen received for ≥50% of the GCV course. Clinical endpoints assessed were composite clinical failure and various safety measures (Figure 1). Additional secondary outcomes included rationale for HD GCV and indications for transition from SD to HD GCV. Results. One hundred patients (74 PSD vs. 26 PHD) were included with baseline characteristics shown in Figure 2. Initial regimens were 66% SD, 26% HD, and 8% alternative (below SD). The most common rationales for HD GCV were increasing or persistent viremia (IPV) (52%), provider preference (18%), and documented GCV resistance (14%). There was no significant difference in the rate of clinical failure (44.6% PSD vs. 46.2% PHD, p=0.89) or reasons for clinical failure between groups (Figure 3). Among the 10 patients who escalated from SD to HD GCV for IPV, 100% experienced clinical success while on HD, although 90% experienced clinical failure prior to the completion of treatment (Figure 4). New onset neutropenia (15% vs. 23%, p=0.34) and 90-day graft rejection (10% vs. 12%, p=0.72) were similar between groups.
Conclusion. GCV dosing practices within our health system are provider-driven and heterogenous. Assessment of these practices is limited by retrospective chart review. No clear benefit of PHD was observed over PSD, nor were increased adverse effects noted in the PHD group. However, in patients changed from an SD to HD regimen, clinical success was observed on HD GCV, with subsequent failure after transition to an oral regimen. This suggests a need to further study optimal oral treatment strategies for patients with clinical success on HD GCV.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724135&blobtype=pdf
|
# A novel bispecific nanobody protects mice against RSV infection via intranasal administration
Min Zhang, Liuxing Qin, Raoqing Guo, Jinwei Li, Si Huang, Chen Sheng, Yijia Xiang
## Abstract
Respiratory syncytial virus (RSV) is the leading cause of respiratory infectionrelated hospitalizations in children younger than 5 years. Neutralizing nanobody-based interventions represent a promising strategy against RSV. Here, we identify a novel nanobody (4-H1) targeting the RSV prefusion F (pre-F) protein, which demonstrates potent neutralization against both RSV A and B subtypes. Epitope characterization via binning assays, molecular docking, and mutational analyses revealed that 4-H1 interacts with a unique region within antigenic site Ø by engaging critical residues L207, K209, and the K65-N67-C69 cluster. To improve the in vivo efficacy and stability of the 4-H1 nanobody, we engineered a heterotrimeric bispecific nanobody (4-H1-anti-HSA-4-H1). This single-chain molecule contains two anti-RSV F nanobody domains and one anti-human serum albumin (HSA) domain, resulting in a trivalent molecule with dual specificity. This construct demonstrated sub-nanogram per milliliter (sub-ng/mL) neutralization potency against both RSV A and B subtypes, with prolonged in vivo half-life. Notably, intranasal administration of this construct before exposure conferred robust protection against RSV challenge in BALB/c mice. These results underscore the potential of 4-H1-anti-HSA-4-H1 as a respiratory-delivered prophylactic against RSV.IMPORTANCE RSV is the leading cause of infant respiratory hospitalizations, highlight ing the urgent need for effective prophylaxis. Here, we engineered a potent bispecific nanobody (4-H1-anti-HSA-4-H1) that exhibits exceptional neutralization against both RSV A and B subtypes with prolonged serum persistence. Prophylactic intranasal delivery of this construct conferred robust protection against RSV challenge in mice, indicating its potential as a respiratory-delivered prophylactic candidate against RSV. KEYWORDS respiratory syncytial virus, bispecific nanobodies, albumin-binding, intranasal administration, antigenic site Ø R SV is the predominant cause of severe lower respiratory tract infections, including bronchiolitis and pneumonia, in infants and young children (1). Globally, in 2019, RSV caused 33.0 million episodes of acute lower respiratory infections (ALRI) in children younger than 5 years of age, resulting in 3.6 million hospital admissions and 101,400 deaths (2, 3). Currently, no vaccine is approved for the active immunization of infants against RSV. Instead, protection in this population is achieved through passive immuni zation strategies, which include one maternal vaccine that provides passive immunity to infants (4), as well as three prophylactic monoclonal antibodies (mAbs). Palivizumab was the first mAb approved for RSV prevention. However, its use requires multiple doses and is approved only for high-risk infants (5). More recently developed mAbs, includ ing nirsevimab and clesrovimab, offer single-dose prophylaxis and have substantially expanded patient eligibility (6, 7). Despite these advancements, persistent challenges related to cost, equitable access, and real-world implementation underscore the ongoing need for novel interventions.
RSV strains comprise two antigenic subtypes (A and B) and possess a negative-sense single-stranded RNA genome encoding 11 viral proteins. Among them, the surface fusion glycoprotein (F) is highly conserved across RSV isolates from both subgroups, making it a prime target for vaccine and drug development (8,9). This metastable protein exists in two conformational states: a prefusion form (pre-F) that transitions either spontaneously or upon cell attachment to the stable postfusion form (post-F). The pre-F conformation displays six major antigenic sites (Ø, I-V), whereas only sites I-IV persist in the post-F form (10,11). Antibodies targeting pre-F-specific sites Ø or V demonstrate superior neutralizing potency, exemplified by nirsevimab (site Ø-specific), which exhibits >100-fold higher in vitro neutralization potency than palivizumab, a site II-targeting agent, against RSV A2 and RSV B9320 (12). This structural superiority underlies the superior clinical efficacy of nirsevimab (89.6% protection in infants ≤ 3 months), whereas palivizumab is only recommended for high-risk subgroups, including preterm neonates and patients with congenital heart disease or other severe comorbidi ties (6,13). Moreover, all three clinically approved RSV vaccines leverage pre-F antigens to elicit potent neutralizing antibodies (4,14,15).
In addition to standard immunoglobulins, camelids express heavy-chain-only antibodies (HCAbs) featuring single variable domains (VHHs, also called nanobodies) (16,17). Nanobodies exhibit distinctive structural simplicity that confers advantages such as efficient tissue penetration, superior stability, and exceptional solubility (18). Their single-gene encoding also enables low-cost production in microbial hosts (E. coli or yeast), bypassing the complex glycosylation requirements and substantial expenses inherent to mammalian cell expression systems (19). The extended complementaritydetermining region (CDR) loops of nanobodies facilitate the formation of a finger-like or convex paratope, allowing them to penetrate cryptic cavities or bind to concave epitopes inaccessible to conventional antibodies (20,21). Additionally, nanobodies can be readily engineered into multivalent formats, which are devoid of Fc regions and thus reduce the risk of antibody-dependent enhancement (ADE) observed in viral infections such as dengue virus, human immunodeficiency virus (HIV), severe acute respiratory syndrome coronavirus (SARS-CoV), SARS-CoV-2, Middle East respiratory syndrome coronavirus (MERS-CoV), and RSV (22)(23)(24)(25)(26). Multivalency further enhances binding avidity and broadens antigen recognition. These biophysical and economic advantages accelerate the development of potent antiviral nanobodies against RSV and related pathogens (27)(28)(29)(30)(31)(32).
To date, several RSV F-targeting nanobodies have demonstrated in vitro neutraliz ing activity, with select candidates advancing to in vivo evaluation (31,(33)(34)(35)(36)(37). Given the superior neutralizing responses elicited by pre-F compared with post-F in murine and non-human primate models (38), we immunized an alpaca with an engineered pre-F-stabilized trimer, DS-Cav1, which has been demonstrated to be a highly immuno genic antigen. Subsequent phage display screening yielded a potent nanobody (4-H1) with neutralizing activity against both RSV subtypes. Building on this scaffold, we engineered the heterotrimeric bispecific construct 4-H1-anti-HSA-4-H1, which simultane ously targets viral pre-F and HSA, achieving extended half-life and effective intranasal protection against RSV in vivo. Collectively, these results establish 4-H1-anti-HSA-4-H1 as a promising candidate for respiratory-delivered RSV prophylaxis, warranting further translational development.
## MATERIALS AND METHODS
## Cells and viruses
FreeStyle 293 F cells (Gibco, R79007) were maintained in FreeStyle 293 expression medium (Gibco, 12338018). HEp-2 cells and RSV A2 strain (kindly provided by Dr. Zishu Pan, Wuhan University) were cultured in DMEM (Gibco, C11995500BT) supplemented with 10% FBS (Lonsera, S711-001S), penicillin, and streptomycin (Gibco, 15140122). Viral stocks of RSV strains A2 and B1 were propagated and quantified in HEp-2 cells by plaque assay. Briefly, confluent HEp-2 monolayers in 96-well plates were inoculated with serially diluted RSV for 1 h at 37°C, and then, the cells were overlaid with 1.5% carboxymethyl cellulose (Sigma, C4888) in MEM (Gibco, 11900-024) containing 2% FBS. After 72 h of incubation, the cells were fixed and stained with the RB1 mAb specific for RSV F, followed by horseradish peroxidase (HRP)-conjugated goat anti-human IgG (Jackson ImmunoRe search, 109-035-088). Plaques were visualized using TrueBlue Peroxidase Substrate (KPL, 5510-0030) and counted using ImmunoSpot Analyzer (CTL).
## Production and purification of RSV F mAbs
The variable regions of the heavy and light chains from mAbs were synthesized and cloned into a pcDNA3.1 expression vector containing the constant domains of the human IgG1 heavy chain (HC) and κ or λ light chain (LC). To generate the nirsevimab HC variant, YTE mutations (M252Y/S254T/T256E) were introduced into the HC constant region. Equimolar amounts of the HC and LC expression plasmids were transiently transfected into 293 F cells using polyethylenimine (Polysciences, 24765). After 5 days of culture, the cell supernatants were harvested and clarified by centrifugation (10,000 × g, 10 min), filtered through a 0.45 µm membrane, and purified with rProtein A Magarose Beads (Smart-Lifesciences, SM003025). The antibodies were dialyzed against PBS and concentrated using 30 kDa Amicon Ultra centrifugal filters (Millipore).
## Production and purification of RSV F proteins
RSV post-F proteins were expressed in 293 F cells via transient transfection of expression plasmids. Pre-F proteins (DS-Cav1 for RSV A2 and B1) were produced using a lentiviral system. The wild-type and mutated DS-Cav1 sequences were synthesized and cloned into the PLVX-EF1α-IRES-Puro lentivirus expression vector (designated lenti-DS-Cav1). Lentiviral particles were generated by co-transfecting 293T cells with lenti-DS-Cav1, VSV-G envelope plasmid pMD2.G, and packaging plasmid psPAX2 using polyethyleni mine. Supernatants harvested at 48 h post-transfection were used to transduce fresh 293T cells in the presence of 6 µg/mL polybrene (Beyotime Biotechnology, C0351). After 72 h, the cells were trypsinized and subjected to puromycin selection (1 µg/mL, Sigma, P8833) for 7 days to eliminate non-resistant cells. Puromycin-resistant pools were adapted to serum-free FreeStyle 293 expression medium without puromycin and cultured for 3-4 days. Subsequently, the supernatants were collected, clarified, sterilefiltered, and purified by Ni-NTA affinity chromatography (Cytiva, 17531802). The proteins were dialyzed against PBS, concentrated to 1-2 mg/mL using 30 kDa Amicon Ultra centrifugal filters, flash-frozen in liquid nitrogen, and stored at -80°C until use.
## Isolation of nanobodies
An adult alpaca was immunized subcutaneously with 0.5 mg of DS-Cav1 A2 antigen emulsified in camelid adjuvant (AlpVHHS, 600-000-001). Booster doses (0.25 mg) were administered every 3 weeks. Pre-and post-immunization sera were collected for titer monitoring by ELISA. Briefly, 96-well plates coated with DS-Cav1 A2 and B1 antigens were probed with serially diluted serum. Twenty milliliters of peripheral blood were collected 14 days after the final immunization for peripheral blood mononuclear cells (PBMCs) isolation via Ficoll-Paque density gradient centrifugation. Total RNA was then extracted, reverse-transcribed, and nanobody genes amplified by nested PCR were ligated into SfiI-digested pADL-23c phagemid vectors. These constructs were sequenceverified and electroporated into TG1 electrocompetent cells (Weidi, DE1055). Phage display libraries were generated by superinfection with M13KO7 helper phage (NEB, N0315S), followed by phage precipitation using 20% PEG/2.5 M NaCl. To isolate highaffinity neutralizing nanobodies targeting the Ø and V epitope, competitive biopan ning was performed. In rounds 1 and 2, biotinylated DS-Cav1 A2 was immobilized on Dynabeads M-280 Streptavidin beads (Invitrogen, 11205D) and incubated with the phage library. Bound phage particles were eluted with trypsin. In rounds 3 and 4, biotinylated DS-Cav1 was pre-incubated with site I-, II-, III-, and IV-targeting antibod ies, followed by incubation with phage particles enriched from the round 2 and 3 eluates. After four rounds of selection, enriched phage clones were screened by phage ELISA. DS-Cav1 was captured on NeutrAvidin-coated plates (Thermo, 31007), and bound phage was detected using an HRP-conjugated anti-M13 antibody (Sino Biological, 11973-MM05T-H). High-affinity clones were expressed in WK6 cells induced with 1 mM IPTG (24 h, 28°C). Cells were pelleted (10,000 × g, 10 min, 4°C) post-induction, resuspen ded in TES buffer (200 mM Tris, 0.5 mM EDTA, 500 mM sucrose, pH 8.0), and lysed at 4°C for 16 h. After centrifugation, the supernatant was purified by Ni-NTA affinity chromatography. Eluted nanobodies were dialyzed against PBS and concentrated using 3 kDa Amicon filters.
## Enzyme-linked immunosorbent assay (ELISA)
ELISAs were performed using 96-well plates coated with 2 µg/mL of respective antigens (RSV A2 and B1 DS-Cav1 or post-F, mutated DS-Cav1, HSA, MSA) overnight at 4°C. After washing with PBST buffer (PBS containing 0.05% Tween-20), plates were blocked with 5% skim milk in PBS for 2 h at room temperature (RT). For serum binding ELISA, blood samples were centrifuged to isolate serum, and serially diluted serum samples were added to the plates and incubated for 2 h at RT. Bound nanobodies were detected using an HRP-conjugated anti-camelid VHH antibody (GenScript, A01861). For antibody binding ELISA, serially diluted test antibodies (motavizumab, 4-H1-Fc, nirsevimab, 5C4, RSD5, anti-HSA, and 4-H1-anti-HSA-4-H1) were added and incubated for 2 h at RT. Detection was performed using HRP-conjugated goat anti-human IgG (Invitrogen, A18817), anti-mouse IgG (Elabscience, E-AB-1001), and anti-camelid VHH antibody, as appropriate. For phage ELISA, phage supernatants were pre-blocked with 20% skim milk in PBS (50 µL skim milk solution mixed with 200 µL phage supernatant) for 1 h at RT before incubation with coated antigens. Bound phages were detected using an HRP-conjugated anti-M13 antibody. All assays were developed using TMB substrate (Biohao, N0160) and stopped with 1 M H₂SO₄; absorbance was measured at 450 nm using a microplate reader. The half-maximal effective concentration (EC₅₀) was calculated by four-parameter logistic regression in GraphPad Prism 9.5.0, defined as the compound concentration producing 50% of the maximal stimulatory effect relative to baseline and maximum response controls.
## Surface plasmon resonance (SPR) binding characterization
SPR binding kinetics were analyzed using a Biacore T200 system. Ligands (anti-HSA, 4-H1, and 4-H1-anti-HSA-4-H1) were immobilized via amine coupling. Briefly, CM5 sensor chips were activated with a 1:1 mixture of 0.4 M 1-ethyl-3-(3-dimethylaminopropyl)carbodii mide (EDC) and 0.1 M N-hydroxysuccinimide (NHS), followed by injection of ligands in 10 mM sodium acetate (pH optimized individually) for 420 s. Remaining active groups were blocked with 1 M ethanolamine-HCl for 420 s (pH 8.0). Analytes (wild-type and mutated DS-Cav1, HSA, MSA) were prepared in running buffer (PBS containing 0.005% Tween-20) and injected over the sensor surface. The association phase was monitored for 100 or 400 s, followed by a 200 or 600 s dissociation phase in running buffer. Sensor surfaces were regenerated with 10 mM glycine-HCl (pH 2.0). Data were processed, and kinetic parameters were calculated using Biacore T200 evaluation software. All supplies, instrumentation, and analytical software were sourced from Cytiva unless otherwise specified.
## Plaque reduction neutralization test (PRNT)
Antibody neutralizing activity was assessed by PRNT using HEp-2 cells. Cells were detached by trypsinization and seeded in 96-well plates at 1.5 × 10⁴ cells/well 24 h prior to infection. Serially diluted antibodies in serum-free DMEM were mixed 1:1 (vol/vol) with RSV inoculum (also prepared in serum-free DMEM) and incubated at 37°C for 1 h. After removing the culture medium, 50 µL of the antibody-virus mixture was added to the cell monolayers and incubated at 37°C for 1 h. Unbound material was aspirated, and the monolayers were overlaid with semi-solid MEM. Following a 72 h incubation, the cells were fixed with paraformaldehyde and stained with the RB1 mAb specific for RSV F. After washing, HRP-conjugated goat anti-human IgG was applied. Plaques were visualized using TrueBlue Peroxidase Substrate and counted using ImmunoSpot Analyzer. The half-maximal inhibitory concentration (IC₅₀) was calculated by four-parameter logistic regression in GraphPad Prism 9.5.0, defined as the antibody concentration reducing plaque counts by 50% relative to virus-only controls.
## Epitope binning by bio-layer interferometry (BLI)
Epitope binning analyses were performed using a ForteBio Octet instrument equipped with Ni-NTA biosensors (ForteBio). DS-Cav1 antigen was diluted to 10 µg/mL in 0.02% PBST buffer (PBS containing 0.02% Tween-20). Fc-conjugated nanobodies and mAbs targeting distinct epitopes were prepared at final concentrations of 500 nM or 1 µM. Prior to the assay, biosensors were pre-wetted in PBST for 10 min. An initial baseline was taken in PBST (60 s), followed by antigen loading with 10 µg/mL DS-Cav1 (60 s). A post-loading baseline was recorded in PBST (60 s). Association kinetics were meas ured by exposing antigen-loaded biosensors to nanobody-Fc solution for 150-250 s. For competition assessment, the sensors bearing antigen-nanobody complexes were immersed in solutions containing mAbs targeting distinct known epitopes for 100-200 s. The reverse binding order was also tested, in which sensors were first exposed to competitor mAbs and then to nanobody-Fc. Biosensors were regenerated by three cycles, each consisting of a 5 s immersion in 500 mM imidazole and a 5 s immersion in PBST. Percentage inhibition was calculated as inhibition (%) =100 × [1 -(response with competitor mAb/response with isotype control)], where response denotes the binding signal (nm). Data analysis was performed using GraphPad Prism software.
## Computer-guided epitope mapping of RSV F protein
The interaction mechanism between 4-H1 and DS-Cav1 was investigated through integrated computational approaches. Homology modeling of 4-H1 was performed using Insight II 2000 (MSI Co.) based on its amino acid sequence, with explicit definition of framework regions (FRs) and CDRs. The model was optimized by energy minimiza tion under a consistent valence force field (CVFF) via sequential steepest descent and conjugate gradient algorithms, followed by geometric validation via Ramachandran plot analysis. The DS-Cav1 structure was predicted with AlphaFold 3 and subjected to limited energy minimization under CVFF to relieve steric clashes. Molecular docking simulations employed the crystal structure of the DS-Cav1-D25 complex (PDB: 4JHA) as a topological template to predict binding modes for the 4-H1-DS-Cav1 complex. Using the refined complex structure, 50-ns molecular dynamics simulations were conducted with the Discovery Studio 3.5 module under CVFF to sample conformational space. All calculations were performed on IBM workstations using Insight II 2000 and Discovery Studio suites.
## Pharmacokinetic evaluation of 4-H1-anti-HSA-4-H1 in mice
To extend the half-life of 4-H1, we engineered a trivalent construct, 4-H1-anti-HSA-4-H1, by fusing two 4-H1 copies to an anti-HSA nanobody using (GS)₆ linkers. As a control, a 4-H1 trimer consisting of three 4-H1 copies linked in the same manner was prepared in parallel. The anti-HSA nanobody was developed through alpaca immunization and phage display screening against HSA. All constructs contained an N-terminal 6 × His tag for purification. After cloning into pcDNA3.1(+) vectors, constructs were expressed in 293 F cells and purified by Ni-NTA affinity chromatography. For in vivo evaluation, 6to 8-week-old female BALB/c mice (Guangdong Yaokang Biotechnology Co., Ltd.) were randomly assigned to two groups (n = 3) and administered intranasally at a dose of 2 mg/kg of either 4-H1-anti-HSA-4-H1 or 4-H1-trimer. Blood samples were collected via retro-orbital puncture at 4, 24, 48, 96, 168, and 240 h post-administration. Serum was isolated by allowing blood to clot at RT for 4 h, followed by centrifugation at 2,000 × g for 15 min at 4°C. Serum nanobody concentrations were quantified by sandwich ELISA. Diluted serum samples were added to DS-Cav1-coated plates, and nanobody concen trations were quantified against standard curves generated with purified nanobodies. Serum concentration-time data were analyzed by non-linear regression fitting using GraphPad Prism version 9.5.0. The data were fitted to a constrained one-phase decay model defined by the equation: Y = (Y0 -Plateau)exp(-KX) + Plateau, where Y0 is the initial concentration, K is the rate constant, and the Plateau was constrained to zero. The elimination half-life (T 1/2 ) was calculated as ln (2)/K. The area under the curve (AUC) was calculated using the trapezoidal rule, and the maximum serum concentration (Cmax) was determined directly from the observed concentration-time data.
## In vivo efficacy of 4-H1-anti-HSA-4-H1 against RSV in mice
The in vivo prophylactic efficacy of the bispecific 4-H1-anti-HSA-4-H1 against RSV was evaluated in 6-week-old female BALB/c mice (n = 6). Mice were randomly divided into two groups and administered 2 mg/kg 4-H1-anti-HSA-4-H1 or PBS vehicle control intranasally in a 50 µL volume 24 h prior to viral challenge. Following compound delivery, mice were anesthetized with isoflurane and inoculated intranasally with 4 × 10⁶ PFU of RSV A2 in 50 µL PBS. Lungs were harvested at 4 days post-infection for viral load quantification and histopathological analysis.
## H&E staining of RSV-infected cells in tissues
Resected left lung tissue specimens were fixed in 4% paraformaldehyde at RT for 24 h to preserve morphological integrity. Following fixation, tissues were dehydrated through a graded ethanol series, cleared in xylene, and then embedded in paraffin blocks. Serial sections of 5 µm thickness were cut using a precision microtome (Leica). For histological evaluation, sections were stained with H&E (Servicebio, G1076) according to standar dized protocols. Histopathological scoring was evaluated as described in the study (39,40).
## Viral load measurement by quantitative reverse transcription PCR (RT-qPCR)
RNA extraction from right lung tissues was performed using the EZBio Viral RNA Kit (EZBioscience, EZB-VRN1). Prior to extraction, tissues were mechanically homogenized in a Freezer Mill (Guangzhou Luka Sequencing Instrument Co., Ltd., Models LUKYM-I/II/III) under the following conditions: -35°C (the actual temperature reached -65°C during operation), 70 Hz, for 150 s, repeated twice. First-strand cDNA was then synthesized from the extracted RNA using HiScript III Reverse Transcriptase (Vazyme, R312). RT-qPCR reactions (20 µL final volume) contained 2 µL of cDNA template, 10 µL of ChamQ SYBR qPCR Master Mix (Vazyme, Q311), and 0.4 µM primers against the RSV nucleoprotein (NP) gene (GenBank accession no. NC_ 038235.1) (Forward: 5′-AGATCAACTTCTGTCATCCA GCAA-3′, Reverse: 5′-AACATGCCACATAACTTATTGAT-3′). Amplification was conducted on an ABI 7900HT system (Applied Biosystems) under standardized cycling conditions. Viral NP RNA levels were normalized to endogenous mRPL13A mRNA expression using the 2⁻ΔΔCt method.
## Statistical analyses
The graphical and statistical evaluations were conducted using Prism software (Graph Pad Prism 9.5.0). The data are presented as the mean ± standard error of the mean. Comparisons between experimental and control groups were made using unpaired two-tailed t-tests. Statistical significance was defined as P values < 0.05.
## RESULTS
## DS-Cav1 immunization elicits serum with potent neutralizing activity against RSV
Based on the enhanced neutralizing potency of antibodies targeting pre-F-specific epitopes, we generated the pre-F-stabilized RSV F variant DS-Cav1, incorporating the stabilizing mutations and a C-terminal "foldon" trimerization domain (38). An alpaca was immunized with eukaryotic-expressed DS-Cav1 at weeks 0, 3, and 6 (Fig. 1A). Serum collected after both second and third immunizations exhibited high nanobody titers against RSV A2 and B1 DS-Cav1 antigens, confirming robust humoral immunity (Fig. 1B). Compared with post-second immunization samples, serum from the third immuniza tion demonstrated significantly enhanced neutralizing activity against RSV A2 and B1, achieving ∼10-fold higher 50% inhibitory dilution (ID₅₀) titers for both strains (Fig. 1C). These results establish that DS-Cav1 immunization elicits potent neutralizing antibody responses against both RSV subtypes, providing a solid foundation for isolating anti-RSV neutralizing nanobodies that neutralize both RSV subgroups.
## Isolation and characterization of anti-RSV F nanobodies
PBMCs were isolated for the construction of a phage display library to screen for nanobodies. Biotinylated DS-Cav1 was immobilized on streptavidin-coated magnetic beads via biotin-streptavidin conjugation to preserve conformational integrity. In rounds 1 and 2, phage pools were directly panned against immobilized DS-Cav1 to enrich binders. Beginning in round 3, phage libraries were pre-incubated with a cocktail of site I-IV antibodies before antigen exposure. This competitive blocking step depleted binders to low-neutralization epitopes, resulting in preferential enrichment of clones recognizing high-neutralization-potency epitopes. Following four rounds of biopanning, single phage clones from the enriched library were screened by phage ELISA. The results showed that 41 out of 96 clones from the third round and 32 out of 96 clones from the fourth round were positive binders (Fig. 2A). Based on ELISA signals and sequence analysis, eight nanobodies were selected for further characterization. These nanobodies were expressed as nanobody-human IgG1 Fc fusion proteins in 293 F cells (Fig. 2B). All eight nanobodies exhibited neutralizing activity against RSV infection, with 4-H1-Fc showing the highest potency against both subtypes (Fig. 2C). Subsequent bio-layer interferometry (BLI)-based mAb competition assays revealed that all nanobod ies competed with D25 (the precursor to nirsevimab targeting site Ø) for antigen binding (41). Binding curves demonstrated that pre-incubation of DS-Cav1 with any selected nanobodies inhibited D25 binding. In contrast, pre-saturation of DS-Cav1 with the non-neutralizing antibody, referred to as NC, did not impair D25 binding (Fig. S1). Overall, these findings demonstrate that our biopanning strategy successfully isolated potent neutralizing nanobodies targeting site Ø.
## Characterization of 4-H1
Based on superior neutralizing potency against RSV A2 and B1, 4-H1 was selected for further characterization (Fig. 2C). Epitope mapping via BLI-based cross-competition assays using site-specific mAbs revealed that pre-saturation of DS-Cav1 with site Ø mAbs (nirsevimab, 5C4, RSD5 [12,42,43]) abolished subsequent 4-H1-Fc binding. In contrast, mAbs targeting sites I (ADI-14359 [44]), II (motavizumab, 14N4 [45,46]), III (ADI-19425, MPE8 [44,47]), IV (RB1 [48]), and V (CR9501, ADI-14442, hRSV90, AM14 [49][50][51][52]) showed minimal competition, confirming exclusive specificity of 4-H1 for site Ø (Fig. 3A). Given the site Ø specificity observed by BLI, we assessed the conformational dependence of 4-H1 by ELISA. 4-H1-Fc demonstrated potent binding to pre-F (DS-Cav1, EC₅₀ =5.575 ng/mL for RSV A2 and 4.188 ng/mL for RSV B1) but minimal reactivity toward post-F, confirming preferential recognition of the pre-F conformation (Fig. 3B). For structural modeling, energy minimization was performed using the consistent valence force field (CVFF). The 4-H1 framework underwent 30,000 iterations with steepest descent followed by conjugate gradient algorithms, converging at an energy gradient threshold of 0.02 kcal/mol (Fig. 3C). Ramachandran analysis confirmed >99% of residues in allowed regions. DS-Cav1 was refined identically. Using the experimental DS-Cav1-D25 complex as a topological scaffold, the theoretical 3D complex structure of 4-H1 and DS-Cav1 was generated by constrained docking methods (Fig. 3D). Analysis of interac tions within a 0.6 nm radius of 4-H1 CDRs, including van der Waals forces, hydrogen bonds, polar contacts, and electrostatic potentials, revealed two high-affinity binding clusters on DS-Cav1, that is, residues 65/67/69 and 204-212, respectively (Fig. 3E). The spatial adjacency of these clusters to critical binding residues of established site Ø mAbs (nirsevimab, RSD5, 5C4) suggests potential overlap in their binding interfaces (12,42,43).
## Epitope mapping of RSV F bound to 4-H1
To map functional residues, we performed alanine scanning mutagenesis on DS-Cav1 at positions L204, P205, I206, L207, K209, Q210, S211, and C212, based on the volume and character of amino acid residues. We also generated the K65A-N67A-C69A triplemutated protein. Control antigens included DS-Cav1 variants encoding the F protein sequences from a nirsevimab-escape mutant (harboring K68E-N208Y substitutions) and a palivizumab binding-attenuated mutant (carrying N262R-K272E-S275R mutations) (39,45). ELISA screening at 2 µg/mL coating concentration revealed that only K65A-N67A-C69A exhibited marginally reduced binding to 4-H1-Fc. By contrast, this mutation severely impaired 5C4 and RSD5 binding, whereas nirsevimab binding remained unaffected. Importantly, both 4-H1-Fc and RSD5 retained binding to the K68E-N208Y, suggesting their epitope divergence with nirsevimab and 5C4 (Fig. S2A). When antigen coating was reduced to 0.5 µg/mL to enhance sensitivity, the L207A, K209A, and K65A-N67C-C69A mutations disrupted 4-H1-Fc binding, resulting in EC₅₀ values > 100 ng/mL. In contrast, motavizumab binding remained unaffected under these conditions (Fig. 4A andB).
To precisely define the binding kinetics underlying these disruptions, we performed SPR analysis. Using His-tagged 4-H1, we observed that compared with wild-type DS-Cav1, the L207 and K65A-N67A-C69A mutated DS-Cav1 showed reduced association rates (k a , 1.18 × 10⁵/Ms and 9.50 × 10 4 /Ms) and increased dissociation rates (k d , 3.12 × 10⁻ 3 /s and 5.22 × 10⁻ 3 /s) to 4-H1, resulting in K D values of 26.4 nM and 54.9 nM, respec tively (Fig. 4D andE). This represents a ∼6-fold and 13-fold affinity decrease relative to wild-type (K D = 4.37 nM, Fig. 4C). The K209A mutated protein displayed mild affinity loss (K D = 8.67 nM) solely through increased k d (2.15 × 10⁻ 3 /s) (Fig. 4F). These SPR results correlated with ELISA EC₅₀ data. Additionally, the C212 structural disulfide bond stabilizes F protein folding, but we found it did not contribute to functional epitope-paratope interactions, as evidenced by minimal changes in the binding kinetics of C212A despite reduced ELISA signals (Fig. 4G). All SPR binding kinetics parameters are listed in Fig. 4H. Structural modeling quantified atomic distances between paratope residues (D30, I101) and antigenic site Ø residues (K65, N67, C69, L207, and K209). Specific atoms within the side chains of C69, L207, and K209 were positioned within 3 Å of D30 or I101, suggesting potential hydrogen bonding or hydrophobic interactions. In contrast, K65 and N67 exhibited longer interaction distances, indicating weaker energetic contributions to complex stabilization (Fig. S2B). Collectively, these data define a shared dependency on the K65-N67-C69 cluster among 5C4, RSD5, and 4-H1, indicating that this cluster constitutes a critical component of their overlapping functional epitopes. In contrast, nirsevimab binding is independent of this cluster. Although 4-H1 and nirsevimab share no critical binding residues, their epitopes spatially neighbor each other (12), suggesting that steric hindrance, not epitope identity, may mediate the observed complete competition.
## 4-H1-anti-HSA-4-H1, heterotrimeric and bispecific for F and HSA, exhibits robust neutralization against RSV
To enhance the in vivo efficacy and stability of 4-H1, we engineered a heterotrimeric bispecific construct, 4-H1-anti-HSA-4-H1. This molecule fuses two 4-H1 copies to a novel anti-HSA nanobody via (GS)₆ linkers (Fig. 5A). The underlying design principle is supported by a prior study in which an HSA-fused construct demonstrated longer serum half-life and higher peak serum concentration (Cmax) than the 3 × Nb15 control. Notably, it achieved effective lung delivery with sustained localization after intranasal administration, providing significant protection against SARS-CoV-2 when administered 24 h pre-exposure (53).
The novel anti-HSA nanobody, developed in our laboratory, exhibits a binding affinity for HSA of 5.47 nM (Fig. S3A) and cross-reacts with mouse serum albumin (MSA) (Fig. S3B). ELISA confirmed that this construct binds simultaneously to both HSA and DS-Cav1 (Fig. 5B). Building on these results, SPR assays were subsequently performed to characterize binding alterations resulting from dual 4-H1 integration.
Assessment of 4-H1-anti-HSA-4-H1 binding showed a modest decrease in HSA affinity (K D = 8.65 nM, Fig. 5C) compared with the monomeric anti-HSA control (5.47 nM, Fig. S3A). Conversely, DS-Cav1 binding was enhanced (K D = 1.44 nM versus 4.37 nM for monomeric 4-H1, Fig. 5D and4C). This enhancement in DS-Cav1 affinity may stem from the bivalently bound avidity facilitated by the dual 4-H1 architecture, potentially stabilizing the antigen-antibody complex and increasing functional valency. Neutralization capacity assessed via PRNT demonstrated exceptional potency against both RSV subtypes, with IC 50 values of 0.5641 ng/mL for A2 and 0.3571 ng/mL for B1, as shown in Fig. 5E andF. Building on the potent in vitro neutralization activity of 4-H1-anti-HSA-4-H1, we evaluated its pharmacokinetics following intranasal administration in mice, using a monospecific 4-H1 trimer (lacking anti-HSA) as a control (Fig. 6A). Blood samples were obtained at six post-administration time points (4,24,48,96,168, and 240 h) for serum isolation, and the nanobody concentrations were subsequently quantified via the developed sandwich ELISA. Our analysis revealed that 4-H1-anti-HSA-4-H1 exhibits significantly extended serum persistence, demonstrating a half-life (T 1/2 ) of 48.78 h versus 12.82 h for the control trimer, a ∼4-fold higher Cmax, and a ∼7-fold greater serum area under the curve (AUC, Fig. 6B). These enhanced pharmacokinetic properties suggest enhanced retention in respiratory mucosa, supporting its progression to challenge models based on robust neutralization capacity and favorable pharmacokinetics.
## In vivo anti-RSV activity of 4-H1-anti-HSA-4-H1
Finally, we evaluated the in vivo antiviral efficacy of 4-H1-anti-HSA-4-H1 against RSV in mice. Animals received intranasal administration of 2 mg/kg 4-H1-anti-HSA-4-H1 or an equal volume of PBS, followed by intranasal challenge with 4 × 10⁶ PFU RSV A2 24 h later (Fig. 7A). Four days after the challenge, lungs were harvested for viral titer determi nation and histopathological examination. Viral RNA was quantified by an RSV NP-spe cific RT-qPCR, and the lungs of mice treated with 4-H1-anti-HSA-4-H1 were borderline positive for RSV RNA, whereas the relative amount of RNA indicated much higher viral load in samples from PBS-treated mice (Fig. 7B). Next, we performed H&E staining to assess viral lung pathology. The PBS-treated mice exhibited severe pulmonary inflammation, characterized by inflammatory infiltration across multiple tissue compartments. In contrast, a mild lung infiltration was observed in the prophylactic 4-H1-anti-HSA-4-H1 treated group (Fig. S4). Quantitative pathology scoring confirmed markedly lower injury in the 4-H1-anti-HSA-4-H1 treated group (Fig. 7C), with significantly reduced alveolitis and interstitial pneumonia (Fig. 7D andE). Collectively, these findings demonstrate that 4-H1-anti-HSA-4-H1 significantly mitigated RSV-induced pulmonary damage, indicating its value as a prophylactic candidate against RSV infection.
## DISCUSSION
RSV imposes a substantial global burden of severe respiratory disease (54)(55)(56). To develop high-potency therapeutics, we pursue molecules with exceptional neutralizing activity against RSV. Given the documented superiority of pre-F-specific antibodies over those targeting epitopes present in both pre-F and post-F conformations (57), an alpaca was immunized with the engineered pre-F-stabilized F trimer DS-Cav1. Through an epitope-blocking depletion strategy, we isolated a panel of site Ø-directed nanobodies, with 4-H1 exhibiting the most potent inhibitory activity. Critically, epitope characteriza tion established that 4-H1 engages site Ø residues K209, C212, and the K65-N67-C69 cluster, as demonstrated by impaired binding upon alanine substitution. Notably, in ELISA binding assays at 2 µg/mL antigen coating, none of the mutated proteins markedly impaired 4-H1-Fc binding. By contrast, other site Ø-directed mAbs showed substan tial binding reduction for specific mutations. Smaller binding perturbation by point mutations suggests distributed energy contributions across the epitope. We propose that 4-H1 recognizes an epitope with remarkable mutational resilience, requiring multiple substitutions for escape. Notably, no site V-targeting antibodies were isolated during initial screening. We attribute this to steric occlusion by prebound site I-IV antibodies, which may have physically masked site V through spatial overlap or induced conformational masking via allosteric effects. Additional unidentified factors could also contribute to the absence of site V binders. In subsequent efforts, we will employ post-F negative selection to replace the epitope-blocking strategy, thereby potentially enabling isolation of site V-specific binders. Following successful isolation, biparatopic agents incorporating both site Ø-and V-targeting nanobodies will be developed. This design is expected to enhance neutralizing potency and limit viral escape.
Owing to their small size (~15 kDa), nanobodies undergo rapid renal clearance (58), necessitating half-life extension strategies. Albumin-binding approaches represent a clinically validated method, exemplified by the trivalent bispecific nanobody Ozorali zumab (59). This molecule incorporates two TNF-α-targeting domains and an HSA-bind ing module that extends half-life via neonatal Fc receptor (FcRn)-mediated recycling (59,60). Despite multiple reported anti-RSV F nanobodies, none have achieved clinical approval (31,(33)(34)(35)(36)(37). Notably, ALX-0171-a trivalent anti-RSV F nanobody developed for therapeutic use-progressed to phase II trials but was terminated due to failure in reducing hospitalization duration (31). Other previously described nanobodies also exhibit certain limitations. For instance, pre-F-specific nanobodies such as m17 and m35 recognize a novel prefusion epitope, termed site VI, and effectively prevent conforma tional rearrangements during the fusion process. However, their neutralization potency against both RSV A and B subtypes is markedly less potent than that of other nanobod ies (35). Another nanobody, F-VHHb, which targets regions IV-VI, has been shown to effectively inhibit viral entry and suppress replicative virus to undetectable levels in mice when administered prior to challenge. Nevertheless, its efficacy against RSV B was not evaluated, limiting the assessment of its broad-spectrum potential (34,37). Similarly, F-VHH-Cl184, which primarily binds antigenic site I with additional interactions in sites III and IV, demonstrates markedly reduced neutralization against RSV B compared to RSV A (33). To our knowledge, this represents the first report of a pre-F-specific anti body predominantly targeting antigenic site I. In contrast, the pre-F-specific nanobody F-VHH-4, developed by the same team, exhibits superior neutralizing activity and is among the most potent nanobodies reported to date. It provides effective prophy laxis at doses as low as 0.5 mg/kg in mice. Structural studies reveal that F-VHH-4 targets a conserved cavity formed between two F protomers, engaging residues from antigenic sites II, III, IV, and V (36). However, its application is constrained by rapid clearance, necessitating administration shortly before viral exposure, as demonstrated in the referenced study, where it was administered 4 h prior to challenge (36). To address these limitations, we engineered heterotrimeric 4-H1-anti-HSA-4-H1, synergiz ing potent neutralization via site Ø with albumin-mediated half-life extension. This molecule potently neutralizes both RSV A2 and B1, exhibiting neutralizing activity that is comparable with, if not greater than, any nanobody reported to date. Leveraging albumin-mediated recycling, the construct achieved a 48.78 h serum half-life in mice, markedly surpassing that of the unconjugated control trimer (12.82 h). Consequently, a single intranasal administration (2 mg/kg) conferred robust protection, as evidenced by reduced lung viral load via RT-qPCR and attenuated lung pathology in H&E-stained sections.
Regrettably, no infectious viral titers were detected by plaque assay in any group. We reason that this is likely due to technical limitations in our sample processing protocol. The core issue appears to be that the entire workflow, in our hands, was suboptimal for preserving the infectivity of the labile RSV, with the intensive mechanical homogeniza tion step being a potential critical point. Nevertheless, the liberated viral RNA remained quantifiable by RT-qPCR. Prior studies have validated RT-qPCR-based quantification of RSV in tissues (36,(61)(62)(63)(64)(65), with some relying on it without additional viral titrations of the lung (61,63,64), thus supporting its use as a reliable surrogate metric.
Critically, the < 45 kDa size enables efficient inhalation delivery, achieving high local drug concentrations at respiratory infection sites while minimizing systemic barriers. This confers potential for faster and more potent local antiviral effects compared with systemic delivery of the identical drug (66,67). Such rapid efficacy is vital for RSV management, as most infected infants are hospitalized at or after peak viral replication and disease progression, missing the early intervention window. Hence, immediate respiratory mucosal drug delivery is essential to arrest pathogenesis. Accordingly, our future work will focus on formulating the 4-H1-anti-HSA-4-H1 molecule for nebulized delivery and rigorously evaluating its prophylactic and therapeutic potential in relevant respiratory infection models. Despite achieving an extension in serum half-life versus the control (48.78 h vs 12.82 h), this half-life remains insufficient for long-term prophylaxis. The reduced binding affinity of 4-H1-anti-HSA-4-H1 for MSA, as indicated by its ∼5-fold higher EC₅₀ against MSA than HSA (Fig. S3C), likely underlies this limitation. Beyond this, optimization of HSA binding affinity remains critical to enhance pharmacokinetic properties and enable robust clinical application.
In summary, respiratory-administered bispecific nanobodies represent a next-gener ation therapeutic platform. By integrating potent neutralization, extended durability, and targeted delivery, 4-H1-anti-HSA-4-H1 may provide a promising strategy for RSV prevention and treatment.
## References
1. Chanock, Roizman, Myers (1957) "Recovery from infants with respiratory illness of a virus related to chimpanzee coryza agent (CCA). I. Isolation, properties and characterization" *Am J Hyg*
2. Nair, Nokes, Gessner et al. (2010) "Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: a systematic review and meta-analysis" *Lancet*
3. Shi, Mcallister, Brien et al. (2017) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study" *Lancet*
4. Kampmann, Madhi, Munjal et al. (2023) "Bivalent prefusion F vaccine in pregnancy to prevent RSV illness in infants" *N Engl J Med*
5. (1998) "Prevention of respiratory syncytial virus infections: indications for the use of palivizumab and update on the use of RSV-IGIV" *Pediatrics*
6. Raghuveer, Zackula (2024) "Nirsevimab for prevention of RSV hospitalizations in infants" *N Engl J Med*
7. Syed (2025) "Clesrovimab: first approval" *Drugs (Abingdon Engl)*
8. Jenkins, Hoet, Hochrein et al. (2023) "The quest for a respiratory syncytial virus vaccine for older adults: thinking beyond the F protein" *Vaccines (Basel)*
9. Fuentes, Hahn, Chilcote et al. (2020) "Antigenic fingerprinting of respiratory syncytial virus (RSV)-A-infected hematopoietic cell transplant recipients reveals importance of mucosal anti-RSV G antibodies in control of RSV infection in humans" *J Infect Dis*
10. Gilman, Castellanos, Chen et al. (2016) "Rapid profiling of RSV antibody repertoires from the memory B cells of naturally infected adult donors"
11. Taleb, Thani, Ansari et al. (2018) "Human respiratory syncytial virus: pathogenesis, immune responses, and current vaccine approaches" *Eur J Clin Microbiol Infect Dis*
12. Zhu, Mclellan, Kallewaard et al. (1928) "A highly potent extended half-life antibody as a potential RSV vaccine surrogate for all infants" *Sci Transl Med*
13. Ngwuta, Chen, Modjarrad et al. (2015) "Prefusion F-specific antibodies determine the magnitude of RSV neutralizing activity in human sera" *Sci Transl Med*
14. Papi, Ison, Langley et al. (2023) *Full-Length Text Journal of Virology*
15. "Respiratory syncytial virus prefusion F protein vaccine in older adults" *N Engl J Med*
16. Wilson, Goswami, Baqui et al. (2023) "Efficacy and safety of an mRNA-based RSV PreF vaccine in older adults" *N Engl J Med*
17. Alexander, Leong (2024) "Discovery of nanobodies: a comprehen sive review of their applications and potential over the past five years" *J Nanobiotechnology*
18. Hamers-Casterman, Atarhouch, Muyldermans et al. (1993) "Naturally occurring antibodies devoid of light chains" *Nature*
20. Salvador, Vilaplana, Marco (2019) "Nanobody: outstanding features for diagnostic and therapeutic applications" *Anal Bioanal Chem*
21. Liu, Huang (2018) "Expression of single-domain antibody in different systems" *Appl Microbiol Biotechnol*
22. Kunz, Durandy, Seguin et al. (2023) "NANOBODY molecule, a giga medical tool in nanodimensions" *IJMS*
23. Jovčevska, Muyldermans (2020) "The therapeutic potential of nanobodies" *BioDrugs*
24. Halstead, Chow, Marchette (1973) "Immunological enhancement of dengue virus replication" *Nat New Biol*
25. Yip, Leung, Li et al. (2016) "Antibody-dependent enhancement of SARS coronavirus infection and its role in the pathogenesis of SARS" *Hong Kong Med J*
26. Wan, Shang, Sun et al. (2020) "Molecular mechanism for antibody-dependent enhancement of coronavirus entry" *J Virol*
27. Wang, Wang, Yu et al. (2022) "Antibody-dependent enhancement (ADE) of SARS-CoV-2 pseudoviral infection requires FcγRIIB and virus-antibody complex with bivalent interaction" *Commun Biol*
28. Ponnuraj, Springer, Hayward et al. (2003) "Antibody-dependent enhancement, a possible mechanism in augmented pulmonary disease of respiratory syncytial virus in the Bonnet monkey model" *J Infect Dis*
29. Chen, Huang, Wang et al. (2025) "Influenza A virus H7 nanobody recognizes a conserved immunodominant epitope on hemagglutinin head and confers heterosubtypic protection" *Nat Commun*
30. Zhu, Huang, Wang et al. (2024) "Highly potent and broadly neutralizing anti-CD4 trimeric nanobodies inhibit HIV-1 infection by inducing CD4 conformational alteration" *Nat Commun*
31. Xiang, Xu, Mcgovern et al. (2024) "Adaptive multi-epitope targeting and avidity-enhanced nanobody platform for ultrapotent, durable antiviral therapy" *Cell*
32. Koenig, Das, Liu et al. (2021) "Structureguided multivalent nanobodies block SARS-CoV-2 infection and suppress mutational escape" *Science*
33. Detalle, Stohr, Palomo et al. (2016) "Generation and characterization of ALX-0171, a potent novel therapeutic nanobody for the treatment of respiratory syncytial virus infection" *Antimicrob Agents Chemother*
34. Feng, Li, Zhang et al. (2025) "A sharkderived broadly neutralizing nanobody targeting a highly conserved epitope on the S2 domain of sarbecoviruses" *J Nanobiotechnology*
35. Rossey, Hsieh, Sedeyn et al. (2021) "A vulnerable, membrane-proximal site in human respiratory syncytial virus F revealed by a prefusion-specific singledomain antibody" *J Virol*
37. Schepens, Ibañez, Baets et al. (2011) "Nanobodies specific for respiratory syncytial virus fusion protein protect against infection by inhibition of fusion" *J Infect Dis*
38. Xun, Song, Hu et al. (2021) "Potent human single-domain antibodies specific for a novel prefusion epitope of respiratory syncytial virus F glycoprotein" *J Virol*
39. Rossey, Gilman, Kabeche et al. (2017) "Potent single-domain antibodies that arrest respiratory syncytial virus fusion protein in its prefusion state" *Nat Commun*
40. Hultberg, Temperton, Rosseels et al. (2011) "Llama-derived single domain antibodies to build multivalent, superpotent and broadened neutralizing anti-viral molecules" *PLoS One*
41. Mclellan, Chen, Joyce et al. (2013) "Structure-based design of a fusion glycoprotein vaccine for respiratory syncytial virus" *Science*
42. Sun, Liu, Qiang et al. (2024) "A potent broad-spectrum neutralizing antibody targeting a conserved region of the prefusion RSV F protein" *Nat Commun*
43. Yang, Xue, Liu et al. (2024) "Farnesyltransferase inhibitor lonafarnib suppresses respiratory syncytial virus infection by blocking conforma tional change of fusion glycoprotein" *Signal Transduct Target Ther*
44. Mclellan, Chen, Leung et al. (2013) "Structure of RSV fusion glycoprotein trimer bound to a prefusion-specific neutralizing antibody" *Science*
45. Tian, Battles, Moin et al. (2017) "Structural basis of respiratory syncytial virus subtype-dependent neutralization by an antibody targeting the fusion glycoprotein" *Nat Commun*
46. Jones, Battles, Lin et al. (2019) "Alternative conformations of a major antigenic site on RSV F" *PLoS Pathog*
47. Goodwin, Gilman, Wrapp et al. (2018) "Infants infected with respiratory syncytial virus generate potent neutralizing antibodies that lack somatic hypermuta tion" *Immunity*
48. Mclellan, Chen, Kim et al. (2010) "Structural basis of respiratory syncytial virus neutralization by motavizumab" *Nat Struct Mol Biol*
49. Mousa, Sauer, Sevy et al. (2016) "Structural basis for nonneutralizing antibody competition at antigenic site II of the Full-Length Text Journal of Virology December"
50. "respiratory syncytial virus fusion protein" *Proc Natl Acad Sci*
51. Corti, Bianchi, Vanzetta et al. (2013) "Cross-neutralization of four paramyxo viruses by a human monoclonal antibody" *Nature*
52. Tang, Chen, Cox et al. (2019) "A potent broadly neutralizing human RSV antibody targets conserved site IV of the fusion glycoprotein"
53. Gilman, Furmanova-Hollenstein, Pascual et al. (2019) "Transient opening of trimeric prefusion RSV F proteins" *Nat Commun*
54. Mukhamedova, Wrapp, Shen et al. (2021) "Vaccination with prefusion-stabilized respiratory syncytial virus fusion protein induces genetically and antigenically diverse antibody responses" *Immunity*
55. Mousa, Kose, Matta et al. (2017) "A novel prefusion conformation-specific neutralizing epitope on the respiratory syncytial virus fusion protein" *Nat Microbiol*
56. Gilman, Moin, Mas et al. (2015) "Characterization of a prefusion-specific antibody that recognizes a quaternary, cleavagedependent epitope on the RSV fusion glycoprotein" *PLoS Pathog*
57. Wu, Cheng, Fu et al. (2021) "A potent bispecific nanobody protects hACE2 mice against SARS-CoV-2 infection via intranasal administration" *Cell Rep*
58. Li, Wang, Blau et al. (2022) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis" *The Lancet*
59. Shi, Denouel, Tietjen et al. (2020) "Global disease burden estimates of respiratory syncytial virus-associated acute respiratory infection in older adults in 2015: a systematic review and meta-analysis" *J Infect Dis*
60. Savic, Penders, Shi et al. (2020) "Respiratory syncytial virus disease burden in adults aged 60 years and older in high-income countries: a systematic literature review and metaanalysis" *Influenza Resp Viruses*
61. Jumapili, Zivalj, Barthelmess et al. (2023) "A few good reasons to use nanobodies for cancer treatment" *Eur J Immunol*
62. Tanaka, Kawanishi, Nakanishi et al. (2023) "Efficacy and safety of the anti-TNF multivalent NANOBODY compound ozoralizumab in patients with rheumatoid arthritis and an inadequate response to methotrexate: a 52-week result of a phase II/III study (OHZORA trial)" *Mod Rheumatol*
63. Roopenian, Akilesh (2007) "FcRn: the neonatal Fc receptor comes of age" *Nat Rev Immunol*
64. Torres, Gomez, Khokhar et al. (2010) "Respiratory syncytial virus (RSV) RNA loads in peripheral blood correlates with disease severity in mice" *Respir Res*
65. Ogonczyk-Makowska, Brun, Vacher et al. (2024) "Mucosal bivalent live attenuated vaccine protects against human metapneumovirus and respiratory syncytial virus in mice" *NPJ Vaccines*
66. Goritzka, Durant, Pereira et al. (2014) "Alpha/beta interferon receptor signaling amplifies early proinflammatory cytokine production in the lung during respiratory syncytial virus infection" *J Virol*
67. Dai, Ruan, Mao et al. (2022) "The antiviral efficacies of small-molecule inhibitors against respiratory syncytial virus based on the F protein" *J Antimicrob Chemother*
68. Olszewska, Ispas, Schnoeller et al. (2011) "Antiviral and lung protective activity of a novel respiratory syncytial virus fusion inhibitor in a mouse model" *Eur Respir J*
69. Gai, Ma, Li et al. (2020) "A potent neutralizing nanobody against SARS-CoV-2 with inhaled delivery potential" *MedComm*
70. Shaibie, Faizal, Buang et al. (2025) "Inhaled biologics for respiratory diseases: clinical potential and emerging technologies" *Drug Deliv Transl Res*
71. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12553294&blobtype=pdf
|
Yong Poovorawan
## Abstract
chikungunya virus (CHIKV) [3]. Since the clinical symptoms of disease associated with arbovirus infection often overlap, comprehensive molecular surveillance is crucial for detecting emerging or lesser-known pathogens, such as Oropouche virus (OROV). This overlap in clinical features poses significant challenges for accurate diagnosis and proper treatment.OROV is an RNA virus belonging to the family Peribunyaviridae, genus Orthobunyavirus [4]. The virus is primarily transmitted to humans through the bite of an infected biting midge Culicoides paraensis, but certain mosquito species, including Culex quinquefasciatus, Aedes serratus, and Coquillettidia venezuelensis [5], can also serve as vectors. OROV is a significant cause of arboviral illness in South and Central American
## Introduction
Acute febrile illness, characterized by a sudden onset of fever and non-specific symptoms such as fatigue, headache, arthralgia, retro-orbital pain, and myalgia, can result from a wide range of infectious etiologies [1,2]. In Thailand, significant arboviruses affecting humans include dengue (DENV), Zika virus (ZIKV), and Oropouche virus screening among acute febrile illness cases in Thailand: no evidence of circulation in arboviral-endemic regions Sarawut Khongwichit 1 , Piyada Linsuwanon 2 , Sirima Wongwairot 2 , Erica Lindroth 2 and Yong Poovorawan 1* countries, particularly in the Amazon basin, including Brazil, Columbia, Cuba, and Peru [6]. Although the overall incidence of OROV infections has decreased in recent years, the ongoing outbreak across South America and the Caribbean in 2024 has resulted in over 8,000 reported cases, including two fatal cases and five cases of vertically transmitted infections associated with congenital abnormalities or fetal death. Additionally, travelassociated OROV infections have been identified in the United States and Europe among individuals returning from Brazil and Cuba [7], suggesting that the virus may have a wider geographic distribution than previously anticipated. The lack of global recognition and routine monitoring of OROV may further contribute to underestimating its true burden.
While OROV has been well-documented in Latin America, its presence in other tropical and subtropical areas including regions already burdened by arboviruses, remains unknown. Given the public health significance and the presence of competent vectors of OROV within Thailand, it is crucial to assess the potential introduction of this virus into the country. In this study, we retrospectively screened serum samples from patients with acute febrile illness who were clinically suspected of arboviral infections but had previously tested negative for DENV, CHIKV, and ZIKV using quantitative RT-PCR to detect OROV RNA.
## Method
## Specimen collection
This study was approved by the Institutional Review Boards of Chulalongkorn University (IRB378-59, IRB710/64) and WRAIR (WRAIR2386). Informed consent was waived due to the use of anonymized residual specimens. A total of 1,374 serum samples collected from patients with acute febrile illness who visited hospitals between 2018 and 2024 were included in the study. These samples were from the hospitals across five provinces in three regions: Central region (Bangkok, n = 761; Samut Prakan, n = 227; Samut Sakhon, n = 43), Eastern region (Chon Buri, n = 13), and Northeastern region (Khon Kaen, n = 330). Samples with laboratory-confirmed CHIKV (positive nucleic acid test and/or IgM), DENV, or ZIKV infections (positive nucleic acid test) were excluded from the study, as the primary focus was on OROV infection.
## RNA extraction and OROV detection using quantitative reverse transcriptase polymerase chain reaction (RT-qPCR) amplification
RNA was extracted from patient sera using the Mag-DEA® Dx SV according to the manufacturer's protocol (Precision System Science Co., Ltd., Japan). The real-time RT-qPCR method was performed using TaqMan™ Fast Virus 1-Step master mix (Applied Biosystems) to detect OROV with primer and probe sequences as previously described [8]. A chimeric plasmid containing a partial gene encoding the OROV S segment was used as a synthetic, non-infectious positive control. In brief, primers
) were designed to target conserved regions in the OROV small (S) segment. The RT-qPCR cycling conditions were reverse transcription at 50 °C for 10 min, RT inactivation/initial denaturation at 95 °C for 20 s, followed by 45 cycles of 95 °C for 3 s and 60 °C for 30 s. A Ct value < 38 was considered positive in this study. The assay has been reported to demonstrate an analytical sensitivity of 2-20 RNA copies per reaction, with amplification efficiency >98% and R² close to 1, broad inclusivity across genetically diverse OROV strains, and specificity confirmed against a wide panel of 17 arboviruses with no cross-reactivity [8]. Inclusivity analysis confirmed primer/probe alignment with >100 OROV sequences from multiple global lineages, supporting detection should the virus be introduced into Southeast Asia.
$$(OROV_FNF: 5′-T C C G G A G G C A G C A T A T G T G-3′ and OROV_FNR: 5′-A C A A C A C C A G C A T T G A G C A C T T-3′) and labeled fluorescent probe (OROV_FNP: 5′-(FAM) C A T T T G A A G C T A G A T A C G G-3′$$
## Results and discussion
Thailand, located on the mainland of Southeast Asia and bordered by Myanmar, Laos, Cambodia, and Malaysia, serves as a crucial zoogeographic crossroads because of its geographic location and natural gateways. Recognized as a biodiversity hotspot, Thailand borders the Andaman Sea to the west and the Gulf of Thailand to the east, harboring a wide range of potential animal reservoirs and arthropod vectors. It is also known as an area of circulation for arboviruses such as DENV, CHIKV, and ZIKV. The symptoms of these arboviral illnesses are similar to those of patients suffering from Oropouche fever [9]. While the primary vector for OROV in the Americas, Culicoides paraensis, has not been reported in Southeast Asia, the presence of Cx. quinquefasciatus [10], one of the experimentally proven competent vectors of OROV, in many Southeast Asia countries raises concerns about the potential presence of OROV in Thailand.
To monitor the presence of OROV in Thailand, we retrospectively screened 1,374 serum samples from acute febrile illness patients who had previously tested negative for DENV, CHIKV, and ZIKV. Using highly sensitive and specific primers and probes, previously validated by Naveca et al. [8], we did not detect OROV RNA in any of the samples. To further confirm assay specificity in our setting, we additionally tested representative samples positive for DENV (all four serotypes), CHIKV (ECSA genotype), and ZIKV (Asian genotype), which are the major arboviruses circulating in Thailand [3], as well as Bandavirus dabieense (genotypes B) [11], which has also been reported in the country, and no cross-reactivity was observed. Most patients in this study were from urban and suburban areas and none had a travel history to areas where OROV is endemic. Given the absence of OROV RNA in this cohort, it is plausible to conclude that the virus was not circulating in Thailand during the study period. The widespread distribution of Cx. quinquefasciatus, which has been shown experimentally to be a competent OROV vector [5], together with the presence of several Culicoides species in Thailand, whose competence for OROV transmission has not yet been established, suggests that ecological conditions could support OROV transmission should the virus be introduced.
Current evidence together with the negative findings in the present study suggests that the risk of OROV transmission in Thailand is potentially low. Nevertheless, it is imperative to maintain continuous surveillance and implement effective preventive strategies to minimize the potential for future outbreaks, particularly in regions with high levels of international travel. Beyond these findings, surveillance should be expanded to rural and border regions, as well as traveler populations, where introduction risks may be higher. Incorporating OROV assays into existing arboviral diagnostic panels and prioritizing vector surveillance of Culex and Culicoides species would further enhance preparedness. These measures, together with regional collaboration, will strengthen early warning capacity for potential OROV emergence. Furthermore, the development of rapid and accurate diagnostic capabilities remains essential to early diagnosis, appropriate treatment, and preventing worsen disease progression. Raising public and healthcare professional awareness about OROV can further enhance early detection, and quarantine, and reduce the overall impact of the disease.
Although our findings indicate no OROV infection in the Thai population, it is important to acknowledge the several limitations of our study. First, the specimens were residual anonymized sera from patients who were clinically suspected of dengue, chikungunya, or Zika virus infection but tested negative. Because only anonymized samples were available, detailed demographic and clinical information such as age, sex, and precise symptom onset could not be uniformly retrieved, which restricted patient-level analysis and comparison with other cohorts. Second, most patients were from urban and suburban hospitals, while rural and border populations and traveler groups were under-represented, limiting generalizability across Thailand. Third, samples positive for DENV, CHIKV, or ZIKV were excluded by design; however, co-infections between OROV and other arboviruses have been reported [12,13], and future studies should include parallel OROV testing in arbovirus-positive cases. Fourth, this study relied exclusively on molecular detection, and serological testing for OROV was not performed due to the limited residual sample volumes and the absence of validated serological assays in Thailand. Although all specimens were collected after symptom onset, it is possible that some were obtained outside the peak of viremia, when viral RNA levels were already undetectable, leading to false negatives. Future studies should incorporate seroepidemiological approaches to better assess past exposure and population immunity. Fifth, serum specimens were stored at -80 °C under biobank protocols prior to testing, but prolonged storage may still pose a theoretical risk of RNA degradation. Finally, the retrospective design, limited temporal coverage of sampling (2018-2024), and relatively small sample size from each site may not fully capture seasonal, interannual, or nationwide variability in OROV circulation. Future research should employ prospective longitudinal surveillance covering multiple years and seasons, with larger sample sets from both urban and under-represented rural and border populations, to better characterize the risk of OROV emergence.
Further research, particularly in areas where Cx. quinquefasciatus is prevalent, will be important to comprehensively assess the potential risk of OROV in Thailand. Such studies should incorporate vector surveillance and competence testing, alongside expanded human surveillance. Although Culicoides paraensis has not been reported in Thailand, various other Culicoides species have been documented [14,15]. Given the high rate of tourism in Thailand and the potential for importation of OROV, further research should also determine the competence of local Thai vector species for OROV transmission. At present, no entomological studies in Thailand have specifically investigated OROV vectors. Nonetheless, such investigations are critical for preparedness and should be prioritized in future research. This work was supported by the Center of Excellence in Clinical Virology of the Faculty of Medicine of Chulalongkorn University and Hospital and Armed Forces Health Surveillance Division-Global Emerging Infections Surveillance Program (AFHSD-GEIS). SK is supported by the Second Century Fund (C2F) of Chulalongkorn University.
## Data availability
Data is provided within the manuscript.
## References
1. Sippy, Farrell, Lichtenstein et al. (2020) "Severity index for suspected arbovirus (SISA): machine learning for accurate prediction of hospitalization in subjects suspected of arboviral infection" *PLoS Negl Trop Dis*
2. Moreira, Bressan, Brasil et al. (2018) "Epidemiology of acute febrile illness in Latin America" *Clin Microbiol Infect*
3. Khongwichit, Chuchaona, Vongpunsawad et al. (2018) "Molecular surveillance of arboviruses circulation and co-infection during a large Chikungunya virus outbreak in Thailand" *Sci Rep*
4. Adams, Lefkowitz, King et al. (2017) "Changes to taxonomy and the International Code of Virus Classification and Nomenclature ratified by the International Committee on Taxonomy of Viruses"
5. Zhang, Liu, Wu et al. (2024) "Oropouche virus: A neglected global arboviral threat" *Virus Res*
6. Scachetti, Forato, Claro et al. (2025) "Re-emergence of Oropouche virus between 2023 and 2024 in Brazil: an observational epidemiological study" *Lancet Infect Dis*
8. (2024) "Increased oropouche virus activity and associated risk to travelers 2024"
9. Naveca, Nascimento, Souza et al. (2017) "Vasconcelos P. Multiplexed reverse transcription real-time polymerase chain reaction for simultaneous detection of Mayaro, Oropouche, and Oropouche-like viruses" *Mem Inst Oswaldo Cruz*
10. Endy (0323) "Hunter's Tropical Medicine and Emerging Infectious Diseases"
11. Thongsripong, Green, Kittayapong et al. (2013) "Mosquito vector diversity across habitats in central Thailand endemic for dengue and other Arthropod-Borne diseases" *PLoS Negl Trop Dis*
12. Rattanakomol, Khongwichit, Chuchaona et al. (2023) "Severe fever with thrombocytopenia syndrome virus genotype B in Thailand" *Arch Virol*
13. Salvato (2025) "Re-emergence of oropouche virus as a novel global threat" *Curr Res Microb Sci*
14. Zé-Zé, Laranjinha, Borges et al. (2025) "Dengue and Oropouche virus co-infection in a traveller from Cuba to Portugal" *J Travel Med*
15. Thepparat, Bellis, Ketavan et al. (2015) "Ten species of Culicoides Latreille (Diptera: Ceratopogonidae) newly recorded from Thailand" *Zootaxa*
16. Kamyingkird, Choocherd, Chimnoi et al. (2023) "Molecular identification of Culicoides species and host preference blood meal in the African horse sickness outbreak-affected area in Hua Hin District" *Insects*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792536&blobtype=pdf
|
# P-1815. Epidemiological Patterns of Dengue Fever in Children aged 0-14 years in Kathmandu, Nepal: A Seven-Year Analysis in a High-Altitude Setting (2018-2024) Saugat Bhandari, MBBS 1 ; Meeru Gurung, Masters in International Health 1 ; Sanjeev Man Bijukchhe, PHD 1 ; Bhishma Pokhrel, n/a 1 ; Puja Amatya, Fellowship in Pediatric Critical Care 1 ; Anil Raj Ojha, Developmental Pediatrician Fellowship in Development & Behavior Pediatrics 1
Ruby Basi, Aayush Rizal, Rusy Shrestha, Shrijana Shrestha
Background. Dengue fever is typically prevalent in low-lying, tropical regions. However, in recent years, cases have been observed in higher-altitude areas as well. The Kathmandu Valley, located at an elevation of approximately 1,400 meters above sea level, was previously considered a low-risk region for dengue. Despite this, an increasing number of paediatric cases have been documented in clinical settings. This study examines the trend of laboratory-confirmed non-structural protein 1 (NS1) positive dengue cases in children presenting to outpatient and emergency departments to a tertiary healthcare centre in Kathmandu over the past seven years.
Methods. A retrospective review was conducted of all laboratory-confirmed NS1 positive dengue cases in children presenting to outpatient and emergency departments to Patan Hospital, a tertiary healthcare facility in Nepal, from 2018 to 2024. Data were collected on the number of cases per year and the gender distribution of patients to analyze trends and changes over time. A significant rise in cases was observed from 2019 onwards, coinciding with a major dengue outbreak in the Kathmandu Valley.
Results. Between 2018 and 2024, a total of 857 paediatric dengue cases were recorded. Although no paediatric cases were reported cases in 2018, the number of cases began increasing in 2019 (33 cases) and surged sharply to a peak in 2022 (646 cases). Another noticeable increase was observed in early 2024 (150 cases). Male children were affected more than female children, with 528 cases among boys compared to 329 among girls. The majority of cases occurred during and after the monsoon season.
Conclusion. Paediatric dengue cases have exhibited a rising trend in Kathmandu. Although the region was previously considered at low risk, the emerging threat now necessitates enhanced clinical vigilance and preparedness. Addressing the interplay between climate change and disease transmission is essential to mitigating future outbreaks and protecting paediatric populations in high-altitude settings.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792914&blobtype=pdf
|
Marissa Nicolas-Cagigal, Jonathan Pavia, Patrik Hornak
Background. West Nile Virus (WNV) is the most prevalent mosquito-borne illness in the United States, with 20-30% of infected individuals experiencing symptomatic illness. Previous studies have suggested that temperature and precipitation strongly influence mosquito survival and therefore WNV transmission. Since 1999 the CDC has tracked U.S. cases, of particular significance is the state of Texas, which has consistently reported some of the highest case numbers. As weather patterns inevitably change it is crucial to identify environmental conditions that promote WNV transmission to inform prevention efforts. Results. In our analysis between the years 2002-2024, we found a significant relationship between the frequency of WNV incidence and the average summer temperatures (p = 0.038). Specifically we found for every 1°F increase in average summer temperature, the incidence of WNV decreases by ∼44.11 cases, holding all else constant. Per our analysis ∼20% of variability in WNV incidence is explained by average summer temperature alone (R 2 = 0.198) with a 95% confidence interval of [-85.58, -2.65].
## Average
Conclusion. Arthropod-borne diseases are inevitably influenced by weather patterns. Understanding the extent of this influence is of utmost importance, as it can inform effective prevention strategies. Our analysis of patterns in Texas found a statistically significant relationship between average summer temperatures and incidence of WNV. Specifically, as summer temperatures rise, the incidence of WNV decreases. The significance in relationships investigated in this study are likely underestimated due to other variables affecting transmission as well as the clinical presentation of WNV.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12617374&blobtype=pdf
|
# Genome sequence of Mycobacterium smegmatis phage Mao1
Mackenzie Kluttz, Catherine Griffin, Nathan Simpson, Lydia Suplita, Alison Kanak
## Abstract
Mao1, a temperate bacteriophage isolated in North Georgia, was found to have a 65,240 base pair genome with 101 confirmed genes, no tRNA genes, and a 66.3% guanine-cytosine content. It shares 99.79% nucleotide identity with phage Sejanus. Bacteriophages that share over 50% identities are grouped into clusters, with Mao1 being in cluster AD.
KEYWORDS bacteriophages, genomesB acteriophages can potentially fight antibiotic drug resistance by using phage therapy (1). To do this, phage development and diversity must be researched. This aims to contribute to the knowledge of phage by presenting bacteriophage Mao1, a temperate phage in cluster AD that infects Mycobacteria smegmatis mc 2 155.Mao1 was isolated from enriched soil at the University of North Georgia in Dahlo nega, Georgia (34.527776 N, 83.986272 E) in dry, sandy soil. The phage was isolated using protocols provided by the Science Education Alliance-Phage Hunting Advancing Genomics and Science (2). Briefly, 7H10 liquid medium was added to the soil sample, incubated at 37°C for 24 h, and filtered using a 0.22 µm filter. This filtrate was incubated with Mycobacteria smegmatis mc 2 155 at 37°C and refiltered. Once plaques were detected via traditional plaque assay, three rounds of purification were performed. The titer was amplified to extract the genomic DNA for sequencing. Electron microscopy using phosphotungstic acid stain imaged Mao1. Mao1 was identified as having a siphovirus morphology with a capsid and tail measuring approximately 67.603 and 278.974 nm, respectively (Fig. 1A). Mao1's plaques are approximately 3 mm with distinct, circular edges (3) (Fig. 1B).The phage genomic DNA was isolated from lysate and created with a web plate using the Wizard DNA Extraction Kit (Promega) per manufacturer instructions. An NEB Ultra II Library Kit 9 with v3 Reagents and 150-base reads was used to synthesize a library. Mao1 was found to consist of 66.3% guanine-cytosine and 65,240 base pairs long and circularly permuted ends and was assigned with cluster AD. There were 325,500 singleend reads from the library. These raw reads were assembled using Newbler v2.9 with default settings. The resulting single-phage contig was checked for completeness, accuracy, and phage genomic termini using Consed v29 (4).Mao1's genome was annotated using Glimmer v3.02 (5), GeneMark v2.5p (6), DNA Master v4.2.1.11, Phamerator v557 (7), Starterator v557, NCBI BLASTp v2.15.0 (8), HHPred v2.08 (9), TMHMM v.1.0.24 (9), and PECAAN v20240320 (10). Software used was run with default parameters. Hits with E values of 10 e-10 or less were considered acceptable. Phamerator and GeneMark indicated that Mao1 has 101 open reading frames (ORFs) with the ability to assign function to 36. Genes 1-38, 49-90, 93-95, and 97-99 read in the forward direction, while genes 39-48, 91-92, 96, and 100-101 read in the reverse direction. Mao1 is predicted to be temperate, as a tyrosine integrase (ORF43) was identified. Mao1 is mostly like Sejanus (GenBank accession no. OP172873), with 99.79% nucleotide identity via BLAST nucleotide alignment. Of interest, there is one base pair insertion found within ORF3, causing a frameshift resulting in a single long ORF rather
than the two observed in other cluster members. The latter half of Mao1's ORF 3 aligns 1:1 with Sejanus' ORF 4 and the front half aligns almost 1:1 with Sejanus ORF 3 via BLASTp. Mao1's genome has synteny with its cluster members in the front half of the genome, while the latter half loses most of its synteny, demonstrating mosaicism commonly found in phage genomes (11).
## References
1. Petrovic Fabijan, Iredell, Danis-Wlodarczyk et al. (2023) "Translating phage therapy into the clinic: recent accomplishments but continuing challenges" *PLoS Biol*
2. Poxleitner, Pope, Jacobs-Sera et al. (2018) "HHMI SEA-PHAGES phage discovery guide"
3. Kluttz (2023) "Mycobacterium phage Mao1. The actinobacteriophage database"
4. Da ; Clokie, Kropinski, Lavigne (2018) "Sequencing, assembling, and finishing complete bacteriophage genomes"
5. Delcher, Bratke, Powers et al. (2007) "Identifying bacterial genes and endosymbiont DNA with Glimmer" *Bioinformatics*
6. Besemer, Borodovsky (2005) "GeneMark: web software for gene finding in prokaryotes, eukaryotes and viruses" *Nucleic Acids Res*
7. Cresawn, Bogel, Day et al. (2011) "Phamerator: a bioinformatic tool for comparative bacteriophage genomics" *BMC Bioinformatics*
8. Altschulsf, Miller, Myersew et al. (1990) "Basic local alignment search tool" *J Mol Biol*
9. Söding, Biegert, Lupas (2005) "The HHpred interactive server for protein homology detection and structure prediction" *Nucleic Acids Res*
10. Hallgren, Tsirigos, Pedersen et al. (2022) "DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks" *BioRxiv*
11. Hatfull, Hendrix (2011) "Bacteriophages and their genomes" *Curr Opin Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12570490&blobtype=pdf
|
# mSphere of Influence: The perfect slice-from pizza to proteases
Nicholas Lennemann
## Abstract
Nick Lennemann studies the intracellular interactions of viral and host proteins. In this mSphere of Influence article, he discusses his employment history and the mentors that promoted his training and transition to an independent research program focused on proteolytic determinants of virus infection. He highlights how "A novel interaction between dengue virus nonstructural protein 1 and the NS4A-2K-4B precursor is required for viral RNA replication but not for formation of the membra nous replication organelle" by A.
to pursue research opportunities that would promote a strong independent program merging my graduate training in molecular virology with new concepts and techniques in cell biology. During this time, there were two publications that influenced my research, which combined the techniques and interests I developed throughout my training from undergraduate to postdoctoral research.
In 2019, Płaszczyca et al. identified an essential intermediate of the orthoflavivirus polyprotein for viral replication (1). This study utilized innovative reverse genetics to perform a forward genetic screen that identified a novel viral polyprotein intermediate required for infection. The mechanistic experiments of this study identified that the interaction between the secreted nonstructural protein 1 (NS1) interacts in the lumenal space of the endoplasmic reticulum (ER) with a multi-transmembrane proteolytic intermediate of the polyprotein (NS4A-2K-4B). This study used molecular virology and microscopy techniques to show that the interactions between NS1 and the NS4A-2K-4B precursor, not the individual proteolytic products, are required for viral genome replication, but not the formation of the "peas in a pod" membranous replication structures, which are indicative of DENV infection. Their data piqued my interest in the complexities of viral polyproteins and the role of viral proteases in the regulation of replication and manipulation of the host cell. The "simplest" of positive-strand RNA virus genomes are translated as single polyproteins that are proteolytically cleaved into individual functional protein subunits by virus-encoded proteases and, in certain instances (e.g., flaviviruses), host proteases. Some of these viruses have well-defined proteolytic intermediates with differential functions of the individual protein compo nents (e.g., picornaviruses). Thus, viruses have evolved to encode complex mechanisms within a minimal genome that dictate the efficacy of polyprotein processing in order to efficiently perform the colossal number of tasks required for replication. While these intermediates are defined for picornaviruses, the study by Płaszczyca et al. ( 1) identified a functional precursor in orthoflaviviruses. This study provokes investigation into the presence of other functional orthoflavivirus polyprotein intermediates and the potential importance of proteostasis in other poorly investigated positive-strand RNA viruses. It also raises questions regarding the proteolytic program of polyprotein processing. What regulates the differential cleavage that allows for the production of both specific intermediates and individual protein products?
In 2018, Ding et al. expanded on previous studies that identified the ER-localized stimulator of interferon genes (STING) host protein as a restriction factor for orthoflavivirus replication (2)(3)(4). In this publication, they identified that STING is cleaved during Zika virus infection to subvert the antiviral activity. This was further confirmed to be mediated by multiple orthoflavivirus protease complexes, which highlighted several intriguing results. First, this study showed that orthoflaviviral proteolytic cleavage of STING occurs at a non-consensus cleavage site (Y-R|G), rather than the dibasic-glycine/ser ine consensus motif present throughout the viral polyprotein. To my knowledge, this was the first example of the intracellular substrate flexibility of orthoflavivirus protea ses. Thus, the study by Ding et al. indicated that orthoflavivirus proteases can cleave non-consensus sequences in antiviral host proteins during infection. What other host proteins are targeted for cleavage to promote infection? What is the degree of inherent substrate flexibility of orthoflavivirus proteases during infection? Interestingly, this study also showed that the yellow fever virus protease, which targets the same consensus motif in the polyprotein, does not cleave STING upon exogenous expression. Thus, there must be additional orthoflavivirus-specific determinants for protease cleavage that have yet to be defined.
Collectively, these studies have influenced a major focus of my research on viral proteases encoded by orthoflaviviruses, enteroviruses, and astroviruses. I am fortunate to have built a career in academic science with the opportunity to train the next generation of energetic scientists, which will be among the most resilient and perseverant scientists in modern history. Thus, it is imperative that we continue to support and encourage their progress by contributing to innovative science that can inspire the foundations for their independent careers-hopefully to be discussed in an mSphere of Influence article during the early years of their careers.
## References
1. Płaszczyca, Scaturro, Neufeldt et al. (2019) "A novel interaction between dengue virus nonstructural protein 1 and the NS4A-2K-4B precursor is required for viral RNA replication but not for formation of the membranous replication organelle" *PLoS Pathog*
2. Ding, Gaska, Douam et al. (2018) "Species-specific disruption of STING-dependent antiviral cellular defenses by the zika virus NS2B3 protease" *Proc Natl Acad Sci*
3. Yu, Chang, Liang et al. (2012) "Dengue virus targets the adaptor protein MITA to subvert host innate immunity" *PLoS Pathog*
4. Aguirre, Maestre, Pagni et al. (2012) "DENV inhibits type I IFN production in infected cells by cleaving human STING" *PLoS Pathog*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12691673&blobtype=pdf
|
# Structural insight into RNA encapsidation by the severe fever with thrombocytopenia syndrome virus nucleocapsid protein
Yong Wang, Hao Wu, Jiawen Sun, Wenhua Kuang, Hualin Wang, Yun-Jia Ning, Zengqin Deng
## Abstract
Severe fever with thrombocytopenia syndrome virus (SFTSV), an emerging highly pathogenic bunyavirus, poses a significant public health threat with a case fatality rate of up to 30%. The viral nucleocapsid protein (NP) encapsidates the genomic RNA to form a ribonucleoprotein (RNP) complex, which is critical for transcription and replica tion. However, the molecular mechanism underlying SFTSV RNA encapsidation remains poorly understood, largely due to the lack of structural information on the NP-RNA complex. Here, we report a cryo-electron microscopy structure of the SFTSV NP in complex with single-stranded RNA. The structure reveals a pentameric NP assembly that sequesters RNA along the inner surface of the oligomeric ring in a sequence-independ ent manner. Strikingly, all the RNA bases face the protein, rendering them inaccessible for transcription and replication. Each NP subunit accommodates four nucleotides within an evolutionarily conserved hydrophobic cleft, with an additional two to three nucleo tides bound at the inter-subunit interface. The functional importance of the NP-RNA interactions is further corroborated by a minigenome-based assay. This work provides structural insight into RNA encapsidation by SFTSV NP and offers a foundation for the rational design of antiviral therapeutics targeting this essential viral protein.
IMPORTANCESevere fever with thrombocytopenia syndrome virus (SFTSV) is a highly pathogenic bunyavirus that causes severe hemorrhagic fever, leukopenia, thrombocy topenia, and multi-organ failure, with a case fatality rate of up to 30%. No licensed vaccines or specific antiviral therapies are currently available. The viral nucleocapsid protein (NP) is essential for viral transcription and replication, forming a ribonucleopro tein complex (RNP) by encapsidating viral genomic RNA. However, the structural basis of RNA recognition and encapsidation by SFTSV NP remains poorly understood. In this study, we determined a cryo-electron microscopy structure of the SFTSV NP-RNA complex. Structural comparisons and evolutionary conservation analysis of NPs across the family Phenuiviridae uncovered a conserved RNA-binding mode among phenuivi ruses, suggesting a shared RNA encapsidation mechanism among related viruses. Our findings provide critical structural insights into SFTSV RNA encapsidation and will aid future efforts to develop antivirals against SFTSV and related pathogenic viruses. KEYWORDS SFTSV, RNA encapsidation, Cryo-EM, nucleocapsid protein, drug targets T he class Bunyaviricetes is a large group of segmented negative-strand RNA viruses, encompassing numerous pathogenic viruses. Among them, the family Phenuiviridae includes two highly pathogenic human pathogens: severe fever with thrombocytopenia syndrome virus (SFTSV) and Rift Valley fever virus (RVFV) (1, 2). Currently, there are no licensed human vaccines or specific antiviral treatments available against SFTSV and RVFV, underscoring the urgent need for further research to develop effective countermeasures against these two highly pathogenic bunyaviruses. SFTSV, an emerging tick-borne virus, has rapidly become endemic in several East Asian countries, including
South Korea, Japan, China, Vietnam, Pakistan, and Thailand, since its first identification in China in 2009 (3). Infection with SFTSV leads to severe hemorrhagic fever, leukope nia, thrombocytopenia, and multi-organ failure, with a case fatality rate of up to 30%. Although SFTSV infection remains primarily confined to Asia, its potential for global spread is a growing concern. The Asian longhorned tick (Haemaphysalis longicornis), the primary vector for SFTSV (4), has already been detected in the United States, Russia, Australia, and the Western Pacific (5)(6)(7), raising alarms about SFTSV's possible emergence beyond its current endemic regions.
Like other bunyaviruses, SFTSV contains three single-stranded RNA genome segments designated as large (L), medium (M), and small (S). The L segment encodes an RNA-dependent RNA polymerase, also referred to as the L protein, that is responsible for viral genome transcription and replication. The M segment encodes the glycoproteins Gn and Gc, which form heterodimers covering the viral surface and facilitate cellular attachment, entry, and fusion. The S segment encodes the nucleocapsid protein (NP) that encapsidates the viral genomic RNA to form the ribonucleoprotein (RNP) complex. The S segment also encodes an important virulence factor, nonstructural protein (NSs), via an ambisense coding strategy. RNP plays critical roles in protecting the viral RNA from host RNase degradation and serving as functional templates of viral polymerase to synthesize viral mRNA and genome RNA.
Crystallographic studies revealed that the wild-type SFTSV NP forms pentameric and hexameric rings, while a SFTSV NP mutant forms a tetramer, mediated by the flexible N-terminal arm (8,9). The C-terminal core harbors a putative RNA-binding cleft on the inner surface of the oligomeric ring. Notably, the inhibitor suramin binds to this putative RNA-binding site and suppresses viral replication (8), underscoring the potential for developing therapeutic strategies targeting this viral protein. RNA co-purified with recombinant SFTSV NP is resistant to degradation by external nuclease, further supporting NP's protective role (9). Despite these advances in the biochemical and structural characterization of SFTSV NP, key questions remain unresolved, including the detailed architecture of the NP-RNA complex, the specificity and stoichiometry of RNA binding, and the mechanistic basis of RNP assembly.
In this study, we determined the cryo-electron microscopy (cryo-EM) structure of the SFTSV NP-RNA complex. By combining structural analysis, structure-guided mutagenesis, and functional assays, we provide insights into RNA recognition and encapsidation, offering a foundation for rational antiviral design targeting SFTSV NP.
## RESULTS
## Recombinant SFTSV NP oligomerizes and binds cellular RNA
We successfully expressed the full-length SFTSV NP in Escherichia coli. During purification by size-exclusion chromatography (SEC), the protein was eluted as a major peak at a retention volume of ~16.9 mL, corresponding to a molecular weight exceeding 100 kDa. SDS-PAGE analysis confirmed that the peak fractions contain a highly pure protein with the expected size for SFTSV NP (27 kDa) (Fig. 1A). These results suggest that the recombinant SFTSV NP forms a higher-order oligomer in solution, consistent with previous studies (8,9). Moreover, the ratio of the optical density (OD) at 260 nm to the OD at 280 nm of the major protein peak is ~1.1, indicating that the purified SFTSV NP encapsidates nucleic acids. Denaturing urea PAGE after protease treatment resolved two major nucleic acid bands with molecular weight slightly larger than a 28-nt single-stranded RNA (ssRNA). These nucleic acids were completely degraded upon RNase A treatment but remained unaffected by DNase I treatment, indicating that the recombinant SFTSV NP predominantly binds RNA (Fig. 1B).
A previous negative-staining study reported heterogeneity in the oligomeric state of the recombinant SFTSV NP in solution, with assemblies ranging from tetramers to hexamers (9). To further elucidate its oligomerization, we performed cryo-EM analysis. From 550 micrographs, we selected 193,082 particles for 2D classification. After two rounds of reference-free 2D classification, 47,381 particles were retained for analysis.
These particles exhibited a predominant pentameric (20,776 particles, 43.8%) and hexameric (14,249 particles, 30.0%) organization, with minor populations of heptamers and trace amounts of tetramer and octamer (Fig. 1C; Fig. S1).
## Structure determination of the SFTSV NP-RNA complex
Initial attempts to solve the cryo-EM structure of the SFTSV NP-RNA complex were unsuccessful due to the extreme preferred orientation of the sample (Fig. S1). To address this issue, we treated the sample with methyl-PEG4-NHS ester to PEGylate the protein (10). The PEGylated sample showed diverse views after 2D classification, allowing for further 3D reconstruction (Fig. S2A andB). Despite five distinct oligomeric states of the recombinant SFTSV NP existing in solution, likely due to preferred orientation issues of the other states, we were only able to determine the cryo-EM structure of the pentameric assembly at an overall resolution of 3.98 Å (Fig. S2B). Notably, imposing C5 symmetry during data processing did not improve the resolution and density quality of the reconstruction, indicating intrinsic conformational flexibility among protomers within the NP pentamer. Local resolution analysis revealed that the NP-RNA interaction interfaces were better resolved (around 3 Å) than the solvent-exposed regions, likely due to their higher structural rigidity. Using the crystal structure of SFTSV NP as an initial model, we docked and refined the model against the cryo-EM density. The density corresponding to the co-purified RNA was clearly visualized, allowing unambiguous tracing of a 28-nucleotide single-stranded RNA bound to the NP pentamer (Fig. S2F). A poly(U) RNA sequence was arbitrarily modeled within the complex structure. The final refined model has good geometry and fits well into the electron density map (Table S1).
## Overall structure of the SFTSV NP-RNA complex
The pentameric SFTSV NP-RNA complex adopts a ring-like architecture with overall dimensions of 99 Å × 107 Å ×60 Å. Within this structure, the resolved 28-nt ssRNA is bound along the inner circumference of the ring (Fig. 2A andB). Each NP subunit consists of three distinct regions: the N-terminal arm region (N-arm), the middle N-lobe, and the C-terminal C-lobe (Fig. 2C). The middle N-lobe and C-terminal C-lobe together form the subunit core. While the subunit cores exhibit only minor conformational variations among protomers, the N-arm displays significant positional flexibility relative to the core, with an all-atom root-mean-square deviation (RMSD) ranging from 2.7 to 3.6 Å across different subunits (Fig. 2D). The N-arm inserts into the surface groove of the adjacent subunit, forming extensive arm-to-core interactions that mediate NP oligomerization. Each NP subunit contains a deep, narrow RNA-binding cleft that collectively forms a continuous groove along the inner surface of the pentameric ring. This groove features a hydrophobic interior lined with multiple positively charged residues at its rim. RNA binds with the bases inserted into the cleft and the sugar-phosphate backbone oriented toward the center of the ring (Fig. 3). Detailed analysis reveals two distinct RNA-binding stoichiometries per NP subunit: each can accommodate either 6 or 7 nucleotides (Fig. S3A). Specifically, four nucleotides are bound within the primary RNA-binding cleft, while an additional 2-3 nucleotides are located at the subunit interface (Fig. S3B andC). Thus, the number of nucleotides accommodated at the subunit interface determines whether an NP protomer binds 6 or 7 nucleotides in total. In the complex structure, we can unambiguously resolve the nucleotides at the subunit interface for four protomers, revealing an alternating pattern of 6 and 7 nucleotides around the pentamer. Structural comparison with the RNA-free SFTSV NP core structure demonstrates that RNA binding induces a conformational transition from a closed to an open state, involving slight rotations of both the N-lobe and C-lobe (Fig. S3D).
## Detailed interactions between SFTSV NP and RNA
The RNA segments (6 or 7 nt) bound to each NP subunit exhibit similar conformations, with the primary difference occurring at the 3′ terminus. In the 7-nt segment, three bases are stacked at this position, whereas only two are stacked in the 6-nt segment (Fig. S3B andC). Here, we focus on the detailed interactions observed in the 7-nt binding mode (Fig. 4A andB). In each NP subunit, the 5′-most base (B1) is stacked against residue L33. The B2 and B3 bases are stacked within the central compartment of the RNA-binding cleft, which is lined with the side chains of hydrophobic residues M147, F177, I181, F197, and P200. Base B4 inserts into a hydrophobic pocket formed by residues G65, V105, P127, M147, F177, and I181. The three 3′-most bases (B5-7) form a stacking interaction, with B7 stacking with residue Y30 from a neighboring subunit. In addition to hydrophobic and base-stacking interactions, NP forms a network of polar contacts with the RNA 5′ phosphates (P1 to P7) of all nucleotides except P6: residue Y30 from a neighboring subunit with P1; R106 and Q109 with P2; R95 with P3 and P4; K67 with P4; K70 with P5; K74 with P7. Furthermore, Q174 forms hydrogen bonds with RNA bases 2 and 3. Residues N182 and R186 form hydrogen bonds with the 2′-OH of nucleotide 3 and nucleotide 5, respectively. The majority of the interactions involve polar contacts with the RNA backbone and hydrophobic interactions with the nucleobase, suggesting a sequence-independent RNA-binding mode.
To corroborate our structural findings, we examined the function of NP mutants using a cell-based minigenome system. An M RNA analog containing the M untranslated regions (UTRs) and a negative-sense enhanced green fluorescent protein (eGFP) gene sequence was subcloned into a pRF42 vector under the control of pol I promoter to allow the generation of a viral genome-like RNA segment. After co-transfection with the NP and L genes under the control of the CAG promoter into cells, NP recognizes the RNA UTRs and assembles with the minigenome into functional RNPs, then replication and transcription by the L protein can occur, resulting in reporter gene expression. To disrupt RNA-binding interactions, polar residues implicated in RNA binding were substituted with alanine. Mutating most basic residues (lysine and arginine), as well as residue Y30, abolished the eGFP signal. Interestingly, alanine substitution of residues N182 and R186, which recognize 2′-OH group of RNA, also completely abolished the expression of eGFP. By contrast, mutating residue Q174 had only a moderate effect. Both residues R106 and Q109 interact with the 5′ phosphate of nucleotide 5. While a single mutation of either residue had a limited impact, the double mutation significantly reduced eGFP expression (Fig. 4C and Fig. S4). These results underscore the functional significance of the identified RNA-interacting residues of SFTSV NP.
## Structural convergence and divergence of bunyavirus NP-RNA complexes
Structures of the NP-RNA complexes have been reported for several bunyaviruses, including RVFV, Toscana virus, La Crosse Virus (LACV), Leanyer virus (LEAV), Schmallen berg virus (SBV), Bunyamwera viruses (BUNV), Lassa fever virus (LASV), Hantaan virus (HTNV), and Tomato spotted wilt virus (TSMV) (11)(12)(13)(14)(15)(16)(17)(18)(19). All of these viral NPs form oligomers in complex with ssRNA, except for LASV NP. The LASV NP alone forms a trimer with N-and C-terminal domains arranged in a head-to-tail manner (20), while the N-terminal domain of the LASV NP forms a complex with ssRNA as a monomer. The NP-RNA complexes of HTNV and TSMV bind only short RNA fragments with their sugar-phosphate backbones oriented toward a positively charged cleft. By contrast, RVFV, Toscana virus, and SFTSV, all members of the family Phenuiviridae, exhibit an RNA-binding mode where RNA bases are sequestered in a deep hydrophobic groove in a sequence-independent manner (Fig. 5). Structural superposition of the NP subunit of SFTSV NP-RNA with those of RVFV and Toscana virus highlights their high struc tural similarity, yielding RMSDs of 1.3 Å over 208 and 199 aligned Cα atoms per NP subunit, respectively (Fig. S5A). Additionally, evolutionary conservation analysis of 154 phenuiviral NPs using the consurf server indicates that the RNA-binding cleft is the most conserved region, formed by highly conserved residues (21) (Fig. S5B). These findings suggest that NPs within the Phenuiviridae family may share a highly conserved RNA-bind ing mode.
LACV, BUNV, LEAV, and SBV, all belonging to the Orthobunyavirus genus, adopt a similar mode for RNA sequestration. For example, each LACV NP subunit binds 11 nucleotides, with U3 and U8-11 positioned at the base of the RNA-binding cleft, facing inward, while the remaining bases remain solvent-exposed (Fig. 5). In contrast, LASV (a member of Arenaviridae) exhibits a distinct RNA-binding mode. Each LASV NP binds six nucleotides with the sugar-phosphate backbone directed into an RNA-binding pocket. Five of six bases are oriented outward and fully exposed to the solvent, a striking difference compared to SFTSV and LACV (Fig. 5). These findings underscore that viruses within the same genus or family might employ a conserved structural fold and man ner for genome encapsidation, whereas NPs from different genera or families exhibit significant divergence in RNA-binding mechanisms.
## DISCUSSION
The SFTSV NP-RNA complex structure, determined in this study, and the RVFV and Toscana virus NP-RNA complexes, along with the evolutionary conservation analysis of NPs across the family Phenuiviridae, potentially provide a clue to the mechanism of viral genome packaging. In each NP subunit, four RNA bases are deeply embedded within a conserved hydrophobic cleft, while an additional two to three bases at the subunit interface remain protein-facing. This extensive sequestration sterically prevents RNA bases from Watson-Crick pairing, rendering the encapsidated RNA inaccessible for transcription or replication unless actively remodeled by the viral polymerase or unidentified factors. The remarkable stability of these NP-RNA interactions further explains why obtaining RNA-free NPs of SFTSV, RVFV, and Toscana virus requires either denaturing conditions or extensive nuclease digestion (8,9,19,22).
The genomic RNA of bunyaviruses is encapsidated by NP to form RNPs, which not only facilitate viral RNA replication and transcription but also protect the viral genome from degradation by host RNases. In addition to these essential roles, NPs from certain bunyaviruses contribute to suppression of the host innate immune response and mediate RNP packaging into virus-like particles through interactions with the viral envelope glycoproteins (23)(24)(25). Despite the conserved functional role, NP structures exhibit remarkable diversity across different bunyavirus families. Structural studies have revealed that the recombinant NP-RNA complexes from various bunyaviruses typically form oligomers in vitro. However, whether these oligomeric states represent the functional RNP organization in authentic virions remains unclear. RNPs of RVFV and another phenuivirus, Uukuniemi virus, isolated from infected cells display a string-like morphology that lacks higher-order symmetry (22,26). The measured width of these authentic virus RNPs suggests that monomeric NP-RNA units, rather than the resolved NP or NP-RNA oligomers, serve as the fundamental building blocks (11,27). Given the similar NP-NP and highly conserved NP-RNA interactions, we hypothesize that SFTSV assembles its RNP through a mechanism analogous to that of RVFV.
In the cryo-EM map, an unmodeled density connects the 5′ and 3′ ends of the RNA, likely corresponding to unmodeled terminal regions (Fig. S6A). If these termi nal sequences are complementary, they could form an RNA duplex, a scenario well documented in bunyaviruses. The 5′ and 3′ termini of each genomic segment are highly conserved within a given bunyavirus species and exhibit strong complementarity, enabling the formation of a duplex stem (28). This interaction allows each segment to potentially adopt a circularized panhandle structure. We propose a model in which SFTSV RNPs adopt a circular organization facilitated by both RNA panhandles and NP-NP interactions (Fig. S6B). The viral RNA polymerase recognizes the panhandle structure to initiate RNA synthesis. During elongation, the genomic RNA, which is encapsidated by the NPs near the replication center, becomes transiently uncoated, making it accessible to the RNA polymerase. Meanwhile, the NPs remain held in position by the flexible N-arm. After the RNA polymerase passes through, the temporarily dislodged genomic RNA is re-encapsidated by the NPs.
The evolutionarily conserved hydrophobic RNA-binding cleft in the NP represents a promising drug target for developing broad-spectrum therapeutics against medically important phenuiviruses. Indeed, the inhibitor suramin binds within this conserved groove with micromolar affinity and effectively suppresses SFTSV replication (Fig. S7). Notably, suramin also exhibits comparable binding affinity to NPs from other Phenuivir idae members, including RVFV, Buenaventura virus, and Granada virus, but shows no detectable binding to NPs from orthobunyavirus, nairovirus, or hantavirus (8). Given the striking structural conservation of this RNA-binding pocket in phenuiviruses, further investigation is warranted to determine whether suramin exhibits antiviral activity against additional phenuiviruses by targeting NPs.
In summary, our SFTSV NP-RNA complex structure reveals a conserved RNA seques tration mechanism within the family Phenuiviridae, advancing our understanding of viral RNA encapsidation. These findings provide a foundation for developing antivirals against SFTSV, RVFV, and related pathogenic viruses.
## MATERIALS AND METHODS
## Protein expression and purification
DNA sequence encoding SFTSV NP (MN510206.2) was cloned into a pET28a vector containing a C-terminal His tag and transformed into E. coli BL21(DE3) cells. Protein expression was induced by the addition of isopropyl β-D-thiogalactopyranoside to a final concentration of 0.4 mM at 16°C when the OD 600 reached 0.6-0.8, and the cells were cultured for another 16 h. Cells were harvested via centrifugation at 5,000 × g for 10 min at 4°C. The collected cells were then resuspended in lysis buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl) with protease inhibitor (1 mM PMSF) and lysed using a high-pressure cell crusher (Union-Biotech) at 600 bar. The resulting lysate was centrifuged at 30,000 × g for 60 min at 4°C. The supernatant was collected, loaded onto Ni-charged Resin FF (GenScript), and washed with 50 mL of wash buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 100 mM imidazole). The target protein was eluted with the elution buffer containing 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 500 mM imidazole. Subsequent purification was performed using a Superose 6 Increase 10/300 Gl SEC column (Cytiva) pre-equilibrated with buffer containing 20 mM Tris-HCl, pH 8.0, and 150 mM NaCl. Peak fractions containing the SFSTV NP were pooled and concentrated to ~10 mg/mL and stored at -80°C for future use.
## Denaturing polyacrylamide gel electrophoresis
To identify the type of nucleic acids co-purified with SFTSV NP, 20 µL of SFTSV NP (10 mg/mL) was treated with proteinase K at a final concentration of 0.1 mg/mL and incubated at 37°C for 30 min. The mixture was then divided equally into two aliquots: one was treated with DNase I (20 µg/mL), and the other with RNase A (20 µg/mL), followed by incubation at 37°C for another 30 min. Loading buffer (50% glycerol) was added to each sample, which was then denatured at 95°C for 5 min. The samples were resolved on a 12% (wt/vol) polyacrylamide gel containing 7 M urea and stained with Stains-all. Two ssRNA markers were included as references (28-nt ssRNA: CUGUGCUCU UUUUUUCACAGUUUUUGAU, 59-nt ssRNA: GCUUUAUCAGAAGCCAGACAUUAACGCUUCU GGAGAAACUCAACGAGCUGGACGCGGAU). The gel was visualized using a gel imaging system, ChemiDoc Imaging System (Biorad).
## PEGylation of SFTSV NP
SFTSV NP was diluted to 1-2 mg/mL in 20 mM HEPES-NaOH, pH 8.0, and 150 mM NaCl, then incubated with methyl-PEG4-NHS ester (PEG4) at a final concentration of 4 mM for 30 min at 37°C. The reaction was quenched by adding 1/10 (vol/vol) 1 M Tris-HCl, pH 8.0, and the proteins were further purified using a Superose 6 Increase 10/300 Gl column (Cytiva) to remove excess PEG4. Peak fractions containing the PEGylated SFTSV NP were pooled and concentrated to ~5 mg/mL and stored at -80°C for future use.
## Cryo-EM sample preparation
Purified protein (3.5 µL) at a concentration of 0.6 mg/mL (SFTSV NP) or 2.4 mg/mL (PEGylated SFTSV NP) was applied to the glow-discharged Cu 200 mesh R1.2/1.3 holey carbon grids (Quantifoil). After a 20 s incubation, the grids were blotted for 2 s with a blot force of 0 at 100% humidity and 4°C, then plunge-frozen using Vitrobot Mark IV (FEI, Thermo Fisher Scientific).
## Cryo-EM data collection and image processing
Cryo-EM data were collected with a CRYO ARM 300 electron microscope (JEOL, Japan) operating at 300 kV, with a K3 direct electron detector (Gatan, United States). Data were collected at a nominal magnification of 50,000× in a super-resolution counting mode, with a pixel size of 0.475 Å/pixel. Movies were automatically collected using Serial-EM software (29) at a frame rate of 40 frames per second, accumulating a total dose of 40 e/Å 2 within a defocus range of -0.5 to -2.5 µm. Patch-based motion correction and CTF estimation of the recorded movies were performed in cryoSPARC v4.2.0 (30). Particles were automatically picked by Topaz picking from 4850 micrographs and extracted with a particle box size of 240 pixels and subsequently subjected to 2D classification. In all, 788,291 particles from good classes were selected for further heterogeneous refinement with six initial ab initio models generated by ab initio reconstruction in cryoSPARC. One good class was selected for subsequent 3D classification, requesting three classes. One class containing 63,930 particles was selected for further NU-refinement, yielding a map with an overall resolution of 3.98 Å.
## Model building and refinement
Crystal structure of the SFTSV NP in RNA-free (PDB: 4J4U) was placed into the cryo-EM density map using UCSF Chimera. The RNA was manually built as poly(U) to fit the densities. Cycles of model building in COOT (31) and refinement using real_space_refine in Phenix (32) were performed to obtain the final refined model. The quality of the final models was analyzed with MolProbity in Phenix (33). Refinement statistics are summarized in Table S1.
## SFTSV minigenome reporter assay
SFTSV minigenome reporter assay was conducted as previously described (34,35). Briefly, BHK21 cells seeded in 96-well plates were co-transfected with a PolI-MUTR-eGFP transcription plasmid producing the M RNA analog containing M UTRs and a negativesense eGFP gene sequence, together with the SFTSV L protein expression plasmid and the plasmids encoding SFTSV wild-type NP (as positive control) or one of the NP mutants, or the empty vector (as negative control), using Lipofectamine 3000 transfection reagent (Invitrogen, Cat#L3000015). At 48 h post-transfection, cells were fixed with 4% parafor maldehyde, permeabilized with 0.5% Triton X-100, and blocked with 5% BSA in PBS. To visualize the expression of wild-type NP or mutants, cells were incubated with an anti-SFTSV NP polyclonal antibody at 4℃ overnight, followed by staining with Alexa Fluor 647-conjugated anti-rabbit IgG antibody (Abcam, Cat#ab150077) for 1 h at room temperature. To visualize the nuclei, cells were stained with Hoechst33258 (Beyotime, Cat#C1011) for 15 min at room temperature. Imaging and quantification of eGFP-positive and total cells per well were performed using the Operetta CLSTM high-throughput system (PerkinElmer). Relative minigenome reporter activity (%) = [(number of eGFP-pos itive cells of sample group)/(number of total cells of sample group)] ÷ [(number of eGFP-positive cells of positive control group)/(number of total cells of positive control group)] × 100.
## References
1. Casel, Park, Choi (2021) "Severe fever with thrombocytopenia syndrome virus: emerging novel phlebovirus and their control strategy" *Exp Mol Med*
2. Gibson, Noronha, Tubbs et al. (2023) "The increasing threat of Rift Valley fever virus globalization: strategic guidance for protection and preparation" *J Med Entomol*
3. Yu, Liang, Zhang et al. (2011) "Fever with thrombocytopenia associated with a novel bunyavirus in China" *N Engl J Med*
4. Zhang, Zhao, Cheng et al. (2022) "Rapid spread of severe fever with thrombocytopenia syndrome virus by parthenogenetic asian longhorned ticks" *Emerg Infect Dis*
5. Egizi, Bulaga-Seraphin, Alt et al. (2020) "First glimpse into the origin and spread of the Asian longhorned tick, Haemaphysalis longicornis, in the United States" *Zoonoses Public Health*
6. Heath (2020) "A history of the introduction, establishment, dispersal and management of Haemaphysalis longicornis Neumann, 1901 (Ixodida: Ixodidae) in New Zealand" *N Zealand J Zool*
7. Miao, Dai, Zhao et al. (2020) "Mapping the global potential transmission hotspots for severe fever with thrombocytopenia syndrome by machine learning methods. Research Article mBio December"
8. *Emerg Microbes Infect*
9. Jiao, Ouyang, Liang et al. (2013) "Structure of severe fever with thrombocytopenia syndrome virus nucleocapsid protein in complex with suramin reveals therapeutic potential" *J Virol*
10. Zhou, Sun, Wang et al. (2013) "The nucleoprotein of severe fever with thrombocy topenia syndrome virus processes a stable hexameric ring to facilitate RNA encapsidation" *Protein Cell*
11. Zhang, Shigematsu, Shimizu et al. (2021) "Improving particle quality in cryo-EM analysis using a PEGylation method" *Structure*
12. Raymond, Piper, Gerrard et al. (2012) "Phleboviruses encapsidate their genomes by sequestering RNA bases" *Proc Natl Acad Sci*
13. Reguera, Malet, Weber et al. (2013) "Structural basis for encapsidation of genomic RNA by La Crosse orthobunyavirus nucleoprotein" *Proc Natl Acad Sci*
14. Niu, Shaw, Wang et al. (2013) "Structure of the Leanyer orthobunyavirus nucleoprotein-RNA complex reveals unique architecture for RNA encapsidation" *Proc Natl Acad Sci*
15. Hastie, Liu, Li et al. (2011) "Crystal structure of the Lassa virus nucleopro tein-RNA complex reveals a gating mechanism for RNA binding" *Proc Natl Acad Sci*
16. Arragain, Reguera, Desfosses et al. (2019) "High resolution cryo-EM structure of the helical RNA-bound Hantaan virus nucleocapsid reveals its assembly mechanisms"
17. Komoda, Narita, Yamashita et al. (2017) "Asymmetric trimeric ring structure of the nucleocapsid protein of tospovirus" *J Virol*
18. Dong, Li, Böttcher et al. (2013) "Crystal structure of Schmallenberg orthobunyavirus nucleoprotein-RNA complex reveals a novel RNA sequestration mechanism" *RNA*
19. Li, Wang, Pan et al. (2013) "Bunyamwera virus possesses a distinct nucleocapsid protein to facilitate genome encapsidation" *Proc Natl Acad Sci*
20. Olal, Dick, Woods et al. (2014) "Structural insights into RNA encapsidation and helical assembly of the Toscana virus nucleoprotein" *Nucleic Acids Res*
21. Qi, Lan, Wang et al. (2010) "Cap binding and immune evasion revealed by Lassa nucleopro tein structure" *Nature*
22. Yariv, Yariv, Kessel et al. (2023) "Using evolutionary data to make sense of macromole cules with a "face-lifted" *ConSurf. Protein Sci*
24. Raymond, Piper, Gerrard et al. (2010) "Structure of the Rift Valley fever virus nucleocapsid protein reveals another architecture for RNA encapsidation" *Proc Natl Acad Sci*
25. Zhang, Yan, Chu et al. (2025) "Bunyavirus SFTSV nucleoprotein exploits TUFMmediated mitophagy to impair antiviral innate immunity" *Autophagy*
26. Jiang, Huang, Dong et al. (2013) "Structures of arenaviral nucleoproteins with triphosphate dsRNA reveal a unique mechanism of immune suppression" *J Biol Chem*
27. Överby, Pettersson, Neve (2007) "The glycoprotein cytoplasmic tail of Uukuniemi virus bunyaviridae interacts with ribonucleoproteins and is critical for genome packaging" *J Virol*
28. Pettersson, Bonsdorff (1975) "Ribonucleoproteins of Uukuniemi virus are circular" *J Virol*
29. Ferron, Li, Danek et al. (2011) "The hexamer structure of Rift Valley fever virus nucleoprotein suggests a mechanism for its assembly into ribonucleoprotein complexes" *PLoS Pathog*
30. Malet, Williams, Cusack et al. (2023) "The mechanism of genome replication and transcription in bunyaviruses" *PLoS Pathog*
31. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
32. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
33. Emsley, Lohkamp, Scott et al. (2010) "Features and development of Coot" *Acta Crystallogr D Biol Crystallogr*
34. Afonine, Poon, Read et al. (2018) "Real-space refinement in PHENIX for cryo-EM and crystallography" *Acta Crystallogr D Struct Biol*
35. Williams, Headd, Moriarty et al. (2018) "MolProbity: More and better reference data for improved all-atom structure validation" *Protein Sci*
36. Ren, Zhou, Deng et al. (2020) "Combinatorial minigenome systems for emerging banyangviruses reveal viral reassortment potential and importance of a protruding nucleotide in genome "Panhandle" for promoter activity and reassortment" *Front Microbiol*
37. Mo, Xu, Deng et al. (2020) "Host restriction of emerging high-pathogenic bunyaviruses via MOV10 by targeting viral nucleopro tein and blocking ribonucleoprotein assembly" *PLoS Pathog*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12451616&blobtype=pdf
|
# Erratum: Summary of taxonomy changes ratified by the International Committee on Taxonomy of Viruses (ICTV) from the Archaeal Viruses Subcommittee, 2025
Mart Krupovic, Diana Baquero, Eduardo Bignon, Ariane Bize, Guillaume Borrel, Mingwei Cai, Lanming Chen, Marion Coves, Changhai Duan, Simonetta Gribaldo, Eugene Koonin, Meng Li, Lirui Liu, Yang Liu, Ying Liu, Sofia Medvedeva, Yimin Ni, Apoorva Prabhu, Christian Rinke, Yongjie Wang, Tianqi Xu, Shuling Yan, Qinglu Zeng, Rui Zhang, Ictv Taxonomy, Summary Consortium, Yang Liu, Yang Liu, Ying Liu
of Innsbruck, Innsbruck, Austria; 10 Entwicklungsgenetik und Zellbiologie der Tiere, Philipps-Universität Marburg, Marburg, Germany; 11 Department of Ocean Science, The Hong Kong University of Science and Technology, Hong Kong, PR China.
The version of record of this article will be updated to reflect the correction to the author list.
The Microbiology Society apologises for any inconvenience caused.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12838443&blobtype=pdf
|
# Virome diversity and molecular characterization of two emerging RNA viruses in mosquito populations from Yantai, China
Meixi Ren, Yumei Liu, Yongqin Wang, Yingxin Tu, Yaqing Guo, Xiaodong Sun, Guoyu Niu, Yanyan Wang, Yanyan Wang, Yongqin Wang
## Abstract
Mosquito-borne viruses represent a major global public health threat, with transmission dynamics governed by climatic, ecological, and anthropogenic factors. Yantai City, Shandong Province, situated in a warm-temperate monsoon climate zone, shares geographical and ecological characteristics with regions where mosquito-borne viruses are endemic, creating potential for virus introduction. We used metagenomics to systematically analyze viral communities in mosquitoes from the Yantai region. We collected 8,111 mosquitoes representing four genera and six species, with Culex being predominant (89.8%). High-throughput sequencing revealed 11 viral species spanning 9 families, including Peribunyaviridae and Picornaviridae. Notably, Serbia mononega-like virus 1 and Biggievirus Mos11 represent the first reports from China, with quantita tive reverse transcription PCR revealing minimum infection rates of 0.34% and 0.68%, respectively. Phylogenetic analysis revealed close relationships to known viral strains, with several isolates potentially representing novel genera or species. Analysis revealed that Culex quinquefasciatus harbored the greatest viral diversity (five species), with significantly higher viral diversity in agricultural versus urban areas (P < 0.001). Several viruses demonstrated cross-species transmission potential, including Zhee mosquito virus, Zhejiang mosquito virus 3, and Culex tritaeniorhynchus rhabdovirus, all detected across multiple mosquito species. While most viruses appear mosquito-specific, several show close phylogenetic relationships to known pathogens, potentially posing public health risks warranting surveillance. This study addresses knowledge gaps regarding mosquito-borne viruses in the Bohai Rim region and provides a scientific foundation for regional viral surveillance and early warning systems. IMPORTANCE Mosquito-borne viruses are a significant global health threat, with the potential to cause widespread disease outbreaks. This study investigated the viral diversity within mosquito populations in Yantai, China, and characterized the molecular features of two emerging RNA viruses. These findings highlight the remarkable viral diversity harbored by Culex mosquitoes and reveal higher viral diversity in agricul tural areas compared to urban settings. Several identified viruses exhibit cross-spe cies transmission potential and close phylogenetic relationships to known pathogens, suggesting that they may pose public health risks. Understanding these interactions is essential for predicting how environmental changes may affect virus transmission and the resilience of surveillance and control strategies. KEYWORDS mosquito-borne viruses, Culex quinquefasciatus, novel viruses, phyloge netic analysis introduction M osquito-borne viruses constitute important pathogens transmitted primarily through mosquito bites that infect vertebrate hosts and cause disease. Mosquitoes
serve as major arthropod vectors, harboring and transmitting diverse viruses that comprise over half of all known vector-borne pathogens. Hundreds of mosquitoborne viruses have been identified globally, with approximately 100 demonstrating pathogenicity and posing significant threats to human and animal health. Mosquito species exhibit distinct vectorial capacities based on their ecological niches: Culex spp. serve as primary vectors for Japanese encephalitis virus (JEV) and West Nile virus (WNV); Aedes spp. primarily transmit dengue virus (DENV), Zika virus (ZIKV), chikungunya virus (CHIKV), and yellow fever virus, whereas Anopheles spp. serve as vectors for malarial parasites. Mosquito-borne viruses represent a major global public health threat, with transmission dynamics governed by climatic, ecological, and anthropogenic factors, including environmental factors such as temperature and rainfall (1)(2)(3)(4)(5).
Yantai City, Shandong Province (119°34′-121°57′E, 36°16′-38°23′N), is situated in the northeastern Shandong Peninsula, bordered by the Yellow and Bohai seas. The region exhibits a warm temperate monsoon climate with pronounced maritime influences. This latitude (~37°N) coincides with several global regions endemic for mosquito-borne viruses. Mediterranean coastal regions (Italy, Greece) represent WNV endemic zones, where outbreak intensity correlates with warm, humid conditions and enhanced vector breeding capacity at elevated temperatures (6). The Midwestern United States (Illi nois, Ohio; 37°N-42°N) constitutes a WNV high-risk region, with transmission cycles dependent on summer temperatures and wetland ecosystems (7). Southern Korea (34°N-38°N) represents a historical JEV endemic region, where Culex tritaeniorhynchus density correlates positively with monsoon precipitation (8). These regions share climatic conditions (mean annual temperature 12°C-15°C, hot humid summers with abundant precipitation) and ecological features (coastal wetlands, migratory flyways, dense port infrastructure) with Yantai, creating favorable conditions for cross-regional viral transmission. As a major Belt and Road Initiative port hub, Yantai's intensive international shipping and trade activities increase the risk of viral introduction through infected travelers or transported mosquito vectors (9).
This study used metagenomics to systematically characterize viral communities in Yantai mosquitoes, identifying 11 viral species across nine families, including Peribu nyaviridae and Picornaviridae, and constructing phylogenetic trees to elucidate their evolutionary relationships. These findings address critical knowledge gaps regarding mosquito-borne viruses in the Bohai Rim region and demonstrate the diverse viral threats present in Yantai, providing a scientific foundation for regional surveillance and early warning systems.
## MATERIALS AND METHODS
## Collection and processing of mosquito samples
This study was conducted in rural agricultural areas and urban zones of Laizhou City, Yantai, Shandong Province. Ten standardized monitoring stations were established across diverse habitats, including pig farms, dairy facilities, goat pastures, and urban green spaces. Sampling was conducted during the peak summer season, with collections performed nightly from dusk to dawn (18:00-06:00). Mosquitoes were collected using UV light traps (365 nm; Kungfu Xiaoshuai, China). Collected specimens were transported in chilled biosafety containers and stored at -20°C until processing. Specimens underwent initial morphological identification and were grouped by collection site (50 individuals per pool) with comprehensive database documentation. Taxonomic identification was confirmed through cytochrome oxidase subunit I (COI) barcoding (10).
## Nucleic acid extraction
Pooled specimens were homogenized in 1,000 µL of pre-chilled DMEM culture medium at 4°C. Homogenization was performed using a cryogenic tissue homogenizer (Tissue lyser, Germany) at 30 Hz for 2 × 4 min with 1-min intervals until complete tissue disruption. Following centrifugation (15,000 × g, 30 min, 4°C), supernatants were collected, and total RNA was extracted using a viral RNA purification kit (TIANamp, TianGen, China). Extracted RNA was used immediately for downstream applications or stored at -80°C until analysis.
## High-throughput sequencing library preparation
Equal volumes (5 µL) of RNA from each pooled sample were combined to gener ate 19 sequencing libraries. RNA samples with concentrations ≥10 ng/µL underwent ribosomal RNA depletion using the FastSelect ribosomal RNA removal system (Vazyme, China), while low-concentration samples proceeded directly to library preparation. cDNA libraries were prepared using the V8 version high-throughput transcriptomics sequenc ing library preparation kit (Vazyme, China) following the manufacturer's protocol. Library quality was assessed using an Agilent 2100 Bioanalyzer and quantified by quantitative (qPCR). Qualified libraries were converted to DNA nanoballs, loaded onto flow cells, and sequenced using paired-end 150 bp reads on a DNBSEQ-T7 platform.
## Sequence assembly and alignment
Raw sequencing data underwent multidimensional quality control using CLC Genomics Workbench, including base quality distribution assessment and adapter contamination screening. Low-quality sequences (Q-value <20) and redundant adapter sequences were removed using dynamic trimming algorithms, with reads <100 bp filtered out. STAR alignment was used to remove host-derived ribosomal RNA and genomic DNA sequences. Remaining high-quality reads were assembled de novo using MEGAHIT with k-mer sizes ranging from 21 to 121 (step size 10). Resulting contigs were identified using dual alignment strategies: BLASTX against the viral protein database (v2025.05) and BLASTN searches against the non-redundant nucleotide database. The Reference Viral Database public version was used (data retrieved through May 2025). Positive identification criteria were E-value <1 × 10⁻⁵ and sequence coverage >70%. Unique sequence clusters with sequencing depth >50 × and <60% homology to known viruses were designated as novel virus candidates for phylogenetic analysis and experimental validation.
## Virus taxonomic identification and functional annotation
Viral classification followed the International Committee on Taxonomy of Viruses (ICTV) taxonomy standards (Release 2025). Species identification was performed using whole-genome nucleotide alignments (NCBI nt database) and RdRp protein alignments (NCBI nr database). Novel species were defined by <80% genome nucleotide iden tity or <90% RNA-dependent RNA polymerase (RdRp) amino acid identity (11). Viral nomenclature followed the format: geographic location +host + taxonomic assignment. Study isolates of previously characterized viruses were designated with suffixes "21W-ZC" or "21W-GJD" to indicate collection sites. Viral genomes were analyzed for open reading frames (ORFs) using NCBI ORF finder (minimum length 300 bp/100aa), and functional domains were predicted using the Conserved Domain Database. Viral protein functions were predicted by comparison with known viral protein databases.
## Phylogenetic analysis
Sequence alignments were performed using the ClustalW algorithm within MEGA v10.2.6, with a gap opening penalty of 15 and a gap extension penalty of 6.66. Two phylogenetic methods were selected based on data characteristics: For known viral sequences, phylogenetic trees were constructed using the neighbor-joining (NJ) method with the Kimura two-parameter model, and node support was assessed through 1,000 bootstrap replicates. For newly discovered viral sequences: maximum likelihood analysis was performed using IQ-TREE 2.0, with the ModelFinder module selecting the best substitution model (BIC criterion), and branch support calculated via 1,000 ultra-fast bootstrap replicates. All phylogenetic trees underwent topological optimization and esthetic adjustment using FigTree v1.4.4, with key nodes labeled when branch support was >70%.
## Detection of newly identified viral RNA in mosquitoes
A one-step quantitative reverse transcription (qRT-PCR) detection system using TaqMan probes was established for viral nucleic acid detection in mosquitoes. The reaction mixture (25 µL total) contained the following: 5 × buffer (5 µL), dNTP mixture (1 µL), enzyme mixture (1 µL), forward and reverse primers (0.5 µL each, 10 µM), probe (0.25 µL, 5 µM), and template RNA (5 µL). The thermal cycling program consisted of the following: reverse transcription at 50°C for 30 min, initial denaturation at 95°C for 15 min, followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 1 min, with final extension at 72°C for 10 min. Positive results were defined by Ct values ≤ 35 and typical S-shaped amplification curves. Each experimental batch included no-template controls and positive controls (in vitro transcribed RNA) for quality assurance. Although metagenomics and qRT-PCR are powerful techniques for identifying viral genetic material, it is important to note that the detection of viral RNA does not provide definitive evidence of the presence of infectious virus particles.
## Data analysis and accession numbers
Nonparametric statistical methods were used to analyze viral carriage characteristics. The Mann-Whitney U test (α = 0.05) was employed to compare differences in viral species numbers between Culex and mosquito species other than Culex, combined with 10,000iteration permutation tests to assess individual-level viral load distributions. Simpson's 1 -D index (1 -∑pi², where pi is the relative abundance of the ith virus) was used to quantify viral diversity within host groups, and 10,000-iteration permutation tests were conducted to compare viral species differences between farming areas and urban areas at the individual level. Community structure analysis was performed using the vegan package for PERMANOVA tests (Bray-Curtis distance matrix, 999 permutations), and transmission networks were visualized using multidimensional chord diagrams constructed with the circlize package. Three-dimensional structures of viral proteins were predicted using AlphaFold, followed by functional domain annotation using PyMOL. Novel virus prevalence was expressed as minimum infection rate (minimum infection rate [MIR] = number of positive pools/total number of tests × 1,000‰), and intergroup differences were assessed using Fisher's exact test (P < 0.05 considered significant).
## RESULTS
## Virome profiles of mosquitoes collected from Shandong
From June to August 2021, we collected 8,111 mosquitoes from 10 sampling sites in Yantai City, Shandong Province (Fig. 1). Morphological identification classified these mosquitoes into four genera and six species: Culex pipiens pallens, Culex quinquefasciatus, Armigeres subalbatus, Culex tritaeniorhynchus, Anopheles sinensis, and Aedes togoi. To confirm species identification accuracy, representative samples from each species were randomly selected for COI gene PCR amplification and sequencing, yielding fragments of approximately 603 bp. Sequencing results were compared with NCBI reference sequences using BLAST. Based on high homology (>98%) with reference sequences, mosquito species were ultimately confirmed. Among these, Culex pipiens pallens was most abundant (n = 3,050, 37.6%), while Aedes togoi was least abundant (n = 128, 1.6%). Mosquitoes were pooled into 165 tubes based on collection site and species, then processed for nucleic acid extraction and library construction, yielding 19 libraries (Table 1). High-throughput sequencing generated 143 Gb of data across the 19 libraries. After quality control and host sequence removal, 5,027,344,916 clean reads were obtained for downstream analysis, of which 53.9 million were viral reads (1.1% of total). De novo assembly yielded 12,804 viral contigs, which were classified into 11 virus species across nine families: Picornaviridae, Peribunyaviridae, Tombusviridae, Narnaviridae, Botourmiavir idae, Solemoviridae, Xinmoviridae, Nodaviridae, and Rhabdoviridae. None of the viruses were found in the host genome, excluding the likelihood of these viruses being present as endogenous viral elements.
## Diversity, genome, and phylogenetic analysis of mosquito viromes
Our investigation identified viral RNA from 11 different viruses, most of which possess complete coding regions (Fig. 2). Notably, two of these viral RNAs-Serbia mononegalike virus 1 and Biggievirus Mos11-represent the first documented cases of these viral species in China. The remaining nine viruses, including Culex tritaeniorhynchus rhabdovirus (CTRV), Hubei picorna-like virus 59, Hubei mosquito virus 3, Hubei sobemolike virus 41, Hubei tombus-like virus 20, the Sichuan mosquito picorna-like virus, Wenzhou sobemo-like virus 4, Zhejiang mosquito virus 3, and Zhee mosquito virus, exhibit close phylogenetic relationships with previously characterized mosquito-associ ated viruses, though they remain officially unclassified by the ICTV. Comprehensive phylogenetic analysis enabled the classification of these viruses into nine distinct families. Bidirectional clustering analysis of the heatmap and phylogenetic tree revealed three key evolutionary patterns among the viruses: (i) A well-defined cluster comprising Biggievirus Mos11, Hubei picorna-like virus 59, and two sobemo-like viruses (Wenz hou sobemo-like virus 4 and Hubei sobemo-like virus 41), with the sobemo-like pair exhibiting particularly close genetic affinity, indicative of shared ancestry; (ii) A loosely associated group containing Zhee mosquito virus, Sichuan mosquito picorna-like virus,
## TABLE 1 Primers and probes used for identification of the two newly identified viruses in this study
## Virus
## Botourmiaviridae
We discovered a member of the Botourmiaviridae family, Hubei mosquito 3 viral RNA. Chinese researchers sequenced this virus in 2016 and submitted the sequence to GenBank (accession KX883460.1), but no detailed taxonomic analysis was carried out.
The viral genome comprises a single-stranded positive-sense RNA of 2,011 nt contain ing two ORFs. ORF1 spans positions 162-1,829 and encodes 555 amino acids; ORF2 The strain 21W-GJD identified in this study shares 99.4% nucleotide sequence identity with strain 3mos6141. The highest amino acid sequence identity (75.5%) was observed with the RdRp of Qianjiang botourmia-like virus 52 within the Botourmiaviridae family.
We then conducted a comparative analysis of representative viral sequences within the Botourmiaviridae family. Phylogenetic analysis based on RdRp sequences showed that Hubei mosquito virus 3 is most closely related to Armillaria mellea ourmia-like virus 1 within the Betamhizoulivirus genus, suggesting these viral sequences share a common origin. Therefore, Hubei mosquito virus 3 may represent a novel member of the Betamhizoulivirus genus (Fig. 4a).
## Narnaviridae
We that Biggievirus Mos11 and Culex Biggie-like virus form an independent clade, exhibiting significant genetic distance from members of the Alphanodavirus and Betanodavirus genera. This suggests these viruses may represent a novel evolutionary lineage within the Nodaviridae family, sharing similar taxonomic status with unclassified members such as Culex Daeseongdong-like virus (Fig. 4c).
## Peribunyaviridae
We detected a virus from the Peribunyaviridae family, specifically the Zhee mosquito viral RNA. Although its genome sequence was submitted to GenBank in 2016 (acces sion KM817705. We then conducted a comparative analysis of representative viral sequences within the Picornaviridae family. In the RdRp phylogeny, Hubei picorna-like virus 59 and Sichuan mosquito picorna-like virus, although belonging to different clades, are most closely related to Ampivirus A1 within the Ampivirus genus, indicating these viral sequences share a common origin. Therefore, these viruses may represent novel members of the Ampivirus genus within the Picornaviridae family (Fig. 4e).
## Rhabdoviridae
We detected a virus from the Rhabdoviridae family, CTRV viral RNA, which was first identified in Japan, and its sequence information was submitted to NCBI in 2014 (accession AB604791. The strain 21W-ZC identified in this study exhibits 90% sequence identity and 89.4% nucleotide sequence identity with known strains. A comparative analysis of representa tive viruses from all genera within the Xinmoviridae family was performed. Phylogenetic analysis based on polyprotein sequences revealed that Serbia mononega-like virus 1 forms a monophyletic group with Aedes albopictus anphevirus within the Culivirus genus. Therefore, this virus likely represents a novel member of the Xinmoviridae family and Culivirus genus (Fig. 4i).
## Viral coexistence across mosquito species
The number of viral species varied among mosquito species. Culex quinquefasciatus harbored the highest viral diversity with five species, while Culex pipiens pallens, Culex tritaeniorhynchus, Armigeres subalbatus, Anopheles sinensis, and Aedes togoi harbored three, three, one, one, and one viral species, respectively (Fig. 5).
In the six Culex quinquefasciatus libraries (L8-L13), viral reads predominantly origina ted from four families: Picornaviridae (30.1%), Narnaviridae (27.4%), Xinmoviridae (15.6%), and Nodaviridae (13.5%). Species-level analysis revealed that these reads primarily comprised Hubei picorna-like virus 59, Zhejiang mosquito virus 3, Serbia mononega-like virus 1, Biggievirus Mos11, and CTRV.
In the seven Culex pipiens pallens libraries (L1-L7), three viruses were identified: Zhee mosquito virus, Hubei mosquito virus 3, and Zhejiang mosquito virus 3. Three viruses were detected in the Culex tritaeniorhynchus libraries (L16 and L17): Hubei sobemo-like virus 41, Wenzhou sobemo-like virus 4, and CTRV. Hubei tombus-like virus 20 was detected in Armigeres subalbatus libraries (L14 and L15), Zhee mosquito virus in the Anopheles sinensis library (L18), and Sichuan mosquito picorna-like virus in the Aedes togoi library (L19).
Several viral species exhibited cross-species detection across different mosquito hosts. Specifically, Zhee mosquito virus was detected in both Culex pipiens pallens and Anopheles sinensis, Zhejiang mosquito virus 3 in both Culex quinquefasciatus and Culex pipiens pallens, and CTRV in both Culex quinquefasciatus and Culex tritaeniorhynchus.
## Viral ecology analysis
Host composition analysis revealed that the Culex genus (Culex pipiens, Culex quinquefas ciatus, and Culex tritaeniorhynchus) comprised 89.8% (6,568/7,311) of the total sample, with Culex pipiens pallens (3,050 individuals) and Culex quinquefasciatus (2,862 individu als) as dominant species. Culex mosquitoes carried nine viral species, whereas each of the three other taxa examined-Aedes, Anopheles, and Armigeres-harbored only one. The Mann-Whitney U test revealed no statistically significant difference in viral species distribution at the species level (U = 8.5, P = 0.157). Permutation testing at the individual level similarly revealed no significant differences (P = 0.089). The Simpson diversity index indicated higher viral community diversity in the Culex genus (1-D = 0.832 vs. 0.507). This pattern may reflect the broad host range of Culex mosquitoes (feeding on birds, live stock, and humans) and high population density (comprising 89.8% of the sample). As anthropophilic vectors, Culex mosquito larvae preferentially develop in organically enriched aquatic habitats, potentially facilitating multi-host viral transmission. Notably, although Aedes togoi harbored only one viral species (Sichuan mosquito picorna-like virus), its small sample size (128 specimens) yielded a detection rate of 7.8/1,000substantially higher than the 1.4/1,000 observed in Culex. This disparity likely stems from our nighttime, livestock-shelter trapping protocol, which favors endophilic species but under-samples day-active or coastal vectors such as Aedes togoi. The resulting samplesize imbalance precludes reliable estimation of viral prevalence in this mosquito.
Viral diversity analysis revealed significantly higher viral prevalence in agricultural areas compared to urban areas (permutation test: P < 0.001). Average viral species density reached 5.02 per thousand individuals in agricultural areas, significantly exceeding 2.94 per thousand in urban areas. PERMANOVA analysis confirmed significant differences in viral community structure between agricultural and urban sites (F = 15.7, P = 0.010). Agricultural sites harbored multiple cross-species viruses (e.g., Zhee mosquito virus and CTRV, both detected across two Culex species). This elevated viral diversity likely reflects multiple ecological factors: livestock operations facilitate cross-species viral transmission, organic-rich wastewater systems enhance viral persistence, and highdensity host populations increase vector-host contact rates. Conversely, reduced host diversity and urban pollutants may limit viral establishment in urban environments (Fig. 6).
## Structural analysis of novel virus-encoded proteins
Structural and functional analyses of Serbia mononega-like virus 1 protein-coding regions were performed using AlphaFold predictions. The virus exhibits a complex multidomain architecture (Fig. 7). The RdRp domain displays a characteristic compact globular topology typical of viral polymerases, with the highly conserved GDD catalytic motif positioned at the structural core for optimal catalytic geometry. The N-terminal and Cterminal regions exhibit distinct folding patterns, suggesting roles in genome replication regulation or host factor interactions. The nucleoprotein domain adopts a compact conformation, suggesting involvement in genome packaging or host factor recruitment. The glycoprotein domain displays unique folding patterns that may facilitate host cell recognition and viral entry. Predicted alignment error (PAE) analysis revealed extremely high model confidence (PAE <10 Å) in the core catalytic region, particularly within the RdRp domain. Analysis revealed extensive interdomain interactions, particularly in the Cterminal region, suggesting potential allosteric regulation of enzyme activity. The Nterminal region exhibited higher predicted errors, suggesting structural flexibility that may be crucial for functional regulation.
Two major functional domains were identified in Biggievirus Mos11 (Fig. 8). The RNA helicase domain contains a typical ATP-binding site and multiple conserved helicase motifs, likely facilitating viral genome replication through ATP-dependent nucleic acid FIG 5 The figure presents the ecological relationships between virus families (y-axis) and their mosquito vectors (x-axis). Virus-mosquito associations are indicated by colored data points, with dashed ellipses highlighting significant vector-virus groupings observed in our study.
unwinding. This activity is essential for viral RNA processing during infection. The RdRp domain contains classic GDD and SDD catalytic motifs, characteristic of RNA-dependent RNA polymerase activity. PAE analysis revealed confidence distribution across the protein structure. Both RNA helicase and RdRp domains exhibit excellent internal structural stability (PAE <10 Å) with distinct interdomain interfaces, suggesting functional complex formation. However, while the helicase domain shows accurate internal conformation predictions, it exhibits moderate PAE values with other domains, suggesting participa tion in viral regulation through dynamic conformational changes.
## Nucleic acid detection of novel viruses in mosquito samples
Nucleic acid detection was performed for two novel viruses across all 165 mosquito sample pools. qRT-PCR analysis revealed 8 and 16 positive pools for Serbia mononegalike virus 1 and Biggievirus Mos11, respectively. Both viruses were detected exclusively in Culex quinquefasciatus. MIRs were 0.34% for Serbia mononega-like virus 1 and 0.68% for Biggievirus Mos11 (Table 2).
## DISCUSSION
Mosquito-borne viruses represent a persistent global public health threat, with out breaks causing a substantial burden of disease in regions, healthcare system strain, and causing socioeconomic losses (14). Climate change has expanded suitable mosquito habitats, while increased international travel and trade facilitate cross-border pathogen dispersal. Rapid urbanization and population densification have enhanced the efficiency of viral transmission, escalating global mosquito-borne virus transmission risk and triggering multiple regional outbreaks (15). Historical examples include the 20th-century DENV outbreak in Southeast Asia, causing over one million severe cases, the transconti nental spread of WNV in North America, and the 2016 ZIKV-associated microcephaly outbreak, all demonstrating the potential for mosquito-borne pathogens to become international public health emergencies (16,17). Current surveillance data indicate that natural mosquito viral diversity far exceeds our understanding. Metagenomic sequenc ing has identified over 500 novel mosquito-associated viruses, some with cross-species transmission potential (18). This approach broadens the temporal and spatial scope of pathogen surveillance while providing molecular foundations for targeted control strategies through viral evolutionary analysis (19,20).
This study systematically identified and analyzed viral genomes from six common mosquito species in Yantai City, Shandong Province, China. High-throughput sequencing yielded complete coding sequences or RdRp gene sequences for 11 viruses, including two viruses newly identified in China. These viruses span nine viral families, confirming extensive viral diversity in mosquito vectors. These findings align with global studies. eight viral families in four common Shandong mosquito species were almost entirely distinct from those identified here (23). This underscores the complexity of mosquito viral communities, which exceeds previous understanding. Most viruses identified here are likely mosquito-specific and pose no direct threat to human or animal health. This aligns with known mosquito-borne virus ecology, as most medically important viruses replicate at low levels in mosquitoes and proliferate only under specific epidemiological conditions. Multiple studies have shown that known human and animal pathogens make up only a small fraction of mosquito viral genomes. One of the viral RNAs detected in our study, Hubei tombus-like virus 20, is likely a plant virus. Given that the Tombusviridae family is predominantly associated with plants, it is highly probable that this virus primarily infects plant hosts rather than mosquitoes. The detection of Hubei tombuslike virus 20 RNA in our mosquito samples is more likely a result of the mosquitoes ingesting plant material containing the virus, rather than the virus replicating within the mosquitoes themselves. This scenario is consistent with the feeding behavior of mosquitoes, which can ingest plant sap or nectar as part of their diet. Therefore, the presence of Hubei tombus-like virus 20 RNA in our samples should be interpreted as the detection of viral RNA fragments originating from plant sources, rather than evidence of active viral replication or infection within the mosquito hosts. While the majority of detected viral RNAs may not pose direct threats to humans or animals, some identified viruses warrant further investigation due to their potential pathogenicity. In addition, Zhee mosquito virus (Peribunyaviridae) may have pathogenic potential, as phylogenetic analysis reveals close relationships to Shangavirus genus members. Shangavirus and Orthobunyavirus (e.g., La Crosse virus and Oropouche virus) share similar genomic structures, with the latter causing human fever and encephalitis (24,25). The Merhavirus member CTRV detected here has been found only in mosquito hosts with no reported vertebrate infections. However, phylogenetic analysis reveals that CTRV is closely related to pathogenic genera within the Alpharhabdovirinae subfamily, including Chandipura virus (Vesiculovirus) and bovine ephemeral fever virus (Ephemerovirus) (26,27). This suggests CTRV may possess cross-species transmission risk or pathogenic evolutionary potential. The pathogenic potential of these viruses requires experimental verification. High-throughput metagenomic sequencing enabled assembly of complete coding sequences for 11 known viruses from mosquito samples. Although all viruses are recorded in public databases, existing sequences predominantly derive from recent publications with limited reference genomic information. Systematic genomic analysis revealed significant differences in nucleic acid type: three viruses (Zhee mosquito virus, CTRV, and Serbia mononega-like virus 1) are single-stranded negative-sense RNA viruses, while eight are single-stranded positive-sense RNA viruses. This positive-sense RNA virus predominance may reflect mosquitoes' ecological role as hosts for plant-and insect-specific viruses. Similar patterns have been reported in comparable international studies (28). Regarding viral classification, CTRV and Serbia mononega-like virus 1 have established genus-level classifications, while nine viruses lack systematic taxonomic positioning. All viruses encode complete RdRp domains. Given RdRp's high conserva tion and sequence length advantages, phylogenetic analysis based on RdRp sequen ces successfully classified these viruses into genus-level taxonomic units. This analysis refines viral taxonomic status while providing insights for future biological studies. This study identified molecular evidence of cross-species transmission. Three viruses (Hubei picorna-like virus 59, Hubei tombus-like virus 20, and Hubei sobemo-like virus 41) were previously reported from arachnid hosts. However, highly homologous sequences were detected in Culex quinquefasciatus, Armigeres subalbatus, and Culex tritaeniorhynchus. This suggests that these viruses may breach host barriers or participate in complex transmission networks within natural ecosystems. However, food chain transmission (e.g., mosquitoes preying on spiders) cannot be excluded based on current data.
This study reports the first identification of two viral RNAs in China: Serbia mono nega-like virus 1 and Biggievirus Mos11. Serbia mononega-like virus 1 was previously reported only from Serbia, with this study expanding its geographical distribution to China. Genomic analysis revealed 96.4% and 95.0% sequence identity for glycopro tein and RdRp between strains, respectively, indicating conserved protein structures despite regional differences. This virus belongs to the Culivirus genus (Xinmoviridae) and may exhibit mosquito-borne transmission, although mammalian infection or human pathogenicity remains unclear. Biggievirus Mos11 has been reported from North America, Western Europe, and South Asia, with this study confirming its first detection in East Asia. RNA helicase and RdRp sequences exhibit >99.1% identity with known strains, indicating high evolutionary conservation. Phylogenetic analysis suggests membership in the Alphanodavirus genus (Nodaviridae), whose members primarily infect insects but not vertebrates. Local mosquito population testing revealed positive rates of 0.34% and 0.68% for both viruses, respectively, indicating local adaptation and stable regional circulation. This aligns with previous findings for Hubei mosquito virus 2 (HMV2) in Shandong Province (29). The viral RNAs of the two novel viruses (Serbia mononega-like virus 1 and Biggievirus Mos11) were detected exclusively in Culex quinquefasciatus in this study. (This finding is specific to these viruses and does not imply that all viru ses discussed in this manuscript were detected in Culex quinquefasciatus), suggesting potential cross-species infectivity. This expands understanding of viral host range, with potential public health implications warranting further investigation.
Mosquitoes serve as important arbovirus vectors and harbor complex viral commun ities. Multiple viruses were detected across mosquito species: Culex quinquefasciatus harbored five viruses from different families, while Culex pipiens pallens and Culex tritaeniorhynchus each harbored three viruses. This viral diversity may reflect mosquito feeding behavior. Female mosquitoes, in addition to feeding on plant juices for energy and nutrients, require blood meals for reproduction, which can expose them to a broader range of potential viral sources. Male mosquitoes, on the other hand, primarily obtain their carbohydrates from plant juices, including nectar and other plant fluids, which provide the necessary energy for their daily activities and survival (30,31). This broad feeding range may facilitate exposure to and acquisition of multiple viruses. While pooled sample detection determines species-level viral profiles, it cannot distinguish simultaneous multi-viral infections in individual mosquitoes. Certain viruses (Zhee mosquito virus and Zhejiang mosquito virus 3) were detected across multiple mosquito species, suggesting broad vector specificity, cross-species transmission potential, and expanded vector ranges. These findings suggest strong environmental adaptability and enhanced transmission capabilities for these viruses. While our study detected viral RNA in several mosquito species, it is notable that some viruses were not detected in others. This could be due to differences in ecological niches and behaviors among mosquito species. For instance, certain mosquito species may have feeding preferences or habitats that reduce their exposure to viral sources. Additionally, variations in vector competence and susceptibility to viral infection could also play a role. Future studies should investigate these factors in more detail to elucidate the reasons behind the observed distribution patterns of viral detection among different mosquito species.
This study has several limitations. First, nighttime collection may have overlooked diurnal species such as Aedes albopictus, potentially resulting in incomplete viral diversity data. Second, without viral enrichment and isolation, sequences may reflect only high-abundance viral populations, precluding comprehensive assessment of low-abun dance virus infectivity and pathogenicity. Third, reliance solely on metagenomic sequencing without viral isolation, cultivation, or animal infection experiments precludes confirmation of viral infectivity and pathogenic potential. While our study detected viral RNA from 11 different viruses, it is important to note that the presence of viral RNA does not definitively confirm the presence of infectious virus particles. For example, previous studies have reported the detection of viral RNA from arboviruses such as Zika, dengue, and CHIKVs in mosquitoes, even when no infectious virus was present (e.g., mosquitoes exposed to inactivated virus). This highlights the need for additional studies to confirm the presence of infectious viruses, including attempts to isolate live virus particles and conduct further experiments to assess their infectivity and pathogenic potential.
## Conclusion
This study reveals complex mosquito-borne viral diversity in Yantai, Shandong Province, with potential public health significance. Metagenomic analysis revealed widespread distribution of multiple mosquito-borne viruses within the Culex genus, with some exhibiting cross-host transmission characteristics that reflect virus-vector co-evolution. Certain viruses share phylogenetic relationships with known pathogens, indicating the need for cross-species transmission monitoring. These findings enhance understanding of mosquito-borne viral ecology while providing evidence for regional vector-borne disease control.
## Highlights:
• This study uses metagenomics to systematically analyze viral communities in mosquitoes from Yantai, Shandong, identifying 11 viral species across 9 families, with Serbia mononega-like virus 1 and Biggievirus Mos11 being first reported in China.
• Phylogenetic analysis based on RdRp sequences clarifies the taxonomic status of the identified viruses, with several potentially representing novel genera or species, enhancing understanding of their evolutionary 8 relationships. • Host distribution analysis shows Culex quinquefasciatus harbors the 10 greatest viral diversity, and some viruses exhibit cross-species transmission potential, indicating host specificity and transmission complexity. • Viral ecology analysis reveals significantly higher viral diversity in agricultural areas than urban areas, and qRT-PCR confirms the prevalence of the two novel viruses in local Culex quinquefasciatus populations with minimum infection rates of 0.28% and 0.56% respectively.
## References
1. Weaver, Reisen (2010) "Present and future arboviral threats" *Antiviral Res*
2. Weaver, Barrett (2004) "Transmission cycles, host range, evolution and emergence of arboviral disease" *Nat Rev Microbiol*
3. Bhatt, Gething, Brady et al. (2013) "The global distribution and burden of dengue" *Nature*
4. Liu, Liu, Nie et al. (2016) "Flavivirus NS1 protein in infected host sera enhances viral acquisition by mosquitoes" *Nat Microbiol*
5. Cheng, Liu, Wang (2016) "Mosquito defense strategies against viral infection" *Trends Parasitol*
6. Paz (2015) "Climate change impacts on West Nile virus transmission in a global context" *Phil Trans R Soc B*
7. Cdc (2023) "West Nile virus: statistics & maps"
8. Zheng, Li, Wang et al. (2012) "Japanese encephalitis and Japanese encephalitis virus in mainland China" *Rev Med Virol*
9. Zhu, Liu, Cheng (2023) "Progress towards research on mosquitoborne arboviral transmission and infection" *Sci Bull (Beijing)*
10. Aung, Bawm, Chel et al. (2023) "Molecular identification of aedes, armigeres, and culex mosquitoes (diptera: culicidae) using mitochondrial cytochrome oxidase subunit I genes in Myanmar" *Acta Parasitol*
12. Babaian, Edgar (2022) "Ribovirus classification by a polymerase barcode sequence" *PeerJ*
13. Shi, Lin, Tian et al. (2016) "Redefining the invertebrate RNA virosphere" *Nature*
14. Dudas, Huber, Wilkinson et al. (2021) "Polymorphism of genetic ambigrams" *Virus Evol*
15. Franklinos, Jones, Redding et al. (2019) "The effect of global change on mosquito-borne disease" *Lancet Infect Dis*
16. Ryan, Carlson, Mordecai et al. (2019) "Global expansion and redistribution of Aedes-borne virus transmission risk with climate change" *PLoS Negl Trop Dis*
17. Kramer, Styer, Ebel (2008) "A global perspective on the epidemiology of West Nile virus" *Annu Rev Entomol*
18. De Araújo, De, De et al. (2018) "Association between microcephaly, zika virus infection, and other risk factors in Brazil: final report of a casecontrol study" *Lancet Infect Dis*
19. Pan, Zhao, Gou et al. (2024) "Metagenomic analysis of individual mosquito viromes reveals the geographical patterns and drivers of viral diversity" *Nat Ecol Evol*
20. Mohebbi, Zelikovsky, Mangul et al. (2024) "Early detection of emerging viral variants through analysis of community structure of coordinated substitution networks" *Nat Commun*
21. Struelens, Ludden, Werner et al. (2024) "Real-time genomic surveillance for enhanced control of infectious diseases and antimicrobial resistance" *Front Sci*
22. Williams, Paradkar, Karl (2021) "Arbovirus detection in vectors"
23. Schilling, Jagdev, Thomas et al. (2025) "Metagenomic analysis of mosquitoes from Kangerlussuaq, Greenland reveals a unique virome" *Sci Rep*
24. Wang, Lin, Li et al. (2024) "Metagenomic sequencing reveals viral diversity of mosquitoes from Shandong Province" *China. Microbiol Spectr*
25. Hughes, Adkins, Alkhovskiy et al. (2020) "ICTV virus taxonomy profile: peribunya viridae"
26. Elliott (2014) "Orthobunyaviruses: recent genetic and structural insights" *Nat Rev Microbiol*
27. Walker, Firth, Widen et al. (2015) "Evolution of genome size and complexity in the rhabdoviridae" *PLoS Pathog*
28. Walker, Klement (2015) "Epidemiology and control of bovine ephemeral fever" *Vet Res*
29. Wu, Zhang, Feng et al. (2024) "An evolutionarily conserved ubiquitin ligase drives infection and transmission of flaviviruses" *Proc Natl Acad Sci*
30. Wu, Liu, Feng et al. (2023) "Identification and molecular characteristics of a novel single-stranded RNA virus isolated from Culextritaeniorhynchus in China" *Microbiol Spectr*
31. Douglas (2006) "Phloem-sap feeding by animals: problems and solutions" *J Exp Bot*
32. Foster (1995) "Mosquito sugar feeding and reproductive energetics" *Annu Rev Entomol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817925&blobtype=pdf
|
# Conformational dynamics of the HIV-1 envelope glycoprotein from CRF01_AE is associated with susceptibility to antibodydependent cellular cytotoxicity
Marco Díaz-Salinas, Mehdi Benlarbi, Debashree Chatterjee, Manon Nayrac, Megane Robidas, Suteeraporn Pinyakorn, Nittiya Phanuphak, Carlo Sacdalan, Halima Medjahed, Jérémie Prévost, Lydie Trautmann, Marzena Pazgier, Andrés Finzi, James Munro
## Abstract
The HIV-1 envelope glycoprotein (Env) is expressed at the surface of infected cells and, as such, can be targeted by non-neutralizing antibodies (nnAbs) that mediate antibody-dependent cellular cytotoxicity (ADCC). Previous single-molecule Förster resonance energy transfer (smFRET) studies demonstrated that Envs from clinical isolates predominantly adopt a "closed" conformation (State 1), which is resistant to nnAbs. After interacting with the cellular receptor CD4, the conformational equilibrium of Env shifts toward States 2 and 3, exposing the coreceptor-binding site (CoRBS) and permitting targeting by CD4-induced (CD4i) antibodies. We showed that the binding of anti-CoRBS Abs enables the engagement of other nnAbs that target the cluster A epitopes on Env. Anti-cluster A nnAbs stabilize an asymmetric Env conformation, State 2A, and have potent ADCC activity. CRF01_AE strains were suggested to be intrinsically susceptible to ADCC mediated by nnAbs. This may be due to the presence of a histidine at position 375, known to shift Env toward more "open" conformations. In this work, through adaptation of an established smFRET imaging approach, we report that native, unliganded CRF01_AE HIV-1 Envs frequently sample the State 2A conformation. This is in striking contrast with Envs from clades A and B, for example HIV-1 JR-FL , which do not transition to State 2A in the absence of ligands. These findings inform on the confor mational dynamics of CRF01_AE Env, which are relevant for structure-based design of both synthetic inhibitors of receptor binding and enhancers of ADCC as therapeutic alternatives.IMPORTANCE A concerning increase in infections with HIV-1 from CRF01_AE has occurred globally and regionally in recent years, especially in Southeast Asia. Despite the advances made in understanding HIV-1 envelope glycoprotein (Env) conformational dynamics, the knowledge about Env from CRF01_AE HIV-1 is limited. Here, we demon strate that the unliganded CRF01_AE Env readily samples an "open" conformation (State 2A), which is susceptible to antibody-dependent cellular cytotoxicity (ADCC). This is in contrast with the subtypes previously studied from HIV-1 group M that rely on anti-clus ter A antibodies to adopt State 2A. These findings are relevant for the structure-based design of novel synthetic inhibitors of CD4 binding and enhancers of ADCC for the elimination of infected cells.
specific to the HIV-1 envelope glycoprotein (Env) in a subset of individuals with low plasma IgA (1,2). This suggests that ADCC may have contributed to the protec tion observed in the RV144 trial. HIV-1 strains of the circulating recombinant form AE (CRF01_AE) predominate the AIDS epidemic in Southeast Asia (3,4). Therefore, the RV144 trial used an ALVAC-HIV prime and AIDSVAX B/E boost regimen, which included one CRF01_AE (A244) and one clade B (MN) gp120 glycoprotein. Moreover, the prevalence of HIV-1 CRFs has risen in recent years, most significantly in Southeast Asia (5). For these reasons, a detailed investigation of Env from HIV-1 CRFs is warranted. While advances in the understanding of Env conformational dynamics have been achieved using virological and biophysical approaches, these studies have focused on HIV-1 subtypes A and B (6)(7)(8)(9)(10)(11)(12). A similar elucidation of the dynamics of Env from HIV-1 CRFs has not been reported. However, prior studies demonstrated an inherent susceptibility of CRF01_AE HIV-1 to ADCC, which begins to explain the results of the RV144 trial (6,13). Subsequent structural investigation of CRF01_AE Env indicated features that are distinct from other subtypes and perhaps enable conformations related to recognition by Abs with ADCC activity (14).
In the present study, we explore the conformational features of Envs from CRF01_AE and their relationship to ADCC mediated by plasma from people living with HIV (PLWH).
The first step in the replication of HIV-1 is the binding of Env to the cellular receptor CD4. Env is synthesized as the gp160 precursor, which is trimerized and glycosylated in the endoplasmic reticulum of infected cells (15,16), followed by proteolytic processing by host furin-like proteases in the Golgi apparatus (17)(18)(19). The resulting cleaved and mature Env trimer is composed of three gp120 subunits, which are non-covalently associated with three transmembrane gp41 subunits [(gp120-gp41) 3 ] (20 -22). Mature Env is present on virions as well as exposed on the surface of infected cells, making it the primary target of host Abs. Some Abs neutralize the virus (NAbs) by blocking Env's interaction with receptors or inhibiting conformational changes needed to promote fusion of the viral and cellular membranes. Other Abs that are frequently elicited during HIV-1 infection, including in PLWH, are non-neutralizing (nnAbs) since they recognize Env epitopes occluded within "closed" Env conformations. Certain classes of nnAbs, however, can induce the death of infected cells through ADCC, provided Env samples an "open" conformation.
Single-molecule Förster resonance energy transfer (smFRET) imaging studies demonstrated that Env is highly dynamic, transitioning from a "closed" conformation (State 1) to an "open" conformation (State 3), which is promoted through the interaction with CD4. An asymmetric intermediate (State 2) of Env can be observed during the transition from State 1 to State 3 (10,12). The Env conformational equilibrium from primary HIV-1 isolates of clades A and B favors State 1 in the absence of ligands, which confers resistance to most Abs, especially those that target CD4-induced (CD4i) epitopes (12,23). Nonetheless, some broadly neutralizing Abs (bNAbs) preferentially bind this "closed" conformation (8,12). However, after interacting with cellular CD4, Env adopts State 3, exposing cryptic epitopes including the coreceptor-binding site (CoRBS) and gp120 cluster A region, which can be targeted by nnAbs to promote ADCC (6,(23)(24)(25)(26)(27)(28). CD4-mimetic compounds (CD4mcs) are small molecules designed to target specifically the CD4 binding cavity within HIV-1 Env. CD4mcs can induce conformational changes in Env that sensitize it to recognition by nnAbs (25,26). In the presence of soluble CD4 (sCD4) or CD4mcs, anti-CoRBS and anti-cluster A Abs stabilize State 2A, which is an asymmetric Env conformation associated with increased ADCC responses in vitro and Fc-effector functions in vivo (9,25,(29)(30)(31).
The findings presented here indicate that native HIV-1 Envs from tier-1 and tier-2 CRF01_AE strains intrinsically sample the State 2A conformation, which is susceptible to ADCC even in the absence of CD4 or CD4mcs. This contrasts with clade-B HIV-1 JR-FL Env, which depends on interaction with CD4 or CD4mcs and antibodies targeting the CoRBS to adopt State 2A (9,25). Interaction of the tier-1 CRF01_AE Env with CD4 and CoRBS Abs further stabilized State 2A. The conformational features of CRF01_AE Envs warrant further research to identify the structural determinants or elements that govern its dynamic equilibrium shift to more "open" conformations. Targeting CRF01_AE HIV-1infected cells with cocktails of antibodies (32) or molecules able to recognize down stream Env conformations represents promising strategies to decrease the reservoir of individuals living with this viral strain (33).
## RESULTS
## CRF01_AE HIV-1-infected cells are more susceptible to ADCC than a represen tative subtype B strain
We made a direct comparison of the susceptibility of infected cells to ADCC using representative infectious molecular clones (IMCs) from CRF01_AE (strain 703357) and subtype B (strain JR-FL). First, we evaluated the binding capacity of plasma from 10 PLWH infected by clade B viruses (Table 1). No significant differences between the two strains were observed (Fig. 1A). However, the ADCC responses to HIV-1 CRF01_AE were approximately twofold higher than those observed with the HIV-1 JR-FL strain (Fig. 1B). To evaluate whether the infecting HIV subtype could impact the functional differences observed, we also tested plasma from 10 PLWH infected by CRF01_AE viruses from the Thai RV304 cohort (Table 1). Interestingly, we observed a significant increase in both plasma binding and ADCC responses to HIV-1 CRF01_AE compared with HIV-1 JR-FL strain (Fig. S1).
To further explore the intrinsic Env conformational landscape of HIV-1 CRF01_AE and HIV-1 JR-FL , we decided to evaluate the binding capacity of plasma from both cohorts (Table 1) in a CD4-negative cell line. Briefly, HEK293T cells were transfected with plasmids encoding the Env from CRF01_AE (strain 92TH023) or subtype B (strain JR-FL), and plasma binding was measured using flow cytometry. In concordance with our results obtained in primary CD4 T cells, we observed for both cohorts a significant increase in plasma binding against HIV-1 CRF01_AE compared to HIV-1 JR-FL (Fig. 2). Because activation of the ADCC response has been associated with a conformation of HIV-1 Env that enables binding of a specific class of Abs, these results suggest that HIV-1 CRF01_AE Env may have distinct conformational features that confer sensitivity to ADCC (6,7).
## Modifications in HIV-1 CRF01_AE Env that enable site-specific fluorescent labeling do not affect viral infectivity
With the aim of visualizing the conformational dynamics of HIV-1 CRF01_AE Env, we adapted a previously validated smFRET imaging assay. We investigated CRF01_AE strains 92TH023 and CM244, which are tier-1 and tier-2 isolates, respectively. Insertion of the A4 peptide (DSLDMLEW) and incorporation of non-natural amino acids (nnAAs) into HIV-1 Env facilitate fluorophore attachment. These methods have been applied with minimal effect on functionality to subtype-B HIV-1 strains NL4-3 and JR-FL, as well as the subtype-A strain BG505 (8-10, 12, 30). As for previous applications, we attached site-specifically fluorophores in the V1 and V4 loops of a single gp120 domain within CRF01_AE Env on the surface of pseudovirions (Fig. 3A). To this end, we inserted the A4 peptide next to V135 in V1 (V1-A4), which enabled enzymatic attachment of the LD650 fluorophore. We also substituted an amber stop codon for amino acid N398 in V4 of gp120 (V4-N398 TAG ). Suppression of the amber stop codon incorporates the nnAA TCO*, which facilitated Cy3 fluorophore attachment through copper-free click chemistry (Fig. 3B) (34). Mann-Whitney t-test, and P values <0.05 were considered statistically significant. In both panels, a dotted line indicates the limit of detection, which was determined using five plasmas from uninfected individuals. The characteristics of the cohort of plasma donors are shown in Table 1. We next confirmed full-length translation of HIV-1 92TH023 and HIV-1 CM244 Env containing the V1-A4 and V4-N398 TAG mutations (tagged) and its incorporation into virions. We evaluated through immunoblots the abundance of both full-length gp120 and the HIV-1 core capsid protein p24 in purified viral preparations (Fig. 3C andD). As expected, tagged gp120 was not detected in virions produced in the absence of the nnAA TCO* and the corresponding aminoacyl tRNA synthetase and suppressor tRNA, which codes for the amber stop codon. This indicates that readthrough of the amber codon in the V4 loop did not occur, resulting in the lack of Env incorporation into viral particles (Fig. 3C andD, top immunoblots, lane 4). However, in the presence of TCO*, the synthetase, and the suppressor tRNA, tagged gp120 was detected in virions at a comparable level as wild-type Env (Fig. 3C andD, top immunoblots, lane 5). We next verified that V1-A4/V4-N398 TAG modifications in the Envs do not alter virus infectivity. Virus preparations bearing wild-type or tagged Env showed no statistically significant difference in their infectivity in TZM-bI cells (Fig. 3E), suggesting that both incorporation of the A4 peptide in V1 and the nnAA TCO* in V4 do not affect the function of Env. Altogether, these data demonstrate that tagged HIV-1 92TH023 and HIV-1 CM244 Envs are incorporated into pseudovirions and maintain native function during infection of cells.
## Native CRF01_AE HIV-1 Envs intrinsically sample "open" conformations
We next sought to evaluate the conformational dynamics in real-time of individual HIV-1 92TH023 and HIV-1 CM244 Env molecules on the surface of virions using smFRET imaging. To this end, we prepared virions bearing a single fluorescently labeled gp120 domain as described for Env from other HIV-1 strains (Fig. 3A) (8)(9)(10)12). Labeled virions were immobilized on passivated quartz microscope slides and imaged using prism-based TIRF microscopy. We used the well-characterized clade-B HIV-1 JR-FL Env as a point of comparison. Consistent with previous reports, the application of hidden Markov modeling (HMM) for analysis of the smFRET trajectories enabled the identification of four FRET states, as indicated by the four Gaussian fits overlaid on the FRET histograms (Fig. 4A through C). HMM also allowed us to determine the fraction of time (occupancy) each molecule spent in each state (Table 2), which are presented as violin plots (Fig. 4D through F). The violin plot representation displays the heterogeneity of the population and allows for statistical comparisons across the data sets. For all three strains under consideration-HIV-1 JR-FL , HIV-1 92TH023 , and HIV-1 CM244 -a low-FRET value (0.22 ± 0.1 FRET [mean ± standard deviation], State 1) was predominant, which is associated with a "closed" Env conformation (Fig. 4A through C). Quantification of the mean occupancies in State 1 across the populations of molecules indicated 72% ± 2%, 42% ± 2%, and 45% ± 2% for HIV-1 JR-FL , HIV-1 92TH023 , and HIV-1 CM244 , respectively (Fig. 4D). We also 2.
observed State 3 (0.45 ± 0.1 FRET) for all strains, which is associated with an "open" Env conformation. We determined State 3 occupancies of 26% ± 2%, 27% ± 2%, and 28% ± 2% for HIV-1 JR-FL , HIV-1 92TH023 , and HIV-1 CM244 , respectively (Table 2). Consistent with previous reports, we detected minimal occupancy for HIV-1 JR-FL Env in States 2 and 2A (0.70 ± 0.1 and 0.85 ± 0.1 FRET, respectively). In contrast, both unliganded CRF01_AE strains displayed modest but statistically significant occupancies in State 2 and State 2A. HIV-1 92TH023 displayed 19% ± 1% occupancy in State 2 and 12% ± 1% in State 2A; HIV-1 CM244 displayed 14% ± 1% and 13% ± 1% in State 2 and State 2A, respectively. These data demonstrate that Envs from both tier-1 and tier-2 strains of CRF01_AE have greater intrinsic access to "open" conformations than HIV-1 JR-FL Env.
We next asked if sCD4 consisting of domains 1 and 2 (sCD4 D1D2 ) or the anti-CoRBS mAb 17b further stabilize "open" conformations. For all three Envs, the addition of sCD4 D1D2 destabilized State 1 and promoted transition to the higher FRET states, with the effect being the greater for HIV-1 JR-FL . Although no statistically significant difference was seen in the State 1 occupancy across the three stains (Fig. 4E; Table 2). For HIV-1 JR-FL Env, we observed increased occupancy in States 2 and 3, as previously reported (12). sCD4 D1D2 had only a modest effect on the conformations of both CRF01_AE Envs, with only a slight stabilization of State 3. The State 2A occupancy of HIV-1 92TH023 and HIV-1 CM244 Envs changed minimally in the presence of sCD4 D1D2 but remained greater than seen for HIV-1 JR-FL .
The addition of both sCD4 D1D2 and 17b further promoted State 3 for HIV-1 JR-FL Env, as expected (Fig. 4F; Table 2). In contrast, the predominant effect of sCD4 D1D2 /17b on the tier-1 HIV-1 92TH023 Env was to stabilize State 2A, increasing the occupancy to 22% ± 2%, as compared to 8% ± 2% for HIV-1 JR-FL . In contrast, the tier-2 HIV-1 CM244 showed no significant change in the State 2A occupancy with sCD4 D1D2 /17b, although it again remained greater than HIV-1 JR-FL . Access to State 2A correlates with the greater inherent sensitivity to ADCC seen for HIV-1 from CRF01_AE.
## DISCUSSION
During HIV-1 infection, the humoral response against Env mainly produces antibodies that are non-neutralizing. Despite the lack of neutralization, nnAbs can still trigger ADCC to clear infected cells, provided that Env is exposed in an "open" conformation (35). Env glycoproteins from most HIV-1 strains naturally adopt State 1, which is associated with a closed conformation (12), and confers resistance to nnAbs (23,36). In contrast, previous functional studies suggested that Env glycoproteins from CRF01_AE strains intrinsically adopt "open" conformations even in the absence of CD4, CD4 mimetics, or anti-CoRBS mAbs (6,7,13). Recent insights from structural data further support this idea (14). Here, we have shown that plasma obtained from PLWH triggers ADCC against CRF01_AE HIV-1-infected cells to a greater extent than the clade-B HIV-1 JR-FL -infected cells. We therefore sought to directly test the conformational equilibrium of CRF01_AE a Data are presented as mean ± standard error determined from the total population of traces analyzed.
HIV-1 Env using smFRET imaging. We have demonstrated through real-time analysis of HIV-1 92TH023 and HIV-1 CM244 Env conformational dynamics that these glycoproteins intrinsically sample "open" conformations in the absence of bound ligands. The tier-1 HIV-1 92TH023 appeared to be slightly less stable in State 1 as compared to the neutral ization-resistant tier-2 HIV-1 CM244 , consistent with greater exposure of antigenic sites in tier-1 isolates. However, this difference failed to reach statistical significance. Both CRF01_AE Envs intrinsically adopted State 2A, which was previously linked to exposure of both the CoRBS and cluster A epitopes that are targeted by Abs with potent ADCC activity (9). This provides a mechanistic rationale for the enhanced sensitivity to ADCC seen in the presence of plasma from PLWH. Only in the presence of both sCD4 D1D2 and 17b was State 2A stabilized on HIV-1 92TH023 Env, whereas the State 2A occupancy on HIV-1 CM244 Env remained similar to unbound Env. Here again, this probably reflects the increased concealment of antigenic sites on the tier-2 isolate. The data presented here provide a means of interpreting the inherent sensitivity of CRF01_AE HIV-1 to ADCC in terms of the conformation of Env. These data also provide a new understanding of the role of vaccine-induced Abs that mediated ADCC during the RV144 trial in Thailand, where HIV-1 CRF01_AE predominates (6). These data underscore the importance of considering Env conformational diversity across different HIV-1 clades when designing more effective HIV-1 interventions and vaccine strategies. This is of particular importance for the development of tailored strategies for enhancing ADCC against CRF01_AE HIV-1, which offers promising avenues for the elimination of cells infected with this prevalent strain in Southeast Asia.
## MATERIALS AND METHODS
## Plasma samples
The FRQS-AIDS and Infectious Diseases Network supports a representative cohort of newly HIV-infected subjects with clinical indication of primary infection (the Montreal Primary HIV Infection Cohort). Plasma samples from 10 deidentified PLWH donors were heat-inactivated and stored at -80°C (24,26). The RV304/SEARCH 013 (NCT00796263) supports a cohort from Bangkok, Thailand who initiated antiretroviral therapy during the chronic phase of infection (CHI) (37). Plasma samples from 10 PLWH donors were heat-inactivated and stored at -80°C until use.
## Cell lines and primary cells
ExpiCHO-S cells (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) were cultured in ExpiCHO Expression media (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) at 37°C, 8% CO 2 with orbital shaking according to the manufacturer's instructions. HEK293T human embryonic kidney cells (obtained from ATCC) were grown as previously described (38). The cell line HEK293T-FIRB with enhanced furin expression was a kind gift from Dr. Theodore C. Pierson (Emerging Respiratory Virus section, Laboratory of Infectious Diseases, NIH, Bethesda, MD, USA), and was cultured at 37°C, 5% CO 2 in complete DMEM made of DMEM (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% (vol/vol) cosmic calf serum (Hyclone, Cytiva Life Sciences, Marlborough, MA, USA), 100 U/mL penicillin, 100 µg/mL streptomycin, and 1 mM glutamine (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) (39). The use of HEK293T-FIRB cells minimizes the presence of uncleaved Env on the virions. The HeLa-derived TZM-bl cell line stably expressing high levels of CD4 and CCR5 receptors and bearing an integrated copy of the luciferase gene under the control of the HIV-1 long-terminal repeat (LTR) was obtained from the former NIH AIDS Reagent Program (BEI catalog HRP-8129) and cultured in the same conditions as HEK293T-FIRB cells (40).
Human peripheral blood mononuclear cells (PBMCs) from three HIV-negative individuals (three males, age range: 40-66 years) obtained by leukapheresis and Ficoll density gradient isolation were cryopreserved in liquid nitrogen until further use. Primary CD4+ T cells were purified from resting PBMCs by negative selection using immunomagnetic beads per the manufacturer's instructions (StemCell Technologies, Vancouver, BC) and were activated with phytohemagglutinin-L (PHA-L, 10 µg/mL) for 48 h and then maintained in RPMI 1640 (Thermo Fisher Scientific, Waltham, MA, USA) complete medium supplemented with 20% FBS, 100 U/mL penicillin/streptomycin and with recombinant IL-2 (rIL-2, 100 U/mL). All cells were maintained at 37°C under 5% CO 2 .
## Plasmids and proviral constructs
The plasmid encoding the soluble CD4 domains 1 and 2 (sCD4 D1D2 ) fused to an anti-6×-His tag, as well as the molecular clones of the heavy and light chains of the anti-HIV-1 Env monoclonal antibodies 17b and 2G12, were kindly provided by Dr. Peter Kwong (NIAID, NIH). Plasmids for expression of NESPylRS AF /hU6tRNA Pyl and eRF1-E55D for the amber codon suppression system were previously described (34). The pNL4-3 ΔRT Δenv plasmid has been previously described (12). pNL4-3.Luc.R-E-provirus was obtained from the former NIH AIDS Reagent Program (BEI catalog HRP-3418). The stop codon in the tat gene of this plasmid was substituted with an ochre stop codon as described (41). Plasmids for the expression of full-length HIV-1 JR-FL Env wild type, which was engineered to have an amber (TAG) stop codon at position N135 in the V1 loop of gp120 and the A1 peptide (GDSLDMLEWSLM) in the V4 loop of gp120 (V1-N135 TAG /V4-A1) have been previously described (30). The HIV-1 CRF01_AE Env expressors from strains 92TH023 and CM244 have been described (13). These plasmids were engineered to insert the A4 peptide (DSLDMLEW) after residue V135 in the V1 loop of gp120, and substitute an amber codon at position N398 in the V4 loop of gp120 (A4-V1/V4-N398 TAG , Fig. 3B). All the indicated residues in HIV-1 JR-FL and HIV-1 CRF01_AE Env are numbered according to the HIV-1 HXBc2 Env sequence.
The IMC of HIV-1 JR-FL was kindly provided by Dr. Dennis Burton (The Scripps Research Institute). The CRF01_AE IMC was previously reported (6). The sequence of HIV-1 CRF01_AE transmitted-founder (T/F) clone 703357 was derived by using a singlegenome amplification strategy. The entire DNA sequence, including both LTRs, was cloned into pUC57 to generate a full-length IMC (GenBank accession numbers: JX448154 and JX448164). The vesicular stomatitis virus G (VSV-G)-encoding plasmid was previously described (24).
## Recombinant sCD4 D1D2 and antibodies
Expression of soluble CD4 domains D1-D2 (sCD4 D1D2 ) fused to an anti-6×-His tag was performed by transfection of ExpiCHO-S cells with plasmid using the ExpiFectamine CHO Transfection Kit (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Purification and preparation of this protein were performed with a previously described strategy (42). Briefly, supernatant containing soluble sCD4 D1D2 was harvested 9 days post-transfection and adjusted to 1 mM NiSO 4 , 20 mM imidazole, and pH 8.0 before binding to the Ni-NTA resin (Invitrogen, Waltham, MA, USA). The resin was washed, and sCD4 D1D2 was eluted from the column with 300 mM imidazole, 500 mM NaCl, 20 mM Tris-HCl pH 8.0, and 10% (vol/vol) glycerol. Elution fractions containing sCD4 D1D2 were pooled and concentrated by centrifugal concentra tors (Sartorius AG, Göttingen, Germany). Final purification was performed through size exclusion chromatography on a Superdex 200 Increase 10/300 GL column (GE Health care, Chicago, IL, USA) followed by concentration as described above.
Expression and preparation of monoclonal antibodies 2G12 and 17b have been described before (42,43). Briefly, ExpiCHO-S cells were co-transfected with plasmids encoding heavy and light chains using the ExpiFectamine CHO Transfection Kit (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instruc tions. Both antibodies were purified from the cell culture supernatant 12 days post-trans fection using protein G affinity resin (Thermo Fisher Scientific, Waltham, MA, USA), subjected to buffer exchange with phosphate-buffered saline (PBS) pH 7.4 (Fisher Bioreagents, Thermo Fisher Scientific, Waltham, MA, USA) and concentrated as described above. Mouse monoclonal antibody targeting HIV-1 p24 capsid protein (anti-p24, catalog no. GTX41618) was purchased from Genetex (Irvine, CA, USA). Anti-6×-His-tag polyclonal antibody (catalog no. PA1-983B), horseradish peroxidase (HRP) conjugated anti-human IgG Fc (catalog no. A18823), and anti-mouse IgG Fc (catalog no. 31455) were purchased from Invitrogen (Waltham, MA, USA). Goat anti-rabbit IgG antibody conjugated to HRP (catalog no. ab205718) was purchased from Abcam (Cambridge, UK).
## Virus production and fluorescent labeling
Non-replicative HIV-1 CRF01_AE Env pseudoviruses for infectivity assays were produced by co-transfecting HEK293T-FIRB cells with either a 1:0.005 or 1:1 mass ratio of plas mid pNL4-3.Luc.R-E-tat_ochre to wild-type or V1-A4/V4-N398 TAG tagged version of HIV-1 CRF01_AE Env expressors, respectively. Plasmids encoding NESPylRS AF /hU6tRNA Pyl and eRF1-E55D were also included along with 0.5 mM TCO* (SiChem GmbH, Bremen, Germany) as previously described (30,41,44,45). Virus was collected 48 h post-transfec tion and pelleted over a 10% sucrose cushion at 25,000 RPM for 2 h at 4°C using an SW32Ti rotor (Beckman Coulter Life Sciences, Brea, CA, USA). Pellets were resuspended in DMEM (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), aliquoted, and stored at -80°C until use.
For smFRET imaging, non-replicative HIV-1 JR-FL and HIV-1 CRF01_AE Env pseudovirions with a single gp120 domain bearing the above-mentioned modifications in the V1 and V4 loops were also produced in the presence of TCO* as previously described (30). Briefly, HEK-293T FIRB cells were co-transfected with plasmids NESPylRS AF /hU6tRNA Pyl and eRF1-E55D, in addition to pNL4-3 ΔRT ΔEnv, and a 20:1 mass ratio of HIV-1 JR-FL or HIV-1 CRF01_AE Env wild-type expressor to the corresponding tagged version. The virus was collected 48 h post-transfection and pelleted as above. Virus pellets were then resuspended in labeling buffer (50 mM HEPES pH 7.0, 10 mM CaCl 2 , 10 mM MgCl 2 ), and incubated overnight at room temperature with 5 µM LD650-coenzyme A (Lumidyne Technologies, New York, NY, USA), and 5 µM acyl carrier protein synthase (AcpS), which labels the A1 (or A4) peptide. The virus was then incubated with 0.5 µM Cy3-tetra zine (Jena Biosciences, Jena, Germany) for 30 min at room temperature, followed by incubation with 60 µM DSPE-PEG2000-biotin (Avanti Polar Lipids, Alabaster, AL, USA) for an additional 30 min at room temperature. Finally, the labeled virus was purified through ultracentrifugation for 1 h at 35,000 RPM using a rotor SW40Ti (Beckman Coulter Life Sciences, Brea, CA, USA), at 4°C in a 6%-30% OptiPrep (Sigma-Aldrich, MilliporeSigma, Burlington, MA, USA) density gradient. Labeled pseudovirions were collected, analyzed by anti-p24 Western blot, aliquoted, and stored at -80°C until their use in imaging experiments.
## Immunoblots
HIV-1 gp120 and p24 proteins, or sCD4 D1D2 , were detected through immunoblot assays as follows. Samples were mixed with 4× Laemmli sample buffer (Bio-Rad, Hercules, CA, USA) supplemented with 2-mercaptoethanol (Fisher Chemical, Hampton, NH, USA) and heated for 5 min at 98°C. Proteins were then resolved by denaturing PAGE using 4%-20% acrylamide gels (Bio-Rad, Hercules, CA, USA). Proteins were then transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA, USA) according to the manufacturer's instructions. After blocking for 1 h at room temperature with 5% (wt/vol) skim milk in PBS-T buffer (PBS and 0.1% [vol/vol] Tween-20, Fisher Scientific, Hampton, NH, USA), membranes were incubated overnight at 4°C with the indicated primary antibodies diluted in blocking buffer. Detection of gp120 was achieved by using a 3 µg/mL dilution of 2G12, while detection of p24 and sCD4 D1D2 was performed with 2 µg/mL dilutions of anti-p24 mAb (GeneTex, Irvine, CA, USA) or rabbit anti-6×-His-tag polyclonal antibody (Invitrogen, Waltham, MA, USA), respectively. Membranes were washed three times with PBS-T and incubated for 1 h at room temperature with a 1/10,000 dilution (vol/vol) in 0.5% (wt/vol) skim milk/PBS-T of HRP-conjugated anti-human IgG Fc or anti-mouse IgG Fc (Invitrogen, Waltham, MA, USA) antibodies for membranes incubated with 2G12 or anti-p24 mAbs, respectively, or a 1/50,000 dilution of HRP-conjugated anti-rabbit IgG antibody (Abcam, Cambridge, UK) was used for membranes incubated with anti-6×-His antibody. After three washes with PBS-T, membranes were developed using SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Due to the use of HEK293T-FIRB cells for virus production, we detected no uncleaved Env (gp160) for either of the CRF01_AE strains evaluated.
## Infectivity assays
TZM-bl cells (2.5 × 10 4 /well) were seeded 24 h before the assay in 24-well plates. Cells were then washed once with DMEM (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) and inoculated with pseudo-typed lentiviruses bearing wild-type or tagged HIV-1 CRF01_AE Env. After 2 h of virus adsorption at 37°C, viral inoculums were removed, and cells were washed with DMEM, followed by the addition of fresh complete phenol red-free DMEM (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Cell supernatants were removed 48 h post-infection. The cells were lysed with Glo Lysis Buffer (Promega, Madison, WI, USA) according to the manufacturer's instructions. Luciferase activity in cell lysates was detected by mixing equal volumes of lysate and Steady-Glo Luciferase Assay System reagent (Promega, Madison, WI, USA) and measured on a Synergy H1 microplate reader (Biotek, Winooski, VT, USA). The luminescence signal from mock-infected cell lysates was subtracted from the signal obtained from infected cells and normalized by the abundance of both envelope gp120 and p24 proteins in viral inoculums, which were determined through densitometric analysis of protein bands observed in immunoblots using ImageJ software v1.52q (NIH, Bethesda, MD, USA). Infectivity was expressed as the percentage of that seen in cells inoculated with wild-type HIV-1 CRF01_AE Env pseudoviri ons.
## smFRET imaging
Labeled HIV-1 JR-FL or HIV-1 CRF01_AE Env pseudovirions were immobilized on streptavi din-coated quartz slides and imaged on a custom-built wide-field prism-based TIRF microscope (41,46). Where indicated, pseudovirions were incubated with 50 µM sCD4 D1D2 and 50 µg/mL 17b mAb for 1 h at room temperature prior to surface immo bilization. Imaging was performed in PBS pH ~ 7.4, containing 1 mM trolox (Sigma-Aldrich, St. Louis, MO, USA), 1 mM cyclooctatetraene (Sigma-Aldrich, St. Louis, MO, USA), 1 mM 4-nitrobenzyl alcohol (Sigma-Aldrich, St. Louis, MO, USA), 2 mM protocatechuic acid (Sigma-Aldrich, St. Louis, MO, USA), and 8 nM protocatechuate 3,4-dioxygenase (Sigma-Aldrich, St. Louis, MO, USA) to stabilize fluorescence and remove molecular oxygen. When indicated, concentrations of sCD4 D1D2 and mAb 17b were maintained during imaging. smFRET data were collected using Micromanager v2.0 at 25 frames/s, processed, and analyzed using SPARTAN software in Matlab (Mathworks, Natick, MA, USA) (47). smFRET traces were identified according to criteria previously described (9); traces meeting those criteria were verified manually. FRET histograms were generated by compiling traces from each of three technical replicates, and the mean probability per histogram bin ± standard error was calculated. Traces were idealized to a five-state HMM (four nonzero-FRET states and a zero-FRET state) using the maximum point likelihood algorithm (48). The idealizations were used to construct Gaussian distributions of each FRET state, which were overlaid on the FRET histograms to visualize the results of the HMM analysis. The HMM analysis was also used to determine the occupancies (fraction of time until photobleaching) in each FRET state for each individual Env molecule. The distributions in occupancies were used to construct violin plots in Matlab, as well as the calculation of mean occupancies and standard errors. Whereas the FRET histograms reflect an estimate of the thermodynamics of the Env conformational equilibrium, the violin plots display the heterogeneity of the population and permit statistical compari sons (P values).
## Viral production and infection of primary CD4+ T cells
VSV-G-pseudotyped HIV-1 viruses were produced by co-transfection of 293T cells with the HIV-1 JRFL or HIV-1 CRF01_AE proviral construct and a VSV-G-encoding vector at a ratio of 3:2 using the polyethylenimine method. Two days post-transfection, cell supernatants were harvested, clarified by low-speed centrifugation (300 × g for 5 min), and concentra ted by ultracentrifugation at 4°C (100,605 × g for 1 h) over a 20% sucrose cushion. Pellets were resuspended in fresh RPMI 1640 complete medium, aliquoted, and stored at -80°C until use.
Primary CD4+ T cells from HIV-1-negative individuals were isolated from PBMCs, activated for 2 days with PHA-L, and then maintained in RPMI 1640 complete medium supplemented with rIL-2. Five to seven days after activation, the cells were spinoculated with the virus at 800 × g for 1 h in 96-well plates at 25°C. All viral productions were titrated on primary CD4+ T cells to achieve similar levels of infection (around 15% of infected cells).
## Flow cytometry analysis of cell-surface staining
Forty-eight hours after infection, HIV-1-infected primary CD4+ T cells were collected, washed with PBS, and transferred to 96-well V-bottom plates. The cells were then incubated for 45 min at 37°C with plasma (1:1,000 dilution). Cells were then washed twice with PBS and stained with anti-human IgG Alexa Fluor 647-conjugated secondary antibody (2 µg/mL), FITC-conjugated mouse anti-human CD4 (Clone OKT4) antibody (1:500 dilution) and AquaVivid viability dye (1:1,000 dilution) (Thermo Fisher Scientific, Cat# L43957) for 20 min at room temperature. Cells were then washed twice with PBS and fixed in a 2% PBS-formaldehyde solution. The cells were then permeabilized using the Cytofix/Cytoperm Fixation/Permeabilization Kit (BD Biosciences, Mississauga, ON, Canada) and stained intracellularly using PE-conjugated mouse anti-p24 mAb (clone KC57; Beckman Coulter, Brea, CA, USA; 1:100 dilution). Samples were acquired on a Fortessa cytometer (BD Biosciences), and data analysis was performed using FlowJo v10.5.3 (Tree Star, Ashland, OR, USA). The percentage of productively infected cells (p24 + , CD4 low) was determined by gating on the living cell population according to viability dye staining (AquaVivid; Thermo Fisher Scientific). Dotted lines represent the limit of detection calculated using five plasmas from uninfected individuals.
Cell surface staining of Env-expressing 293T cells was performed as previously described (7). Briefly, 2 × 10 6 cells were transfected with 7 µg of Env expressor and 1 µg of a green fluorescent protein (GFP) expressor (pIRES-GFP) with the calcium-phosphate method. At 48 h post-transfection, 293T cells were stained with anti-Env antibodies (5 µg/mL) or plasma at a 1:1,000 dilution. To normalize Env expression, we used the anti-CD4BS bNabs N6, known to potently neutralize 98% of circulating isolates, including CRF01_AE and clade B strains (49). As such, plasma binding (median fluorescent intensity [MFI]) was normalized to the MFI of N6 bNabs obtained for each respective Envs. Dotted lines represent the limit of detection calculated using five plasmas from uninfected individuals.
## ADCC assay
ADCC activity was measured using a FACS-based infected cell elimination assay 48 h after infection. The HIV-1-infected primary CD4+ T cells were stained with AquaVivid viability dye and cell proliferation dye eFluor670 (Thermo Fisher Scientific) and used as target cells. Resting autologous PBMCs were stained with cell proliferation dye eFluor450 (Thermo Fisher Scientific) and used as effector cells. The HIV-1-infected primary CD4+ T cells were co-cultured with autologous PBMCs (effector:target ratio of 10:1) in 96-well V-bottom plates in the presence of plasma from PLWH (dilution 1:1,000) for 5 h at 37°C. After the 5 h incubation, cells were then washed once with PBS and stained with FITC-conjugated mouse anti-human CD4 (Clone OKT4) antibody for 10 min at room temperature. Cells were then washed twice with PBS and fixed in a 2% PBS-formaldehyde solution. The cells were then permeabilized and stained intracellularly for p24 as described above. Samples were acquired on a Fortessa cytometer (BD Biosciences), and data analysis was performed using FlowJo v10.5.3 (Tree Star, Ashland, OR, USA). The percentage of infected cells (p24 + , CD4 low) was determined by gating on the living cell population according to viability dye staining (AquaVivid; Thermo Fisher Scientific). The percentage of ADCC was calculated with the following formula: (% of p24 + CD4 low cells in Targets plus Effectors) -(% of p24 + CD4 low cells in Targets plus Effectors plus plasma)/(% of p24 + CD4 low cells in Targets) × 100. Dotted lines represent the limit of detection calculated using five plasmas from uninfected individuals.
## Statistical analysis
Statistics for infectivity assays were determined using GraphPad Prism version 10.2.3 (GraphPad, San Diego, CA, USA). Every data set was tested for statistical normality, and this information was used to apply the appropriate (parametric or nonparametric) statistical test. Statistical significance measures (P values) of FRET state occupancies were determined by one-way ANOVA followed by multiple comparison testing in Matlab. In all cases, P values <0.05 were considered statistically significant.
## References
1. Haynes, Gilbert, Mcelrath et al. (2012) "Immune-correlates analysis of an HIV-1 vaccine efficacy trial" *N Engl J Med*
2. Tomaras, Ferrari, Shen et al. (2013) "Vaccineinduced plasma IgA specific for the C1 region of the HIV-1 envelope blocks binding and effector function of IgG" *Proc Natl Acad Sci*
3. Dai, Peng, Sun et al. (2024) "Distinct clusters of HIV-1 CRF01_AE in Zhejiang, China: high-risk transmission cluster 4 requires heightened surveillance" *Infect Drug Resist*
4. Khairunisa, Indriati, Megasari et al. (2024) "Spatial-temporal transmission dynamics of HIV-1 CRF01" *_AE in Indonesia. Sci Rep*
5. Hemelaar, Elangovan, Yun et al. (2020) "Global and regional epidemiology of HIV-1 recombinants in 1990-2015: a systematic review and global survey" *Lancet HIV*
6. Prévost, Zoubchenok, Richard et al. (2017) "Influence of the envelope gp120 phe 43 cavity on HIV-1 sensitivity to antibody-dependent cellmediated cytotoxicity responses" *J Virol*
7. Prévost, Tolbert, Medjahed et al. (2020) "The HIV-1 env gp120 inner domain shapes the Phe43 cavity and the CD4 binding site" *mBio*
8. Lu, Ma, Castillo-Menendez et al. (2019) "Associating HIV-1 envelope glycoprotein structures with states on the virus observed by smFRET" *Nature*
9. Alsahafi, Bakouche, Kazemi et al. (2019) "An Asymmetric Opening of HIV-1 Envelope Mediates Antibody-Dependent Cellular Cytotoxicity" *Cell Host Microbe*
10. Ma, Lu, Gorman et al. (2018) "HIV-1 env trimer opens through an asymmetric intermediate in which individual protomers adopt distinct conformations"
11. Herschhorn, Ma, Gu et al. (2016) "Release of gp120 restraints leads to an entrycompetent intermediate state of the HIV-1 envelope glycoproteins" *mBio*
12. Munro, Gorman, Ma et al. (2014) "Conforma tional dynamics of single HIV-1 envelope trimers on the surface of native virions" *Science*
13. Zoubchenok, Veillette, Prévost et al. (2017) "Histidine 375 modulates CD4 binding in HIV-1 CRF01_AE envelope glycoproteins" *J Virol*
14. Prévost, Chen, Zhou et al. (2023) "Structure-function analyses reveal key molecular determinants of HIV-1 CRF01_AE resistance to the entry inhibitor temsavir" *Nat Commun*
15. Checkley, Luttge, Freed (2011) "HIV-1 envelope glycoprotein biosynthesis, trafficking, and incorporation" *J Mol Biol*
16. Willey, Bonifacino, Potts et al. (1988) "Biosynthesis, cleavage, and degradation of the human immunodeficiency virus 1 envelope glycoprotein gp160" *Proc Natl Acad Sci*
17. Earl, Doms, Moss (1990) "Oligomeric structure of the human immunodeficiency virus type 1 envelope glycoprotein" *Proc Natl Acad Sci*
18. Freed, Myers, Risser (1989) "Mutational analysis of the cleavage sequence of the human immunodeficiency virus type 1 envelope glycoprotein precursor gp160" *J Virol*
19. Mccune, Rabin, Feinberg et al. (1988) "Endoproteolytic cleavage of gp160 is required for the activation of human immunodeficiency virus" *Cell*
20. Wyatt, Sodroski (1998) "The HIV-1 envelope glycoproteins: fusogens, antigens, and immunogens" *Science*
21. Allan, Coligan, Barin et al. (1985) "Major glycoprotein antigens that induce antibodies in AIDS patients are encoded by HTLV-III" *Science*
22. Robey, Safai, Oroszlan et al. (1985) "Characterization of envelope and core structural gene products of HTLV-III with sera from AIDS patients" *Science*
23. Veillette, Coutu, Richard et al. (2015) "The HIV-1 gp120 CD4bound conformation is preferentially targeted by antibody-dependent cellular cytotoxicity-mediating antibodies in sera from HIV-1-infected individuals" *J Virol*
24. Veillette, Désormeaux, Medjahed et al. (2014) "Interaction with cellular CD4 exposes HIV-1 envelope epitopes targeted by antibodydependent cell-mediated cytotoxicity" *J Virol*
25. Jonathan, Pacheco, Gohain et al. (2016) "Co-receptor binding site antibodies enable CD4-mimetics to expose conserved anti-cluster A ADCC epitopes on HIV-1 envelope glycoproteins" *EBioMedicine*
26. Richard, Veillette, Brassard et al. (2015) "CD4 mimetics sensitize HIV-1-infected cells to ADCC" *Proc Natl Acad Sci*
27. Prévost, Richard, Medjahed et al. (2018) "Incomplete downregulation of cd4 expression affects HIV-1 env conformation and antibody-dependent cellular cytotoxicity responses" *J Virol*
28. Prévost, Richard, Ding et al. (2018) "Envelope glycoproteins sampling states 2/3 are susceptible to ADCC by sera from HIV-1-infected individuals" *Virology (Auckl)*
29. Rajashekar, Richard, Beloor et al. (2021) "Modulating HIV-1 envelope glycoprotein conformation to decrease the HIV-1 reservoir" *Cell Host Microbe*
30. Marchitto, Richard, Prévost et al. (2026) *Full-Length Text Journal of Virology*
31. Hahn, Munro, Pazgier "Smith AB 3rd, Finzi A. 2024. The combination of three CD4-induced antibodies targeting highly conserved Env regions with a small CD4-mimetic achieves potent ADCC activity" *J Virol*
32. Anand, Prévost, Baril et al. (2019) "Two families of env antibodies efficiently engage fc-gamma receptors and eliminate HIV-1-infected cells" *J Virol*
33. Marchitto, Richard, Prévost et al. (2024) "The combination of three CD4-induced antibodies targeting highly conserved Env regions with a small CD4-mimetic achieves potent ADCC activity" *J Virol*
34. Richard, Nguyen, Tolbert et al. (2021) "Across functional boundaries: making nonneutralizing antibodies to neutralize HIV-1 and mediate fc-mediated effector killing of infected cells"
35. Nikić, Kang, Girona et al. (2015) "Labeling proteins on live mammalian cells using click chemistry" *Nat Protoc*
36. Richard, Prévost, Alsahafi et al. (2018) "Impact of HIV-1 envelope conformation on ADCC responses" *Trends Microbiol*
37. Decker, Bibollet-Ruche, Wei et al. (2005) "Antigenic conservation and immunogenicity of the HIV coreceptor binding site" *J Exp Med*
38. Schuetz, Deleage, Sereti et al. (2014) "Initiation of ART during early acute HIV infection preserves mucosal Th17 function and reverses HIV-related immune activation" *PLoS Pathog*
39. Finzi, Xiang, Pacheco et al. (2010) "Topological layers in the HIV-1 gp120 inner domain regulate gp41 interaction and CD4-triggered conformational transitions" *Mol Cell*
40. Mukherjee, Dowd, Manhart et al. (2014) "Mechanism and significance of cell type-dependent neutralization of flaviviruses" *J Virol*
41. Platt, Wehrly, Kuhmann et al. (1998) "Effects of CCR5 and CD4 cell surface concentrations on infections by macrophage tropic isolates of human immunodeficiency virus type 1" *J Virol*
42. Jain, Govindan, Berkman et al. (2023) "Regulation of ebola GP conformation and membrane binding by the chemical environment of the late endosome" *PLoS Pathog*
43. Díaz-Salinas, Li, Ejemel et al. (2022) "Conformational dynamics and allosteric modulation of the SARS-CoV-2 spike" *Elife*
44. Díaz-Salinas, Jain, Durham et al. (2024) "Single-molecule imaging reveals allosteric stimulation of SARS-CoV-2 spike receptor binding domain by host sialic acid" *Sci Adv*
45. Das, Govindan, Nikić-Spiegel et al. (2018) "Direct visualization of the conformational dynamics of single influenza hemagglutinin trimers" *Cell*
46. Das, Bulow, Diehl et al. (2020) "Conformational changes in the Ebola virus membrane fusion machine induced by pH, Ca2+, and receptor binding" *PLoS Biol*
47. Blakemore, Burnett, Swanson et al. (2021) "Stability and conformation of the dimeric HIV-1 genomic RNA 5'UTR" *Biophys J*
48. Juette, Terry, Wasserman et al. (2016) "Single-molecule imaging of non-equilibrium molecular ensembles on the millisecond timescale" *Nat Methods*
49. Qin, Auerbach, Sachs (2000) "A direct optimization approach to hidden Markov modeling for single channel kinetics" *Biophys J*
50. Huang, Kang, Ishida et al. (2016) "Identification of a CD4-bindingsite antibody to HIV that evolved near-pan neutralization breadth" *Immunity*
51. (2026) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645932&blobtype=pdf
|
# Mouse-adapted SARS-CoV-2 Omicron BA.5 infection induces post-acute lung fibrosis in BALB/c mice
John Powers, Sarah Leist, Naveenchandra Suryadevara, Seth Zost, Elad Binshtein, Anfal Abdelgadir, Michael Mallory, Caitlin Edwards, Kendra Gully, Miranda Hubbard, Mark Zweigart, Alexis Bailey, Timothy Sheahan, James Crowe, Stephanie Montgomery, Jack Harkema, Ralph Baric, Ralph Ai110700, Baric, Ralph Ai158571
## Abstract
Following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron BA.1, subsequent Omicron sub-lineages have continued to emerge, challeng ing the development of intervention and prevention strategies, including monoclonal antibodies and vaccines. To better understand the pathogenic effects caused by Omicron BA.5 infection, we developed a mouse-adapted virus with overt disease burden in BALB/c mice. Acute disease was characterized by significant weight loss and lung dysfunction following high-dose challenges. In survivor animals that were followed through 107 days post-infection, subpleural fibrosis with associated tertiary lymphoid structures was noted. Serum from these mice demonstrated potent neutralization against BA.5, with substantially reduced neutralization titers against early epidemic, zoonotic, and more recent contemporary XBB.1.5 variants. Intervention with pre-clin ical monoclonal antibodies revealed that robust protection from BA.5-induced lung disease was possible after prophylactic administration. Together, this model enables the investigation of therapeutic approaches for both acute and post-acute sequelae of COVID-19.
IMPORTANCETo best combat the evolving landscape of SARS-CoV-2 variants of interest and variants of concern, the development of effective small animal models is of critical importance. Herein, we describe the development of a model system in BALB/c mice to study the effects of SARS-CoV-2 BA.5 S gene in both acute and chronic disease manifes tations. Intriguingly, we determined that fibrotic lung disease with tertiary lymphoid structures was a prominent feature in the lungs of mice that survived through the acute phase of infection. This is a prominent concern in human patients that survive the initial infection insult. As such, and most critically, the model system presented here provides researchers with an effective pathway in which long COVID manifestations and potential interventions can be studied. KEYWORDS sarbecovirus, fibrotic lung disease, mouse models, monoclonal antibodies, SARS-CoV-2 Omicron BA.5 S evere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virions contain an approximately 30 kb positive-sense single-stranded viral genome that encodes numerous structural proteins (spike, membrane, envelope, and nucleocapsid), non-struc tural proteins (nsp1-16), and several accessory genes (1, 2). Following its emergence in late 2019, SARS-CoV-2 rapidly evolved into numerous variants of interest and variants of concern (VOC). The Omicron B.1.1.529 VOC (BA.1) was of significant interest because it encoded a substantial number of amino acid changes within the main antigenic site, the spike gene (S), especially within the receptor-binding domain (RBD), and the receptorbinding motif (RBM). BA.5, which emerged in early 2022, is a descendant lineage of
Omicron that contained additional alterations within the RBD/RBM protein sequence. Due to the role of the spike protein (S) in cellular attachment to angiotensin-con verting enzyme 2 (ACE2) and subsequent viral entry following its cleavage by transmem brane protease serine 2 (TMPRSS2), the expanded spike variation was associated with enhanced receptor binding, antibody evasion, and reduced efficacy of both natural and vaccine-induced immunity (3). Amino acid mutations in these domains enabled escape from neutralizing antibodies (4)(5)(6)(7)(8)(9).
Inconsistent results in mouse models using BA.5 isolates were observed by different research groups regarding the degree of pathogenicity as compared to BA.1 or previous ancestral viruses (10,11). However, in both cases, disease severity was overall attenuated compared to early VOCs, like Alpha and Delta, likely mitigated by differences in the clinical isolate, inoculum dose, mouse strain, sex, and age of animals (12)(13)(14). Conse quently, we generated recombinant SARS-CoV-2 viruses to study the effect of the BA.5 S protein in pathogenicity. As such, we utilized a reverse genetics infectious clone system in order to replace the spike gene of our previously described MA10 virus with the spike gene sequence from BA.5 (designated BA.5 MA) (15)(16)(17). In parallel, we also recovered a derivative strain without mouse adaptations in which the viral ORF7a was replaced with a gene encoding the reporter protein nanoluciferase (nLuc) to be used as an indicator virus in live-virus neutralization assays (BA.5 nLuc) (15)(16)(17)(18)(19)(20). BA.5 MA infection in young (14-to 16-week-old) or aged (10-to 12-month-old) female BALB/c mice caused severe diffuse alveolar damage with accompanying mortality and high virus titers in the lungs and nasal turbinates, which was significantly attenuated by prophylactic administration of monoclonal antibodies (mAbs). In contrast, we also produced viruses expressing the BA.2 Omicron S protein on the same viral backbones to be used in in vivo and in vitro assays as were performed with BA.5. Infection of aged female BALB/c mice with BA.2 MA failed to produce overt disease symptoms such as weight loss and mortality but did replicate to high viral titers in both the lungs and nasal turbinates.
SARS-CoV-2 symptomatic or asymptomatic infection may progress to post-acute sequelae of SARS-CoV-2 (PASC), a frequent chronic disease syndrome that includes a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems and lasts for weeks to years in about ~10% of SARS-CoV-2 survi vors (21)(22)(23). In the respiratory tract, shortness of breath, cough, persistent fatigue, post-exertional malaise, and diagnosable conditions like interstitial lung disease and hypoxemia are among the most frequent chronic disease phenotypes. Omicron-related infections are reported to cause less frequent PASC disease in humans as compared to ancestral VOCs (24). To begin to dissect potential virus-centric differences in PASC disease potential, we first sought to develop Omicron-based models of acute and post-acute lung pathogenesis using mouse-adapted SARS-CoV-2 bearing BA.5 or BA.2 S proteins. We show that S is a major driver of acute pathogenesis and that severe acute disease can drive post-acute lung sequelae. The humoral immune response was dominated by homotypic responses, and the magnitude of heterotypic neutralizing responses correlated with genetic relatedness to Omicron. We also demonstrate these models can be used to evaluate antiviral therapies. Altogether, these models can be leveraged to study pathogenic mechanisms of acute and long COVID and whether pre-existing long COVID conditions impact acute/chronic disease potential for future SARS-CoV-2 VOC or even other respiratory virus infections.
## RESULTS
## Recovery of Omicron recombinant viruses
Using reverse genetics, we generated a mouse-adapted SARS-CoV-2 MA10 recombinant bearing the BA.2 or BA.5 subvariant spike (BA.2 MA and BA.5 MA) and a related reporter virus where we replaced ORF7a with nanoluciferase (BA.2 nLuc and BA.5 nLuc) (15)(16)(17). Genome schematics of the infectious clone constructs are shown in Fig. S1. To assess replication kinetics, the growth of the two BA.5 recombinant viruses was compared to that of the ancestral strains SARS-CoV-2 MA10 and D614G nLuc, at an MOI of 0.001 on Vero E6 cells. The growth kinetics of BA.5 MA and BA.5 nLuc were the same, indicating that replication was not impacted by the inclusion of the reporter gene. The ancestral pandemic strain (SARS-CoV-2 D614G) had a growth advantage over both BA.5 viruses at most times assessed (Fig. 1). Importantly, parental SARS-CoV-2 MA10 virus grew similarly to both BA.5 viruses early in the kinetic, yet parental MA10 titers exceeded those of BA.5 viruses at later timepoints (Fig. 1). Altogether, these data demonstrate that our BA.5. spike recombinant viruses grew efficiently in Vero cells.
## Pathological features of BA.5 MA infection in young mice
We have previously established a model of SARS-CoV-2 pathogenesis using mouse-adap ted SARS-CoV-2 MA10 in female BALB/c mice (15). To evaluate BA.5 MA pathogenesis in a similar model, we infected 14-to 16-week-old adult female BALB/c mice intranasally with different doses, 10 4 or 10 5 plaque-forming units (PFUs), evaluating dose-depend ent disease outcomes. While both cohorts experienced weight loss, the low-dose cohort only experienced a 10.7% group mean weight loss nadir at 4 days post-infec tion (dpi). In contrast, the high-dose cohort continued to lose weight throughout the study, eventually exceeding 20% of starting group mean weights (Fig. 2A). Gross lung discoloration (GLD) score is a semi-quantitative measure of acute lung damage associated with emerging CoV replication, indicative of edema and diffuse alveolar damage (15,25). Like weight loss, GLD scores were significantly increased in the high-dose cohort as compared to the low-dose group (Fig. 2B). Similarly, mortality was only observed in the high-dose group (Fig. 2C). Regardless of inoculum dosage, the levels of viral replication within the lungs and nasal turbinates showed no significant difference between the two cohorts (Fig. 2D andE). Acute lung injury (Fig. 2F) and diffuse alveolar damage (Fig. 2G) were assessed using histological scoring schema that have been utilized for multiple emerging CoVs (15,25). Both scoring metrics (Fig. 2F andG) revealed that acute lung injury was significantly elevated regardless of virus dose, which was evident in photomicrographs of representative mock-infected (phosphate buffered saline [PBS] inoculation) (Fig. 2H), 10 4 PFU (Fig. 2I), and 10 5 PFU (Fig. 2J) infected animals sacrificed at 4 dpi.
## Pathological features of infection in aged mice
Viral pathogenesis was then evaluated in 10-to 12-month-old mice infected with 10 4 or 10 5 PFU of BA.5 MA. Both high-and low-dose cohorts rapidly lost weight approaching 80% of starting weight by 4 dpi (Fig. 3A). GLD scores were elevated in the high-dose cohort as compared to the low-dose group (Fig. 3B). Although trends in body weight loss were similar up to 5 dpi, high-dose infection was uniformly lethal (Fig. 3C), while the majority of low-dose infected animals survived out to 15 dpi. As observed in the younger mice above, lung and nasal titers between the infected groups did not differ significantly, regardless of dose (Fig. 3D andE). As expected, infection was associated with increases in the histopathological measures of acute lung injury (Fig. 3F) and diffuse alveolar damage (Fig. 3G). Lung tissue sections were evaluated for fibrotic lesions by Picrosirius red staining (26). Like acute lung injury scores, the frequency of profibrotic lesions increased with infection (Fig. 3H). At later times post-infection (7, 15 dpi), we evaluated tissue sections for the prevalence of pro-fibrotic regions (Fig. 3H). Like before, only the low-dose group had elevated fibrosis scores over mock-infected animals, and importantly, no high-dose animals survived to 15 dpi and thus could not be included in this analysis. Counterintuitively, the low-dose group had elevated histological scores (Fig. 3F through H) as compared to the high-dose group, but this was likely driven by survivor bias since the few high-dose infected animals that survived to later times post-infection likely had diminished disease severity compared to those that perished at earlier times. Histologic manifestations of acute lung injury, including accumulation of proteinaceous debris in the airways, inflammatory cell infiltration, and alveolar septal thickening are shown in Fig. 3I through K.
## Monoclonal antibody treatment abrogates BA.5 MA-inflicted disease
From a selected panel of representative SARS-CoV-2 reactive human monoclonal antibodies (mAbs), we identified two that showed robust neutralization titers against BA.5 nLuc (Fig. 4A). Two mAbs (COV2-3605 and COV2-3678) were compared against a recombinant version of a neutralizing SARS-CoV-2 monoclonal antibody (rLY-CoV1404) that served as a positive control, and an isotype-matched recombinant version of a human mAb antibody recognizing the unrelated dengue virus envelope protein (rDENV-2D22) that served as a negative control. In addition, a mock treatment group received an equivalent volume of PBS. In a neutralization assay against BA.5 nLuc, all three SARS-CoV-2 antibodies potently neutralized the virus, with IC 50 titers of 9.7 ng/mL for COV2-3605, 51.7 ng/mL for COV2-3678, and 3.2 ng/mL for rLY-CoV1404 (Fig. 4A). Both, COV2-3605 and COV2-3678 are encoded by the human antibody variable gene segment IGHV3-53. Many IGHV3-53and IGHV3-66-encoded mAbs constitute a public clonotype, and this class of antibodies has been recurrently isolated from human subjects following SARS-CoV-2 infection or vaccination (27)(28)(29)(30). Negative-stain electron microscopy (EM) revealed COV2-3605 Fab and COV2-3678 Fab bound to RBDs in the open conformation of the spike, consistent with the fact that members of this public clonotype target the semicryptic class I antigenic site (Fig. 4B) (31). Negative-stain EM data collection statistics are provided in Table S1.
To assess the prophylactic efficacy of mAbs, 10-to 12-month-old female BALB/c mice were treated with antibodies via intraperitoneal injection of 200 µg of mAb, and 12 hours later, mice were inoculated with a lethal dose of 10 5 PFU of BA.5 MA. Mice treated with COV2-3605 or rLY-CoV1404 showed minimal or no weight loss. In contrast, ~10% transient weight loss was observed at 5 and 6 dpi for the COV2-3678-treated cohort (Fig. 4C). Importantly, negative control group mice (i.e., PBS or isotype-matched control mAb rDENV-2D22) showed rapid weight loss through 7 dpi, with all animals succumbing to infection or reaching humane endpoints for euthanasia by 7-8 dpi (Fig. 4D). Similarly, only negative control group animals had elevated GLD scores (Fig. 4E). Minimal break through infection was noted in the COV2-3678-treated mice, with one of five mice exhibiting low replicating virus at 4 dpi (Fig. 4F). In contrast, no live virus was detected in the COV2-3605 and rLY-CoV1404 mAb-treated cohorts. Viral titers in mAb-treated cohorts were significantly reduced compared to PBS-treated cohorts, which had titers approaching 10 5 PFU/lobe (P < 0.0001). Modeling Omicron-associated post-acute lung pathology with BA.2 and BA.5 MA
We have previously developed a model of PASC-like lung pathology using a mouseadapted ancestral pandemic strain SARS-CoV-2 MA10 in aged mice (26). Omicron-related infections are reported to cause less frequent PASC disease in humans as compared to ancestral VOCs (24). To understand the potential role of spike protein variation in this disease process, we generated a mouse-adapted recombinant BA.2 Omicron spike virus (BA.2 MA) that is different in 4 amino acids in the spike gene from the BA.5 MA spike-containing virus. To understand BA.2 MA PASC potential, we infected cohorts of 10-to 12-month-old female BALB/c mice with PBS (mock) or 10 5 PFU of BA.2 MA and followed animals through 120 dpi (BA.2 MA) to evaluate virologic and pathologic outcomes. Unlike with the BA.5 MA infection noted above, BA.2 MA-infected mice exhibited minimal weight loss (Fig. 5A), despite viral replication in both the lung and nasal turbinates achieving comparable levels to those seen in the BA.5 MA-infected mice described above, demonstrating the role of spike protein variation on disease progres sion and severity (Fig. 5B). Congruent with body weight loss, gross pathology (Fig. 5C), and histologic measures of lung fibrosis (Fig. 5D) were largely absent and nonremarkable post-BA.2 MA infections. We next measured the magnitude and durability of the neutralizing antibody response in BA.2 MA challenged mice using antigenically homologous (BA.2 nLuc) or heterologous (SARS-CoV-1 nLuc, SARS-CoV-2 D614G nLuc, SARS-CoV-2 BA.1 nLuc, SARS-CoV-2 BA.5 nLuc, and SARS-CoV-2 XBB.1.5 nLuc) Sarbecovi ruses (20). As expected, the homotypic neutralization response against BA.2 was the most robust (~10,000 IC 50 ) (Fig. 5E). The magnitude of the heterotypic responses varied by genetic proximity to BA.2, where responses were 10-fold lower to ancestral pandemic strains (D614G) and were also reduced against more future emerging Omicron variants (e.g., BA.5, XBB1.5) (Fig. 5E). To understand the long-term consequences of BA.5 MA infection, we infected similarly aged mice noted above with PBS or 10-fold less virus (10 4 PFU) to ensure disease with the majority of animals surviving. Congruent with Fig. 3, BA.5-infected animals lost an average of 17.4% ± 7.6% by 7 dpi but recovered by 30 dpi (Fig. 6A) with approximately 60% survival (Fig. 6B). Gross pathology GLD scores, which are most evident during the acute phase of infection (Fig. 3B), were low but measurable at 15 dpi and waned over time (Fig. 6C). Neither replication-competent virus via plaque assay nor viral RNA via quantitative reverse transcription-PCR was detected in lung tissue, and viral nucleocapsid antigen was not detected in lung, liver, kidney, or spleen tissue sections at any times assessed (data not shown). Lung fibrosis was evident by Picrosirius red staining with mean scores averaging between ~2 and 3 over time, indicating fibrotic lesions involving 6% to 50% of the lung parenchyma (Fig. 6D). By day 107, about 20% of the BA.5 MA virus-infected mice had resolved the most prominent long COVID lesions. Unlike mock-infected animals, BA.5 infection was associated with subpleural chronic alveolitis with scattered tertiary lymphoid structures and areas of PSR-stained interstitial fibrosis in the alveolar parenchyma at both 30 and 107 dpi (Fig. S2), thus confirming an association between chronic inflammation and fibrosis. Like BA.2 infection, homotypic neutralizing antibody responses were most robust at 30 dpi and waned through 107 dpi (Fig. 6E). As before, BA.5 serum poorly neutralized ancestral and more contemporary SARS-CoV-2 VOC and zoonotic strains. Overall, the magnitude of the heterotypic responses was driven by genetic proximity to the homotypic antigen like BA.2 above and demonstrated that Omicron spike variation can have a profound impact on acute and post-acute pathogenesis in our mouse model.
## DISCUSSION
SARS-CoV-2, the etiological agent responsible for the COVID-19 pandemic, has caused significant global morbidity and mortality, with excess mortality estimates approaching 20-25 million or more (32). In most populations, Omicron acute disease severity is reduced compared to ancestral VOCs but nevertheless continues to cause a significant disease burden with an estimated 32,000 to 50,000 deaths from October 2024 through June 2025 primarily concentrated among the elderly and certain immunocompromised populations (33)(34)(35). Both ancestral and contemporary SARS-CoV-2 VOC infections can cause chronic multiorgan system-level disease phenotypes, which have been termed long COVID or PASC. Long COVID disease symptoms can persist for months after resolution of acute infection (36,37). Although long COVID symptoms are more likely to occur after primary infection and the overall rates of long COVID diagnoses are decreas ing over time, millions of people have been impacted by long COVID, and there remains a risk of post-acute sequelae associated with reinfection (36,38,39). While some human studies suggest that persistence of viral RNA is a principal driver of PASC, the vast minority of chronic cases have little if any detection of persistent viral RNA, and a 15-day antiviral treatment with Paxlovid did not improve outcomes suggesting that at least persistent replication was not playing a major role in PASC (40,41). Our data, and those of others, suggest that aberrant epithelial-immune cell interactions and/or lung repair defects drive chronic post-acute lung disease (42,43). Altogether, these data show that important questions remain regarding the pathogenic mechanisms of long COVID and whether pre-existing long COVID conditions impact acute/chronic disease potential for future SARS-CoV-2 VOC or even other respiratory virus infections. Our ancestral and contemporary SARS-CoV-2 acute and chronic disease models presented herein provide a platform to systematically address these issues and assess the impact of medical countermeasures in models with more contemporary strains. BA.5 MA infection in mice causes severe acute infection with histologic manifesta tions of acute lung injury (e.g., diffuse alveolar damage) reminiscent of infection with SARS-CoV-2 MA10, a mouse-adapted virus based on the original pandemic strain. Here, we find that BA.5 and BA.2 spike glycoprotein genes attenuate MA10 pathogenesis in vivo in aged mice, which is uniformly lethal at 10 4 or greater dose in aged animals (15). We show an age-related exacerbation of pathogenesis like that observed in humans. Nevertheless, in humans, females are more likely to develop long COVID, and we aimed to generate data sets directly comparable to those from SARS-CoV-2 MA10-infected female mice (26,44). Importantly, we demonstrate here that lower doses of BA.5 MA result in post-acute pathogenesis in the lung as evidenced by organizing pneumonia with tertiary lymphoid structures and fibrotic lesions through 107 dpi. Thus, this model could be leveraged to understand the impact of post-acute sequelae on subsequent viral infections, vaccine efficacy, and medical countermeasures to treat acute and chronic pathogenesis (17). In contrast, the BA.2 spike fully attenuated acute and chronic disease phenotypes in aged mice, despite replicating to similar titers at early times post-infec tion. The discordance among BA.2 and BA.5 MA outcomes in our model demonstrates the potential for S variation to impact acute and chronic pathogenesis. Our spike recombinant viruses were constructed within an isogenic SARS-CoV-2 background, thus genetically are only different in S at four positions (Δ69-70, L452R, F486V, and the wild-type Q493 in BA.5) all of which could be potential drivers of disease (45)(46)(47)(48). Three of these mutations fall within the RBM of S. Previous studies have demonstrated that the L452R mutation in RBD enhanced ACE2 binding, fusogenicity, and infectivity, suggesting this mutation could be responsible for driving the differences in pathogenicity amongst BA.2 and BA.5 MA (49,50). Notably, our group has previously demonstrated that a single amino acid polymorphism in S protein at position F486 can drive widely disparate disease outcomes in mice infected with XBB.1 MA versus XBB.1.5 MA (20). Despite BA.2 and BA.5 MA achieving similar titers in mice at 2 dpi, S variation could impact the kinetics of replication prior to 2 dpi and/or differences in epithelial cell tropism leading to the disparate outcomes in our data. Regardless of the mechanism, our data indicate that changes in ACE2 binding affinity could drive differential pathogenesis further implicating S protein as a major driver of disease.
COVID-19 humoral immunity acquired from natural infection is highly variable in magnitude and durability compared to vaccine-elicited immunity. In general, more severe infections oftentimes elicit higher and more durable neutralizing responses as compared to mild infections in humans (51)(52)(53). While the pathogeneses of BA.2 and BA.5 MA differed in our model, the levels of viral replication were similar at early times, and as such, the magnitude of the homotypic neutralizing responses was similar. However, neutralization breadth was limited, especially against heterotypic ancestral strains (e.g., SARS-CoV, SARS-CoV-2 D614G, and Omicron BA.1) and future, more distantly evolved Omicron-related VOC strains (i.e., XBB.1.5). Nevertheless, detailed study of the antibody repertoire after natural infection or vaccination provides an opportunity for the discovery of novel therapeutic monoclonal antibodies which have demonstrated clinical utility in treating COVID-19 (54,55). Unfortunately, the emergence of the Omicron lineage was associated with a significant decline in Food and Drug Administrationapproved mAb performance due to the accrual of mutations in S and especially those in the RBD and RBM (56)(57)(58). To understand the breadth of efficacy and mechanism of action of two mAbs (COV2-3605 and COV2-3678) isolated from an individual following a BA.1 breakthrough infection, we evaluated these mAbs in our BA.5 models. These mAbs target the RBM of the SARS-CoV-2 RBD and are encoded by antibody variable gene IGHV3-53, making them members of a previously described public clonotype using IGHV3-53/IGHV3-66 genes. COV2-3605 and COV2-3678 potently neutralized BA.5 in vitro and in vivo, demonstrating that continued surveillance of patients with break through infections can facilitate the identification of novel monoclonal antibodies with great breadth and efficacy. The use of in vivo models in this context is of importance as multiple factors including biodistribution, route of administration, and therapeutic windows can impact the efficacy of mAbs and their neutralizing competency compared to in vitro data. Importantly, the class I antigenic site targeted by members of this public clonotype continues to be a major target of human immunity and has accumulated further antigenic substitutions (59). However, some mAbs with this gene usage have been described that show extraordinary breadth of reactivity and resilience to escape (60).
In summary, we show that the Omicron spike can differentially elicit acute and chronic pathogenesis in BALB/c mice, including progression to long-COVID-like pulmonary disease. Disease severity is tunable based on the nature of the S protein and animal age. These findings could reflect the impact of initial viral infection inoc ulum on immediate response to the insult, perhaps suggesting that cytokine storm phenomena may be driving enhanced disease profiles at higher doses. Most critically, the model enables study of acute and chronic disease mechanisms in pulmonary and extra-pulmonary compartments and provides a platform for the development of medical countermeasures that target both acute and chronic disease manifestations. We show that homotypic immunity dominated the primary Omicron humoral response and that cross-protection prominently waned against ancestral pandemic strains. Since immunity to SARS-CoV-2 S has limited durability and S evolution is likely to continue to retain endemicity, our global immunity in decades to come is unlikely to effectively combat zoonoses from ancestral SARS-like viruses which circulate among reservoir species, providing opportunities for future emergence events (Fig. S3) (61)(62)(63)(64)(65)(66)(67)(68)(69). A weakness of this study is the focus on disease in female mice, as studies with MA10 have revealed more significant virulent acute and chronic disease phenotypes in the lungs of males (70). The study of acute and chronic CoV disease mechanisms as well as vaccine and countermeasure development remains essential for pandemic preparedness, providing a rationale for the continuation of CoV model development like those described herein.
## MATERIALS AND METHODS
## Virus and cells
Using reverse genetics, we recovered the BA.5 and BA.2 wild-type spike gene (S) sequence in the background of previously described mouse-adapted mutations (BA.5 MA and BA.2 MA) (GenBank under accession numbers PV800150 and PV800152) (20,21).
A second set of viruses encoding the BA.5 or BA.2 S protein sequence that expressed nanoluciferase gene in place of ORF7a as previously described (BA.5 nLuc and BA.2 nLuc) (GenBank under accession numbers PV800151 and PV800153) (20,21). Infectious virus recovery was performed as previously described (17,20). Viral growth curve analysis was performed by infecting Vero E6 cells at an MOI of 0.001.
For recombinant protein expression, Expi293F cells (Thermo Fisher Scientific; cat # A1435101) were maintained at 37°C in 8% CO 2 in Expi293F Expression Medium (Thermo Fisher Scientific; catalog number A1435102), while ExpiCHO cells were maintained at 37°C in 8% CO 2 in ExpiCHO Expression Medium (Thermo Fisher Scientific, cat # A2910001).
## Mice and in vivo infections
Female 14-to 16-week-old or 10-to 12-month-old BALB/c mice were obtained from Envigo (Inotiv) (strain 047). Mice were inoculated intranasally under ketamine/xylazine anesthesia with either 1 × 10 4 or 1 × 10 5 PFU BA.5 MA, or 1 × 10 5 PFU BA.2 MA in 50 µL PBS as indicated, as previously described (15,17,20).
## mAb and antigen production and purification
cDNAs encoding mAbs of interest were synthesized (Twist Bioscience) and cloned into an IgG1 monocistronic expression vector (designated as pTwistmCis_G1) and used for production in mammalian cell culture. This vector contains an enhanced 2A sequence and GSG linker that allows for the simultaneous expression of mAb heavy and light chain genes from a single construct upon transfection. For antibody production, we performed transfection of ExpiCHO cell cultures using the Gibco ExpiCHO Expression System as described by the vendor (71). IgG molecules were purified from culture supernatants using HiTrap MabSelect SuRe (Cytiva) columns on a 24-column parallel protein chromatography system (Protein BioSolutions).
To express SARS-CoV-2 S proteins for ELISA binding and EM studies, we introduced the mutations of the BA.2 variant into the context of a previously described stabilized S protein construct (VFLIP) (72). In addition to a C-terminal T4 fibritin foldon domain, an 8 × His tag, and a TwinStrep tag, this construct contains an inter-protomer disulfide bond, a shorter glycine-serine-rich linker between the S1 and S2 domains, and five proline substitutions relative to the native SARS-CoV-2 S sequence. Plasmid encoding the BA.2_VFLIP antigen was transiently transfected into Expi293F cells, and culture supernatants were collected 4 to 5 days following transfection. After clarification by centrifugation and the addition of BioLock (IBA LifeSciences), antigen was purified using affinity chromatography with StrepTrap XT columns.
## EM sample preparation
EM imaging was performed with COV2-BA2 spike protein in complex with either COV2-3605 or COV2-3678. A recombinant form of the COV2-BA.2 spike was expressed and purified by affinity. Fabs were generated from purified, recombinantly expressed mAbs via enzymatic digestion using a FabALACTICA kit (Genovis, cat # A2-AFK-005). Antigen-Fab complexes were generated by incubating BA.2 S_VFLIP antigen with COV2-3605 Fab or COV2-3678 Fab in a 1:4 (antigen:Fab) molar ratio for 2 hours at room temperature.
## Negative-stain grid preparation, imaging, and processing
Three microliters of the complex sample at ~10 µg/mL was applied to a glow-discharged grid with continuous carbon film on 400 square mesh copper EM grids (Electron Microscopy Sciences). Grids were stained with 2% uranyl formate (73). Images were recorded on a Gatan US4000 4 k ´ 4 k CCD camera using an FEI TF20 (TFS) transmission electron microscope operated at 200 keV and controlled with SerialEM (74). All images were taken at 50,000 magnification with a pixel size of 2.18 Å/pixel in low-dose mode at a defocus of 1.5 to 1.8 µm. The total dose for the micrographs was ~33 e/Å2. Image processing was performed using the cryoSPARC software package (75). Images were imported, contrast transfer function estimated, and particles were picked automatically. The particles were extracted with a box size of 256 pix and binned to 128 pix (4.36 Å/ pixel), and multiple rounds of 2D class averages were performed to achieve clean data sets. The final data set was used to generate an initial 3D volume, and the volume was refined for a final map at the resolution of ~18 Å. Fab Model docking to the EM map was done in Chimera. PDB: 12E8 was used for the Fab. ChimeraX software was used to make all the figures (76). Data collection statistics are provided in Table S1.
## Assessment of mAbs in vivo
Antibody studies were conducted with mice treated prophylactically (12 hours prior to infection) with 200 µg of the indicated mAbs or isotype-matched IgG controls. Virus-chal lenged mice were inoculated with BA.5 MA at 1 × 10 5 PFU intranasally. A mock-infected control cohort received an equivalent volume of phosphate-buffered saline intranasally.
## Nanoluciferase-based neutralization assays
Moderate-throughput nanoluciferase assays in a 96-well format were conducted as previously described (20).
## Histopathology and antigen staining
Following harvest, mouse lungs were fixed for ≥7 days in 10% phosphate-buffered formalin at 4°C and prepared for histological examination and scoring as previously described (17,20).
## Statistical analysis
All statistical analyses were performed using GraphPad Prism 10. Statistical significance was determined by two-way analysis of variance (ANOVA) with Tukey's multi ple comparison test for weight loss, Kruskal-Wallis non-parametric test with Dunn's correction for lung discoloration, and one-way ANOVA with Tukey's multiple compari son test for tissue titers. Symbols denote statistically significant relationships at the respective levels: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. All cohorts started with five mice per harvest timepoint at the time of infection. Error bars represent standard error of the mean.
## References
1. Kim, Lee, Yang et al. (2020) "The architecture of SARS-CoV-2 transcriptome" *Cell*
2. Jungreis, Sealfon, Kellis (2021) "SARS-CoV-2 gene content and COVID-19 mutation impact by comparing 44 Sarbecovirus genomes" *Nat Commun*
3. Lan, Ge, Yu et al. (2020) "Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor" *Nature*
4. Tsai, Lee, Tseng (2021) "Comprehensive deep mutational scanning reveals the immune-escaping hotspots of SARS-CoV-2 receptor-binding domain targeting neutralizing antibodies" *Front Microbiol*
5. Greaney, Loes, Crawford et al. (2021) "Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies" *Cell Host Microbe*
6. Starr, Greaney, Addetia et al. (2021) "Prospective mapping of viral mutations that escape antibodies used to treat COVID-19" *Science*
7. Surie, Bonnell, Adams et al. (2021) "Effectiveness of monovalent mRNA vaccines against COVID-19-associated hospitalization among immunocompetent adults during BA.1/BA.2 and BA.4/BA.5 predominant periods of SARS-CoV-2 Omicron variant in the United States -IVY Network, 18 states" *MMWR Morb Mortal Wkly Rep*
8. Takashita, Yamayoshi, Simon et al. (2022) "Efficacy of antibodies and antiviral drugs against omicron BA.2.12.1, BA.4, and BA.5 subvariants" *N Engl J Med*
9. Suryawanshi, Ma, Syed et al. (2022) "Limited cross-variant immunity from SARS-CoV-2 Omicron without vaccination" *Nature*
10. Rizvi, Dandotiya, Sadhu et al. (2023) "Omicron sub-lineage BA.5 infection results in attenuated pathology in hACE2 transgenic mice" *Commun Biol*
11. Uraki, Halfmann, Iida et al. (2022) "Characterization of SARS-CoV-2 Omicron BA.4 and BA.5 isolates in rodents" *Nature*
12. Stewart, Yan, Ellis et al. (2023) "SARS-CoV-2 omicron BA.5 and XBB variants have increased neurotropic potential over BA.1 in K18-hACE2 mice and human brain organoids" *Front Microbiol*
13. Hoffmann, Wong, Arora et al. (2023) "Omicron subvariant BA.5 efficiently infects lung cells" *Nat Commun*
14. Shuai, Chan, Hu et al. (2023) "The viral fitness and intrinsic pathogenicity of dominant Full-Length Text Journal of Virology November"
15. "SARS-CoV-2 Omicron sublineages BA.1, BA.2, and BA" *EBioMedicine*
16. Leist, Dinnon, Iii et al. (2020) "A mouse-adapted SARS-CoV-2 induces acute lung injury and mortality in standard laboratory mice" *Cell*
17. Hou, Okuda, Edwards et al. (2020) "SARS-CoV-2 reverse genetics reveals a variable infection gradient in the respiratory tract" *Cell*
18. Dinnon, Iii, Leist et al. (2020) "A mouse-adapted model of SARS-CoV-2 to test COVID-19 countermeas ures" *Nature*
19. Feng, Yuan, Powers et al. (2023) "Broadly neutralizing antibodies against sarbecoviruses generated by immunization of macaques with an AS03-adjuvanted COVID-19 vaccine" *Sci Transl Med*
20. Martinez, Schäfer, Leist et al. (2021) "Prevention and therapy of SARS-CoV-2 and the B.1.351 variant in mice" *Cell Rep*
21. Powers, Leist, Mallory et al. (2024) "Divergent pathogenetic outcomes in BALB/c mice following Omicron subvariant infection" *Virus Res*
22. Parotto, Gyöngyösi, Howe et al. (2023) "Post-acute sequelae of COVID-19: understanding and addressing the burden of multisystem manifestations" *Lancet Respir Med*
23. Huang, Yao, Gu et al. "2021. 1-year outcomes in hospital survivors with COVID-19: a longitudinal cohort study" *The Lancet*
24. Mandel, Yoo, Allen et al. (2025) "Long COVID incidence proportion in adults and children between 2020 and 2024: an electronic health record-based study from the RECOVER initiative" *Clin Infect Dis*
25. (2024) "A long COVID definition: a chronic, systemic disease state with profound consequences"
26. Sheahan, Sims, Leist et al. (2020) "Comparative therapeutic efficacy of remdesivir and combination lopinavir, ritonavir, and interferon beta against MERS-CoV" *Nat Commun*
27. Dinnon, Iii, Leist et al. (2022) "SARS-CoV-2 infection produces chronic pulmonary epithelial and immune cell dysfunction with fibrosis in mice" *Sci Transl Med*
28. Kim, Noh, Kim et al. (2021) "Stereotypic neutralizing V H antibodies against SARS-CoV-2 spike protein receptor binding domain in patients with COVID-19 and healthy individuals" *Sci Transl Med*
29. Yuan, Liu, Wu et al. (2020) "Structural basis of a shared antibody response to SARS-CoV-2" *Science*
30. Li, Chen, Wang et al. (2023) "Breakthrough infection elicits hypermutated IGHV3-53/3-66 public antibodies with broad and potent neutralizing activity against SARS-CoV-2 variants including the emerging EG.5 lineages" *PLoS Pathog*
31. Tan, Yuan, Kuzelka et al. (2021) "Sequence signatures of two public antibody clonotypes that bind SARS-CoV-2 receptor binding domain" *Nat Commun*
32. Greaney, Starr, Barnes et al. (2021) "Mapping mutations to the SARS-CoV-2 RBD that escape binding by different classes of antibodies" *Nat Commun*
33. Wang, Paulson, Pease et al. (2022) "Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020-21" *The Lancet*
34. Navarrete, Barone, Qureshi et al. (2021) "SARS-CoV-2 infection and death rates among maintenance dialysis patients during Delta and early Omicron waves -United States" *MMWR Morb Mortal Wkly Rep*
35. Hedberg, Parczewski, Serwin et al. (2024) "-hospital mortality during the wild-type, alpha, delta, and omicron SARS-CoV-2 waves: a multinational cohort study in the EuCARE project"
36. (2024) "Preliminary estimates of COVID-19 burden for 2024-2025"
37. Davis, Mccorkell, Vogel et al. (2023) "Long COVID: major findings, mechanisms and recommendations" *Nat Rev Microbiol*
38. Su, Zhao, Zeng et al. (2023) "Epidemiology, clinical presentation, pathophysiology, and management of long COVID: an update" *Mol Psychiatry*
39. Hadley, Yoo, Patel et al. (2024) "Insights from an N3C RECOVER EHR-based cohort study characterizing SARS-CoV-2 reinfections and Long COVID" *Commun Med*
40. Bowe, Xie, Al-Aly (2022) "Acute and postacute sequelae associated with SARS-CoV-2 reinfection" *Nat Med*
41. Cohen, Jaudon, Schurman et al. (2025) "Impact of extended-course oral nirmatrelvir/ ritonavir in established Long COVID: a case series" *Commun Med*
42. Geng, Bonilla, Hedlin et al. (2024) "Nirmatrelvir-ritonavir and symptoms in adults with postacute sequelae of SARS-CoV-2 infection: The STOP-PASC randomized clinical trial" *JAMA Intern Med*
43. Wei, Qian, Narasimhan et al. (2025) "Macrophage peroxisomes guide alveolar regeneration and limit SARS-CoV-2 tissue sequelae" *Science*
44. Schäfer, Leist, Powers et al. (2024) "Animal models of long Covid: a hit-and-run disease" *Sci Transl Med*
45. Sudre, Murray, Varsavsky et al. (2021) "Attributes and predictors of long COVID" *Nat Med*
46. Tegally, Moir, Everatt et al. (2022) *Emergence Full-Length Text Journal of Virology*
47. "SARS-CoV-2 Omicron lineages BA.4 and BA.5 in South Africa" *Nat Med*
48. Yajima, Nomai, Okumura et al. (2024) "Molecular and structural insights into SARS-CoV-2 evolution: from BA.2 to XBB subvariants" *mBio*
49. Tsueng, Mullen, Alkuzweny et al. (2023) "Outbreak.info Research Library: a standardized, searchable platform to discover and explore COVID-19 resources" *Nat Methods*
50. Gangavarapu, Latif, Mullen et al. (2023) "Outbreak.info genomic reports: scalable and dynamic surveillance of SARS-CoV-2 variants and mutations" *Nat Methods*
51. Zhang, Zhang, Fang et al. (2022) "SARS-CoV-2 spike L452R mutation increases Omicron variant fusogenicity and infectivity as well as host glycolysis" *Sig Transduct Target Ther*
52. Motozono, Toyoda, Zahradnik et al. (2021) "SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity" *Cell Host Microbe*
53. Choe, Kang, Suh et al. (2020) "Antibody responses to SARS-CoV-2 at 8 weeks postinfection in asymptomatic patients" *Emerg Infect Dis*
54. Yan, Liu, Li et al. (2021) "Neutralizing antibodies and cellular immune responses against SARS-CoV-2 sustained one and a half years after natural infection" *Front Microbiol*
55. Sancilio, Schrock, Demonbreun et al. (2022) "COVID-19 symptom severity predicts neutralizing antibody activity in a community-based serological study" *Sci Rep*
56. Yu, Choi, Cheong et al. (2023) "Clinical efficacy and safety of SARS-CoV-2-neutralizing monoclonal antibody in patients with COVID-19: a living systematic review and meta-analysis" *J Microbiol Immunol Infect*
57. Kip, Mccreary, Collins et al. (2023) "Evolving real-world effectiveness of monoclonal antibodies for treatment of COVID-19: a cohort study" *Ann Intern Med*
58. Cox, Peacock, Harvey et al. (2023) "SARS-CoV-2 variant evasion of monoclonal antibodies based on in vitro studies" *Nat Rev Microbiol*
59. Focosi, Mcconnell, Casadevall et al. (2022) "Monoclonal antibody therapies against SARS-CoV-2" *Lancet Infect Dis*
60. Wang, Guo, Iketani et al. (2022) "Antibody evasion by SARS-CoV-2 Omicron subvariants BA.2.12.1, BA.4 and BA" *Nature*
61. Jian, Wang, Yisimayi et al. (2025) "Evolving antibody response to SARS-CoV-2 antigenic shift from XBB to JN.1" *Nature*
62. Jian, Wec, Feng et al. (2024) "Viral evolution prediction identifies broadly neutralizing antibodies to existing and prospective SARS-CoV-2 variants" *Nat Microbiol*
63. Goraichuk, Arefiev, Stegniy et al. (2021) "Zoonotic and reverse zoonotic transmissibility of SARS-CoV-2" *Virus Res*
64. Markov, Ghafari, Beer et al. (2023) "The evolution of SARS-CoV-2" *Nat Rev Microbiol*
65. Yen, Sit, Brackman et al. (2022) "Transmission of SARS-CoV-2 delta variant (AY.127) from pet hamsters to humans, leading to onward human-tohuman transmission: a case study" *Lancet*
66. Mcbride, Garushyants, Franks et al. (2023) "Accelerated evolution of SARS-CoV-2 in free-ranging white-tailed deer" *Nat Commun*
67. Caserta, Martins, Butt et al. (2023) "White-tailed deer (Odocoileus virginianus) may serve as a wildlife reservoir for nearly extinct SARS-CoV-2 variants of concern" *Proc Natl Acad Sci*
68. Pickering, Lung, Maguire et al. (2022) "Divergent SARS-CoV-2 variant emerges in white-tailed deer with deerto-human transmission" *Nat Microbiol*
69. Ge, Li, Yang et al. (2013) "Isolation and characteriza tion of a bat SARS-like coronavirus that uses the ACE2 receptor" *Nature*
70. Xiao, Zhai, Feng et al. (2020) "Isolation of SARS-CoV-2-related coronavirus from Malayan pangolins" *Nature*
71. Ou, Xu, Li et al. (2023) "Host susceptibility and structural and immunological insight of S proteins of two SARS-CoV-2 closely related bat coronaviruses" *Cell Discov*
72. Davis, Voss, Turnbull et al. (2022) "A C57BL/6 mouse model of SARS-CoV-2 infection recapitulates age-and sex-based differences in human COVID-19 disease and recovery" *Vaccines (Basel)*
73. Chng, Wang, Nian et al. (2015) "Cleavage efficient 2A peptides for high level monoclonal antibody expression in CHO cells" *MAbs*
74. Olmedillas, Rajamanickam, Avalos et al. (2025) "Structure of a SARS-CoV-2 spike S2 subunit in a pre-fusion, open conformation" *Cell Rep*
75. Ohi, Li, Cheng et al. (2004) "Negative staining and image classification -powerful tools in modern electron microscopy" *Biol Proced Online*
76. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
77. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
78. Pettersen, Goddard, Huang et al. (2021) "UCSF ChimeraX: structure visualization for researchers, educators, and developers" *Protein Sci*
79. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724344&blobtype=pdf
|
# Functionality of potato virus Y coat protein in cell-to-cell movement dynamics is defined by its N-terminal region
Anže Vozelj, Tjaša Povalej, Katja Stare, Magda Žnidarič, Katarina Bačnik, Valentina Levak, Ion Gutiérrez- Aguirre, Marjetka Podobnik, Kristina Gruden, Anna Coll, Tjaša Lukan
## Abstract
Potato virus Y (PVY) is one of the top 10 economically most important plant viruses and responsible for major yield losses. We previously suggested the involvement of the N-terminal region of PVY coat protein (CP) in PVY spread. By constructing different N-terminal deletion mutants of the PVY N605 strain, we here show that deletions of 40 or more amino acid residues from the N-terminal region of the CP resulted in the PVY multiplication limited to primary infected cells in Nicotiana clevelandii plants. Deletion of 26 residues profoundly impaired PVY cell-to-cell movement and prevented systemic PVY spread, while deletions of 19-23 residues allowed delayed systemic PVY spread. Introduced point mutations in the identified region prevent (S21G) or delay (G20P) PVY movement. In summary, this work shows the significance of the CP N-terminus for movement of the PVY. IMPORTANCE Potato virus Y (PVY) is one of the most economically important plant viruses worldwide, since it causes major yield losses in Solanaceae, especially in potato, where it is the causal agent of potato tuber necrosis ringspot disease, negatively impacting tuber quality and significantly reducing potato yield. By constructing different PVY N-terminal deletion mutants, we identified the regions of the coat protein that are important for efficient PVY cell-to-cell movement. Understanding these regions, responsible for cell-to-cell as well as long-distance movement, is crucial for economi cally important plant viruses such as PVY, since dissolving such complex processes will enhance our understanding of the PVY infective cycle and contribute to the develop ment of effective tackling strategies against PVY infections in potato and prevent yield losses.KEYWORDS potato virus Y, potato (Solanum tuberosum), coat protein (CP), viral movement, point mutations P lants undergo constant exposure to various pathogens and pests, with viruses being the most severe amongst them. Potato virus Y (PVY) is classified among the top 10 economically most important plant viruses infecting solanaceous crops including potato, tomato, tobacco, and pepper (1). PVY is the causal agent of potato tuber necrosis ringspot disease (2), a devastating disease that negatively impacts potato tuber quality and yield (3).PVY belongs to the Potyvirus genus. Its genome consists of 9.7 kb positive sense single-stranded RNA (+ ssRNA) encoding for single open reading frame (ORF) that is translated into 350 kDa long polyprotein which is cleaved into 10 mature proteins (3). Furthermore, PVY genome contains an additional ORF pipo ("pretty interesting Potyviri dae ORF"), embedded within the P3 cistron of the polyprotein. Expression of pipo occurs through transcriptional slippage by the viral RNA polymerase and results in production of the P3N-PIPO fusion protein, which is involved in cell-to-cell movement (4, 5).
The majority of potyviral proteins are multifunctional and they altogether contrib ute to the establishment of successful viral infection that encompasses multiple steps including transmission by aphids, penetration of viral particles into the cell, virion disassembly, translation, replication, suppression of host defense mechanisms, virion assembly, and virus movement from the primary infected cells to neighboring cells and systemic spread (6).
Viral movement is facilitated through plasmodesmata, plant-specific structures connecting two adjacent cells that serve as gateways for cell-to-cell movement and contribute to the establishment of systemic infection (7)(8)(9)(10). At least four potyviral proteins, coat protein (CP), cylindrical inclusion protein (CI), P3N-PIPO, and the second 6 kDa protein (6K2) protein are essential for cell-to-cell movement of potyviruses (11). P3N-PIPO is localized in the plasmodesmata and directs CI to form conical structures that are crucial for assistance of potyviral intercellular movement through plasmodes mata, which was shown for turnip mosaic virus (TuMV) in Nicotiana benthamiana (12). Accumulation of CP in plasmodesmata during infection is also crucial since together with the potyviral Helper-Component Proteinase (HCPro) it increases the size exclusion limit (SEL) of plasmodesmata, which was determined by microinjection studies for bean common mosaic necrosis potyvirus in N. benthamiana (13). Systemic infection encom passes viral movement through phloem and less often through xylem to reach and infect plant distant tissues (14,15). However, before reaching vasculature, plant viruses need to cross various cellular barriers including bundle sheath, vascular parenchyma, and companion cells, to reach sieve elements, through which they are transported to distant tissues (14). The invasion of distant tissues requires unloading from sieve elements into companion cells and cell-to-cell movement into bundle sheath and mesophyll cells (14). Despite PVY being one of the most extensively studied potyviruses on a molecular scale, information about crucial factors governing cell-to-cell as well as systemic spread is scarce (6,11).
The cryo-EM structures of some potyviruses including PVY, watermelon mosaic virus, and TuMV virions have been determined and offered a detailed insight into the conserved helical arrangement of CPs assembled around viral ssRNA as well as the conserved three-dimensional structure of CP among them (16)(17)(18). CP is the only potyviral structural protein and more than 2,000 copies of CP encapsidate the viral ssRNA molecule to form potyviral virions (6,10). Individual CP is built of three distinct regions. The N-terminal region with residues up to Val44 in PVY, which are exposed at the outer surface of the virus, is flexible and thus its 3D structure was not determined. The rest of the N-terminal region, from Val44 to Gln76, is structured. It has an extended structure with one short alpha helix in the middle (17). The central core has a globular shape. The C-terminal region is in the lumen of the virus, and its extended structure is supported by the viral ssRNA scaffold (17). Beyond its primary role in viral encapsidation, the CP is indispensable for aphid transmission, potyviral RNA amplification, and cell-to-cell movement (10,13,19). Functional studies of the CP protein indicate crucial roles of its N-and C-terminal regions and the core region in potyvirus infectivity. In the case of PVY, the absence of the flexible CP N-terminal region hinders the formation of filaments and cell-to-cell viral movement (17). For soybean mosaic virus, the CP C-terminal region is involved in CP intersubunit interactions, viral cell-to-cell and long-distance movement, and virion assembly (20). In potato virus A (PVA), phosphorylation of the Thr243 residue on CP C-terminal region is crucial for PVA replication in planta (21). These results and results of several other studies suggest that different CP regions are important for cell-to-cell movement of potyviruses (22)(23)(24)(25)(26)(27). It is not known yet in what structural shape the viruses are propagated through the plant, as assembled virions or viral ribonucleo protein complexes associated with CP (14,15).
The so far generated evidence-based knowledge differs from one potyvirus to another, and thus, more research is needed to comprehensively elucidate these complex processes, particularly for those economically relevant potyviruses such as PVY. By constructing different PVY N-terminal deletion mutants, we identified the regions of the CP that are important for efficient PVY cell-to-cell movement.
## RESULTS
## Deletion of 40 or more amino acid residues from the PVY CP N-terminal region affects cell-to-cell movement but does not prevent replication
Previously, we showed that PVY with deleted 50 N-terminal amino acid residues was still able to replicate, but based on the low and spatially limited levels of RNA accumu lation, we suggested that N-terminal region is required for cell-to-cell movement (17). To confirm this hypothesis, we here constructed two green fluorescent protein (GFP) tagged PVY (PVY N605-GFP) mutants lacking either 40 or 50 amino acid residues at the CP N-terminal region (hereafter ΔN50-CP and ΔN40-CP). In all constructed mutants, GFP was inserted between NIb and CP coding sequence, flanked by protease recognition sites, allowing GFP excision from the polyprotein after translation (28). We followed the spread of ΔN50-CP and ΔN40-CP with confocal microscopy upon bombardment of Nicotiana clevelandii leaves (Fig. 1B). Neither mutant was able to move cell-to-cell, remaining limited to bombarded cells (Fig. 1B). We detected viral RNA replication of both mutants in the bombarded leaves, albeit at significantly lower levels in comparison to the non-mutated clone (hereafter WT-CP) (Fig. S1 and Datasets S1 and S2). Despite the lower RNA levels observed in leaves bombarded with constructed mutants, confocal microscopy image analysis showed that the accumulation of GFP in individual infected cells was the same in WT and ΔN40-CP mutant, confirming unhindered replication (Table S2). These findings corroborate our previous results showing that N-terminal region CP truncations still allow viral replication (17) and further show that the absence of either 50 or 40 N-terminal residues abolishes cell-to-cell viral movement.
## The deletions of 19-26 amino acid residues of the PVY CP N-terminal region are influencing dynamics of viral cell-to-cell spread
To further assess which region of CP N-terminal is crucial for cell-to-cell virus spread, we constructed GFP-tagged mutants lacking 26, 23, 20, 19, and 14 amino acid residues at the CP N-terminal region (Fig. 1A) and followed their localization under confocal microscope at different time points after virus inoculation (Fig. 1C andD, Table S1, and Dataset S2). Mutants lacking 26 (ΔN26-CP) and 23 (ΔN23-CP) residues on CP N-terminal region had the first non-deleted amino acid replaced with glycine (G) during the mutagenesis.
At 5 days post-bombardment (dpb), ΔN14-CP and ΔN19-CP mutants did not show any differences in cell-to-cell spread compared to WT-CP, while a lower spread level was noticed for ΔN20-CP and ΔN23-CP (Fig. 1C). The most pronounced difference in comparison to other mutants and to WT-CP was observed in the case of ΔN26-CP mutant, which remained confined to single cells at five dpb (Fig. 1C andD). However, at later time points (10 and 14 dpb), in approximately half of the infected plants, we also observed cell-to-cell viral movement, although substantially delayed compared to WT-CP (Fig. 1D and Table S1). Also, the measured viral RNA load for ΔN26-CP mutant was significantly lower than that of WT-CP at all 5, 10, and 14 dpb (Fig. S2 and Dataset S1).
Whole plant imaging was conducted to more precisely examine the dynamics of cell-to-cell virus spread of WT-CP, ΔN23-CP, ΔN19-CP, and ΔN14-CP through time on inoculated N. clevelandii leaves (Fig. 2A). As expected, the results revealed statistically significant differences in the viral multiplication area between N23-CP and WT-CP, and ΔN19-CP compared to WT-CP (Fig. 2B andC and Datasets S3 and S4). A similar, though less evident, trend was observed in the case of ΔN14-CP compared to WT-CP, which was not statistically significant (Fig. 2C and Dataset S4).
These results show that the dynamics of viral cell-to-cell spread is dependent on the length of the CP N-terminal region (Fig. 1 and Dataset S2). More precisely, as the number of CP N-terminal region deleted amino acid residues increases, cell-to-cell movement speed is decreased (Fig. 2B andC and Datasets S3, S4, and S6). We conclude that amino acids in the regions 19-26 on CP N-terminal region are essential for an optimal PVY cell-to-cell movement. Furthermore, with electron microscopy, we observed that viral assembly at all tested deletion mutants is feasible (Fig. S10).
## Systemic viral spread dynamics is affected by the cell-to-cell movement
Since viral cell-to-cell spread is a prerequisite for systemic viral spread, we further tested if the observed delay in cell-to-cell viral spread of mutants (Fig. 1C, 2B andC) affects the systemic viral spread. In contrast to the systemic spread of WT-CP, which is detectable already at seven dpb, the spread of ΔN23-CP, ΔN19-CP, and ΔN14-CP mutants to systemic leaves occurs with a delay (Fig. 3A). If compared to WT-CP, ΔN14-CP was observed in systemic tissue with 1 day delay (8 dpb), ΔN19-CP with 2 days delay (9 dpb), and ΔN23-CP with approximately 3 weeks delay (25 dpb) (Fig. 3A). The mutant ΔN26-CP did not reach systemic leaves even at the highest tested dpb (Fig. S4). Quantitative analysis of viral multiplication signal in systemic tissue confirmed that WT-CP achieved the largest multiplication area, followed by ΔN14-CP and lastly ΔN19-CP in all observed time points (Fig. 3 and Dataset S7). When comparing ΔN14-CP and WT-CP, we observed a trend of slower spread in the case of ΔN14-CP, albeit not statistically significant (Fig. 3B and Dataset S7). ΔN23-CP was not included in the comparison, due to the considerable delay in the systemic spread exhibited by this mutant (Fig. 3A), which Raw and normalized data, number of plants, and statistics are specified in Datasets S3 and S4. The same trend was observed in a replicate experiment (Fig. S3 and Dataset S5). For comparison between two independent experiments, data measured for mutant viruses were normalized to median of values obtained for WT virus (Dataset S6). between G20P mutant and WT-CP are represented with asterisks (*), while there were no statistically observed differences between P24A and WT-CP (see Dataset S8 for all results of statistical testing).
Vertical lines present all points except outliers. The results were confirmed in additional experiments (Fig.
## S5 and Dataset S9).
made it difficult to measure the viral multiplication area. In conclusion, systemic viral spread is abolished for ΔN26-CP and decreasingly hampered in the case of ΔN23-CP and ΔN19-CP, while the ΔN14-CP mutant is able to move similarly as WT-CP, albeit with a short delay.
## Substitution of serine with glycine on position 21 of CP N-terminal region abolished PVY cell-to-cell viral movement and virion assembly
To pinpoint amino acids important for cell-to-cell virus movement, we generated point mutations in the part of CP N-terminal region that was observed as being important for viral movement (Fig. 1 to 3). We point mutated charged residues Asp14, Glu18, potential phosphorylation site at Ser21, and 3D structure breakers Gly20 and Pro24. All, except Pro24, are highly conserved across sequenced PVY strains. At position 24, 50% of PVY strains carry Pro and the others Ser (Fig. S9). Selected residues were changed to Ala (D14A, E18A, and P24A), except for Ser21 where the sequence coding for Ala and Thr substitution was unstable in Escherichia coli; thus, we mutated it to Gly (S21G) and for Gly20, where we intentionally introduced Pro (G20P) to affect the fold of N-terminal region (Fig. 4A). Viral cell-to-cell spread of mutants D14A, E18A, and P24A was comparable to WT-CP (Fig. 4B and Fig. S6). On the other hand, G20P mutant showed delayed cell-to-cell spread, while S21G mutant remained limited to just single cells at all observed time points (Fig. 4B and Fig. S6).
Next, we performed whole plant imaging to study systemic viral spread of mutants (Fig. 4C and Fig. S8). In accordance with the results of cell-to-cell movement, there were no statistically significant differences in viral multiplication area in systemic tissue when comparing D14A, E18A, or P24A with WT-CP (Fig. 4D, Fig. S5 and S7, and Datasets S8 to S11) at all observed time points. On the other hand, the viral multiplication area of G20P in systemic tissue was statistically significantly lower compared to WT-CP at all observed time points (Fig. 4D and Dataset S8). As expected, S21G did not spread systemically (Fig. 4C). Furthermore, viral assembly was confirmed with electron microscopy in all constructed point mutants, with S21G as an exception, where only oligomeric rings were observed (Fig. S10).
## DISCUSSION
The biological function of C and N-terminal regions of PVY CP has already been initiated in our previous study, where we hypothesized that N-terminal amino acid residues are necessary for an efficient cell-to-cell movement (17). Here, we provide additional insights into the mechanism of PVY spread by identifying the regions of PVY CP N-terminal region that are important for efficient viral cell-to-cell movement and systemic infection.
Despite the conserved flexible structure of potyviruses N-terminal region of CP, there are notable differences observed regarding its involvement in potyviral cell-to-cell and systemic movement. In the case of zucchini yellow mosaic virus (ZYMV), deletion of the entire CP N-terminal region did not affect systemic infection (23). Similarly, a deletion of six to 50 amino acids in the CP N-terminal domain did not compromise TuMV cell-to-cell and systemic movement (19). On the other hand, in tobacco etch potyvirus (TEV), CP N-terminal region deletions of five to 29 amino acid residues delayed cell-to-cell movement and prevented systemic spread (22). We observed an inability to move cell-to-cell for PVY when 40 or more amino acid residues were deleted from the CP N-terminal region (Fig. 1). However, shorter deletions in the range of 19 to 23 did not prevent systemic spread but resulted in delayed systemic and cell-to-cell PVY movement (Fig. 2B andC, Fig. 3B, and Datasets S3, S4, and S7). The observed differences among potyviral species may stem from the high phylogenetic divergence inherent to the genus, which at the CP level is, among other variations, observed as different lengths and nucleotide variations in the N-terminal region across different potyvirus species (23,(29)(30)(31)(32). It is also well-known that the potyvirus movement is facilitated by various protein components, both from the virus and the host plant (10,11,(33)(34)(35). In addition, potyviruses exhibit a divergent host range, majority of them being naturally limited to narrow host range, such as TEV and ZYMV (36,37), while TuMV stands out with its remarkably broad natural host range (38). These features could also be contributing to the differences in the influence that CP N-terminal domain deletions have on the movement of different potyvirus species, because different components involved in potyviral movement, viral and host, could be differently compensating the effect of CP N-terminal deletions.
The analysis of viral multiplication area revealed important insights into the cell-tocell movement dynamics of deletion mutants, as increasing number of deleted amino acid residues at CP N-terminal region resulted in slower cell-to-cell viral spread (Fig. 2B andC). Furthermore, slower cell-to-cell virus spread also impacted infection of systemic tissue (Fig. 3A). The connection between cell-to-cell movement and systemic spread was expected, as the virus reaches the veins, enters the veins, and is transported in systemic tissue with a delay due to decelerated cell-to-cell movement (39).
Results of several studies showed the role of charged amino acids in the CP core and C-and N-terminal regions of different potyviruses in cell-to-cell and systemic virus spread (20,26,40,41). In contrast, substitutions of charged amino acids in the CP N-terminal region of PVY in our study (D14A and E18A mutants) did not affect cell-to-cell spread (Fig. S6). It was also reported (42) that aromatic residues, located on the core CP region, contribute to the formation of π-stackings that are importantly involved in TVBMV and other potyvirus movement, since they maintain CP accumulation by stabilizing α-helixes and β-sheets (43,44). Our region of interest does not possess aromatic amino acid residues; therefore, we did not test this hypothesis.
However, we observed that replication of S21G mutant was limited to single cells (Fig. 4B). We further checked, and the intensity of replication of this mutant within this single cell was not affected (Table S2). Interestingly, electron microscopy imaging showed impaired viral assembly, since only oligomeric rings were observed (Fig. S10). Correct viral particle assembly is, however, not a prerequisite for cell-to-cell movement, as ΔN40-CP and ΔN50-CP mutants did assemble but still could not move between cells (Fig. 1 and Fig. S10). We hypothesize that this might be due to loss of potential phosphorylation site or disruption of N-terminus structure, due to introduction of G. Previous studies suggested that post-translational modifications such as phosphoryla tions have an important role in viral cell-to-cell movement and virion assembly (45,46). In PVA, phosphorylation of the Thr243 residue on the CP C-terminal region is crucial for PVA replication in planta (21). Note that to confirm the hypothesis about the role of phosphorylation of Ser21, we aimed to replace Ser21 also with alanine and threonine, but the design of such point mutants was not achievable, due to their instability in E. coli. It has been previously suggested that PVY cDNA is unstable in E. coli. Stability of full-length PVY plasmids was achieved by inserting introns into putatively toxic genes, but still, for some mutants recombinations can occur (47).
Furthermore, we also showed the important role of the G on position 20 of CP N-terminal region in PVY cell-to-cell as well as systemic spread, since G20P mutant spreads slower than WT-CP (Fig. 4B through D). We hypothesize that exchanging Gly with Pro in the G20P mutant increased rigidity in the N-terminal region. Consequently, the interaction of the CP with other host and viral proteins in the cell may be impaired, resulting in the observed delay. Moreover, alignment of the first 50 amino acid resi dues from PVY CP N-terminal region across all PVY isolates revealed a high degree of conservation (at least 92.9%) at the mutated residues 20 (G20P) and 21 (S21G) across all PVY isolates (Fig. S9), further supporting biological significance of these amino acids in PVY.
The results identifying crucial domains and amino acids of CP N-terminal region in viral cell-to-cell movement provided in our study are important basis to elucidate these complex processes, particularly in most economically important viruses such as PVY.
## MATERIALS AND METHODS
## Plant material
N. clevelandii plants were used to follow virus spread. Plants were grown from seeds and kept in growth chambers under controlled conditions as described previously (48).
## Construction of PVY-N605(123)-GFP CP N-terminal deletion and point mutants
As a template to construct mutants with deletions on the CP N-terminal region, we used a GFP-tagged infectious clone PVY-N605(123) (28). The GFP coding sequence was inserted between NIb and CP coding sequences, flanked by protease target sequences, so that GFP would not bother viral movement, since it is excised from the polyprotein as other viral proteins (28). Megaprimers were designed in accordance with previ ously prepared instructions (49). Detailed information about mutagenic PCR reactions conditions and mixtures can be found in Text S1. After adding 4 µL of DpnI enzyme, the mutagenesis mixtures were transformed into E. coli XL 10 Ultracompetent Cells in accordance with manufacturer's instructions (Agilent Technologies). Transformation mixtures were plated on LB agar with ampicillin selection and incubated overnight at 37°C. Transformants were screened with colony PCR using PVY GFP_F and PVY uni_R primers (Dataset S12) and KAPA2G Robust HotStart Kit (Agilent Technologies). Prior to bombardment, polyprotein and its promoter in PVY-N605(123) lacking 50 and 14 amino acids on CP N-terminus were Sanger-sequenced (Dataset S12). Additionally, restriction analysis was performed to confirm the correctness of the PVY coding region.
## Biolistic bombardment
Constructed PVY mutants and wild-type plasmids were amplified in E. coli. Plasmids were then isolated and coated onto gold microcarriers that were used for N. clevelandii bombardment using a Helios gene gun (Bio-Rad), as described previously (49).
## Plant material sampling and RNA isolation
Plant leaf samples were collected from bombarded N. clevelandii plants by excising 0.3 g plant tissue near the bombardment site. Sampled plant material from bombarded N. clevelandii leaves was homogenized in 700 µL of RNeasy kit RLT lysis buffer with FastPrep (MP Biomedicals) 1 min 6.5 m/s. Total RNA was isolated using RNeasy Plant Mini Kit (Qiagen). Residual DNA was digested with deoxyribonuclease I (DNase I, Qiagen) in solution, using 1.36U DNase/µg RNA. Quality control of isolated RNA was performed as described (48).
## RT-qPCR
Relative PVY RNA concentration was determined by single-step reverse transcription quantitative PCR (RT-qPCR; AgPath-ID One-Step RT-PCR, Thermo Fisher Scientific) for the ΔN50-CP mutant, while two step RT-qPCR was performed for other mutants. RNA was reverse transcribed with High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) and analyzed by qPCR using FastStart Universal Probe Master (Roche).
A qPCR assay targeting the PVY CP encoding region was used as a target gene, while the gene encoding for cytochrome oxidase was used as a reference gene. To avoid false positive results, plants bombarded with plasmid encoding for blue fluorescent protein were sampled and analyzed with our qPCR assay targeting PVY CP encoding region. RT-qPCR amplification program and reaction composition mixtures were the same as previously described (28). Primer and probe details are listed in Dataset S12.
After conducting RT-qPCR reactions, the QuantGenius software (50) was used for relative quantification of PVY RNA using the standard curve approach. Data containing information about the relative RNA quantity were averaged for each group and then normalized to the lowest group average number (Dataset S1). excitation channel and emission filter F-550. Images were taken by EvolutionCapt edge software. For viral multiplication area analysis in inoculated leaves, all three ΔN14-CP-, ΔN19-CP-, ΔN23-CP-, and WT-CP-inoculated leaves of N. clevelandii were imaged using exposure time 1 min 46 s 700 ms, 10 × 10 field of view (FOV) and 1,331-1,337 focus in four time points: 3 days post-inoculation (dpi) a.m. and p.m. and four dpi a.m. and p.m. Measurements for a.m. time point were carried out 9 a.m, while for p.m. time point at 1 p.m. Four plants inoculated with each mutant or WT-CP were imaged. For studying systemic viral spread, at least three N. clevelandii plants bombarded with ΔN26-CP, ΔN23-CP, ΔN14-CP, ΔN19-CP, and WT-CP were imaged, using 50 s exposure time, 20 × 20 FOV, and focus in the range between 1,871 and 1,905 in different time points (7-12 dpb). Note that N. clevelandii plants bombarded with ΔN26-CP and ΔN23-CP were not taken into analysis, due to their inability or big delay in systemic movement. For studying systemic viral spread of point mutations, at least three plants bombarded with construc ted point mutants were imaged at different time points (6-13 dpb) with exposure time 50 s and other settings as for studying systemic viral spread of deletion mutants (see above). In the case of the experiment with D14A and WT-CP at 13 dpb, exposure time was only 5 s to avoid saturation due to a high signal. Raw images were deposited on Zenodo (doi: https://doi.org/10.5281/zenodo.17643798).
## Image analysis and signal quantification following whole plant imaging
Viral multiplication area analysis on inoculated leaves and on upper systemic leaves was performed using Kuant software (Vilber, France). In the case of viral multiplication area, surfaces of three viral multiplication areas per leaf were measured, while in the case of viral multiplication area in systemic leaves, total count of the virus-affected area was measured. Total count is determined as a sum of gray values within region of interest (ROI) and is proportional to the intensity and spread of the signal. All measured data from Kuant software were exported to Excel, where total count in ROIs was averaged for each time point and subsequently normalized on the lowest group average number. Normalized viral multiplication areas were statistically evaluated using Welch's t-test and were used for graphic representation.
## References
1. Scholthof, Adkins, Czosnek et al. (2011) "Top 10 plant viruses in molecular plant pathology"
2. Quenouille, Vassilakos, Moury (2013) "Potato virus Y: a major crop pathogen that has provided major insights into the evolution of viral pathogenicity" *Mol Plant Pathol*
3. Ivanov, Eskelin, Lõhmus et al. (2014) "Molecular and cellular mechanisms underlying potyvirus infection" *J Gen Virol*
4. Olspert, Chung, Atkins et al. (2015) "Transcriptional slippage in the positive-sense RNA virus family Potyviridae" *EMBO Rep*
5. Chung, Miller, Atkins et al. (2008) "An overlapping essential gene in the Potyviridae" *Proc Natl Acad Sci*
6. Yang, Li, Wang (2021) "Research advances in potyviruses: from the laboratory bench to the field" *Annu Rev Phytopathol*
7. Benitez-Alfonso, Faulkner, Ritzenthaler et al. (2010) "Plasmo desmata: gateways to local and systemic virus infection" *Mol Plant Microbe Interact*
8. Folimonova, Tilsner (2018) "Hitchhikers, highway tolls and roadworks: the interactions of plant viruses with the phloem" *Curr Opin Plant Biol*
9. Reagan, Burch-Smith (2020) "Viruses reveal the secrets of plasmodesmal cell biology" *Mol Plant Microbe Interact*
10. Revers, García (2015) "Molecular biology of potyviruses" *Adv Virus Res*
11. Xue, Arvy, German-Retana (2023) "The mystery remains: how do potyviruses move within and between cells?" *Mol Plant Pathol*
12. Wei, Zhang, Hong et al. (2010) "Formation of complexes at plasmodesmata for potyvirus intercellular movement is mediated by the viral protein P3N-PIPO" *PLoS Pathog*
13. (2025) *Full-Length Text Journal of Virology*
14. Rojas, Zerbini, Allison et al. (1997) "Capsid protein and helper component-proteinase function as potyvirus cell-tocell movement proteins" *Virology (Auckl)*
16. Navarro, Sanchez-Navarro, Pallas (2019) "Key checkpoints in the movement of plant viruses through the host" *Adv Virus Res*
17. Mäkinen, Hafrén (2014) "Intracellular coordination of potyviral RNA functions in infection" *Front Plant Sci*
18. Zamora, Méndez-López, Agirrezabala et al. (2017) "Potyvirus virion structure shows conserved protein fold and RNA binding site in ssRNA viruses" *Sci Adv*
19. Kežar, Kavčič, Polák et al. (2019) "Structural basis for the multitasking nature of the potato virus Y coat protein" *Sci Adv*
20. Cuesta, Yuste-Calvo, Gil-Cartón et al. (2019) "Structure of turnip mosaic virus and its viral-like particles" *Sci Rep*
21. Dai, He, Bernards (2020) "The cis-expression of the coat protein of turnip mosaic virus is essential for viral intercellular movement in plants" *Mol Plant Pathol*
22. Seo, Vo Phan, Kang et al. (2013) "The charged residues in the surface-exposed C-terminus of the soybean mosaic virus coat protein are critical for cell-to-cell movement" *Virology (Auckl)*
23. Lõhmus, Hafrén, Mäkinen (2017) "Coat protein regulation by CK2, CPIP, HSP70, and CHIP Is required for potato virus A replication and coat protein accumulation" *J Virol*
24. Dolja, Haldeman, Robertson et al. (1994) "Distinct functions of capsid protein in assembly and movement of tobacco etch potyvirus in plants" *EMBO J*
25. Arazi, Shiboleth, Gal-On (2001) "A nonviral peptide can replace the entire N terminus of zucchini yellow mosaic potyvirus coat protein and permits viral systemic infection" *J Virol*
26. Kimalov, Gal-On, Stav et al. (2004) "Maintenance of coat protein N-terminal net charge and not primary sequence is essential for zucchini yellow mosaic virus systemic infectivity" *J Gen Virol*
27. Tatineni, Kovacs, French (2014) "Wheat streak mosaic virus infects systemically despite extensive coat protein deletions: identification of virion assembly and cell-to-cell movement determinants" *J Virol*
28. Tatineni, French (2014) "The C-terminus of Wheat streak mosaic virus coat protein is involved in differential infection of wheat and maize through host-specific long-distance transport" *Mol Plant Microbe Interact*
29. Tatineni, Elowsky, Graybosch (2017) "Wheat streak mosaic virus coat protein deletion mutants elicit more severe symptoms than wildtype virus in multiple cereal hosts" *Mol Plant Microbe Interact*
30. Lukan, Županič, Povalej et al. (2023) "Chloroplast redox state changes mark cell-to-cell signaling in the hypersensitive response" *New Phytol*
31. Yuste-Calvo, Ibort, Sánchez et al. (2020) "Turnip mosaic virus coat protein deletion mutants allow defining dispensable protein domains for "in planta" eVLP formation" *Viruses*
32. Voloudakis, Malpica, Aleman-Verdaguer et al. (2004) "Structural characterization of tobacco etch virus coat protein mutants" *Arch Virol*
33. Gibbs, Ohshima (2010) "Potyviruses and the digital revolution" *Annu Rev Phytopathol*
34. Gibbs, Hajizadeh, Ohshima et al. (2020) "The potyviruses: an evolutionary synthesis is emerging" *Viruses*
35. Ala-Poikela, Rajamäki, Valkonen (2019) "A novel interaction network used by potyviruses in virus-host interactions at the protein level" *Viruses*
36. Martínez, Rodrigo, Aragonés et al. (2016) "Interaction network of tobacco etch potyvirus NIa protein with the host proteome during infection" *BMC Genomics*
37. Vijayapalani, Maeshima, Nagasaki-Takekuchi et al. (2012) "Interaction of the trans-frame potyvirus protein P3N-PIPO with host protein PCaP1 facilitates potyvirus movement" *PLoS Pathog*
38. García-Arenal, Fraile (2013) "Trade-offs in host range evolution of plant viruses" *Plant Pathol*
39. Desbiez, Lecoq (1997) "Zucchini yellow mosaic virus" *Plant Pathol*
40. Li, Lv, Zhang et al. (2019) "TuMV management for Brassica crops through host resistance: retrospect and prospects" *Plant Pathol*
41. Wang (2015) "Dissecting the molecular network of virus-plant interactions: the complex roles of host factors" *Annu Rev Phytopathol*
42. Atreya, Lopez-Moya, Chu et al. (1995) "Mutational analysis of the coat protein N-terminal amino acids involved in potyvirus transmission by aphids" *J Gen Virol*
43. López-Moya, Pirone (1998) "Charge changes near the N terminus of the coat protein of two potyviruses affect virus movement" *J Gen Virol*
44. Yan, Xu, Fang et al. (2021) "Multiple aromatic amino acids are involved in potyvirus movement by forming π-stackings to maintain coat protein accumulation" *Phytopathol Res*
45. Salonen, Ellermann, Diederich (2011) "Aromatic rings in chemical and biological recognition: energetics and structures" *Angew Chem Int Ed Engl*
46. Chatterjee, Tripathi, Das (2019) "A conserved and buried edge-toface aromatic interaction in small ubiquitin-like modifier (SUMO) has a role in SUMO stability and function" *J Biol Chem*
47. Ivanov, Puustinen, Gabrenaite et al. (2003) "Phosphorylation of the potyvirus capsid protein by protein kinase CK2 and its relevance for virus infection" *Plant Cell*
48. Hervás, Navajas, Chagoyen et al. (2020) "Phosphorylation-related crosstalk between distant regions of the core region of the coat protein contributes to virion assembly of plum pox virus" *Mol Plant Microbe Interact*
49. Bukovinszki, Götz, Johansen et al. (2007) "The role of the coat protein region in symptom formation on Physalis floridana varies between PVY strains" *Virus Res*
50. Baebler, Krecic-Stres, Rotter et al. (2009) "PVY(NTN) elicits a diverse gene expression response in different potato genotypes in the first 12 h after inoculation" *Mol Plant Pathol*
51. Stare, Coll, Gutiérrez-Aguirre et al. (2020) "Generation and in planta functional analysis of potato virus Y mutants" *Bio Protoc*
52. Baebler, Svalina, Petek et al. (2017) "quantGenius: implementation of a decision support system for qPCR-based gene quantification" *BMC Bioinformatics*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12671160&blobtype=pdf
|
# Superinfection interference of alpha and orthoflavivirus pathogens by homologous non-pathogenic vaccine strains
Julie Hicks, Dennis Brown, Richard Anthony, Hsiao-Ching Liu, Raquel Hernandez
## Abstract
We have developed host-restricted, live attenuated virus (HR-LAV) vaccine strains of six tropical disease pathogens (chikungunya and dengue 1-4) that are non-pathogenic in vertebrates. The HR-LAV vaccine strains replicate efficiently in insect cells but are severely impaired in their replication within vertebrate cells, and the large deletions in these vaccine strains effectively prevent reversion to the wild type (WT) in the restrictive host. Here, we describe a novel approach that uses these homologous, but non-pathogenic HR-LAV vaccine strains to disrupt replication of secondary infections of WT pathogenic virus through the mechanism of Superinfection Exclusion, an intrinsic insect mechanism of virus replication control. Our results demonstrate that persistent primary infections of HR-LAV strains show a significant degree of pathogen interference in several mosquito cell lines, which may also be used to limit replication of WT virus in the whole mosquito.
IMPORTANCEWe describe experiments that suggest a theoretical approach for mitigating the spread of mosquito-vectored diseases of medical importance. This method would exploit the phenomenon of superinfection exclusion, in which the presence of a virus in a cell prevents the expression of a second infecting virus. These data suggest that further studies in live insects are merited. Such studies may reveal that this method may be employed to replace the current mosquito control technologies that are environmentally unfriendly and harmful to humans, animals, and other insects that are important to the ecosystem. KEYWORDS arbovirus, superinfection exclusion, virus spread, co-infections M osquitoes are the world's most predominant carriers of insect-vectored human pathogens. They have been referred to as "the world's deadliest animal" (https:// www.cdc.gov/globalhealth/stories/2019/world-deadliest-animal.html). Of the >3,500 mosquito species, only a few transmit disease (1, 2). Aedes aegypti is the most common and versatile vector of virus pathogens because of its genetic predisposition to adapt to urban and rural environments, diurnal behavior, predilection to feed on humans, and the ability to transmit a variety of virus pathogens (3). It is estimated that 80% of the world's population lives in areas supporting infestations by this mosquito in environments poised to harbor the most serious of these pathogens 4(5, 6). These include dengue virus (DENV), chikungunya virus (CHIKV), Zika virus (ZIKV) (7), and yellow fever virus, which cause febrile, arthritogenic, neuroinvasive, and hemolytic diseases that are of global medical and economic importance (8, 9). Dengue virus circulates in areas in which half of the world's population is exposed (10) and is becoming an increasing threat in the US. The ability to infect both mosquitoes and vertebrates enables maintenance of arboviruses in nature in enzootic cycles with spillover into the human population. A second pervasive vector, the highly invasive Aedes albopictus mosquito (11), has been found to be the primary vector of CHIKV and ZIKV (12), making these
two Ae. aegypti and Ae. albopictus the major vector of arbovirus disease (13)(14)(15). What were once tropical diseases restricted to the warmest regions of the planet, climate change has permitted their insect vectors to spread to their geographic range north, transporting these pathogens along with them (16). In spite of over 50 years of intensive effort to develop prophylactic methods to treat these illnesses, no treatment or globally accepted vaccines exist (17,18).
Mosquito population control (19) has been and continues to be the most effective method to minimize the spread of the vectors and concomitant arbovirus diseases (20,21) but has had significant ecological drawbacks (22).
To date, the most advanced biological mosquito control methods involve genetic or mechanical manipulation of the mosquito population. The most effective methods interfere with mosquito fertility and reproduction to inhibit pathogen transmission. Male mosquitoes sterilized through irradiation or genetic modifications have been used to reduce mosquito populations in the wild. This is the sterile insect technique and requires large and repeated release of mosquitoes to reduce the mosquito population (23). Unfortunately, lowering the mosquito populations in their respective environs affects a primary food source for the beneficial fauna of the ecosystem. A recent, more effec tive bio-control approach involves infecting mosquitoes with Wolbachia, an arthropod Rickettsiales endosymbiont (12). There are two main hypotheses to explain Wolbachia interference in viral replication: the activation of vector-host immunity and competition with the virus for cellular resources (24)(25)(26). Regardless of the mechanism of arbovirus bio-interference, Wolbachia treatment was shown to affect up to 98% of the mosquito population in one study (27). However, Wolbachia is not usually found in the A. aegypti mosquito, so a modified form must be established in the laboratory before releasing infected mosquitoes into the wild, and then this requires sustained maintenance release at a substantial cost as a biocontrol method (28,29). Considering the limitations of the existing mosquito control strategies, it becomes clear that novel arbovirus mitigation methods are necessary.
Nasar et al. have demonstrated that the mosquito-specific alpha virus Eilat virus can produce homologous and heterologous interference of other alphaviruses in vitro and in vivo. In this study, we wish to extend these results to alpha-and flaviviruses of medical significance. Here, we provide evidence that an alternative biological method of mitigating circulating arbovirus pathogens in mosquito populations may be possible through the application of genetically modified, non-pathogenic forms of specific virulent arboviruses to induce targeted superinfection exclusion (SIE) (30)(31)(32). Viral SIE is defined by an initial primary infection preventing the replication of a secondary infection by a genetically identical or closely related virus. This phenomenon manifests as homologous interference or heterologous interference, depending on whether the interference occurs between a homologous or heterologous virus, respectively. This approach to mitigate pathogenic virus spread has been suggested previously, but a means of implementation had not been elaborated (32)(33)(34). To circumvent the risk of disease, we propose to establish SIE through a primary exposure of our non-pathogenic, host-restricted, live attenuated virus (HR-LAV) strains, developed as vaccines, as a means to restrict the growth of a secondary superinfection of a target pathogen, essentially vaccinating the mosquito host against the challenge virus. Our HR-LAV strains have been previously described and include mosquito-restricted host range deletion mutants for DENV 1-4, CHIKV, and Sindbis virus (SINV) (35)(36)(37). These HR-LAV strains for DENV 1-4, CHIKV, and SINV have been constructed and evaluated as vaccines for safety and protective efficacy in appropriate animal models, including non-human primates (35)(36)(37). Attenuation of the wild-type (WT) viruses HR-LAV constructs uses a common platform that deletes large segments of the structural glycoproteins' transmembrane domain (TMD) (38) of the proteins that interact with the virus capsid protein, such as E (orthoflaviviruses) (39) or E2 (alphaviruses). We have found that deleting specific sequence motifs produces insect (host)-restricted LAV. These sequence motifs have been found to be necessary for the replication of the viruses in the vertebrate hosts but are nonessential to function in the mosquito membrane. Once the disruptive motif is identified, the same or similar motif is deleted in genetically related viruses, making HR-LAV construction a streamlined process that can theoretically be applied to any membrane-containing arbovirus. These are large TMD deletions-four amino acid residues in the orthoflaviviruses and nine amino acid deletions in the alphaviruses-are permanent, and no reversion to wild-type virus has been documented. For example, in the orthoflaviviruses, these deletions were not found to revert after six passages in vertebrate cell culture or after 6 days of infection in vervet monkeys (36). The mosquitorestricted HR phenotype has been attributed to the unique physiology and biochemistry of the mosquito membrane (38,40). Distinct from mammals, the mosquito membrane is a structure of reduced vertical dimensions containing shorter lipids, no native choles terol, and is biochemically disparate from mammalian membranes (41).
## MATERIALS AND METHODS
## Cells and viruses
Five mosquito cell lines were used in this study: three Ae. albopictus lines, C6/36, C7-10, and U4.4 (42), and two A. aegypti lines, Aag2 (obtained from Ana Sesma, Icahn School of Medicine at Mount Sinai (ISMMS) and A20 (obtained from Zachary Adelman, Texas A&M University). For the DENV 1-4 and CHIKV experiments, all cells were cultured in minimal essential medium (MEM) (Earle's salts; Thermo Scientific) supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals), 2 mM glutamine (Thermo Scientific), and 10% tryptose phosphate broth (BD), at 28°C with 5% CO 2 . The dengue HR-LAVs (DV1 ILLT, DV2 GVII, DV3 GVLL, and DV4 GFLV) and WT (DENV1 WP74, DENV2 16881, DENV3 CH53489/ UNC3001, and DENV4 341750) strains used in this study have been described previously (37). CHIKV 181/25 was obtained from BEI Resources. The CHIKV TM17 HR-LAV was described previously (35). For SINV experiments, all cells were cultured in MEM (Hanks salts; without sodium bicarbonate; Sigma) supplemented with 10% fetal bovine serum (Atlanta Biologicals), 2 mM glutamine (Thermo Scientific), and 10% tryptose phosphate broth (BD), at 28°C without CO 2 . The SINV TM17 HR-LAV and Sindbis virus heat-resistant (SVHR) WT strain have been described previously (38,43).
## Establishment of infected cell lines and SIE experiments
Cells were infected at a multiplicity of infection (MOI) = 0.01 with the appropriate HR-LAV, infections were maintained, and cells were passaged as needed for at least a month. Infections were monitored weekly via conventional plaque assay. Cells were considered infected once virus concentrations leveled off from acute infection levels and titers remained consistent (44).
For the SIE experiments, cells (either infected with an HR-LAV or non-PI cells) were seeded at a density of 1 × 10 6 in 24-well plates (Corning) in triplicate in serum-free media and allowed to adhere at 28°C for 1 h. Cells were then infected with the appropriate WT strain at MOI = 10 and incubated at room temperature (RT) with gentle rocking for 1 h. The inoculum was removed, cells were washed with 1× phophate buffered saline (PBS), and fresh growth medium was added. Culture media were collected and replaced with fresh media at 24 h, 48 h, and 72 h post-secondary infection (45) and centrifuged at 1,000 rpm for 10 min to remove any cellular debris. Glycerol was added at a final concentration of 10%, and samples were snap-frozen in liquid nitrogen and stored at -80°C until analysis.
## Fluorescent in situ hybridization determination of virus titers
In order to differentiate between the HR-LAVs and their WT counterparts, fluorescent in situ hybridization (FISH) (46) was employed using biotinylated probes designed to span the genomic deletion site of the HR-LAVs. Probe sequences are as follows: DV1 ILLT-5ʹ-G TTTAATCCTAGCCACCCTATTCCTATCTTC-3ʹ; DV1 WP74-5ʹ-TCCTAGCCATGTCAGCAGAATCCC TATTCC-3ʹ; DV2 GVII-5ʹ-ATTCCTATCCATGTTATGAGGATTTTCATA-3ʹ; DV2 16681-5ʹ-TCCATG TGATAATGACTCCTATGAGGATTT-3ʹ; DV3 GVLL-5ʹ-CAACCCTATCCAAGTTATTCCAATTTTCAT -3ʹ; DV3 CH-UNC-5ʹ-CCCTATCCAAGTCAAGAGAACAACTATTCC-3ʹ; DV4 GFLV-5ʹ-CGTGCCAA TCCACAAAATTAGGATTCTAAT-3ʹ; DV4 341750-5ʹ-CCACAACACTAAGAACCCAATTAGGATTC T-3ʹ; CHIKV TM17-5ʹ-TGTGCCCACCGACACAATGACTACAGTCAT-3ʹ; CHIKV 181/25-5ʹ-CACA CCCACCATCGACAGGAGTACGAACGA-3ʹ; SINV TM17-5ʹ-TAACACTGCAACAGTTACGCCGACG GCTAA-3ʹ; SVHR-5ʹ-CAATCATCATCGCCACGGTAGCTGATGCGA-3ʹ. All probes were 5ʹ-biotin conjugated and HPLC purified (Sigma). Probes were confirmed to be strain specific, i.e., the probes for the HR-LAVs did not bind to their WT counterpart, and vice versa.
C6/36 cells were seeded at 2 × 10 5 cells per well in 96-well plates in serum-free media and allowed to adhere at 28°C for 1 h. For each SIE sample, 10-fold serial dilutions were made in 1× Hanks Buffered Salt Solution (Corning) containing 3% FBS. Cells were infected in triplicate for each dilution and incubated at room temperature for 1 h. The inoculum was then removed, and fresh growth medium was added. For the dengue SIE samples, cells were fixed 4 days post-infection, and for the CHIK and SINV SIE samples, cells were fixed 2 days post-infection. Cells were washed twice in 1× PBS and then fixed in 1× PBS containing 3.7% formaldehyde (Sigma) at RT for 20 min. Cells were washed twice with 1× PBS and then incubated in permeabilization buffer (1× PBS and 0.1% Triton X-100 [Sigma]) at room temperature for 20 min. Cells were washed twice with 1× PBS and twice with hybridization buffer (10% formamide [EMD Millipore], 2× SSC [Thermo Scientific], 0.1% Tween-20 [Sigma], 2 mM vanadyl ribonucleoside complex [VRC; New England Biolabs], 250 ng/µL sonicated/sheared salmon sperm DNA [Takara Bio], 250 ng/µL C6/36 Cot DNA, and 10% dextran sulfate [Sigma]). Cells were incubated in hybridization buffer at 37°C for 2 h. Plates were heated at 80°C for 10 min (using a slide warmer). Probes were denatured at 100°C for 5 min and then quickly added at a final concentration of 5 µM. Plates were sealed using aluminum foil sealing film and incubated at 37°C for 24 h. Cells were consecutively treated with: (i) one wash: 2× SSC, (ii) one wash: 1× SSC, and (iii) three washes: 1× PBS 0.1% Tween-20. All washes were incubated at 37°C for 5 min. Blocking buffer (2× SSC; 0.1% Tween-20; 1% BSA [heat-shock fraction; Sigma]; 1% egg white; 2 mM VRC) was heated at 37°C and added to each well, and plates were incubated at 37°C for 1 h. The buffer was removed, and a blocking buffer containing 5 µg/mL of Fluorescein (DTAF) Streptavidin (Jackson ImmunoResearch), heated at 37°C, was added. Plates were incubated (protected from light) at 37°C for 2 h. Cells were washed three times in 1× PBS containing 0.1% Tween-20 for 5 min each wash. Cells were counterstained with Hoechst 33342 (MP Biomedicals) and washed once with 1× PBS. The number of fluorescent foci in each well was determined using a Leica DMIL fluorescent microscope and then used to calculate virus titers. All statistical analyses and graphs were produced using GraphPad Prism.
C6/36 Cot DNA was generated as follows. One confluent T75 flask of C6/36 cells was washed three times with 1× PBS, then 10 mL of 1× TE (pH 8.0) was added, and a cell scraper was used to detach cells. Cells were centrifuged for 10 min at 1,000 rpm, 10°C. The cell pellet was resuspended in 3 mL of cell lysis buffer (10 mM Tris-Cl [pH 8.0], 0.1 M EDTA [pH 8.0], 0.5% SDS, 20 µg/mL DNase-free pancreatic RNase, and 100 µg/mL Proteinase K) and incubated at 50°C 3 h. Lysate was cooled to room temperature, and an equal volume of phenol (equilibrated with 0.1M Tris-Cl, pH 8.0; Sigma) was added and mixed on a tube rotator for 10 min. Lysate was centrifuged at 6,500 rpm for 15 min at 22°C. The aqueous phase was transferred to a new tube. The phenol extraction was repeated, and the aqueous phases were pooled. Then an equal volume of phenol:chloro form:isoamyl alcohol (IAA; 25:24:1; Sigma) was added to the aqueous phase, and the tube was gently inverted and centrifuged at 6,500 rpm for 15 min at 22°C. The aqueous phase was transferred to a new tube, and 0.2 vol of 10M ammonium acetate and 2.5 vol of 100% EtOH were added and mixed and left at RT until the DNA precipitant formed (~5 min). DNA was centrifuged for 5 min at 6,500 rpm at 22°C. The pellet was washed twice with 70% EtOH and air-dried. DNA was dissolved in 50-100 µL (depending on the size of the pellet) of 1× TE (pH 8.0) for 24 h at 4°C. Electrophoresis was used to confirm the extraction of high molecular weight DNA. The C6/36 genomic DNA was diluted to 10 µg/µL in 1.2× SSC. For each Cot fraction, 500 µL of DNA was boiled for 2 min then placed in a preheated 60°C heat block. For the Cot2 fraction, the DNA was removed after 1 min of incubation and placed on ice for 2 min. For the Cot3 fraction, the DNA was removed after 1.5 min of incubation and placed on ice for 2 min. To each fraction, 55 µL of preheated (42°C) 10× S1 nuclease buffer and 5 µL S1 nuclease (100 U/µL; Promega) were added, and the reaction was incubated at 42°C for 1 h. Then 0.1 vol of 3M sodium acetate (pH 5.2) and 1 vol of isopropanol were added and mixed and left at room temperature until the DNA precipitant formed (~5 min). DNA was centrifuged at 14,000 rpm for 20 min at 4°C. The pellet was washed twice with 70% EtOH, air-dried, and dissolved in 100 µL of water. Confirmation of fractionation was done using electrophoresis. Equal concentrations of the Cot2 and Cot3 fractions were pooled for use in the FISH buffer.
## RESULTS
Insect cell culture is a simple method of evaluating virus infection of a whole mosquito because many features of the virus replication are indistinguishable between in vitro and in vivo infection. As our goal is to manipulate the SIE phenomenon in vivo, whole mosquitoes, we began by examining the SIE effects of our HR-LAV strains that were infected in three cell lines of Ae. albopictus (U4.4 C7-10 and C6/36), and two cell lines of Ae. aegypti (Aag2 and A20), from the two most relevant mosquito vectors for DENV1-4 and CHIKV, and then challenge these with their pathogenic homolog. We present these data here and make the argument that the preliminary data in vitro is predictive of a robust SIE response in vivo. We examined this phenomenon in cell lines from different mosquito species, cell lifecycle, and tissue lineages because it is thought that each culture may display different infection and physical profiles that may impact the SIE phenomenon (47). Of the arbovirus group, SINV, the prototypical Alphavirus, is the best studied virus system for which the mechanism by which SIE has been elucidated (48)(49)(50). Therefore, SINV serves as the WT control and reference model in these studies.
In the literature, previous data reporting SIE in alphavirus and orthoflaviviruses, whether in vitro and in vivo, are incomplete, and the methods used to measure interfer ence are inconsistent, difficult to interpret, or reproduce (32,33,51,52 reviewed in [42]). Nonetheless, several consistent features of the superinfection interference phenomenon are known for the alpha and orthoflavivirus genera and have been applied to our studies. In the alphavirus system, SIE is established within 15 min of infection (48). Briefly, in SINV, SIE is the result of the RNA polymerase auto-processing into non-structural replicase complexes that can only produce +RNA. Replication of the genomic RNA is fast, and SIE is established early during the infection. Any incoming homologous virus +RNA cannot be transcribed into the negative strands required for a second homologous virus to be replicated because the necessary replicase is no longer available. In our proposed biocontrol approach in mosquitoes in the wild, SIE would be established in HR-LAV-infec ted mosquitoes. Thus, upon a primary infection of HR-LAV, we cultivated infected cells, allowing the primary infection to proceed until firmly established for 30 days. The 30-day period of infection allows the cells to proceed from the acute, lytic phase of infection into the persistent state (44).
It was also necessary to establish the MOI required for >95% of the cells to be infected to produce a single cycle of infection for the second challenge infection to provide an accurate assessment of interference. This ensures that the secondary infection is uniformly infecting all the cells to correctly quantify the virus load during the SIE period analyzed. The effective MOI determined by immunofluorescence was MOI = 10 pfu/mL for orthoflavivirus and ~1 pfu/mL for alphaviruses. The establishment of SIE in orthofla vivirus is slower than that of the alphavirus system, with interference seen around 10 h post-infection (hpi) (34). However, no systematic study of SIE kinetics in the orthoflavi viruses could be found. Although the mechanism of interference in the orthoflavivirus system is not well understood, it is also thought to involve RNA replication components and limited cellular co-factors; in other words, both virus and host participate in the establishment of SIE (33,53).
Previous studies of SIE in SINV infections have shown that SINV interferes with different alphaviruses in Ae. albopictus cells C6/36, C7-10, and U4.4 by several orders of magnitude, depending on the virus strain. The time points previously published were from the 48 h post-infection period (42). We took time points at 24, 48, and 72 hpi to allow the comparison of alphavirus SIE kinetics to that of orthoflavivirus kinetics. We did not see the same level of interference in our SINV control as previously published (45) for two possible reasons. In the report cited above, the acute infections were done at an MOI = 100 pfu/mL, and the resulting virus was titered on chick embryo fibroblast (CEF) secondary cultures, while ours were done at MOI = 10 and titered on C6/36 cells. Also, we previously demonstrated that SINV interfered with Aura, Semliki Forest, and Ross River viruses, showing heterologous interference. SINV infection did not interfere with a secondary infection of yellow fever virus, which illustrates a lack of SIE heterolo gous interference in a related arbovirus (orthoflavivirus). Although this early publication reported that SINV did not support heterologous SIE of an orthoflavivirus, our data have shown that SINV can interfere with DENV 4 in C6/36 cells, demonstrating a heterologous interference system.
A second caveat necessary to measure SIE correctly was to substitute the RT-qPCR measurements of RNA for quantitation of genomic equivalents. Initially, the viral load of select infections in these cell lines was determined by RT-qPCR, using established and validated RT-qPCR assays. However, we found that cells PI with SINV in a superinfected culture produce more WT RNA than is produced by a single acute infection, precluding direct comparison. This was determined by measuring the amount of HR-LAV RNA compared to the WT virus RNA (data not shown). To replicate the findings of Condreay (42), SINV was used as the challenge virus to Ae. albopictus cells PI with SINV TM17. As shown in Fig. 1, the amount of virus suppressed was comparable to the amount of WT virus collected at 24 hpi in that study, ~2-3 orders of magnitude. Additionally, a secondary infection of SINV reduced WT virus production by ~3 orders of magnitude in all cells tested in that study. This indicates that the SINV HR-LAV is capable of establishing SIE at comparable levels of exclusion as seen previously (42). Therefore, the conditions used in the present study for the establishment of SIE are suitable for the examination of the ability of the HR-LAV system to produce SIE. The areas showing the levels of virus collected during this 3-day period are comparable to the SIE values reported herein for SINV and discussed below. For PI infections with SINV TM17, the WT infection was reduced 99.7% in C6/36 (Fig. 1A), 93.3% in C7-10 (Fig. 1B), and 99.8% in U4.4 cells (Fig. 1 and3C). From Aag2 cells, inhibition was 99.8% and (Fig. 1E) 93% from A20 cells. SVHR is the prototype of the alphaviruses and served together with the companion HR-LAV strain SVHR TM17 as the control persistent infection to establish the correct conditions for these experiments in a reference strain model.
The establishment of persistence with each of the arboviruses in the in vitro system was monitored by a modified FISH plaque assay (discussed above) developed in this lab. For the SVHR strain studied, each of the cell clones was superinfected with SINV TM17, a well-studied alphavirus HR-LAV; CHIKV HR-LAV or DENV 1-4 HR-LAV was also studied and quantified using the FISH assay. Clear patterns of virus interference can be observed for these tests in SINV, CHIKV, and the DENV 1-4. These experiments serve as pilot experiments to refine a method to mitigate the circulation of pathogenic arbovi ruses. The establishment of an HR-LAV-immunized mosquito population that persists by natural means is predicted to impede pathogenic virus expansion. This method of pathogen regulation is expected to abate the circulation of pathogenic arbovirus while maintaining the ecology of the mosquito vectors in wild populations under natural conditions.
As detailed in Materials and Methods, each of the Aedes spp. tested was infected with HR-LAV, and the infection was allowed to mature for 30 days into a PI culture. Each of the individual PI-infected cultures was then superinfected, and samples were taken at 24, 48, and 72 hpi. The experiment contained PI cells infected with CHIKV and DENV 1-4; each of the samples was collected and measured by FISH plaque assay. The result for DENV-1 is shown in Fig. 2. SIE in Ae. albopictus and Ae. aegypti mosquito cell lines infected with DV-1 ∆ILLT HR-LAV and challenged with WT DENV 1 WP74 sampled at 24, 48, and 72 hpi. At 24 hpi, significant levels of DENV-1 were suppressed in C6/36, U4.4, and Aag2 cells (Fig. 2A). All cells sampled at 48 hpi demonstrated significant SIE (Fig. 2B). When measured at 72 hpi, significant SIE was restricted to the Ae. aegypti cells (Fig. 2C).
When DENV 1 SIE was measured as an area under the curve, the following levels of interference were determined. All cell lines produced significant levels of SIE. C6/36 cells produced 64% levels of interference, followed by C7-10 at 58%, U4,4 cells at 57%, Aag2 cells at 84%, and A20 cells at 70% all compared to the control infection. The charts are shown in Fig. 3.
The results for the DV2 HR-LAV and DENV-2 16,684 WT virus are shown in Fig. 4. The levels of SIE in Ae. albopictus and Ae. aegypti mosquito cells infected with DV-2 ∆GVII HR-LAV and challenged with DENV 2 16681 are similar to those of DENV 1. At 24 hpi, significant levels of DENV-2 were suppressed in Aag2 cells (Fig. 4A). At 48 hpi, as with DENV 1, all cell lines demonstrated significant SIE induced by HR-LAV DV2 ∆GVII when challenged by DENV 2 16681 (Fig. 4B). At 72 hpi, significant SIE was restricted to the C6/36 cells (Fig. 4C).
When DENV 2 SIE data were calculated, the following levels of interference were determined. All cell lines produced significant levels of SIE. C6/36 cells produced 64% levels of interference (Fig. 5A) followed by C7-10 58% (Fig. 5B), U4,4 cells 57% (Fig. 5C), Aag2 cells 84% (Fig. 5D), and A20 cells 70% (Fig. 5E) all compared to the control infection.
The charts are shown in Fig. 5.
The results for the DV-3 HR-LAV ∆GVLL and DENV-3 CH/UNC WT virus are shown in Fig. 6 and represent the data obtained for all the cells tested, sampled at 24, 48, and 72 h post-superinfection with the WT virus. At 24 hpi, a significant amount of SIE was detected from Aag2 cells (Fig. 6A). For this virus, all cells produced significant SIE at 48 (Fig. 6B) and 72 (Fig. 6C) hpi. This level of SIE was not seen for any of the other viruses tested.
When DENV 3 SIE was calculated as the area under the curve, the following levels of interference were determined. All cell lines produced significant levels of SIE, shown in Fig. 7. In C6/36, 75% interference was observed (Fig. 7A). Interference was observed in C7-10 (73%; Fig. 7B), U4.4 (73%; Fig. 7C), and Aag2 (66%; Fig. 7D). In Fig. 7E, A20, 68% interference was observed. All interference was measured as a percentage of the WT control.
The final DENV tested was DENV 4 HR-LAV ∆GFLV. The experimental protocol was as detailed above. As shown in Fig. 8, the cell line that produced significant SIE at 24 h was U4.4 (Fig. 8A). Again, as with all the previous DENV strains, the 48 h time point produced the most significant time PI, which interfered with WT virus production (Fig. 8B). At 72 hpi, the virus continued to interfere in C6/36, Aag3, and A20 cells to different levels (Fig. 8C).
When the area under the curve is computed, the amount of interference is 91% for C6/36 cells (Fig. 9A), 84% for C7-10 cells (Fig. 9B), 75% for U4.4 cells (Fig. 9C), 50% for Aag2 cells (Fig. 8D), and 53% for A20 cells (Fig. 9E). Also tested for the establishment of SIE was the HR vaccine strain of CHIKV, CHIK TM17, in a superinfection of the CHIK 181/25 strain used as WT. The same five Aedes spp. clones were made PI with HR CHIK TM17, superinfected with CHIKV 181/25, and media harvested at 24, 48, and 72 hpi, as with the dengue samples. CHIKV was found to be a superior suppressor of the second infection, giving 99% suppression in all cell lines tested up to 72 hpi. Results are shown in Fig. 10.
## DISCUSSION
Our superinfection exclusion experiments in cultured mosquito cells are a systematic study of the amount of interference produced in five mosquito cell lines from two mosquito species Ae. aegypti and Ae. albopictus. The date presented above (summarized in Fig. 11) shows that a variety of insect cell lines can be infected with both alpha and flaviviruses. These results further show that SIE can be demonstrated when these cells are secondarily infected with wild-type homologous virus, with one very notable exception. When cells infected with the vaccine strain of dengue 2 are subsequently superinfected with wild-type dengue, suppression of wild-type replication is initially demonstrated; however, wild-type virus recovers and by 72 h grows to normal levels. We have no explanation for this result except to say that this is not the only instance where dengue 2 has shown its own peculiarities. We have made vaccines for all four serotypes of dengue. When the dengue 2 vaccine was injected into African Green Monkeys, it produced a strong immune response with large amounts of neutralizing antibody. It resulted in strong protection upon challenge. When the dengue 2 vaccine was injected in the tetravalent context (all vaccines present), it also produced a strong immune response with high levels of neutralizing antibody; however, when challenged with wild type, it did not protect. Again, this result cannot be explained but suggests that dengue 2 may have properties not shared by other flaviviruses.
The mechanism by which alphaviruses mediate SIE is well understood. In the normal course of infection, the RNA-dependent RNA polymerase (RDRP) initially copies the incoming plus-strand RNA to the negative strand to allow the production of progeny plus strands. At some point in the first 15-30 min of infection, a protease modifies the RDRP so that the production of negative strand is no longer possible. A superinfecting plus-strand RNA encounters this protease as it is translated, and the RDRP is one capable only of producing plus strands (48, 54; reviewed in reference 42). The mechanism by which orthoflaviviruses produce SEI is poorly understood but appears to be the result of a competition for resources where the initially infecting virus has the advantage (32; reviewed in reference 42). These data also support the notion that SIE may provide a mechanism for control of these viruses in nature. In that regard, it is essential that these experiments be repeated on whole insects. We are currently seeking a collaboration to do that.
## References
1. Hawkes, Hopkins (2022) "The mosquito: an introduction"
2. Hall, Tamir "Mosquitopia: the place of pests in a healthy world"
3. Clem (2016) "Arboviruses and apoptosis: the role of cell death in determining vector competence" *J Gen Virol*
4. Alonso-Palomares, Moreno-García, Lanz-Mendozah et al. (2019) "Molecular basis for arbovirus transmission by Aedes aegypti mosquitoes" *Intervirology*
5. Piantadosi, Kanjilal (2020) "Diagnostic approach for arboviral infections in the United States" *J Clin Microbiol*
6. Carrasquilla, Ortiz, León et al. (2021) "Entomological characterization of Aedes mosquitoes and arbovirus detection in Ibagué, a Colombian city with co-circulation of Zika, dengue and chikungunya viruses" *Parasit Vectors*
7. De Almeida, Aguiar, Armache et al. (2021) "The virome of vector mosquitoes" *Curr Opin Virol*
8. Teixeira, De Brito, Correia et al. (2021) "Simultaneous circulation of zakat, dengue, and chikungunya viruses and their vertical co-transmission among Aedes aegypti" *Acta Trop*
9. Srichawla, Manan, Kipkorir et al. (2024) "Neuroinvasion of emerging and re-emerging arboviruses: a scoping review" *SAGE Open Med*
10. Khongwichit, Chuchaona, Vongpunsawad et al. (2018) "Molecular surveillance of arboviruses circulation and co-infection during a large chikungunya virus outbreak in Thailand" *Sci Rep*
11. Bhatt, Gething, Brady et al. (2013) "The global distribution and burden of dengue" *Nature*
12. Benedict, Levine, Hawley et al. (2007) "Spread of the tiger: global risk of invasion by the mosquito Aedes albopictus" *Vector Borne Zoonotic Dis*
13. Ogunlade, Meehan, Adekunle et al. (2021) "A review: aedes-borne arboviral infections, controls and Wolbachia-based strategies" *Vaccines (Basel)*
14. Lwande (2020) "Globe-trotting Aedes aegypti and Aedes albopictus: risk factors for arbovirus pandemics" *Vector-Borne Zoonotic Dis*
15. Näslund, Ahlm, Islam et al. (2021) "Emerging mosquito-borne viruses linked to Aedes aegypti and Aedes albopictus: global status and preventive strategies" *Vector Borne Zoonotic Dis*
16. Kraemer, Sinka, Duda et al. (2015) "The global distribution of the arbovirus vectors Aedes aegypti and Ae"
17. Van Bree, Visser, Duyvestyn et al. (2023) "Novel approaches for the rapid development of rationally designed arbovirus vaccines" *One Health*
18. Wilder-Smith (2020) "Dengue vaccine development by the year 2020: challenges and prospects" *Curr Opin Virol*
19. Principi, Esposito (2024) "Development of vaccines against emerging mosquito-vectored arbovirus infections" *Vaccines (Basel)*
20. Chand, Godbole, Shivlata et al. (2021) "Molecular xenomonitoring of Dengue, Chikungunya and Zika infections: a year-round study from two Dengue endemic districts of central India" *J Vector Borne Dis*
21. Edenborough, Flores, Simmons et al. (2021) "Using Wolbachia to eliminate dengue: will the virus fight back?" *J Virol*
22. Ferreira, Fairlie, Moreira (2020) "Insect vectors endosymbionts as solutions against diseases" *Curr Opin Insect Sci*
23. Wienhues (2022) "The innocent mosquito? The environmental ethics of mosquito eradication"
24. Sarita, Arunima (2022) "Chapter 3, Advances in mosquito control: a comprehensive review"
25. Terradas, Mcgraw (2017) "Wolbachia-mediated virus blocking in the mosquito vector Aedes aegypti" *Curr Opin Insect Sci*
26. Zug, Hammerstein (2015) "Wolbachia and the insect immune system: what reactive oxygen species can tell us about the mechanisms of Wolbachia-host interactions" *Front Microbiol*
27. Pimentel, Cesar, Martins et al. (2020) "The antiviral effects of the symbiont bacteria Wolbachia in insects"
28. Noman, Das, Nesa et al. (2023) "Importance of Wolbachiamediated biocontrol to reduce dengue in Bangladesh and other dengue-endemic developing countries" *Biosaf Health*
29. Ant, Mancini, Mcnamara et al. (2023) "Wolbachia-Virus interactions and arbovirus control through population replacement in mosquitoes" *Pathog Glob Health*
30. Ross (2020) "An elusive endosymbiont: does Wolbachia occur naturally in Aedes aegypti" *Ecol Evol*
31. Igarashi (1979) "Characteristics of Aedes albopictus cells persistently infected with dengue viruses" *Nature*
32. Eaton (1979) "Heterologous interference in Aedes albopictus cells infected with alphaviruses" *J Virol*
33. Laureti, Paradkar, Fazakerley et al. (2020) "Superinfection exclusion in mosquitoes and its potential as an arbovirus control strategy" *Viruses*
34. Goenaga, Kenney, Duggal et al. (2015) "Potential for co-infection of a mosquito specific flavivirus, Nhumirim virus, to block West Nile virus transmission in mosquitoes" *Viruses*
35. Nasar, Erasmus, Haddow et al. (2015) "Eilat virus induces both homologous and heterologous interference"
36. Piper, Ribeiro, Smith et al. (2013) "Chikungunya virus host range E2 transmembrane deletion mutants induce protective immunity against challenge in C57BL/6J mice" *J Virol*
37. Smith, Nanda, Spears et al. (2012) "Testing of novel dengue virus 2 vaccines in African green monkeys: safety, immunogenicity, and efficacy" *Am J Trop Med Hyg*
38. Briggs, Smith, Piper et al. (2014) "Live attenuated tetravalent dengue virus host range vaccine is immunogenic in African green monkeys following a single vaccination" *J Virol*
39. Hernandez, Sinodis, Horton et al. (2003) "Deletions in the transmembrane domain of a sindbis virus glycoprotein alter virus infectivity, stability, and host range" *J Virol*
40. Smith, Nanda, Spears et al. (2011) "Structural mutants of dengue virus 2 transmembrane domains exhibit host-range phenotype" *Virol J*
41. He, Piper, Meilleur et al. (2010) "The structure of Sindbis virus produced from vertebrate and invertebrate hosts as determined by small-angle neutron scattering" *J Virol*
42. Shiomi, Nagao, Yokota et al. (2021) "Extreme deformability of insect cell membranes is governed by phospholipid scrambling" *Cell Rep*
43. Condreay, Brown (1986) "Exclusion of superinfecting homologous virus by Sindbis virus-infected Aedes albopictus (mosquito) cells" *J Virol*
44. Brown (2010) "Host range virus mutants as vaccines for arthropod vectored viruses" *Nova Actda Leopoldina NF*
45. Riedel, Brown (1977) "Role of extracellular virus on the maintenance of the persistent infection induced in Aedes albopictus (mosquito) cells by Sindbis virus" *J Virol*
46. Karpf, Lenches, Strauss et al. (1997) "Superinfec tion exclusion of alphaviruses in three mosquito cell lines persistently infected with Sindbis virus" *J Virol*
47. Shakoori (2017) "Fluorescence in situ hybridization (FISH) and its applications" *Chromos Struct Aberrat*
48. Walker, Jeffries, Mansfield et al. (2014) "Mosquito cell lines: history, isolation, availability and application to assess the threat of arboviral transmission in the United Kingdom"
49. Adams, Brown (1985) "BHK cells expressing Sindbis virus-induced homologous interference allow the translation of nonstructural genes of superinfecting virus" *J Virol*
50. Singer, Ambrose, Danino et al. (2021) "Quantitative measurements of early alphaviral replication dynamics in single cells reveals the basis for superinfection exclusion" *Cell Syst*
51. Reitmayer, Levitt, Basu et al. (2023) "Mimicking superinfection exclusion disrupts alphavirus infection and transmission in the yellow fever mosquito Aedes aegypti" *Proc Natl Acad Sci*
53. Peleg (1965) "Infection of mosquito larvae by arboviruses" *Am J Trop Med Hyg*
54. Benelli, Jeffries, Walker (2016) "Biological control of mosquito vectors: past, present, and future"
55. Goenaga, Goenaga, Boaglio et al. (2020) "Superin fection exclusion studies using West Nile virus and Culex flavivirus strains from Argentina" *Mem Inst Oswaldo Cruz*
56. Folimonova (2012) "Superinfection exclusion is an active viruscontrolled function that requires a specific viral protein" *J Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12590647&blobtype=pdf
|
# The increased expression levels of human endogenous retrovirus-K envelope and human endogenous retrovirus-H polymerase transcripts in laryngeal squamous cell carcinoma
Saeid Jamehdar, Somayeh Jalilvand, Mehrnaz Kaffashian, Amin Tadayoni, Ehsan Khadivi, Mohammad Farahmand, Zabihollah Shoja
## Abstract
Laryngeal squamous cell carcinoma (LSCC) is a significant subtype of head and neck cancers, with tobacco and alcohol being primary risk factors. Many studies have shown that human endogenous retroviruses (HERVs), specifically HERV-K and HERV-H, have been implicated in the development and progression of various cancers, including head and neck cancers; nevertheless, there is a lack of research on the expression levels of HERV-K and HERV-H in LSCC. In this research, the differential expression of HERV-K Rec, Env, Np9, and HERV-H pol transcripts was assessed in 144 laryngeal biopsy specimens (72 polyps and 72 LSCC samples) utilizing quantitative Real-time PCR. The results showed a significant upregulation of HERV-K Env and HERV-H pol in the LSCC group compared to the polyp group (p < 0.001), suggesting their potential role in LSCC progression. The ROC curve analysis further supported the diagnostic significance of HERV-H pol and HERV-K Env transcripts in distinguishing LSCC from noncancerous polyps (HERV-H pol: AUC = 0.86; HERV-K Env: AUC = 0.76). Moreover, being over 50 years old and having an opium addiction were linked to increased expression levels of HERV-H Pol and HERV-K Env, suggesting a potential connection between these elements and laryngeal cancer (p < 0.001). These results highlight the possible utility of HERV-H pol and HERV-K Env as biomarkers for the diagnosis and prognosis of LSCC. Undoubtedly, additional research is imperative to establish the clinical efficacy of these biomarkers.
## Introduction
Laryngeal cancer has exhibited notable epidemiological changes over the past decades. The incidence of laryngeal cancer worldwide rose from 125,175 cases in 1990 to 208,083 cases in 2021 [1]. The global incidence of laryngeal cancer in 2022 was estimated to be 189,191 new cases, as reported by GLOBOCAN 2022 data. According to the Global Cancer Observatory by the International Agency for Research on Cancer (IARC), Laryngeal cancer ranked as the 20th most common cancer worldwide in terms of new cases. If we focus solely on the male gender, laryngeal cancer ranks as the 14th most common cancer, likely attributed to increased alcohol intake and smoking [2]. Laryngeal cancer incidence and deaths are elevated in areas with reduced socioeconomic status, such as certain regions of Eastern Europe, South America, and South Asia [2]. This may be since underdeveloped nations exhibit greater levels of smoking and alcohol consumption, leading to increased laryngeal cancer risk [3,4]. Moreover, other risk factors including an unhealthy diet, long-term health issues like hypertension, and human papillomavirus (HPV) infection. HPV, especially type 16, has been progressively linked to oropharyngeal cancers, notably in younger populations and in developed countries [5].
The development of laryngeal cancer is a multistep process involving genetic mutations, chromosomal instability, and dysregulation of cellular pathways. Research has revealed common chromosomal gains and losses in regions like 3p, 9p, and 17p, linked to tumor advancement and malignant change [6]. Mutations in tumor suppressor genes like RB1 and CDKN2A, which play a critical role in regulating the cell cycle, are common in laryngeal cancer [7]. One of the key points in developing laryngeal cancer is dysregulation of proteins such as p53, p21, p16, and Rb, leading to unchecked cell proliferation [7].
Human endogenous retroviruses (HERVs) are remnants of ancient retroviral infections that have integrated into the human genome through integration processes spanning millions of years. These elements comprise about 8% of the human genome and were previously considered non-functional junk DNA. However, they are now acknowledged for their involvement in diverse biological mechanisms, such as cancer progression [8]. HERV activation may result in the activation of oncogenes by acting as enhancers or alternative promoters. For instance, LTR10 elements have been demonstrated to control the expression of genes like XRCC4 and ATG12 in colorectal cancer [9]. Furthermore, activation of HERV elements in cancer progression involves other mechanisms such as epigenetic regulation and the modulation of specific signaling pathways [10]. HERVs play a complex role in cancer progression, as they can promote oncogenesis and stimulate innate immunity. This dual aspect of HERVs indicates that additional research is essential to completely understand the implications of HERVs in cancer and other diseases [11].
Among HERV families, HERV-K (subgroup HML-2) is the most recently active, with integration events occurring up to 5 million years ago [12]. The expression of HERV-K proteins has been detected in conditions such as amyotrophic lateral sclerosis (ALS), various cancers, and autoimmune disorders [13]. HERV-H is another member of the HERV family that integrated into the genome around 35 million years ago. Unlike HERV-K, HERV-H elements are particularly notable for their high transcriptional activity in human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs) [14]. HERV-K and HERV-H encode structural proteins including Gag, Pol, and Env. Moreover, HERV-K encodes accessory proteins like Rec and Np9 that modulate viral and host cell functions [15,16]. Rec is functionally analogous to HIV-1 Rev, which induces carcinogenesis via interaction with tumor suppressor genes [17,18]. Np9 engages with intracellular transcription factors and proto-oncogenes, such as c-Myc, leading to enhanced oncogenic signaling [19]. HERV-K Env, Np9, and Rec promote oncogenic effects in multiple cancers, including breast, melanoma, prostate, and several others, by enhancing the growth, survival, and invasive traits of tumor cells. The HERV-H pol gene encodes reverse transcriptase and integrase enzymes. In theory, heightened expression of HERV-H pol could amplify retrotranscription and integration occurrences, leading to genomic instability, a characteristic of cancer progression [20].
Limited research specifically focusing on the role of HERV-K or HERV-H in laryngeal cancer exists. However, investigations in head and neck squamous cell carcinomas, a category that encompasses laryngeal cancer, have revealed varying levels of HERV loci expression between tumor tissue and adjacent normal tissue. These findings imply that the activation of HERVs and their subsequent impacts on gene regulation, immune response modulation, and oncogenic signaling could potentially influence the progression of laryngeal cancer, similar to their effects in other cancer types. Consequently, in this research, we aim to evaluate the differential expression of HERV-K Env, Np9, and Rec as well as HERV-H pol genes in laryngeal squamous cell carcinoma (LSCC) samples compared to laryngeal polyp samples.
## Materials and methods
## Research population
This cross-sectional study was conducted with participants from Ghaem Hospital and Kasra Clinic, two prominent referral centers for laryngoscopy in Mashhad, Iran, between 2023 and 2025. A total of 144 biopsy specimens obtained from patients undergoing laryngoscopy were included in the study following the acquisition of informed consent from the participants. Clinical data were obtained via a structured questionnaire. Patients' inclusion criteria were: Histopathologically confirmed diagnosis of LSCC or noncancerous laryngeal polyps (a common benign laryngeal lesion) based on biopsy specimens, age 18 years or older, and no prior treatment with radiotherapy, chemotherapy, or surgical intervention related to laryngeal disease before sample collection. Exclusion criteria were: patients with a history of prior malignancies or concurrent cancers other than LSCC, presence of active infections or inflammatory laryngeal conditions other than polyps at the time of biopsy, and individuals who had received radiotherapy, chemotherapy, or immunotherapy before sample collection.
The samples were divided into two groups based on pathological findings, including 72 laryngeal squamous cell carcinoma (LSCC) and 72 noncancerous laryngeal polyp (a common benign laryngeal lesion) samples. The average age of patients was as follows: 47.49 ± 11.87 (noncancerous polyp group) and 57.72 ± 8.17 (LSCC group).
The methodologies adhered to the ethical principles of the Helsinki Declaration and obtained approval from the Ethics Committee of Tehran University of Medical Sciences (TUMS) (IR.TUMS.SPH.REC.1402.057) under research standards.
## RNA extraction and cDNA synthesis
All biopsy samples were combined into RNA Later (Yekta Tajhiz Azma, Iran; Cat. No YT9085), kept overnight at + 4 °C, and subsequently moved to -80 °C until nucleic acid extraction. Samples were homogenized with a micropestle and RNA extraction was carried out using TRIZOL reagent following the manufacturer's procedure (Yekta Tajhiz Azma, Iran; Cat. No YT9065). The purity and concentration of RNA was assessed utilizing the NanoDrop spectrophotometer and agarose gel electrophoresis. RNase-free DNase I was used to eliminate genomic DNA contamination according to the manufacturer's procedure (Sinaclon, Iran; Cat. No MO5401). cDNA was synthesized from 1 µg of total RNA utilizing a cDNA synthesis kit from ParsTous, Iran, according to the manufacturer's instructions (Cat. No A101162).
## Relative quantification by qRT-PCR
Real-Time PCR assay was conducted using YTA 2×SYBR Green qPCR Mastermix from Yekta Tajhiz Azma, Iran (Cat. No. YT2551) on the Rotor-Gene Q Real-Time PCR system (QIAGEN Co.) to amplify the target genes (HERV-K Env, Np9, Rec, and HERV-H pol) and GAPDH as the reference gene. PCR amplification reactions were conducted in 25 µL reaction mixtures consisting of cDNA, 2x SYBR Green master mix, and 10 pmol of each primer pair under the subsequent conditions: 5 min at 95 °C followed by 40 cycles for 15 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. A melting curve analysis was conducted by heating the sample from 65 °C to 95 °C following the amplification cycles. The sequences and details of the primers are shown in Table 1.
Relative quantitation of target genes was conducted utilizing the 2 -(ΔΔCt) method, with GAPDH serving as the reference control. To compare mRNA levels across groups, the relative value of each target gene was calculated as the fold-change between the case and control groups.
## Statistical analysis
The statistical analysis was conducted using R software (Version: 4.3.3). Paired T-test and Wilcoxon signed-rank statistical tests were used to compare non-normal data in two dependent groups. Normally distributed variables were evaluated using both the Welch ANOVA test and two-tailed Student's t-test. Statistical significance was considered for p-values below 0.05 in this study (*p < 0.05).
## Results
## Expression of HERV-K Rec, Env, Np9, and HERV-H pol mRNAs
A statistically significant increase in the expression of HERV-K Env and HERV-H pol transcripts was observed in the laryngeal squamous cell carcinoma (LSCC) group in comparison to the polyp group (p < 0.001 and p < 0.001, respectively). The mRNA expression levels of HERV-K Rec and Np9 were observed to be slightly elevated in the polyp group compared to the LSCC group. However, it is important to note that this difference did not reach a statistical significance level (p > 0.9 and p = 0.9, respectively) (Fig. 1).
According to the histopathological type of LSCC, there was no significant differential expression detected for HERV-K Env, Np9, Rec, and HERV-H pol genes among the three distinct types of LSCC (p = 0.3, p = 0.6, p = 0.3, and p = 0.2, respectively) (Fig. 2). We also investigated whether HERV-K gene transcripts were correlated with the expression levels of one another or with the HERV-H pol gene, in the two groups studied. As shown in Fig. 3, a significant positive correlation was found in both the LSCC and polyp groups. However, the most robust statistically significant positive correlation was identified between the HERV-H pol and HERV-K Env transcripts (R = 0.74, p = 9.5e-14; Fig. 3A) and between HERV-K Rec and HERV-K Np9 transcripts within the cancer group samples (R = 0.85, P = 8.3e-15; Fig. 3F).
Table 2 displays the association between the expression of HERV-K Rec, Env, Np9, and HERV-H pol mRNAs and the clinicopathological features of the patients. A significant age-related difference was observed in the mRNA levels HERV-H Pol and HERV-K Env, with a higher expression detected in individuals over 50 years old (p < 0.001). Likewise, a notable rise in the mRNA expression levels of HERV-H Pol and HERV-K Env was noted in opium addicts compared to non-addicts (p < 0.001). Furthermore, analysis of the correlation between the expression of HERV-K Rec, Env, Np9, and HERV-H pol mRNAs and the level of SCC differentiation (poorly, moderately, and well differentiated) revealed no significant differences.
## ROC curve analysis of HERV-K Rec, Env, Np9, and HERV-H pol transcripts
The diagnostic accuracy of higher expression levels of HERV-H pol and HERV-K Env in LSCC cancer patients compared to the control group (polyp) was evaluated using a ROC curve. The AUC analysis demonstrated a remarkable distinction between the LSCC and polyp groups based on the expression levels of HERV-H pol and HERV-K Env, indicating excellent diagnostic performance (HERV-H pol: AUC = 0.86; confidence interval [CI]: 0.8 -0.92; HERV-K Env: AUC = 0.76; CI: 0.68 -0.84). There was no significant difference in the predictive values of HERV-K Np9 and Rec mRNA between the two groups (HERV-K Np9: AUC = 0.56; confidence interval [CI]: 0.47 -0.66; HERV-K Rec: AUC = 0.54; CI: 0.42 -0.66) (Fig. 4).
## Discussion
The present study examined the expression levels of HERV-K Rec, Env, Np9, and HERV-H pol transcripts in laryngeal biopsies obtained from patients with polyps (benign laryngeal lesions) or SCC (laryngeal carcinoma). The findings of this study revealed a notable upregulation of the HERV-K Env and HERV-H pol transcripts in the LSCC group compared to the polyp group. In addition, the correlation analysis of HERV-K and HERV-H expression transcripts revealed a positive, robust, and significant correlation between HERV-H pol and HERV-K Env in the LSCC group. This suggests that the findings of our research align with other relevant studies in this area. It is shown that HERV-K and HERV-H, two families of human endogenous retroviruses, are implicated in the development and progression of several cancer types [21,22]. Both HERV-H and HERV-K exhibit hypomethylation and heightened expression in head and neck cancers, suggesting their role in the epigenetic dysregulation of these tumors [18]. Another research indicated that the HERV-K Env protein's immunosuppressive domain (ISD) inhibits CD8 + T-cell cytotoxicity while promoting interleukin-10 (IL-10) secretion, creating an immunetolerant microenvironment. Moreover, HERV-K Env induces epithelial-mesenchymal transition (EMT) by downregulating E-cadherin and activating SNAIL/SLUG transcription factors [23].
Analysis of the expression levels of HERV-K Rec, Env, Np9, and HERV-H pol mRNAs, based on the patients' clinical data, revealed a notably higher and significantly increased expression of HERV-H Pol and HERV-K Env in the age group over 50 years and in the opium addicted group. It can be deduced that, according to the findings of this study regarding the increased expression of HERV-H Pol and HERV-K Env in the LSCC group, and that almost 90% of opium addicts are part of the LSCC group, the combination of these factors (old age, opium addiction, and heightened expression of HERV-H Pol and HERV-K Env) may enhance the likelihood of developing laryngeal cancer [24].
HERV-K is a promising biomarker in several cancers, but current evidence does not support its use as an independent prognostic marker in head and neck cancer. On the contrary, HERV-H shows differential expression in Fig. 2 The mRNA expression levels of HERV-K Rec, Env, Np9, and HERV-H pol genes in poorly, moderately, and well-differentiated LSCC samples. The expression levels were analyzed by One-way analysis. No statistically significant difference in expression was observed for the HERV-K Env, Np9, Rec, and HERV-H pol genes across the three distinct classifications of LSCC. HERV, human endogenous retroviruses; LSCC, laryngeal squamous cell carcinoma Emerging evidence suggests that human endogenous retroviruses (HERVs), particularly HERV-K (HML-2) and HERV-H, may serve as biomarkers for tumor differentiation status in head and neck squamous cell carcinoma (HNSCC). These data demonstrate the increased expression of HERVs in poorly differentiated tumors [17,26]. In opposition to these assertions, our findings revealed no notable differences in the expression of HERV-K Rec, Env, Np9, and HERV-H pol transcripts across the three LSCC differentiated types, which may be attributed to the low sample size in our poorly differentiated tumors.
The main limitations of this study were the modest sample size and the lack of functional confirmation of our findings. Also, our findings should be validated using complementary approaches (e.g. Western blotting, or IHC) to strengthen the robustness of the findings. However, the availability of limited sample material constrained our ability to perform multiple validation techniques, which often require larger tissue sections.
## Conclusion
our results indicated that the expression of HERV-H pol and HERV-K Env mRNAs was increased during the progression of laryngeal squamous cell carcinoma (LSCC). Furthermore, the ROC curve and correlation analysis validated this finding. The expression examination of When the p-value was statistically significant (less than 0.05), it was shown in bold 1 Fisher's exact test; ***p < 0.001 Fig. 4 ROC curves were generated for the expression levels of HERV-K Rec, Env, Np9, and HERV-H pol genes in distinguishing between the LSCC and polyp (benign laryngeal lesions) groups. The AUC values for HERV-H pol and HERV-K Env expression levels were identified as significant indicators for differentiating between the LSCC and polyp groups. HERV, human endogenous retroviruses; LSCC, laryngeal squamous cell carcinoma; ROC, receiver operating characteristic; AUC, area under the ROC curve HERV-K Rec, Env, Np9, and HERV-H pol transcripts revealed no variations among the three LSCC differentiated types. Our study provided important insights into the potential role of HERV-H and HERV-K expression in the progression of LSCC, though further functional assays and complementary approaches are required for confirmation and validation of the results, respectively. It also highlights the possible use of HERV-H pol and HERV-K Env as biomarkers for LSCC diagnosis and prognosis. Certainly, additional research is required to ascertain the clinical usefulness of these biomarkers.
## References
1. Mousavi, Ilaghi, Mirzazadeh et al. (2024) "Global epidemiology and socioeconomic correlates of hypopharyngeal cancer in 2020 and its projection to 2040: findings from GLOBOCAN 2020" *Front Oncol*
2. Sung, Ferlay, Siegel et al. (2021) "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries" *CA Cancer J Clin*
3. Bosetti, Carioli, Santucci et al. (2020) "Global trends in oral and pharyngeal cancer incidence and mortality" *Int J Cancer*
4. Miranda-Filho, Bray (2020) "Global patterns and trends in cancers of the lip, tongue and mouth" *Oral Oncol*
5. Parveen, Kawatra, Maheshwari et al. (2023) "Epidemiology and aetiopathogenesis of oral malignancy: current trends" *Adv Clin Med Res*
6. Guo, Sun, Lv et al. (2008) "Allelic imbalance on chromosomes 3p, 9p and 17p in malignant progression of laryngeal mucosa" *J Laryngol Otol*
7. Volavšek, Glavač, Gale (2002) "CELL CYCLE REGULATING GENES AND THEIR PROTEIN EXPRESSION IN SQUAMOUS CELL CARCINOMA OF THE LARYNX AND HYPOPHARYNX" *Slovenian Med J*
8. Alcazer, Bonaventura, Depil (2020) "Human endogenous retroviruses (HERVs): shaping the innate immune response in cancers" *Cancers (Basel)*
9. Ivancevic, Simpson, Joyner et al. (2024) "Endogenous retroviruses mediate transcriptional rewiring in response to oncogenic signaling in colorectal cancer" *Sci Adv*
10. Cherkasova, Chen, Childs (2024) "Mechanistic regulation of HERV activation in tumors and implications for translational research in oncology" *Front Cell Infect Microbiol*
11. Yang, Dong, You et al. (2024) "Dual roles of human endogenous retroviruses in cancer progression and antitumor immune response" *Biochim Biophys Acta Rev Cancer*
12. Shin, Mun, Han (2023) "Human endogenous Retrovirus-K (HML-2)-related genetic variation: human genome diversity and disease" *Genes*
13. Shaked, Katz, Cohen-Dvashi et al. (2025) "The prefusion structure of the HERV-K (HML-2) Env Spike complex" *Proc Natl Acad Sci*
14. Gemmell, Hein, Katzourakis (2019) "The exaptation of HERV-H: evolutionary analyses reveal the genomic features of highly transcribed elements" *Front Immunol*
15. Medstrand, Dixie (1998) "Human-specific integrations of the HERV-K endogenous retrovirus family" *J Virol*
16. Yi, Kim (2004) "Evolutionary implication of human endogenous retrovirus HERV-H family" *J Hum Genet*
17. Curty, Marston, De Mulder Rougvie et al. (2020) "Human endogenous retrovirus K in cancer: a potential biomarker and immunotherapeutic target" *Viruses*
18. Kitsou, Lagiou, Magiorkinis (2023) "Human endogenous retroviruses in cancer: oncogenesis mechanisms and clinical implications" *J Med Virol*
19. Chen, Foroozesh, Qin (2019) "Transactivation of human endogenous retroviruses by tumor viruses and their functions in virus-associated malignancies" *Oncogenesis*
20. Dolci, Favero, Toumi et al. (2020) "Human endogenous retroviruses long terminal repeat Methylation, Transcription, and protein expression in human colon cancer" *Front Oncol*
21. Downey, Sullivan, Wang-Johanning et al. (2015) "Human endogenous retrovirus K and cancer: innocent bystander or tumorigenic accomplice?" *Int J Cancer*
22. Babaian, Mager (2016) "Endogenous retroviral promoter exaptation in human cancer" *Mob DNA*
23. Bhattacharyya, Dervan, Glynn et al. (2023) "Abstract P4-08-25: unravelling the role of human endogenous retrovirus K (HERV-K) in inducible nitric oxide synthase (iNOS) mediated triple negative breast cancer progression" *Cancer Res*
24. Bakhshaee, Raziee, Afshari et al. (2017) "Opium addiction and risk of laryngeal and esophageal carcinoma" *Iran J Otorhinolaryngol*
25. Agoni (2022) "Alternative and aberrant splicing of human endogenous retroviruses in cancer. What about head and neck? -A mini review" *Front Oncol*
26. Kolbe, Bendall, Pearson et al. (2020) "Human endogenous retrovirus expression is associated with head and neck cancer and differential survival" *Viruses*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12656723&blobtype=pdf
|
# In Memoriam: Leonard Mindich 1936-2024
Paul Gottlieb
Leonard Mindich passed away over one year ago at his home in New York City at age 88, after a lengthy battle with Parkinson's Disease that included cardiovascular complications. His death was reported by his family within a short New York Times Obituary that provided biographical information describing Lenny's background and his wide-ranging interests. At the time, none of the professional journals made a note of his death or provided any type of statement of the loss of a pioneer of molecular virology. The entire early history of cystovirus studies by Lenny and others is currently available in this Special Issue of Viruses. Indeed, this Special Issue likely would not have existed had it not been for Lenny's research, which encouraged other scientists to study this unique viral family. Lenny's career was exceptional, in that it spanned the eras of molecular biology, from the advent in the immediate post-war years until our modern age of structural biology. His studies included all the aspects of these times, with the remarkable ability to readily adapt and employ each new technology into his research; he always provided rich and insightful observations [1].
Lenny came from a humble background; he grew up in the Tremont area of the Bronx New York City and attended the academically selective Bronx High School of Science. He once told his lab team that, according to family lore, his father snuck from Canada to New York State as an illegal immigrant by rowboat and then made his modest success in the USA in the laundry business. Lenny attended the Agricultural School of Cornell University, as this academic avenue offered low tuition. He often spent the summer season as a volunteer at the Cold Spring Harbor Laboratories, and there, he closely interacted with the luminaries of the early bacteriophage group, including Max Delbrück, Salvador Luria, and James Watson. There, Lenny met and was taken under the wing of the legendary physicist-turnedbiologist Leo Szilard, who he greatly admired but described as an eccentric. He then did his graduate research at Rockefeller University, under the direction of Rollin Hotchkiss. The loose and less-structured curriculum of Rockefeller suited Lenny's independent nature and he received his PhD degree in 1962. Lenny performed a post-doctoral year at the Pasteur Institute in Paris and joined the Public Health Research Institute of New York (PHRI) in 1962, where he remained for the rest of his career.
PHRI was then located on Manhattan's Lower East Side, housed on three floors of New York City's Public Health Laboratory Building on First Avenue. There, Lenny ran a small but extremely productive laboratory, where graduate students, post-doctoral fellows, and even the technicians flourished. Jeffrey Strassman, the laboratory's senior member, kept us all on an even keel and made sure all ran smoothly. Lenny frequently hosted visiting fellows from the laboratory of Dennis Bamford in Helsinki, Finland in collaborative efforts that produced friends as well as data (sadly, Dennis's passing was just announced as I am writing this memorial). Lenny's office door was always open to discussions, where ideas, concepts, and experimental plans were hatched. The day usually opened with Lenny asking his team what was up, or new, using the Yiddish word "Nu"? And there was often much new information to describe.
Lenny had an enduring optimism and always brought out the best in all the staff, including myself. I arrived at the lab and certainly was no "wunderkind", but I still sequenced two of the three φ6 double-stranded RNA (dsRNA) segments; at the time, this was performed manually by reverse-transcribing RNA, cloning the cDNA segments into bacteriophage M13, and using dideoxy nucleotide chain termination synthesis with S 35 isotope labeling. The read-out was performed by interpreting the tiny electrophoresis bands on autoradiograms and aligning the contiguous segments on a mainframe computer: the monitor was a bland, green-lit CRT screen. We proofread the results by taping standard graph paper into an approximately one yard by one yard sheet-one of these tapestries for each gene segment-and penciling in the day's results. Needless to say, the entire sequence of the bacteriophage genome took several years to complete, resulting in my first significant publications.
We went on to establish the φ6 genome packaging assay, described in this issue, which was utilized in further studies at PHRI and Helsinki. Lenny encouraged Vesa Olkonnen (visiting from Helsinki) and me to try to recover a recombinant genome segment in a viable φ6 viral particle and we succeeded, setting up the first rescue technology in a segmented dsRNA virus. Lenny, along with the brothers Xueying and Jian Qiao, isolated additional cystoviruses in 1999, greatly expanding the field of study. Lenny cleverly utilized these molecular tools and additional cystovirus types to describe the mechanisms of the viral RNA recombination that is outlined in detail in this Special Issue. His final studies identified host-cell factors that facilitated cystovirus transcription and described with collaborators the crystal structure of capsid protein P1. Lenny was a driving force in the study of virology and is fondly remembered by all those who studied under his guidance and worked alongside him.
## References
1. Gottlieb, Alimova (2022) "RNA Packaging in the Cystovirus Bacteriophages: Dynamic Interactions during Capsid Maturation" *Int. J. Mol. Sci*
2. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817919&blobtype=pdf
|
# Molecular mechanism of resistance to lonafarnib conferred by mutations in the cysteine-rich region of respiratory syncytial virus fusion glycoprotein and discovery of a lonafarnib-derived antiviral PROTAC
Qi Yang, Bao Xue, Xianjie Qiu, Kaixin Yang, Jielin Tang, Anqi Zhou, Jingjing Zou, Yuhan Mao, Jiayi Zhong, Yuan Zhou, Wei Zhang, Qiong Zhang, Qingyu Xiao, Wei Tang, Zhiyu Li, Wencai Ye, Gang Zou, Wei Peng, Jinsai Shang, Xi Xu, Yixue Li, Xinwen Chen
## Abstract
Lonafarnib, an oral antiviral that targets the fusion glycoprotein of respiratory syncytial virus (RSV), has demonstrated efficacy in vitro and in vivo. However, because the RSV has evolved to become resistant to other fusion inhibitors, there is a concern that the same could occur for lonafarnib. Here, we identified resistance to lonafarnib in the RSV A2 strain and a recent clinical isolate, RSV ON1, via in vitro selection at scale. Cell-cell fusion and recombinant live RSV analysis confirmed that the mutations at K394, K399, and T400 of the cysteine-rich region of the fusion protein mediate high-level resistance. Lonafarnib resistance mutations also confer cross-resistance to other fusion inhibitors of clinical interest. All-atom molecular dynamics simulations revealed that these resistance mutations confer reduced stability to the fusion protein, thereby diminishing its binding affinity with lonafarnib. To address this vulnerability proactively and increase the barrier to resistance development, we designed the first potent proteolysis-targeting chimera (PROTAC) fusion protein degrader, compound 0179841, which uses lonafarnib and cereblon as ligands. This PROTAC effectively inhibited RSV replication. Collectively, our findings indicate that RSV develops resistance to lonafarnib in the cysteine-rich region of the fusion protein. This work sheds light on the mechanisms by which RSV evolves resistance to lonafarnib and provides a founda tion for the rational design of antivirals aimed at preventing resistance. IMPORTANCE Respiratory syncytial virus (RSV) infection poses a substantial public health challenge. Resistance to several potent fusion inhibitors, which are currently in various stages of clinical development, can readily emerge. Through a drug repurposing screen, we identified lonafarnib as an RSV fusion inhibitor; however, concerns exist regarding the potential development of resistance. Here, large-scale in vitro selection experiments revealed specific mutations within the highly conserved cysteine-rich region of the fusion (F) protein that confer high-level lonafarnib resistance across diverse RSV strains. These resistance mutations also confer cross-resistance to other clinical-stage fusion inhibitors. Mechanistic investigations demonstrated that these mutations reduce F protein stability, thereby diminishing the binding affinity of lonafarnib. As a proof of concept for an alternative antiviral strategy, we rationally designed the first potent proteolysis-targeting chimera (PROTAC) F protein degrader, compound 0179841, by utilizing lonafarnib and cereblon ligands. This novel antiviral agent effectively inhib its RSV infection by inducing degradation of the F protein. This work elucidates the molecular basis of RSV resistance to lonafarnib and establishes a strategy for developing next-generation antivirals aimed at preempting resistance.
KEYWORDS respiratory syncytial virus, fusion glycoprotein, lonafarnib, resistant mutations, proteolysis-targeting chimera R espiratory syncytial virus (RSV) is a common cause of acute lower respiratory tract infection in infants, older adults, and immunocompromised individuals (1, 2). Despite the clinical benefits of three approved RSV vaccines (Arexvy (3) [GSK]; Abrysvo (4) [Pfizer]; or mResvia (5) [Moderna]) and two monoclonal antibodies (palivizumab (6) [AstraZeneca] and nirsevimab (7) [AstraZeneca, Sanofi]), treatment remains limited to high-risk people, with approval in only a few countries and regions. Arexvy, Abrysvo, and mResvia are available for adults; however, no vaccines are approved for the prevention of RSV infection in infants or children (8). The Food and Drug Administration recently suspended all clinical trials of RSV vaccines in infants and young children following reports of severe illness in infants vaccinated with Moderna's mRNA-based RSV vaccine (9). Palivizumab and nirsevimab are approved for the prevention but not the treatment of RSV disease (10).
There are currently no approved RSV-specific therapeutic small molecules available, but multiple clinical-stage small molecules, including fusion inhibitors targeting the viral fusion glycoprotein (F) (11)(12)(13)(14)(15) and replication inhibitors targeting the viral nucleopro tein (N) (16) or large polymerase subunit (L) (16)(17)(18), are currently available as drug candidates for treating RSV infections. These three viral proteins are essential for the viral life cycle and are well conserved among Orthopneumoviruses (19,20). Nevertheless, the development of mutational escape and therapeutic resistance is still common for RSV (14,15,(21)(22)(23)(24)(25)(26)(27)(28), which is a particular concern in immunocompromised individuals.
The RSV F protein plays a critical role in mediating viral entry into host cells (20,29) and induces the formation of syncytia between infected cells and neighboring uninfec ted cells (30,31). Inhibition of viral entry and spread by blocking the exchange of a metastable state (prefusion) with a stable state (postfusion) of the RSV F protein during the fusion process has emerged as a promising treatment option for patients infected with RSV (19,32). Currently, GS-5806 (14), AK0529 (15), RV521 (12), and JNJ-53718678 (13) are several structurally heterogeneous and potent fusion inhibitors that are being clinically developed for RSV infection treatment. However, all of these inhibitors have the potential to induce the emergence of resistant mutants both in vitro and in vivo (14,15,21,22,24,26), as demonstrated by resistance-associated substitutions for GS-5806 (e.g., T400A, D486N) (21), JNJ-53718678 (e.g., L141W, D489Y) (22), AK0529 (e.g., D486N, D489A/V/Y) (24), and RV521 (e.g., D489Y) (26) in vitro, and the most frequent substitu tions (e.g., F140I, L141F/W, S398L, K399N, T400A/I, D486N, or F488L) in the F protein to GS-5806 or AK0529 have already been identified in real-world cases (14,15). Among the known mutations, D486N and D489Y have shown cross-resistance to various fusion inhibitors, and the emergence of both mutants is a potential concern for human health (33). Thus, new therapeutic options are still needed.
Lonafarnib, an orally active farnesyltransferase inhibitor, has demonstrated high therapeutic efficacy for Hutchinson-Gilford progeria syndrome (HGPS) (34) and is a phase III candidate for hepatitis delta virus (HDV) therapy (35). Recently, drug repurpos ing screens conducted by our team and another research group identified lonafarnib as a potent RSV fusion inhibitor (36,37). In a murine infection model, oral administration of lonafarnib resulted in a dose-dependent reduction in the RSV load and alleviation of lung damage. Thus, the repurposing of lonafarnib has significant potential in combating RSV infection and may offer a more accessible treatment option for patients with HGPS, HDV, or cancer comorbidities. However, whether RSV can evolve resistance mechanisms similar to those observed with other fusion inhibitors remains a concern.
The emergence of novel viruses and drug-resistant strains highlights the urgent need for innovative antiviral therapies (38). Targeted protein degradation (TPD) represents an innovative drug discovery strategy that eliminates proteins of interest (POIs) by harnessing the intracellular ubiquitin-proteasome system (UPS), macroautophagy, and endolysosomal pathways (39). Proteolysis-targeting chimera (PROTAC) molecules serve as classic representatives of TPD and typically comprise three key components: a ligand for the POI, a linker, and a ligand for the E3 ligase. By bridging the POI and the E3 ligase, PROTACs facilitate the ubiquitination of the POI, thereby triggering its degradation through the proteasome system (40). In recent years, substantial research efforts have been devoted to the development of antiviral drugs that target a variety of viruses via PROTAC technology (such as hepatitis B virus [41], hepatitis C virus [42], influenza virus [43,44], and SARS-CoV-2 [45,46]). However, the potential of using PROTAC modalities to target the F protein for degradation and thereby inhibit RSV infection remains largely unexplored.
Here, we identified the conserved cysteine-rich region of the RSV F protein as the critical site associated with resistance to lonafarnib in vitro. To investigate the underlying molecular mechanisms of RSV resistance to lonafarnib, we conducted surface plasmon resonance (SPR) and all-atom molecular dynamics (MD) simulation assays for each resistance mutation in the presence of lonafarnib. Furthermore, we applied PROTAC approaches to develop a lonafarnib-based virus-specific antiviral degrader that targets the viral F protein.
## RESULTS
## Screening for RSV resistance to lonafarnib
To identify resistance mutations against lonafarnib, the RSV A2 strain was passaged in HEp-2 cells under increasing drug concentrations, and viruses from every third passage were subjected to next-generation sequencing (NGS) (Fig. 1A and Methods). Triplicate lineages developed high-level resistance after 18 passages, exhibiting a mean 20.9-fold increase in half-maximal effective concentration (EC 50 ) values compared with those of the control virus (Fig. 1A). NGS revealed that the D392N and K399N mutations appeared in the F gene of the RSV A2 strain in all three lineages, the K399N mutation of which occurred in a stepwise manner (Fig. 1B; Table S1). The D392N and K399N mutations are located in the 392-401 microdomain of the F protein and in immediate proximity to the lonafarnib-binding microdomain (486-489 residues) in prefusion F (Fig. 1C) (36,37). Correlation analysis of the mutation frequencies with >5% mutation across all nine samples revealed a significant negative correlation (Spearman coefficient of -0.88, P < 0.01) (Fig. 1D). χ 2 testing of paired-read distributions confirmed non-independence, indicating mutual exclusivity (Fig. 1E).
Parallel passaging of the clinical isolate RSV ON1 (18 passages, with triplicate lineages) in the presence of increasing concentrations of lonafanib yielded EC 50 increases ranging from 26.9-fold to 37.3-fold (Fig. 2A). Six mutations in the RSV F gene were identified: T72L, V76A, K80N, L119I, T335I, and K394R (Fig. 2B; Table S2). Four mutations (T72L, V76A, K80N, and T335I) presented a low prevalence (<10% mean frequency): T72L (7.2% in lineage 2), V76A (8.0% in lineage 1), K80N (7.8% in lineage 1), and T335I (5.1% in lineage 2). These four mutations occurred in a non-stepwise manner. In contrast, L119I and K394R were conserved across lineages and detected in passages 15-18. The amino acid change at position 119 is in the 27-amino-acid peptide (p27), which is removed by furin during the post-translational processing of the F protein (47,48). The 394th amino acid residue is located in the 392-401 microdomain of the F protein and is in immediate proximity to the lonafarnib-binding microdomain (486-489 residues) in prefusion F (Fig. 2C) (36,37). Owing to the low mutation frequency and skip-like occurrence of these four sites (T72L, V76A, K80N, and T335I), we only conducted correlation analysis on L119I and K394R mutations with frequencies > 5% between all eight samples, which showed a significant positive correlation (Spearman coefficient of 0.87, P < 0.01) (Fig. 2D). Physical distance precluded read-level independence analysis between the L119I and K394R mutations.
Overall, selection at the scale revealed that RSV can lead to the development of lonafarnib resistance. The differences in mutation sites could arise in the RSV A2 and ON1 strains, and all these mutations were not situated in the lonafarnib-binding sites; rather, several mutations in the 392-401 microdomain of the RSV F protein were preferred.
## Characterization of lonafarnib-resistant mutants
To comprehensively investigate which mutations are responsible for lonafarnib resist ance, we constructed eight mutations identified in this study (T72L, V76A, K80N, L119I, T335I, D392N, K394R, and K399N) and other fusion inhibitors reported resistanceassociated substitutions (D392G/N, K394R, S398L, K399I/N, T400A/I, and D401E) in the 392-401 microdomain (15,21,36,(49)(50)(51)(52). Using a GFP-split fusion model for cell-cell fusion mediated by the RSV F protein (Fig. S1), we observed that the substitutions in RSV F resulted in its loss (T72L), decrease (K80N, T335I, and K399I), or increase (V76A, L119I, and K394R) in membrane fusion ability (Fig. 3A andB). Other substitutions (D392G/N, S398L, K399N, T400A/I, D401E, and F488L) of RSV F have the same ability to induce fusion as the wild-type F protein does. F488 serves as a binding site for lonafarnib (37) and acts as an experimental positive control. F488L alteration decreased drug binding and resulted in a loss of inhibitory activity. Compared with wild-type F, lonafarnib significantly inhibited V76A, K80N, L119I, and D392G/N mutant-induced cell-cell fusion activity, suggesting that alterations at these sites are largely not responsible for lonafarnib resistance (Fig. 3A andB). Lonafarnib did not or only slightly inhibit the fusion activity induced by T335I, S398L, K394R, K399I/N, T400A/I, and D401E, indicating that these mutations in the cysteine-rich region might lead to the resistance of RSV to lonafarnib.
Next, we investigated mutations in recombinant live RSV carrying single F protein mutations. Seven mutant viruses (V127G, L141F, K394R, K399N, T400A, T400I, and F488L) were used in this study, with V127G and L141F mutant viruses used as the experimental negative controls and F488L as the positive control. The infectious titer produced by HEp-2 cells infected with wild-type RSV increased faster than did the titer produced by the viruses with mutations (Fig. S2A), and after 90 h, the titer of the wild-type virus was almost 10-fold greater than the titer achieved by viruses with mutations except for V127G (Fig. S2A). Among the individual mutants tested against lonafarnib, the infection of the wild-type and V127G mutant viruses was effectively inhibited; L141F showed only minimal resistance (~1.7-fold to 3.3-fold); K399N (~9.8-fold increase in the EC 50 and ~8.2fold increase in the EC 90 ), and F488L (~15.2-fold increase in the EC 50 and ~10-fold increase in the EC 90 ) presented high-level resistance relative to that of the wild-type virus; the increased resistance caused by T400A (~33-fold increase in the EC 50 and ~24fold increase in the EC 90 ) was approximately equivalent to that caused by T400I; K394R (~111.5-fold increase in the EC 50 and ~79.2-fold increase in the EC 90 ) was the most resistant (Table 1; Fig. S2B). Together, these results indicate that the K394R, K399N, and T400A/I mutations in the cysteine-rich region of the F protein favor RSV resistance to lonafarnib.
Furthermore, we investigated mutant viruses (K394R, K399N, and T400A) against other clinically relevant fusion inhibitors (e.g., GS-5806 (14), AK0529 (15), RV521 (12), and JNJ-53718678 ( 13)) for cross-resistance. All three of these mutant viruses exhibited S1 for exact frequencies). (C) Residues mutated with passaging are overlaid onto the DS-Cav1 structure with lonafarnib bound.
The Cα of each mutated residue is denoted by a colored sphere. The DS-Cav1-lonafarnib complex was downloaded from PDB under accession code 8KG5. (D) Spearman correlation analysis between D392N and K399N mutations in the RSV A2 F protein.
Correlation analysis of D392N and K399N mutations with frequencies > 5% between all nine samples. (E) Linkage relationship between D392N and K399N mutations. The reads of the D392N and K399N mutations were extracted from nine samples simultaneously with two mutations. The number of reads with only the D392N mutation and only the K399N mutation, as well as the number of reads with neither mutation and the simultaneous appearance of two mutations, were counted via a contingency table, and a χ 2 test for independence on this contingency table was performed. The proportions of K399N mutations in the D392 control and D392N mutation reads (left), and the proportions of D392N mutations in the K399 control and K399N mutation reads (right) were calculated. considerable resistance to the above four fusion inhibitors, but K399N resulted in only slight resistance to RV521 (Table 2; Fig. S3). Selection for lonafarnib resistance can clearly yield mutations that confer cross-resistance to other inhibitors of clinical interest as well.
## Mechanism of lonafarnib resistance
To investigate the mechanisms underlying RSV resistance to lonafarnib in the cysteinerich region of the F protein, we expressed prefusion-stabilized RSV F variant DS-Cav1 (19) and its mutants (T335I, D392N, K394R, S398L, K399N, T400A, and D401E) in HEK293F cells (Fig. S4A andB). SPR analysis revealed that the mutants (T335I, K394R, S398L, K399N, T400A, and D401E) significantly decreased the binding affinity for lonafarnib (Fig. 4B and D through H). Compared with DS-Cav1, the D392N mutant retained the ability to bind to lonafarnib but presented slightly lower affinities (Fig. 4A andC). Next, all-atom MD simulations were conducted on DS-Cav1 and these mutants. The simulations were executed for 500 ns and replicated three times (Fig. S5) under two different conditions, where the prefusion F trimer complex was present in the absence or presence of lonafarnib. These residues are located around the central cavity, which includes a cysteine-rich region and heptad repeat B (Fig. 5A andB). The DS-Cav1 variant presented three stable β-sheet conformations in the main cluster structure (Fig. 5A andB), and these β-sheets formed at residues 331-335, 394-398, and 486-488 in the DS-Cav1 cluster structure. As shown, these residues engage in polar contacts: C331 with S398, T335 with M396, K394 with S491, and K399 with S485 (Fig. 5B). The formamide group of the lonafarnib chemical structure is crucial for binding by providing polar interactions, which form hydrogen bonds with residues T397, S398, D486, and E487 (37). The radius of gyration (Rg) directly measures protein stability, showing its folding state and compact ness. The solvent-accessible surface area (SASA) reflects stability by indicating the protection of the hydrophobic core. Both parameters enable a comprehensive assess ment of protein stability. MD simulations revealed that the DS-Cav1 and D392N muta tions were the most stable among all the simulation cases (Fig. 5C). Conversely, the K394R mutation resulted in the highest values of both Rg and SASA, indicating a decrease in stability compared with those of DS-Cav1 and other mutants. Notably, even in the presence of lonafarnib, DS-Cav1 still maintained its status as the most stable (Fig. 5D). Similarly, K394R also presented the highest Rg and SASA values, indicating its relatively lower stability than the other mutations.
To investigate the effects of lonafarnib on various mutations, we analyzed the SASA values of residues surrounding the central cavity, which are residues 333-335, 392-401, and 486-488. The results demonstrated that the DS-Cav1 and D392N mutants presented increased binding affinity for lonafarnib (Fig. 5E). In contrast, other mutations significantly weakened the interaction between the F protein and lonafarnib. Notably, in the RSV F complex model with the K399N mutation, relatively low Rg and SASA values were observed (Fig. 5D), suggesting that lonafarnib may still bind to this mutation. However, the central cavity area of the K399N mutation appears less stable than that of the DS-Cav1 and D392N mutations (Fig. 5E). These findings are consistent with previous experimental data (Fig. 4), validating the role of mutations in modulating the lonafarnib interaction within the molecular mechanisms. Compound 0179841-induced degradation of the F protein confers an anti-
## RSV effect
The resistance mutations in the central cavity region of the RSV F protein confer resistance by structurally destabilizing the protein and reducing the binding affinity of lonafarnib. Crucially, the long-range impact of these mutations on the β-sheet network and hydrophobic cavity conformation underscores how current fusion inhibitors remain susceptible to allosteric resistance mechanisms. As most small-molecule candidates targeting the RSV F protein pose a considerable risk of inducing drug resistance mutations, these findings highlight the pressing need for new approaches to address this vulnerability proactively and overcome the barrier to resistance development. Here, we applied the PROTAC approach to develop a lonafarnib-based antiviral degrader, 0179841, that targets the RSV F protein. The molecule 0179841 anchors to the hydrophobic domain of the RSV F protein through its lonafarnib module and recruits the E3 ubiqui tin ligase complex via the cereblon ligand, thereby enabling ubiquitination-dependent degradation of the target (Fig. 6A; see synthetic details in Fig. S6 andS7). Computational docking predicted the binding conformation between PROTAC 0179841 and both the RSV F protein and the cereblon protein. The best binding poses were identified on the basis of geometric and energy matching, yielding a binding score of -11.38. The results showed that compound 0179841 interacts directly with three residues (Phe137, Phe140, and Arg339) in the monomeric F protein and two residues (Asn351 and Trp380) in the cereblon protein (Fig. 6B). Four sets of experiments were conducted to assess the selective degradation and inhibitory effects of compound 0179841 on the RSV F protein. In HEK293T cells expressing the RSV F protein, we measured the protein levels of F with and without compound 0179841 treatment. We found that compound 0179841induced degradation of the F protein is time-and concentration-dependent (Fig. 6C andD), which appears to occur through proteasomal degradation pathways (Fig. 6E), but lonafarnib did not affect F protein levels (Fig. 6C andD). Notably, compound 0179841 also induced significant degradation of the F protein during RSV infection (Fig. 6F).
Next, using the GFP-split fusion model for cell-cell fusion mediated by RSV, we observed that the compound 0179841 significantly inhibited RSV-induced cell-cell fusion activity (Fig. 7A). Compound 0179841 also effectively inhibited the replication of the RSV A2 and ON1 strains in HEp-2 cells without affecting cell viability, with EC 50 values of 0.45 ± 0.04 µM and 0.46 ± 0.03 µM, respectively (Fig. 7B andC). The selectivity index (SI) of compound 0179841 exceeds 217.39 in HEp-2 cells, indicating its safety at the cellular level. Furthermore, we evaluated the in vitro antiviral efficacy of compound 0179841 against RSV ON1-GFP in primary human bronchial epithelial cells (HBECs). Treatment with compound 0179841 resulted in a significant decrease in the RSV infection level, as evidenced by marked reductions in the GFP intensity (Fig. 7D). Taken together, these results suggest that compound 0179841 can target the F protein for degradation and confer antiviral activity.
## Compound 0179841 inhibits RSV infection in human respiratory organoids
Human respiratory organoids are widely utilized as model systems for investigating human respiratory viral pathogenesis and supporting pharmaceutical research (53)(54)(55)(56)(57).
Here, we differentiated HBECs into mature, physiologically representative apical-out airway organoids. These organoids contain basal, secretory (goblet), and outward-fac ing ciliated cells, mimicking the apical surface of the airway epithelium in vivo (Fig. 8A). Next, we evaluated the efficacy of compound 0179841 against RSV in the human airway organoid model. Compound 0179841 significantly reduced viral loads and protein expression in a dose-dependent manner in lung airway organoids (Fig. 8B through E).
Compared with the DMSO control, compound 0179841 at a concentration of 5 µM reduced viral titers by more than 100-fold (Fig. 8B). Additionally, the GFP intensity demonstrated a pronounced antiviral effect of compound 0179841 against RSV ON1-GFP (Fig. 8C through E). These data demonstrate that compound 0179841 can inhibit RSV infection in a physiologically relevant organoid system.
## DISCUSSION
Fusion inhibitor-based interventions for RSV are facing increasing resistance develop ment in vitro and in vivo, underscoring the need for new antivirals or alternative modes of action. Lonafarnib, a novel antiviral that targets the F protein, has shown efficacy in a BALB/c mouse infection model (36,37). Given the adaptations that RSV has already exhibited to other fusion inhibitors, understanding potential lonafarnib resistance mechanisms is crucial. The results presented herein demonstrate that multiple different mutational residues in the fusion protein of diverse RSV strains are associated with lonafarnib resistance, with a notable preference for mutations in the cysteine-rich region. Mechanistic studies indicate that resistance mutations occupy internal sites within the prefusion RSV F protein, likely destabilizing it and reducing its binding affinity for lonafarnib. Importantly, we designed and identified a lonafarnib-based antiviral degrader, compound 0179841, which inhibits RSV infection by inducing F protein degradation.
The T335I and T400A mutations in the F protein of the RSV Long strain (rHRSV-A-GFP), which were previously linked to lonafarnib resistance, were confirmed (36). Here, large-scale passaging of RSV with lonafarnib selected for resistance mutations identified K399N in the RSV A2 strain and K394R in the clinically dominant RSV ON1 strain as prevalent substitutions (>5% mutation frequency, emerging in a stepwise manner). The differing mutations observed among these strains emphasize the complexity and was further validated via several recombinant live RSV strains carrying single F gene mutations (Table 1). Interestingly, the D392N and K399N mutations in the RSV A2 F protein were significantly negatively correlated, suggesting that they are not independ ent (Fig. 1E). These findings seem to indicate that lonafarnib could induce the D392N mutation to counteract mutations at other sites in the cysteine-rich region. Furthermore, lonafarnib resistance mutations (K394R, K399N, and T400A) confer cross-resistance to other clinically relevant fusion inhibitors (e.g., GS-5806 (14), AK0529 (15), RV521 (12), and JNJ-53718678 ( 13)) (Table 2). GS-5806 (14), AK0529 (24), RV521 (26), and JNJ-53718678 (59) bind to a hydrophobic pocket in the F protein that is rich in phenylalanine residues (Phe488, Phe137, and Phe140). Our recently resolved cryo-EM structure of the F protein-lonafarnib complex revealed that lonafarnib occupies the same hydrophobic cavity and β-sheet network as these inhibitors do and additionally forms hydrogen bonds with phenylalanine residues (37). This binding pocket is proximal to residues K394, K399, and T400. Collectively, these structural insights explain why the K394R, K399N, and T400A mutations confer cross-resistance to multiple fusion inhibitors, including lonafarnib. Consequently, developing effective RSV F protein inhibitors or new therapeutic options against these mutants is urgently needed before the global spread of mutant strains.
Resistance mutations to RSV fusion inhibitors are predominantly located in a central cavity adjacent to the fusion peptide within the prefusion F trimer (32,33). These regions include (i) the fusion peptide (amino acid residues 140-144), (ii) the cysteine-rich region (residues 392-401), and (iii) the heptad repeat B (residues 486-489). Resistance can arise through direct or indirect mechanisms. Direct mechanisms include mutations at the residues that interact directly with the inhibitors (e.g., D486N and F488L) or hindrance of the conformational changes required for inhibitor binding (e.g., D489Y) (32). Indirect mechanisms alter F protein stability and triggering rates (e.g., K394R, T400A, D401E, and D489E), narrowing the window for effective inhibitor binding (33). Mutations associated with lonafarnib-resistant RSV infection are located mainly in the cysteine-rich region of the F protein (Fig. 1 to 3; Table 1). These resistance-associated substitutions result in a low binding affinity by interfering with the binding of the drug to the hydrophobic pocket (Fig. 4) but do not participate directly in the binding of lonafarnib to the F protein (37). Importantly, the SPR experiments utilized a stabilized prefusion F protein; conse quently, mutations that exert allosteric effects may not be detectable in this system. All-atom MD simulations revealed that mutations affect protein stability (Fig. 5). The lower Rg and SASA values in DS-Cav1 and D392N indicate greater stability. In contrast, other mutations, such as K394R, presented the highest SASA and Rg values, indicating decreased stability. This destabilization appears to disrupt the hydrophobic cavity, and the β-sheet network is critical for inhibitor binding, even for compounds that target nonoverlapping sites. Notably, mutations such as K394R in the cysteine-rich region (aa 392-401) induce conformational changes through interconnected structural motifs (e.g., β-sheets at residues 331-335 and 486-488), thereby indirectly altering the fusion peptide and HRB regions. It is worth noting that the K394R mutant confers resistance to lonafarnib by destabilizing the F protein and increasing its membrane fusion activity (Fig. 3 and5), along with cross-resistance to other fusion inhibitors (Table 2) (60). Therefore, it is conceivable that such an RSV mutant carrying the K394R mutation in the F protein may pose a potential future threat to lonafarnib-based therapeutics. This concern necessitates the development of next-generation inhibitors, or their combination with other agents, to treat RSV infection more effectively and prevent the emergence of drug-resistant RSV.
Since the emergence of PROTAC technology in 2001 (61), it has revolutionized drug discovery by enabling the degradation of disease-causing proteins. Over the past two decades, cancer research has undergone significant breakthroughs through the use of PROTACs to target "undruggable" oncoproteins (62). More recently, this approach has been extended to antiviral research, in which unique protein degradation pathways have been utilized to combat viral infections (46,63). Conventional small-molecule inhibitors targeting the RSV F protein carry inherent risks of developing resistance, as demonstrated by the rapid emergence of escape mutations under selective pressure (14,15,(21)(22)(23)(24)(25)(26). To preemptively address this limitation, we pursued a degradation-based strategy using PROTACs. This approach aims to eliminate the target protein rather than modulate its activity, thereby reducing the opportunities for resistance mutations to confer survival advantages. In this study, we explored PROTAC modalities that target the RSV F protein for degradation to achieve virus-specific antiviral effects. The identification of the small-molecule degrader 0179841 demonstrates the potential for developing and rationally designing antivirals via the PROTAC platform. Compound 0179841 is a heterobifunctional small molecule composed of two ligands connected by a linker: lonafarnib (which recruits the F protein) and cereblon (which recruits E3 ubiquitin ligase) (Fig. 6). Simultaneous binding of both the F protein and the E3 ligase by compound 0179841 induces ubiquitylation of the F protein, leading to its subsequent degradation via the ubiquitin-proteasome system. Although compound 0179841 exhibited modest antiviral effects against RSV replication in vitro (Fig. 7 and8), this proof-of-concept research is encouraging. Future studies will focus on optimizing PROTAC degrader sequences through structure-activity relationship (SAR) analysis to enhance anti-RSV efficacy, pharmacokinetic properties, and immunogenicity profiles. These attributes are dictated by sequence architecture, length, and backbone configuration, which critically influence ternary complex formation and target engagement. Additionally, given that lonafarnib binds to the monomeric F protein and trimeric prefusion F protein (37), compound 0179841 may also bind inside the central cavity of the prefusion F protein to block its conformational change in addition to the degradation of the monomeric F protein. Although the dependence of compound 0179841 on lonafarnib binding means that it cannot directly overcome lonafarnib-associated resistance, as resistant mutations reduce the lonafarnib-F protein interaction (Fig. 4), it operates via a distinct antiviral mechanism-targeted degradation. Therefore, the resistance profile of 0179841 against RSV may plausibly differ from that of lonafarnib. Such insights would be invaluable for the development of novel antiviral therapeutics targeting this specific interaction, potentially offering a viable strategy against RSV.
## Limitations of the study
Sequencing of escape populations revealed L119I and K394R mutations in the F gene of RSV ON1 in response to lonafarnib, and a correlation analysis of the two mutation frequencies revealed a significant positive correlation. However, the L119 residue is located in the p27 peptide, which is removed by furin during the post-translational processing of the F protein (47,48); thus, we were unable to verify its role in drug resistance at the protein level. Additionally, our data suggest that at the doses and within the short time frames we utilized, the replication of RSV is sensitive to compound 0179841-mediated F protein degradation in vitro. However, in vivo validation in mice is limited for the following reasons: CRBN exhibits notable species-specific differences between mice and humans, especially in terms of substrate recognition and binding properties. A humanized CRBN (hCRBN) mouse model is therefore needed. Moreover, as a virus-directed strategy targeting proteostasis, further improvement of the EC 50 against RSV in vitro, as well as analyses of the long-term pharmacokinetics, toxicity profiles, and delivery efficiency to lung tissue of compound 0179841 in an in vivo context, is still necessary.
## MATERIALS AND METHODS
## Cells, viruses, and compounds
HEp-2 cells (cat#CCL- CO₂ conditions. Human bronchial epithelial cells (HBECs, cat#CP-H009) were isolated from bronchial tissue, purchased from Procell, and subsequently cultured in specialized medium (Procell, cat#CM-H009). All cell lines were confirmed mycoplasma-free through monthly PCR testing. RSV A2 and RSV ON1-GFP were described elsewhere (37). Recombinant live RSV carrying V127G, L141F, K399N, T400A, T400I, or F488L mutant of F protein was a kind gift from Gang Zou (Shanghai Ark Biopharmaceutical Co., Ltd, China). Recombinant live RSV carrying the K394R mutant of the F protein was a kind gift from Wencai Ye (Jinan University, China) (60,64). All experiments involving live RSV were conducted in the biosafety level 2 (BSL-2) facilities in Guangzhou National Laboratory (GZNL). All viruses were propagated in HEp-2 cells and subsequently subjected to viral titer determination using a fluorescence focus assay (FFA), as previously described (37). Briefly, viral titrations were performed with 10-fold serial dilutions in HEp-2 cells. Twenty-four hours after inoculation, the cell supernatants were discarded, and the cells were fixed in 4% paraformaldehyde (PFA) for 15 min and then permeabilized with 0.3% Triton X-100 for 15 min at room temperature (RT). RSV was probed using anti-RSV-FITC (GeneTex, cat#GTX36375) diluted 1:100, over a 1 h incubation at RT. Fluorescent images were acquired with an Operetta CLS High-Content Screening System (PerkinElmer). Fluores cence dots were quantified and normalized to the total nuclear count using ImageJ analysis software (RRID: SCR_003070; National Institutes of Health, NIH).
Lonafarnib (cat#HY-15136), GS-5806 (cat#HY-16727), AK0529 (cat#HY-109142), RV521 (cat#HY-123475), and JNJ-53718678 (cat#HY-112180) were purchased from MedChemEx press. Compound 0179841 was synthesized as described below.
## In vitro selection for RSV resistance to lonafarnib
To select for the development of drug resistance against lonafarnib, RSV A2 and RSV ON1-GFP strains were cultured in the presence of increasing concentrations of lonafarnib and passaged 18 times, respectively. Virus isolates recovered from culture at various passages were then characterized for their resistance to lonafarnib. To initiate passaging, HEp-2 cells were seeded in a 12-well plate at a density of 2 × 10 5 cells per well, and both lonafarnib and virus were then added the following day. Lonafarnib was initially added at 250 nM, and the virus was added at an MOI of 0.1 per well. Two days post-infection, each well was scored for cytopathic effects (CPE) on an Operetta CLS High-Content Screening System (PerkinElmer), and 500 µL of the supernatant from the well with ~30% CPE was passaged to each well in the next culture plate. The remaining supernatant and cells were stored at -80°C until further analysis. The passage culture was set up in triplicate, and passaging was performed independently. Along with the cultures passaged with lonafarnib, RSV A2 and RSV ON1-GFP strains were passaged with 0.1% DMSO in three independent wells to serve as a passage control, respectively.
## RSV full-genome sequence
The total RNA was extracted from the passaged supernatant and cell stocks using TRIzol LS reagent (Invitrogen, cat#10296028) according to the manufacturer's instructions. Approximately 10 µg of the extracted RNA was then dispatched to Guangdong Magigen Biotechnology Co., Ltd for full-genome sequencing using the Novaseq technique.
Ribosomal RNA was depleted from samples using the Ribo-off rRNA Depletion Kit (Vazyme Biotech Co., Ltd., cat#N406-01/02). Sequencing libraries were prepared with the ALFA-SEQ RNA Library Prep Kit II (Findrop Biosafety Technology [Guangzhou] Co., Ltd., cat#NRI002E) according to the manufacturer's instructions. Constructed amplicon libraries are subjected to PE150 sequencing on the Illumina platform.
A workflow diagram of RSV single-nucleotide variant (SNV) calling is shown in Fig. S8. Briefly, raw sequencing data were processed using Fastp v0.23.2 to remove adapter sequences and low-quality reads. Reads with a quality score below 30 (Q30) and shorter than 36 bp were filtered out to ensure high-quality input for subsequent analyses (65). Filtered reads were aligned to the viral reference genome using BWA-MEM v0.7.17. The reference genome used was the RSV strain A2 or ON1, complete genome, avail able from the NCBI GenBank database (accession no. KT992094 or MW582528). PCR duplicates were marked and removed using Picard Tools v2.26.9. Sequencing depth was calculated using SAMtools v1. 19.2 (66-68). To ensure comprehensive variant detection, we employed a multi-caller approach using five different variant calling algorithms: BCFtools v1.19 (Using htslib 1.19.1) (69), FreeBayes v1.3.6 (70), Mutect2 (GATK v4.2.5.0) (71), Strelka v2.9.10 (72), and VarScan v2.4.6 (73). Additionally, we utilized DeepVariant v1.6.0 (74), a deep learning-based variant caller, to complement our analysis. Amino acid changes resulting from the detected variants were annotated using SnpEff v5.2 (75). A custom annotation database was built using the RSV strain A2 or ON1 from the NCBI GenBank database (accession no. KT992094 or MW582528) to ensure accurate functional predictions. Frequency thresholds for reporting mutations were set at 5% for Illumina sequencing.
## Evaluation of the in vitro antiviral activity
The inhibitory effects of compounds on wild-type RSV, passaged viruses, and recombi nant RSV were evaluated in HEp-2 cells. The cells were seeded into 96-well plates at a density of 1.3 × 10 4 cells per well for 20 h. The virus was then inoculated at a dose of 0.2 MOI per well, and a 3-fold dilution series of the indicated compounds was added in triplicate. Three days post-infection, the level of viral replication was quantitatively assessed by FFA, and EC 50 values were derived by fitting a nonlinear regression curve to the data in GraphPad Prism 8.0 software (RRID: SCR_002798, Graphpad Software Inc.). EC 90 values of lonafarnib for recombinant RSV were calculated by the formula (EC 90 = 9^1/HillSlope*EC 50 ).
The inhibitory effects of compounds on RSV ON1-GFP were evaluated in HBEC. Specifically, 1.3 × 10 4 cells per well were seeded into 96-well plates, infected with RSV ON1-GFP at an MOI of 2, and subsequently incubated with either lonafarnib or compound 0179841. At 72 h post-infection, the infected cells were visualized using an EVOS M5000 Cell Imaging System (Thermo Fisher Scientific). The GFP intensity was quantified in five randomly selected fields per experiment using ImageJ analysis software (RRID: SCR_003070; National Institutes of Health, NIH).
## Cytotoxicity assay
HEp-2 cells (1.3 × 10 4 ) were seeded in 96-well plates for 20 h. The indicated compounds in a 3-fold dilution series were added to the cells, and three wells were performed in parallel. After 72 h, the cells were incubated with DMEM containing 10% FBS and CCK8 reagent (Beyotime Biotechnology, cat#C0038) for 30 min at 37°C. The absorbance at 450 nm and 600 nm was read using the PerkinElmer EnSight reader.
## Cell-cell fusion assay
Cell-cell fusion assays were performed using a split-GFP system (GFP1-10 and GFP11) (76). Briefly, HEK293T cells stably expressing GFP1-10 were transfected with 1,000 ng of plasmids encoding either wild-type F or F mutants containing various amino acid substitutions. Alternatively, the cells were infected with RSV and treated with 0.1% DMSO or the indicated compounds at concentrations of 5, 1, and 0.5 µM. Then, the transfected or infected cells were co-cultured with HEK293T cells stably expressing GFP11 at a 1:1 ratio (8 × 10 4 cells/well) for 24 h in a 96-well plate. Images were acquired per well on an Operetta CLS High-Content Screening System (PerkinElmer). The GFP amounts were quantified using ImageJ analysis software (RRID: SCR_003070; National Institutes of Health, NIH).
## Protein production
The cDNAs encoding prefusion-stabilized RSV F variant DS-Cav1 were cloned into the pcDNA3.1 expression vector as previously described (19,37). The DS-Cav1 mutants were constructed using a site-directed mutagenesis kit. The DS-Cav1 and its mutants were expressed in HEK293F cells with polyethylenimine (PEI, Polysciences Inc., cat#24765) as previously described (37). Briefly, the cell supernatant was harvested, centrifuged, and filtered at 4 days post-transfection. The complex was initially purified with Ni 2+ -NTA resin (Cytiva, cat#17531801) using an elution buffer (1 × PBS pH 7.4, 500 mM imidazole). The proteins were further purified using a Superose 6 10/300 Gl gel filtration column (Cytiva, cat#17517201) with running buffer consisting of 1 × PBS pH 7.4, then concentrated to about 1 mg/mL.
## SPR analysis
SPR assays were performed as previously described (37). Briefly, the interaction between RSV F proteins and lonafarnib was detected using a Biacore 8K (GE Healthcare) at 25°C in a multi-cycle mode. RSV F proteins were immobilized on a Series S Sensor chip CM7 (Cytiva, cat#29147020). Lonafarnib with concentrations of 0.78125, 1.5625, 3.125, 6.25, and 12.5 µM was prepared in buffer containing 10 mM PBS pH 7.4, 0.05% Tween 20, and 5% DMSO when testing interactions with RSV F proteins. The equilibrium dissocia tion constants (K D ) for each pair of interactions were calculated using the Biacore 8K evaluation software (Cytiva).
## Molecular dynamics simulation
The molecular dynamics simulations were performed on a SingleParticle S23TL24 workstation configured with 2× Intel Xeon Gold Platinum 8352 v Processors, 512 GB of RAM, and 4× Nvidia RTX 4090. System preparation was conducted using the Protein Preparation Wizard (RRID: SCR_016749) in Schrödinger Maestro (Schrödinger Release 2020-1, Schrödinger, LLC). The simulation systems were solvated using the SPC water model and neutralized with a 150 mM NaCl buffer. Each system, containing trimeric RSV F proteins and lonafarnib, comprised approximately 125,000 atoms within an ortho rhombic 10 × 10 × 10 Å 3 cubic box. Production simulations were carried out in the NPT ensemble, maintaining a temperature of 300 K and a pressure of 1.01325 bar. The simulations, executed with Desmond (RRID: SCR_014575) using the OPLS4 force field, ran for 500 ns each, with data recorded every 500 ps. The structural representations shown in the figures were rendered using PyMOL v.3.0.4 (RRID: SCR_000305; Schrödinger, LLC).
## Molecular docking
Protein structures of RSV F (PDB: 8KG5) and CRBD (PDB: 6BN7) were prepared for docking by retaining structural monomers, correcting structures, adding hydrogen atoms, and optimizing side chains. The resulting protein complex models were imported into MOE v2022.02 (RRID: SCR_014882) software via HDOCK v1.1 (RRID: SCR_024799). The MOE Dock module was used to dock compound 0179841 with the protein complex cereblon and RSV F. The best binding poses were identified based on geometric and energy matching, yielding a binding score of -11.38. A more negative score indicates stronger binding affinity, and the selected pose was used for interaction analysis.
## Synthesis of compounds
In total, 2-(2,6-Dioxopiperidin-3-yl)-4-fluoroisoindoline-1,3-dione (550 mg, 1.98 mmol) and DIPEA (1.1 mL, 6.54 mmol) were added to a solution of compound 1 (500 mg, 1.8 mmol) in DMSO (8 mL). The mixture was stirred at 90°C for 2.5 h. After being cooled to RT, the mixture was diluted with H 2 O and extracted with ethyl acetate (EA), the organic layer was washed with brine and dried over Na 2 SO 4 . The solvent was evaporated under reduced pressure, then purified by flash column, eluting with a gradient of 0%-2.5% MeOH in Dichloromethane (DCM) to give compound two as an emerald green oil (180 mg, yield 19%
$$tert-butyl 3-(2-(2-(2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4- yl)amino)ethoxy)ethoxy)ethoxy)propanoate (2)$$
## 3-(2-(2-(2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4yl)amino)ethoxy)ethoxy)ethoxy)propanoic acid (3)
TFA (3 mL, 41 mmol) at 0°C was added to a solution of compound 2 (180 mg, 0.35 mmol) in DCM (5 mL). After the addition, the resulting mixture was stirred for 2 h while allowing the temperature to rise slowly to RT. The mixture was spun dry to remove TFA and then purified by flash column, eluting with a gradient of 0%-2.5% MeOH in DCM to give compound three as a light-yellow oil (70 mg, yield 42%), which was used for the next step directly. HRMS (ESI) m/z: calculated for C 22 H 27 N 3 O 9 [M + H] + 478.1820, found 478.1811.
## 4-((2-(2-(2-(3-(4-(2-(4-((R)-3,10-dibromo-8-chloro-6,11-dihydro-5Hbenzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)piperidin-1-yl)-2-oxoethyl)piperidin-1-yl) -3-oxopropoxy)ethoxy)ethoxy)ethyl)amino)-2-(2,6-dioxopiperidin-3-yl)isoindo line-1,3-dione (0179841)
To a solution of compound 3 (70 mg, 0.151 mmol) in DMF (3 mL) was added (R)-1-(4-(3,10-dibromo-8-chloro-6,11-dihydro-5H-benzo [5,6]cyclohepta [1,2-b]pyr idin-11-yl)piperidin-1-yl)-2-(piperidin-4-yl)ethan-1-one (85 mg, 0.138 mmol), then added HATU (83 mg, 0.302 mmol), DIPEA (95 µL, 0.453 mmol). The mixture was stirred for 3 h, then diluted with H 2 O and extracted with EA (20 mL), the organic layer was washed with brine and dried over Na 2 SO 4 . The solvent was evaporated under reduced pressure. The residue was purified by flash column chromatography, eluting with a gradient of 0%-3.3% MeOH in DCM to give compound 0179841 as a light yellow solid (70 mg, yield 46%). HPLC: 98% (t R = 14.319 min); 1 H NMR (300 MHz, CDCl 3 ) δ = 8.86 (s, 1H), 8.45 (s, 1H), 7.56 (s, 1H), 7.54-7.43 (m, 2H), 7.17
## Western blot
HEK293T cells were seeded in 12-well plates (80% confluence) 1 day before F plasmids transfection and compound treatment. On the day of transfection, the cell culture medium was first changed, and then, the cells were transfected with 2 µg of F plasmids.
In the assays for validating the degradation of F, 10 mM DMSO stocks of the indica ted compounds were added into the medium (working concentrations = 1 µM, 0.1% DMSO) for the indicated periods of time post-transfection. Cells were scraped, washed twice with PBS, and lysed with cell lysis buffer (Beyotime Biotechnology, cat#P0013), supplemented with protease inhibitor PMSF (Beyotime Biotechnology, cat#ST505) for 30 min on ice. Cell lysates were clarified by centrifugation at 10,000 × g for 10 min at 4°C. Protein concentrations were determined using Enhanced BCA Protein Assay Kit (Beyotime Biotechnology, cat#P0009). 5 × SDS PAGE Sample Loading Buffer (Beyotime Biotechnology, cat#P0015L) was added to the samples and boiled at 100°C for 10 min. The equal concentrations of protein samples were separated by SDS-PAGE, and proteins were transferred from gels to PVDF membranes (Merck, cat#ISEQ00010). The membranes were incubated in 5% wt/vol milk in 0.1% Tween-20/PBS at RT for 1 h, incubated with the indicated primary antibodies (mouse anti-RSV F antibody, Abcam, cat#ab43812, RRID: AB_777676; mouse anti-β-actin antibody, Proteintech, cat#66009-1-Ig, RRID: AB_2687938) diluted in 0.1% Tween-20/PBS at 4°C overnight and incubated with the indicated secondary antibodies (Peroxidase AffiniPure Goat Anti-Mouse IgG, Jackson Immuno Research, cat#115-035-146, RRID:AB_2307392) diluted in 0.1% Tween-20/PBS at RT for 1 h. The membranes were washed by 0.1% Tween-20/PBS after each incubation. Signal bands were exposed and quantified using the FluorChem HD2 system (Alpha Innotech).
## Antiviral drug assay using lung airway organoids
HBECs-derived airway organoids were kindly provided by Ning Ma (Guangzhou National Laboratory, China). In brief, HBECs were initially cultured in PneumaCult-Ex Plus Medium (STEMCELL Technologies, cat#05040), then seeded at 2 × 10⁵ cells/well into anti-adher ence rinsing solution (STEMCELL Technologies, cat#07010)-treated AggreWell400 24-well plates (STEMCELL Technologies, cat#34415). They were maintained in PneumaCult Apical-Out Airway Organoid Medium (STEMCELL Technologies, cat#100-0620) for 6 days. Subsequently, the cells were transferred to similarly treated flat-bottom 24-well plates and cultured in the same medium for an additional 9 days.
The airway organoids were infected with RSV ON1-GFP at an MOI of 1 for 4 h. After removal of the inoculum, the organoids were washed twice with PBS, and the infected organoids were cultured in medium containing 0.1% DMSO or serially diluted compound 0179841. At 96 h post-infection, cell-free medium was harvested, and a TCID 50 assay was performed to determine the viral titer. The organoids were fixed in parallel in 4% PFA, permeabilized with 0.5% Triton X-100 (1 h), blocked with 5% BSA (Macklin, cat#B824162; 2 h), and incubated overnight at 4°C with primary antibodies: Goat anti-Human p63/ TP73L polyclonal antibody (R&D Systems, cat#AF1916), Rabbit anti-Human Mucin 5AC polyclonal antibody (Abways, cat#CY6826), and Mouse anti-acetylated tubulin monoclo nal antibody (Sigma-Aldrich, cat#T7451). After being washed three times with PBS, the samples were stained with Donkey anti-Goat IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 633 (Thermo Fisher Scientific, cat#A-21082; 1 hour, RT), Donkey anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 568 (Thermo Fisher Scientific, cat#A-10042; 1 h, RT), and Donkey anti-Mouse IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 568 (Thermo Fisher Scientific, cat#A-10037; 1 h, RT), followed by Hoechst 33342 nuclear counterstain (Thermo Fisher Scientific, cat#62249). After being washed three times with PBS, images were acquired on a Nikon A1 confocal microscope (Nikon). The GFP fluorescence intensity was quantified using ImageJ analysis software (RRID: SCR_003070; National Institutes of Health, NIH).
## Statistical analysis
The data in the figures represent means ± SDs and were analyzed using GraphPad Prism 8.0 software (RRID: SCR_002798, Graphpad Software Inc.). The linkage relationship between mutations was analyzed using a χ 2 test for independence. For the cell-cell fusion assay, statistical comparisons between different groups were performed using the non-parametric Mann-Whitney test, combining data from several experiments. The differences between the experimental and control groups were determined by Student's t-test involving two groups or one-way ANOVA, followed by Dunnett's post hoc test for multiple group comparisons. P-values were calculated, and statistical significance was expressed as *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant. All experiments were performed with at least three biological replicates. The details of the statistical analyses were described in the figure legends.
## References
1. Li, Wang, Blau et al. (2022) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis" *The Lancet*
2. Nguyen-Van-Tam, Leary, Martin et al. (2022) "Burden of respiratory syncytial virus infection in older and high-risk adults: a systematic review and meta-analysis of the evidence from developed countries" *Eur Respir Rev*
3. Papi, Ison, Langley et al. (2023) "Respiratory syncytial virus prefusion f protein vaccine in older adults" *N Engl J Med*
4. Walsh, Marc, Zareba et al. (2023) "Efficacy and safety of a bivalent RSV prefusion F vaccine in older adults" *N Engl J Med*
5. Wilson, Goswami, Baqui et al. (2023) "Efficacy and safety of an mRNA-based RSV PreF vaccine in older adults" *N Engl J Med*
6. (1998) "Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants" *Pediatrics*
7. Jones, Ke, Prill et al. (2023) "Use of nirsevimab for the prevention of respiratory syncytial virus disease among infants and young children: recommendations of the advisory committee on immunization practices -United States" *MMWR Morb Mortal Wkly Rep*
8. Anastassopoulou, Medić, Ferous et al. (2025) "Development, current status, and remaining challenges for respiratory syncytial virus vaccines" *Vaccines (Basel)*
9. (2024) "Review of investigational RSV (mRNA-1345) and RSV/ hMPV (mRNA-1365) vaccines in infants and children < 2 Years"
10. Cieslak (2024) "Nirsevimab immunization to prevent respiratory syncytial virus-associated lower respiratory tract infections in infants and children up to 24 months of age" *Nurs Womens Health*
11. (2016) "Safety, Efficacy and Pharmacokinetics of BTA-C585 in a RSV Viral Challenge Study (NCT02718937)"
12. Devincenzo, Tait, Efthimiou et al. (2020) "A random ized, placebo-controlled, respiratory syncytial virus human challenge study of the antiviral efficacy, safety, and pharmacokinetics of RV521, an inhibitor of the RSV-F protein" *Antimicrob Agents Chemother*
13. Martinón-Torres, Rusch, Huntjens et al. (2020) "Pharmacokinetics, safety, and antiviral effects of multiple doses of the respiratory syncytial virus (RSV) fusion protein inhibitor, JNJ-53718678, in infants hospitalized with RSV infection: a randomized phase 1b study" *Clin Infect Dis*
14. Porter, Guo, Perry et al. (2020) "Assessment of drug resistance during phase 2b clinical trials of presatovir in adults naturally infected with respiratory syncytial virus" *Antimicrob Agents Chemother*
15. Zhao, Shang, Yin et al. (2024) "Ziresovir in hospitalized infants with respiratory syncytial virus infection" *N Engl J Med*
16. Ahmad, Eze, Noulin et al. (2022) "EDP-938, a respiratory syncytial virus inhibitor, in a human virus challenge" *N Engl J Med*
17. Devincenzo, Mcclure, Symons et al. (2015) "Activity of oral ALS-008176 in a respiratory syncytial virus challenge study" *N Engl J Med*
18. Devincenzo, Cass, Murray et al. (2022) "Safety and antiviral effects of nebulized PC786 in a respiratory syncytial virus challenge study" *J Infect Dis*
19. Mclellan, Chen, Joyce et al. (2013) "Structure-based design of a fusion glycoprotein vaccine for respiratory syncytial virus" *Science*
20. Griffiths, Bilawchuk, Mcdonough et al. (2020) "IGF1R is an entry receptor for respiratory syncytial virus" *Nature*
21. Perron, Stray, Kinkade et al. (2016) "GS-5806 inhibits a broad range of respiratory syncytial virus clinical isolates by blocking the virus-cell fusion process" *Antimicrob Agents Chemother*
22. Roymans, Alnajjar, Battles et al. (2017) "Therapeutic efficacy of a respiratory syncytial virus fusion inhibitor" *Nat Commun*
23. Zheng, Liang, Wang et al. (2018) "Discovery of benzoazepinequino line (BAQ) derivatives as novel, potent, orally bioavailable respiratory syncytial virus fusion inhibitors" *J Med Chem*
24. Zheng, Gao, Wang et al. (2019) "Discovery of ziresovir as a potent, selective, and orally bioavailable respiratory syncytial virus fusion protein inhibitor" *J Med Chem*
25. Yoshida, Arikawa, Honma et al. (2020) "Pharmacological characterization of TP0591816, a Novel macrocyclic respiratory syncytial virus fusion inhibitor with antiviral activity against F protein mutants" *Antimicrob Agents Chemother*
26. Cockerill, Angell, Bedernjak et al. (2021) "Discovery of sisunatovir (RV521), an inhibitor of respiratory syncytial Full-Length Text Journal of Virology"
27. "virus fusion" *J Med Chem*
28. Coates, Brookes, Kim et al. (2017) "Preclinical characterization of PC786, an inhaled small-molecule respiratory syncytial virus L protein polymerase inhibitor" *Antimicrob Agents Chemother*
29. Rhodin, Mcallister, Castillo et al. (2021) "EDP-938, a novel nucleoprotein inhibitor of respiratory syncytial virus, demonstrates potent antiviral activities in vitro and in a non-human primate model" *PLoS Pathog*
30. Zhao, Singh, Malashkevich et al. (2000) "Structural characteriza tion of the human respiratory syncytial virus fusion protein core" *Proc Natl Acad Sci*
31. Kahn, Schnell, Buonocore et al. (1999) "Recombinant vesicular stomatitis virus expressing respiratory syncytial virus (RSV) glycopro teins: RSV fusion protein can mediate infection and cell fusion" *Virology (Auckl)*
32. Techaarpornkul, Barretto, Peeples (2001) "Functional analysis of recombinant respiratory syncytial virus deletion mutants lacking the small hydrophobic and/or attachment glycoprotein gene" *J Virol*
33. Battles, Langedijk, Furmanova-Hollenstein et al. (2016) "Molecular mechanism of respiratory syncytial virus fusion inhibitors" *Nat Chem Biol*
34. Yan, Lee, Thakkar et al. (2014) "Crossresistance mechanism of respiratory syncytial virus against structurally diverse entry inhibitors" *Proc Natl Acad Sci U S A*
35. Young, Yang, Davies et al. (2013) "Targeting protein prenylation in progeria" *Sci Transl Med*
36. Urban, Neumann-Haefelin, Lampertico (2021) "Hepatitis D virus in 2021: virology, immunology and new treatment approaches for a difficult-to-treat disease" *Gut*
37. Sake, Zhang, Rajak et al. (2024) "Drug repurposing screen identifies lonafarnib as respiratory syncytial virus fusion protein inhibitor" *Nat Commun*
38. Yang, Xue, Liu et al. (2024) "Farnesyltransferase inhibitor lonafarnib suppresses respiratory syncytial virus infection by blocking conforma tional change of fusion glycoprotein" *Sig Transduct Target Ther*
39. Ho, Zhu, Marazzi (2021) "Unconventional viral gene expression mechanisms as therapeutic targets" *Nature*
40. Alabi, Crews (2021) "Major advances in targeted protein degradation: PROTACs, LYTACs, and MADTACs" *J Biol Chem*
41. Békés, Langley, Crews (2022) "PROTAC targeted protein degraders: the past is prologue" *Nat Rev Drug Discov*
42. Montrose, Krissansen (2014) "Design of a PROTAC that antagonizes and destroys the cancer-forming X-protein of the hepatitis B virus" *Biochem Biophys Res Commun*
43. De Wispelaere, Du, Donovan et al. (2019) "Small molecule degraders of the hepatitis C virus protease reduce susceptibil ity to resistance mutations" *Nat Commun*
44. Xu, Liu, Ma et al. (2022) "Discovery of oseltamivir-based novel PROTACs as degraders targeting neuraminidase to combat H1N1 influenza virus" *Cell Insight*
45. Zhao, Wang, Pang et al. (2022) "An anti-influenza A virus microbial metabolite acts by degrading viral endonuclease PA" *Nat Commun*
46. Shaheer, Singh, Sobhia (2022) "Protein degradation: a novel computational approach to design protein degrader probes for main protease of SARS-CoV-2" *J Biomol Struct Dyn*
47. Zhao, Ho, Meng et al. (2023) "Generation of host-directed and virusspecific antivirals using targeted protein degradation promoted by small molecules and viral RNA mimics" *Cell Host & Microbe*
48. Mclellan, Chen, Leung et al. (2013) "Structure of RSV fusion glycoprotein trimer bound to a prefusion-specific neutralizing antibody" *Science*
49. Krzyzaniak, Zumstein, Gerez et al. (2013) "Host cell entry of respiratory syncytial virus involves macropinocytosis followed by proteolytic activation of the F protein" *PLoS Pathog*
50. Cianci, Yu, Combrink et al. (2004) "Orally active fusion inhibitor of respiratory syncytial virus" *Antimicrob Agents Chemother*
51. Roymans, Bondt, Arnoult et al. (2010) "Binding of a potent small-molecule inhibitor of six-helix bundle formation requires interactions with both heptad-repeats of the RSV fusion protein" *Proc Natl Acad Sci*
52. Douglas, Panis, Ho et al. (2005) "Small molecules VP-14637 and JNJ-2408068 inhibit respiratory syncytial virus fusion by similar mechanisms" *Antimicrob Agents Chemother*
53. Lundin, Bergström, Bendrioua et al. (2010) "Two novel fusion inhibitors of human respiratory syncytial virus" *Antiviral Res*
54. Hashimoto, Watanabe, Keshta et al. (2025) "Human iPS cellderived respiratory organoids as a model for respiratory syncytial virus infection" *Life Sci Alliance*
55. Sun, Sun, Yang et al. (2025) "A novel, covalent broad-spectrum inhibitor targeting human coronavirus Mpro" *Nat Commun*
56. Tang, Xue, Zhang et al. (2023) "A multi-organoid platform identifies CIART as a key factor for SARS-CoV-2 infection" *Nat Cell Biol*
57. Rajan, Weaver, Aloisio et al. (2022) "The human nose organoid respiratory virus model: an ex vivo human challenge model to study respiratory syncytial virus (RSV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pathogenesis and evaluate therapeutics"
58. Zhu, Yang, Luo et al. (2025) "Comparison of characteristics and immune responses between paired human nasal and bronchial epithelial organoids" *Cell Biosci*
59. (2026) *Full-Length Text Journal of Virology*
60. Andries, Moeremans, Gevers et al. (2003) "Substituted benzimidazoles with nanomolar activity against respiratory syncytial virus" *Antiviral Res*
61. Cichero, Calautti, Francesconi et al. (2021) "Probing in silico the benzimidazole privileged scaffold for the develop ment of drug-like Anti-RSV agents" *Pharmaceuticals (Basel)*
62. Tang, Li, Song et al. (2021) "Mechanism of cross-resistance to fusion inhibitors conferred by the K394R mutation in respiratory syncytial virus fusion protein" *J Virol*
63. Sakamoto, Kim, Kumagai et al. (2001) "Protacs: chimeric molecules that target proteins to the Skp1-Cullin-F box complex for ubiquitination and degradation" *Proc Natl Acad Sci*
64. Dale, Cheng, Park et al. (2021) "Advancing targeted protein degradation for cancer therapy" *Nat Rev Cancer*
65. Alugubelli, Xiao, Khatua et al. (2024) "Discovery of first-in-class PROTAC degraders of SARS-CoV-2 main protease" *J Med Chem*
66. Song, Zhu, Qiu et al. (2024) "A new mechanism of respiratory syncytial virus entry inhibition by small-molecule to overcome K394R-associated resistance"
67. Chen, Zhou, Chen et al. (2018) "fastp: an ultra-fast all-in-one FASTQ preprocessor" *Bioinformatics*
68. Heng, Handsaker, Wysoker et al. (2009) "1000 Genome Project Data Processing Subgroup"
69. Li (2013) "Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM arXiv"
70. Picard Toolkit (2019) "Broad Institute, GitHub Repository"
71. Danecek, Bonfield, Liddle et al. (2021) "Twelve years of SAMtools and BCFtools. Gigascience 10:giab008"
72. Garrison, Marth (2012) "Haplotype-based variant detection from short-read sequencing"
73. Benjamin, Sato, Cibulskis et al. (2019) "Calling somatic SNVs and Indels with Mutect2" *Bioinformatics*
74. Kim, Scheffler, Halpern et al. (2018) "Strelka2: fast and accurate calling of germline and somatic variants" *Nat Methods*
75. Koboldt, Zhang, Larson et al. (2012) "VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing" *Genome Res*
76. Poplin, Chang, Alexander et al. (2018) "A universal SNP and small-indel variant caller using deep neural networks" *Nat Biotechnol*
77. Cingolani, Platts, Wang et al. (2012) "A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3" *Fly (Austin)*
78. Cabantous, Waldo (2006) "In vivo and in vitro protein solubility assays using split GFP" *Nat Methods*
79. (2026) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12875686&blobtype=pdf
|
# Investigation of a Family Cluster of Human Infections With Highly Pathogenic Avian Influenza A(H5N1) Virus, Clade 2.3.2.1e, in Cambodia, February 2023
Savuth Chin, Chansovannara Sopu, | Heng Seng, Sokly Mom, | Bor Sar, Alyss Finl, Kathrine Tan, Phil Gould, | Ju, Y Siegers, Erik Karlsson, Sonja Olsen, Timothy Uyeki, William Davis, Darapheak Chau, Sovann Ly
## Abstract
In February 2023, an 11-year-old girl in Cambodia developed severe respiratory symptoms and died of pneumonia and respiratory failure after testing positive for influenza A(H5). Contract tracing and testing identified her father as positive for influenza A(H5). Investigations revealed both were likely exposed to the same sick and dead poultry. Control measures included culling sick poultry and providing health education in the community on handling infected birds. Highly pathogenic avian influenza A(H5N1) viruses continue to pose public health risks in Cambodia.
## 1 | Introduction
During 2005 through 2022, 56 human infections with highly pathogenic avian influenza (HPAI) A(H5N1) virus were reported in Cambodia. HPAI A(H5N1) viruses are endemic among poultry in Southeast Asia, including clades 2.3.2.1e (formerly classified as A(H5) clade 2.3.2.1c) and 2.3.4.4b [1][2][3][4]. Clusters of influenza A(H5N1) cases in family members, which may indicate human-to-human transmission, are rare but have been documented [5,6].
On February 14, 2023, an 11-year-old girl from Rumlech commune, Sithor Kandal District, Prey Veng Province, Cambodia, developed cough and fever. She was evaluated as an outpatient by a village nurse and clinicians at two private clinics and received antibiotics and antipyretics, laboratory tests, and intravenous fluids during February 16-20. On February 21, she was brought to the National Pediatric Hospital in Phnom Penh for worsening respiratory distress and was admitted to the intensive care unit. A chest X-ray revealed bilateral pneumonia with diffuse bilateral alveolar opacities and left lung pleural effusion, and the patient was intubated for respiratory failure. Empiric oseltamivir treatment was started for suspected A(H5N1). Testing of upper respiratory tract specimens was negative for SARS-CoV-2 by rapid antigen test. The patient died on February 22. Because the hospital participates in surveillance for severe acute respiratory infection (SARI), nasopharyngeal (NP) and oropharyngeal (OP) swabs were collected just prior to the patient's death; these tested positive for influenza A(H5). Contact tracing and testing revealed that her father was also positive for influenza A(H5). This report describes the investigations and findings.
## 2 | Methods
Staff from Cambodia public health, animal health, and environmental health agencies comprised rapid response teams; agencies included: Cambodia Communicable Diseases Control Department (C-CDC), Ministry of Health (MoH) Field Epidemiology Training Program (FETP), National Institute of Public Health (NIPH), National Animal Health and Production Research Institute (NAPHRI), General Directorate of Animal Health and Production (GDHAP), and Ministry of Agriculture, Forestry and Fisheries. Objectives of the investigations were to identify additional A(H5) cases, determine exposure routes and modes of transmission, and implement control measures. Investigators interviewed family members, including the second case-patient, and teachers to assess dates of illness onset, to identify potentially exposed persons. They interviewed the head of each household in the village to identify respiratory illness in the community, and a mobile clinic was set up in the village for influenza testing. Animal health officials interviewed village leaders about sick and dead poultry in the area, and they sampled poultry and the environment in areas where poultry were sick or died for influenza testing.
A case was defined a person in Sithor Kandal District with laboratory-confirmed influenza A(H5) virus detected in a respiratory specimen by real-time reverse transcription polymerase chain reaction (rRT-PCR). A suspected A(H5) case was defined as a close contact of a case or anyone in Rumlech Commune presenting with fever or cough from February 23 (when the investigation and active surveillance began) to March 10. A close contact was defined as anyone who was within 1 m of a confirmed case for at least 15 min from 1 day before symptom onset in the case and ending with isolation in the hospital or burial after death of the case. Close contacts were interviewed about type and duration of exposure and any personal protective equipment (PPE) worn; they were offered oseltamivir postexposure prophylaxis and monitored daily for 10 days after their last contact with a case and asked to report presence of fever (temperature of ≥ 100°F [37.8°C]) or feeling feverish/chills; cough; sore throat; difficulty breathing/shortness of breath; eye tearing, redness, or irritation; headaches; runny or stuffy nose; muscle or body aches; diarrhea.
Combined NP/OP swabs were collected from symptomatic and asymptomatic close contacts and suspected cases, stored in viral transport medium in cold boxes, and sent to the Virology unit of the National Institute of Public Health laboratories and the Institut Pasteur, both in Phnom Penh. Specimens were tested with rtRT-PCR for influenza A and B viruses and SARS CoV-2, and if positive for influenza A virus, they were tested with CDC subtyping kits for H1, H3, H5a, H5b, and N1 [7]. A(H5)-positive specimens were sequenced using the Illumina MiSeq platform, and sequences were uploaded to GISAID. Viruses from both cases were isolated in viral culture using embryonated eggs.
Serum specimens were collected from 12 close contacts on February 23 and from nine of these again on March 17. Specimens were analyzed at Institut Pasteur by hemagglutination inhibition (HAI) assay to assess antibody reactivity against the isolated virus. The virus isolated from Case 2 A/ Cambodia/2302009/2023 was propagated in embryonated chicken eggs, inactivated, and used as the HAI assay antigen. All sera were treated with receptor-destroying enzyme (RDE) prior to testing, and chicken red blood cells (RBCs) were used for hemagglutination. Control ferret antiserum raised against A/ Duck/Vietnam/NCVD-584/2012 produced a homologous HAI titer of 1:640, confirming antigenic reactivity of the isolate with standard reference reagents. This activity was reviewed by CDC, deemed not to be research, and was conducted consistent with applicable federal law and CDC policy. 1
## 3 | Results
Rumleach Commune comprises 1952 people in 375 households. Contact tracing and collection of upper respiratory specimens for influenza testing was initiated on February 23, including classmates and attendees of Case 1's funeral. On February 24, Case 1's father (Case 2) tested positive for A(H5); cycle threshold (Ct) value was 38.44 for H5a and 36.98 for H5b. Case 2 was a 50-year-old male farmer who lived in a house < 5-min walk from the house where Case 1 lived. He developed a mild sore throat and cough on February 14; he took medication from the local pharmacy and recovered. He was asymptomatic on February 24 when he was isolated at the referral hospital, started on oseltamivir treatment, and placed under observation. He was discharged home on February 28 after NP/OP swabs collected on two consecutive days tested negative.
Viruses isolated from Cases 1 and 2 were identified as influenza A(H5N1) clade 2.3.2.1e in the Virology Unit at the Institut Pasteur du Cambodge; these were similar genetically to A(H5N1) viruses isolated from poultry in Cambodia in 2022 and 2023 [4]. The viral genomes obtained from Case 1 2 and Case 2 3 were nearly identical across all eight gene segments, differing by only a single nucleotide substitution in the PB1 gene (T1793C; L598P). This corresponds to 99.96% nucleotide identity across the PB1 segment. All other gene segments were completely identical. Full genome data for both viruses are publicly available on the GISAID EpiFlu database under the accession numbers listed in the footnotes. Paired serum specimens collected from Case 2, as well as eight other close contacts (including two who were symptomatic), were all seronegative by HAI assay.
Contact tracing of both cases identified 45 close contacts, including 24 healthcare workers (HCWs) and 18 people who attended the funeral, including Case 2; two reported fever, cough, or sore throat over the 14-day period following exposure. Enhanced surveillance identified an additional eight people in the village who had fever and cough. The HCW who conducted the initial 30-min consultation and examination did not wear any PPE; all others (including those who performed the intubation) wore at least surgical facemasks and cleaned their hands with alcohol handrub (Table S1). Of NP/OP swabs collected from 53 close contacts and people identified through enhanced surveillance, only one (Case 2) tested positive for influenza A(H5).
Active surveillance in the village identified six people with acute respiratory symptoms, and 19 people with respiratory symptoms presented to the mobile clinic for respiratory illness; all had respiratory specimens collected, and none tested positive for A(H5).
Investigations revealed that since early February, 22 chickens and three ducks died in the village. In early February 2023, wildfowl died at a lake ~50 km from Rumleach Commune; 9 of 29 of these tested positive for A(H5N1) [8]. Test results from samples of nine healthy chickens in the village taken on February 23 were negative for A(H5); testing results for samples taken on February 24 from two sick, one dead, and two healthy chickens, two ducks, and three environmental samples were not available from animal health authorities. By February 25, approximately 80% of an estimated 7500 chickens in the village had disappeared, likely sold or removed to avoid culling.
On February 8, two chickens died at Case 1's house. Case 2 butchered the dead chickens, cooked, and ate them. Case 1 had exposure to the sick and dead chickens, which were not caged. The earliest known poultry exposures were before February 8 for Case 1 when the chickens first became ill and for Case 2 on February 8 when he butchered the dead chickens on February 8. Both Cases 1 and 2 had illness onset on February 14. Case 2 was exposed to Case 1 on February 14 when he took her to the clinic (Figure 1). The 6-day incubation period after exposure to sick/dead poultry is long, but within the range of what has been previously reported [9,10]. Neither case-patient had recent exposure to ill persons before their illness onset. Paired sera were only available for Case 2, and serology did not identify HAI antibodies to A(H5N1) virus, but not all virologically confirmed cases that experienced mild respiratory illness have serologic evidence of a detectable antibody response [11]. The viruses isolated from the two human cases were virtually identical by sequencing and were the same clade 2.3.2.1e circulating among poultry in Cambodia at the time, and the cases were epidemiologically linked and had the same symptom onset dates; these findings strongly suggest that both cases most likely had the same exposure to sick/dead backyard poultry.
The limited A(H5N1) testing data in poultry and the environment is a gap in the investigation. Although waterfowl at a lake 50 km away from the cases' village died and tested positive for A(H5N1), poultry and environmental samples from the village tested negative or results were not available. However, for some laboratory-confirmed A(H5N1) cases, the source is not always identified or the presumed exposure source did not have A(H5N1) testing confirmation [12]. Additional limitations include the lack of paired acute and convalescent serum specimens for some close contacts of both cases, and the use of serology only to detect HAI antibodies, but not neutralizing antibodies, to the A(H5N1) viruses isolated from both cases.
Investigators concluded that the source of infection for both cases was most likely from direct and close exposure to the same sick/dead backyard chickens and that human-to-human transmission was unlikely because the cases had the same illness onset date and did not have recent exposure to other sick contacts. Starting February 27, a series of health education seminars were conducted in the village on how to handle sick or dead poultry and what to do in case of respiratory illness. In response to the findings, animal health workers culled sick poultry in the village, starting on March 2.
## 4 | Conclusions
A family cluster of confirmed cases of A(H5N1) virus, clade 2.3.2.1e, infection was identified in two blood-related family members who had common exposure to sick/dead backyard poultry and the same date of symptom onset; one pediatric case developed pneumonia, respiratory failure, and died, and the other had mild illness. Clade 2.3.2.1e viruses circulating among poultry in the Mekong Delta pose a significant public health threat. Sixteen cases of clades 2.3.2.1e A(H5N1) virus infections were reported in Cambodia during 2023 and 2024, with six deaths [4].
Soputhy: investigation, supervision, project administration, writingreview and editing, data curation. Heng Seng: investigation, supervision, project administration, writing -review and editing. Sokly Mom: investigation, writing -review and editing, formal analysis. Borann Sar: conceptualization, methodology, investigation, formal analysis, writing -original draft, writing -review and editing. Alyssa Finlay: conceptualization, methodology, investigation, supervision, writingreview and editing. Kathrine R. Tan: conceptualization, methodology, investigation, formal analysis, supervision, project administration, writing -review and editing, writing -original draft, validation. Philip L. Gould: investigation, supervision, writing -review and editing. Jurre Y. Siegers: methodology, formal analysis, writing -review and editing, investigation, data curation. Erik A. Karlsson: methodology, investigation, formal analysis, supervision, writing -review and editing, data curation. Sonja J. Olsen: methodology, investigation, supervision, writing -review and editing. Timothy M. Uyeki: methodology, investigation, writing -review and editing. William W. Davis: methodology, conceptualization, investigation, formal analysis, supervision, writing -original draft, writing -review and editing, validation, data curation. Darapheak Chau: investigation, formal analysis, writingreview and editing. Sovann Ly: methodology, investigation, project administration, writing -review and editing, supervision.
## Funding
Over the past two decades, the US Centers for Disease Control and Prevention (CDC) has partnered with and supported capabilities in Cambodia to prepare and respond to emerging avian influenza threats for timely detection, response, and containment of this novel influenza virus that has global pandemic potential [13]. This activity was funded by a CDC Cooperative Agreement IP21-2101 with the Cambodia National Institute of Public Health, with one of the objectives to "ensure capacity to respond to highly pathogenic viruses transmissible among humans."
## Disclosure
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention.
## References
1. Woah (2023) "WOAH Animal Disease Events Reporting Page"
2. Dung, Dinh, Nam (2011) "Seroprevalence Survey of Avian Influenza A(H5N1) Among Live Poultry Market Workers in Northern Viet Nam"
3. Horwood, Horm, Yann (2023) "Aerosol Exposure of Live Bird Market Workers to Viable Influenza A/H5N1 and A/H9N2 Viruses, Cambodia" *Zoonoses and Public Health*
4. Siegers, Xie, Byrne "Emergence of a Novel Reassortant Clade 2.3.2.1c Avian Influenza A/H5N1 Virus Associated With Human Cases in Cambodia"
5. Olsen, Ungchusak, Sovann (2005) "Family Clustering of Avian Influenza A (H5N1)"
6. Cdc (2023) "Past Examples of Probable Limited, Non-Sustained, Person-to-Person Spread of Avian Influenza A Viruses"
7. Horwood, Karlsson, Horm (2012) "Circulation and Characterization of Seasonal Influenza Viruses in Cambodia" *Influenza and Other Respiratory Viruses*
8. Wahis (2017) "Cambodia-Influenza A Viruses of High Pathogenicity (Inf. With) (Non-Poultry Including Wild Birds"
9. Huai, Xiang, Zhou (2008) "Incubation Period for Human Cases of Avian Influenza A (H5N1) Infection, China" *Emerging Infectious Diseases*
10. Cowling, Jin, Lau (2013) "Comparative Epidemiology of Human Infections With Avian Influenza A H7N9 and H5N1 Viruses in China: A Population-Based Study of Laboratory-Confirmed Cases" *Lancet*
11. Levine, Liu, Bagdasarian (2025) "Neutralizing Antibody Response to Influenza A(H5N1) Virus in Dairy Farm Workers, Michigan, USA" *Emerging Infectious Diseases*
12. Sedyaningsih, Isfandari, Setiawaty (2005) "Epidemiology of Cases of H5N1 Virus Infection in Indonesia" *Journal of Infectious Diseases*
13. Mccarron, Kondor, Zureick (2022) "United States Centers for Disease Control and Prevention Support for Influenza Surveillance, 2013-2021" *Bulletin of the World Health Organization*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12692738&blobtype=pdf
|
# TRIM47 inhibits murine norovirus replication in a straindependent manner
Stacey Crockett, Linley Pierce, Rachel Rodgers, Meagan Sullender, Lawrence Schriefer, Mridula Srinivas, Sanghyun Lee, Megan Baldridge, Robert Orchard, Burroughs Wellcome
## Abstract
Human norovirus is the leading cause of gastroenteritis worldwide. Norovirus exhibits remarkable genetic diversity. Understanding the impact of genetic diversity on infection and immunity has been challenging due to the difficulties in in vitro cultivation and the current lack of a small animal model. Murine norovirus (MNV) has emerged as a premier model system to investigate norovirus biology. Here, we identify TRIM47 as a host restriction factor that potently inhibits MNV infection in a strain-dependent manner. We determine that TRIM47 expression inhibits an early stage of the viral life cycle of the MNV strain CR6 (MNV CR6 ), while the replication of the closely related strain CW3 (MNV CW3 ) is not restricted by TRIM47. MNV CW3 causes an acute infection that spreads beyond intestinal tissues and is lethal to immunodeficient mice. In contrast, MNV CR6 fails to spread to systemic tissues but establishes a persistent infection by infecting intestinal tuft cells. Using a forward genetic screen, we determine that genetic variation within the nonstructural protein 1 (NS1) accounts for this differential sensitivity of MNV strains to TRIM47. While most TRIM-containing proteins promote the ubiquitination and degradation of their targets, TRIM47 does neither. Instead, TRIM47 promotes the deubiquitination of the NS1/2 precursor protein. Our data provide new insight into a potential antiviral gene and mechanistic insight into norovirus evolution that may impact viral tropism.IMPORTANCE Viruses exist as genetically heterogeneous populations. Understanding the contribution of viral genetic variation to infection outcomes is critical in predicting emerging viruses and their variants. Noroviruses are genetically diverse, but human norovirus has been technically challenging to study. In this study, we use the model system murine norovirus to identify a viral strain-specific restriction mechanism where a host gene can specifically restrict one strain of the virus but has no impact on a closely related strain. Dissecting the mechanism of this specificity provides insight into viral diversity and possible host restriction pathways.
as a model system for understanding viral persistence, sterilizing innate immunity, bacterial and viral interactions, development of the immune system, and microbial triggers of inflammatory diseases (5)(6)(7)(8)(9)(10)(11)(12). MNV is an imperfect human norovirus model, as mice lack the capacity to vomit, infection rarely leads to diarrhea, and MNV encodes an additional open reading frame (4,13). However, MNV and human norovirus share many significant features, including genomic organization, molecular mechanisms of RNA expression and transcription, fecal-oral transmission, intestinal replication, and fecal shedding (4). Focusing on these shared properties increases the value of studies in the MNV model, as evidenced by many seminal advances in our understanding of norovirus biology using the MNV model (5-7, 12, 14-20).
While MNV lacks the degree of genetic diversity observed in human norovirus, there is large variation in MNV infection outcomes in vivo (21). For example, the prototypi cal MNV strains MNV CW3 and MNV CR6 share 86% nucleotide identity yet have distinct phenotypes in vivo (7)(8)(9)(20)(21)(22). After oral delivery, MNV CW3 causes an acute infection that spreads beyond intestinal tissues and is lethal to immunodeficient mice (21). MNV CW3 infects a variety of immune cells (23,24). In contrast, MNV CR6 fails to spread to systemic tissues but infects intestinal tuft cells (20,21). MNV CR6 establishes a persistent, lifelong infection in wild-type mice (7,20,21). Viral chimeras between the major capsid protein (VP1) and nonstructural protein 1 (NS1) have been able to account for systemic spread and persistence, respectively (7). However, the molecular basis for these differences in tropism remains unclear. Despite the large differences in MNV cellular tropism in vivo, most MNV strain differences are not observed in vitro. For example, both MNV CW3 and MNV CR6 can infect tuft cells in vitro, although only MNV CR6 infects tuft cells in vivo (25). One exception is that MNV CR6 does not robustly replicate in B cells either in vitro or in vivo (26). Overall, our understanding of MNV strain differences has been largely limited to in vivo studies, creating a challenge to define the molecular mechanisms that distinguish MNV strains.
As CD300lf is a universal MNV receptor required for both in vitro and in vivo infection, we hypothesized that unknown MNV restriction factors contribute to MNV strain-specific differences (27). We mined a previous genome-wide CRISPR activation screen identify ing anti-norovirus restriction factors for MNV CW3 and MNV CR6 infection of HeLa-CD300lfexpressing cells (28). Previous attempts to identify strain-specific restriction factors were unsuccessful, but one untested gene drew our attention: TRIM47. TRIM47 is an E3 ubiquitin ligase that is understudied and scored very strongly as an antiviral gene toward MNV CR6 , but not MNV CW3 (28). Here, we demonstrate that overexpression of TRIM47 specifically restricts MNV CR6 but not MNV CW3 in vitro. We identify that the genetic variation within NS1 accounts for the strain specificity of TRIM47. Surprisingly, TRIM47 recognizes NS2 and binds the NS1/2 precursor of both MNV CW3 and MNV CR6 . TRIM47 expression does not lead to the degradation of NS1/2 but rather promotes the deubiquitination of NS1/2. Thus, our data point to a complicated model in which overexpression of TRIM47 can restrict specific MNV strains in an NS1/2-dependent manner.
## RESULTS
## TRIM47 restricts murine norovirus replication in a strain-specific manner
Our previous data indicated that TRIM47 inhibits MNV CR6 but not MNV CW3 (28). We validated these findings by assessing the ability of MNV to replicate in HeLa cells stably expressing both CD300lf (HeLa-CD300lf ) and TRIM47. We chose HeLa-CD300lf cells because this cell line was used in the initial screen and because HeLa cells are more amenable to genetic manipulation compared to macrophage-like cells. While MNV CR6 replicated robustly over 72 hours in vector control cells, MNV CR6 was unable to replicate above input levels in TRIM47-expressing cells (Fig. 1A). In contrast, MNV CW3 replication is unaffected by TRIM47 expression (Fig. 1A). Mutation of critical cysteine residues in the RING domain of TRIM47 necessary for E3 ligase activity (herein called TRIM47 Mut ) abrogated antiviral activity against MNV CR6 (Fig. 1B). These data indicate that the ubiquitin E3 ligase activity is necessary for TRIM47's strain-specific antiviral activity.
We next wanted to determine at which stage of the MNV life cycle TRIM47 inhibits. We first asked if MNV CR6 could establish a replication complex in TRIM47-expressing cells. In vector control and TRIM47 Mut -expressing cells, both MNV CR6 and MNV CW3 formed structures positive for NS1, a known marker of the replication complex (Fig. 1C). However, in TRIM47-expressing cells, MNV CR6 was unable to form a replication complex, while formation of the MNV CW3 replication complex was not impeded (Fig. 1C andD). Next, we assessed the ability of MNV CR6 and MNV CW3 to synthesize RNA and produce viral proteins. There was a significant reduction in MNV CR6 genomic copies in TRIM47-expressing cells compared to empty vector and TRIM47 Mut controls (Fig. 1E). Similarly, while nonstructural proteins were produced in both control and TRIM47-expressing cells infected with MNV CW3 , MNV CR6 infection had undetectable levels of viral nonstructural proteins in TRIM47-expressing cells (Fig. 1F). These data are consistent with TRIM47 inhibiting an early stage in the MNV CR6 , but not MNV CW3 , life cycle.
## Sensitivity to TRIM47 is linked to genetic variation in NS1
To determine how genetic variation in MNV leads to differential sensitivity to TRIM47 restriction, we performed a directed evolution screen. We passaged MNV CR6 in HeLa-CD300lf-expressing TRIM47 cells (Fig. 2A). In two independent experiments, MNV CR6 populations became resistant to TRIM47 restriction after four passages (Fig. 2B). Using our recently developed method and computational pipeline, we deep-sequenced the viral population, achieving robust coverage of the viral genome (Fig. 2C) (29). While amino acid variants were identified throughout the genome, we focused on a cluster of variants in NS1 found in two independent experiments (Fig. 2D and Table S1). More specifically, two mutations caught our attention due to their abundance in individual experiments: K91R and K119E (Fig. 2D). While position 119 varies between MNV CR6 (Lys) and MNV CW3 (Arg), the identified escape mutant is a more dramatic substitution, a glutamic acid in lieu of a lysine (K119E; Fig. 2D). Additionally, we identified an association with TRIM47 resistance with a substitution of lysine at position 91 for an arginine (K91R), even though both MNV CW3 and MNV CR6 encode lysine yet differ in their sensitivity to TRIM47 restriction (Fig. 2D). To directly test if K91R or K119E confer TRIM47 resistance, we introduced these mutations into our molecular clone of MNV CR6 and generated infectious viruses harboring individual substitutions (MNV CR6 NS1 K91R and MNV CR6 NS1 K119E ). Both MNV CR6 NS1 K91R and MNV CR6 NS1 K119E grew unimpeded in TRIM47expressing cells, confirming both K91R and K119E as bona fide TRIM47 escape mutations (Fig. 2E). Taken together, these data point to a critical role of amino acid variants in NS1 in dictating MNV strain sensitivity to TRIM47-mediated inhibition.
## TRIM47 interacts with NS1/2 and NS2 in a strain-independent manner
While most MNV nonstructural proteins are processed by the viral protease, NS1/2 is cleaved by caspase-3 to generate NS1 and NS2 (30). We next sought to understand the interactions between TRIM47 and NS1, NS2, or NS1/2. For unknown reasons, exchanging the NS2 protein between MNV strains hinders the ability to recover infectious virus from molecular clones (7). Therefore, we generated viral chimeras with the NS1 region of MNV CR6 and MNV CW3 . A virus derived from MNV CW3 but containing the NS1 gene from the CR6 strain (MNV CW3 NS1 CR6 ) is sensitive to TRIM47 inhibition, indicating that the NS1 gene of CR6 is sufficient to convert a resistant virus to being sensitive to TRIM47 (Fig. 2F). Introduction of the NS1 gene from the CW3 strain of MNV into the backbone of MNV CR6 (MNV CR6 NS1 CW3 ) partially rescued viral replication (Fig. 2F). These data highlight that the genetic variation within the NS1 region is a critical determinant of TRIM47 sensitivity. Due to the incomplete rescue of MNV CR6 NS1 CW3 replication in TRIM47-expressing cells, these data also point to a role for other genomic features that may contribute to MNV strain selectivity of TRIM47.
Given the importance of NS1 in determining MNV sensitivity to TRIM47-mediated restriction, we tested whether TRIM47 interacts with NS1. While TRIM47 co-immunopre cipitated with NS1/2, we did not detect any interactions with NS1 (Fig. 3A). Surprisingly, we detected a robust co-immunoprecipitation between NS2 and TRIM47 (Fig. 3A). We also detected a physical interaction between TRIM47 and the NS1/2 protein from both MNV CR6 and MNV CW3 (Fig. 3B). Furthermore, MNV CR6 NS1/2 protein containing the escape mutants K91R and K119E still co-immunoprecipitates with TRIM47 (Fig. 3B). These biochemical interactions point to a role beyond physical binding in determining the sensitivity of MNV strains to TRIM47 antiviral activity.
Many antiviral TRIM domain-containing proteins function via degrading viral proteins (31). However, co-expression of TRIM47 did not promote proteasome-mediated degrada tion of NS1, NS2, or NS1/2 (Fig. 3C). Importantly, as an experimental control, we detected proteasomal degradation of the coxsackievirus protein 2BC by TRIM7 in the same experiment (Fig. 3C) (32). While degradation is a common fate for proteins tagged with ubiquitin, there are degradation-independent functions of antiviral TRIM-containing proteins (31). We set out to determine if TRIM47 ubiquitinates the NS1/2 of MNV CR6 when co-expressed in 293T cells. We found substantial ubiquitination of NS1/2 when coexpressed with the control protein GFP (Fig. 3D). Contrary to our hypothesis, coexpression with TRIM47 decreased NS1/2 ubiquitination (Fig. 3D). This reduction in ubiquitination is largely, but not completely, dependent upon a functional RING domain, as expression of TRIM47 Mut only modestly reduces NS1/2 ubiquitination levels. Unfortu nately, we were unable to perform the degradation and ubiquitination experiments during infection, as MNV CR6 viral protein levels are undetectable in TRIM47-expressing cells (Fig. 1F). Also, we confirmed that the addition of proteasome inhibitors, such as bortezomib, blocks the replication of MNV independently of TRIM47 (33). Thus, it is possible that TRIM47-mediated ubiquitination or degradation of MNV nonstructural proteins only occurs in the context of MNV infection. However, these data suggest that the antiviral mechanism of TRIM47 may be due to the decrease in ubiquitination of NS1/2. Taken together, these data point to a complicated, noncanonical mechanism by which TRIM47 specifically restricts MNV CR6 but not MNV CW3 infection.
## Processing of NS1/2 does not explain MNV strain specificity of TRIM47
We next tested a model in which MNV strains or escape mutants differ in their sensitiv ity to TRIM47 due to different abilities to process NS1/2 into NS1 and NS2. First, we tested the hypothesis that caspase-3 processing of NS1/2 promotes TRIM47 sensitivity. MNV CR6 virus harboring a pair of mutations (D121G and D131G) that do not hinder viral propagation but block caspase-mediated cleavage of NS1/2 (herein called MNV CR6 ΔCasp) was equally inhibited by TRIM47 as the parental virus (Fig. 4A) (18,30,34). Next, we explored the opposite hypothesis, where enhanced processing of NS1/2 by caspase-3 increases resistance to TRIM47 restriction. In support of this hypothesis, the K119E escape mutant falls within the first caspase-3 cleavage site and is predicted to increase cleavage (35). Indeed, in the absence of TRIM47, MNV CR6 harboring the TRIM47 escape mutant K119E has an increase in NS1/2 processing compared to parental MNV CR6 or the other TRIM47 escape mutant MNV CR6 NS1 K91R mutant (Fig. 4B). To directly test if an increase in NS1/2 processing contributes to the resistance of MNV strains to TRIM47, we added the pan-caspase inhibitor z-VAD during infection of control or TRIM47-overex pressing cells. Addition of z-VAD blocked the processing of NS1/2 of all MNV viruses tested, similarly to the MNV CR6 ΔCasp (Fig. 4C). Despite this block in NS1/2 processing, MNV CW3 and both TRIM47 escape mutants (MNV CR6 NS1 K91R or MNV CR6 NS1 K119E ) remained resistant to TRIM47 inhibition in the presence of z-VAD (Fig. 4D). Taken together, our data argue against a major role for NS1/2 processing by host caspases in determining sensitivity to TRIM47.
## DISCUSSION
Despite significant variation in cellular and tissue tropism of MNV strains during in vivo infections, a molecular understanding of these differences has been missing due to the near-uniform growth of MNV strains in vitro. Here, we identify an MNV strain-specific growth phenotype in cells overexpressing TRIM47. Interestingly, we mapped the genetic basis of TRIM47 strain-dependent inhibition of MNV to NS1 (Fig. 2). NS1 is a gene whose genetic variation is necessary and sufficient to explain tuft cell tropism of persistent MNV strains in vivo (7,(18)(19)(20)25). It is important to note that the NS1 variants that promote persistence also render MNV sensitive to TRIM47 (Fig. 2F). We identified a discrepancy between the genetic sensitivity to TRIM47 and the biochemical interactions of TRIM47 and viral proteins. Genetic variation within NS1 largely, but incompletely, accounts for TRIM47 strain-dependent phenotypes (Fig. 2D through F). Biochemical interactions between TRIM47 and NS2 were robust, while an interaction with NS1 and TRIM47 was undetectable (Fig. 3A). TRIM47 did interact with the precursor protein NS1/2. NS1/2 is unique among noroviruses as it is cleaved by host caspases rather than the viral protease (30). We did not detect a significant impact of enhancing or eliminating NS1/2 processing on TRIM47-mediated restriction (Fig. 4). The distinct mechanism of NS1/2 processing suggests some unique relationship between NS1 and NS2. Our data suggest that this relationship might be at the heart of TRIM47-mediated inhibition, as genetics and biochemistry implicate different regions of NS1/2. However, future work defining whether NS1 and NS2 cooperate as individual proteins or as domains in the NS1/2 precursor is necessary to enhance our understanding of norovirus biology. Furthermore, this insight is likely to help explain how TRIM47 selectively restricts MNV CR6 but not MNV CW3 replication. The antiviral activity of TRIM47 requires a functional E3 ligase domain (Fig. 1B). In reconstitution studies, we did not detect ubiquitination of NS1/2 by TRIM47. Rather, NS1/2 is ubiquitinated in the absence of TRIM47 overexpression in 293T cells (Fig. 3D). Surprisingly, TRIM47 co-expression decreased ubiquitination of NS1/2. It remains unclear how TRIM47 expression decreases NS1/2 ubiquitination. One possibility is that TRIM47 catalyzes the addition of a ubiquitin-like molecule rather than ubiquitin. Some TRIM proteins can also function as E3 SUMO ligases (36). For example, TRIM38 can polyubiquitinate some substrates such as TRIF, but TRIM38 SUMOylates MDA5, RIG-I, and cGAS (37,38). Interestingly, the SUMOylation of MDA5 and RIG-I by TRIM38 inhibits their polyubiquitination (37). Whether TRIM47 catalyzes a ubiquitin-like modification to block polyubiquitination is an intriguing possibility but needs experimental validation. TRIM47 may use an alternative method to counter the ubiquitination of NS1/2 by other E3 ligases. This may occur through direct or indirect methods, such as degrading the bona fide NS1/2 ubiquitin ligase or recruiting a deubiquitinase (DUB) to NS1/2. To this latter point, TRAF6, an E3 ubiquitin ligase, also recruits the DUB CYLD to substrates, leading to the loss of ubiquitination on target proteins (39). Interestingly, in addition to PKC-ε and NF-90, CYLD is a reported TRIM47-interacting protein and substrate of TRIM47-medi ated ubiquitination (40,41). However, there is no obvious connection between these host proteins and MNV replication. The impact of NS1/2 ubiquitination for norovirus replication or for the function of NS1/2 is not clear. In the future, determining whether NS1/2 is ubiquitinated during infection will enhance our understanding of MNV biology and the mechanism by which TRIM47 inhibits viral replication.
Our study has several limitations. First, our data rely on overexpression of TRIM47, and we do not assess the role of endogenous TRIM47. The advantages of a near-binary viral phenotype enabled us to define mechanistic differences between MNV strains but may not represent the physiological role of TRIM47. Thus, future studies leveraging genetic loss of function, including in mice, will be necessary to define the physiological role of TRIM47. Additionally, our mechanistic and biochemical studies utilized reconstitution studies since MNV CR6 protein levels were undetectable in TRIM47-expressing cells. The lack of NS1/2 degradation or ubiquitination induced by TRIM47 may be a result of a missing component in our reconstituted system. Lastly, while we identified multiple potential escape mutants within the MNV polypeptide, we only focused on two escape mutants. While each mutant was sufficient to confer resistance to viral infection, we could not detect any biochemical differences between these escape mutants and the wild-type proteins. It is likely that their mechanisms may only be revealed in the context of infection rather than our reconstitution studies. Nevertheless, the strain-dependent restriction of MNV by TRIM47 in vitro provides an opportunity to learn about norovirus genetic diversity and perhaps gain information on the function of the poorly understood viral nonstructural proteins NS1, NS2, and their precursor NS1/2.
## MATERIALS AND METHODS
## Cell culture
293T (ATCC), BV2 cells (kind gift of Dr. Skip Virgin, Washington University), and HeLa cells (ATCC) were cultured in Dulbecco's Modified Eagle Medium with 5% fetal bovine serum. Stable cell lines were generated by lentiviral transduction. Briefly, lentiviral vectors were co-transfected with the packaging vector (psPax2) and the pseudotyping vector (pCMV-VSV-G) into 293T cells using Transit-LT1 (Mirus). Forty-eight hours post-transfec tion, lentivirus was collected, filtered through a 0.45 µm filter, and added to cells. Forty-eight hours post-transduction, media were changed to contain the appropriate antibiotic (5 µg/mL blasticidin and/or 1 µg/mL puromycin). All cell lines are tested regularly and verified to be free of mycoplasma contamination.
## Plasmids
Human TRIM47 cDNA was cloned into the pCDH-MSCV-T2A-Puro vector and pCMV-N-FLAG (Clontech) for lentiviral and transient expression, respectively. For reconstitution studies, maltose-binding protein and eGFP were cloned into pCMV-N-FLAG. MNV CR6 NS1/2 (1-341), NS2 (132-341), and NS1 (1-131) were cloned downstream of eGFP in pcDNA3.1. The NS1/2 sequence of MNV CR6 and NS1/2 of MNV CW3 were cloned into pcDNA3.1 HA (Addgene #128034). Molecular clones for MNV CW3 (GenBank accession EF014462.1) and MNV CR6 (GenBank accession JQ237823) have been described previously (42). Point mutations, including NS1-K91R, NS1-K119E, and TRIM47, C9AC12A were introduced via splicing by overlap extension PCR. All plasmid sequences were verified through Sanger sequencing prior to use.
## Viral assays
MNV CW3 , MNV CR6 , and respective mutants were generated by transfecting molecular clones into 293T cells and amplifying on BV2 cells as described previously (43). For growth curves, 5 × 10 4 indicated HeLa-CD300lf cells were seeded in a 96-well plate and subsequently infected with MNV strains at a multiplicity of infection (MOI) of 0.05.
Samples were subsequently frozen at -80°C at the indicated time points. Viral titers were enumerated via plaque assay as described previously (43).
For immunofluorescence microscopy, HeLa-CD300lf cells expressing either an empty vector or TRIM47 were seeded overnight onto glass coverslips in six-well dishes. The next day, cells were infected with MNV CW3 or MNV CR6 at an MOI of 5. Twenty-four hours post-infection, cells were washed with ice-cold PBS, fixed with 4% PFA, permeabilized with 0.5% Triton X-100, blocked with 1% BSA, stained with mouse anti-NS1 and antimouse AlexaFluor 488 (Thermo Fisher), and mounted using Prolong Gold with DAPI. For each of the three independent experiments, five randomized images containing at least 20 cells were scored for the presence or absence of a replication complex.
For quantification of viral genomes, indicated HeLa-CD300lf cells were infected with MNV CR6 at an MOI of 0.05, and TRI Reagent (Sigma-Aldrich #T9424) was added at the indicated time points. RNA was isolated using the Direct-zol RNA Prep Kit (Zymo Research) following the manufacturer's protocol. Purified RNA was used for cDNA synthesis using the High-Capacity cDNA Reverse Transcription kit (Thermo Fisher Scientific #4368813). TaqMan quantitative PCR (qPCR) for MNV was performed in triplicate on each sample and standard with forward primer 5′-GTGCGCAACACAGAGA AACG-3′, reverse primer 5′-CGGGCTGAGCTTCCTGC-3′, and probe 5′-6FAM-TAGTGTCTCCT TTGGAGCACCTA-BHQ1-3′. TaqMan qPCR for actin was performed in triplicate on each sample and standard with forward primer 5′-GATTACTGCTCTGGCTCCTAG-3′, reverse primer 5′-GACTCATCGTACTCCTGCTTG-3′, and probe 5′-6FAM-CTGGCCTCACTGTCCACCT TCC-6TAMSp-3′.
For detecting viral protein production, cells were infected with the indicated MNV strains at an MOI of 5 and lysed in RIPA buffer containing HALT protease and phospha tase inhibitors at indicated time points. Lysates were clarified via centrifugation prior to Western blot analysis.
## Directed viral evolution
Two independent viral passaging experiments were conducted in HeLa-CD300lf TRIM47-expressing cells using a similar strategy as we have described previously (44). Briefly, 1 × 10 6 HeLa-CD300lf TRIM47-expressing cells were seeded in a 10 cm 2 plate and subsequently infected with MNV CR6 at an MOI of 5. Forty-eight hours post-infection, supernatants from the cultures were harvested and clarified (10 minutes at 3,000 × g). One milliliter of the clarified supernatant was added onto 1 × 10 6 HeLa-CD300lf TRIM47-expressing cells seeded in a 10 cm 2 plate. This passaging was performed four times, and 1 mL of clarified supernatant was used to isolate total RNA using the Direct-zol RNA Prep Kit (Zymo Research). Sequencing and data analysis were performed identically to what we have previously described (29,44).
## Antibodies and Western blotting
Samples were subjected to SDS-PAGE and subsequently transferred to PVDF mem branes. Membranes were blocked in TBS-T supplemented with 5% non-fat dry milk prior to probing with antibodies. Antibodies used include rabbit α-TRIM47 (1:1,000; Abcam: ab72234), mouse α-GAPDH-HRP (1:10,000; Sigma: G9295), rabbit α-GFP (1:2,000; Invitrogen: A-11122), mouse α-FLAG M2-HRP (1:2,500; Sigma: A8592), rabbit α-HA (1:1,000; Cell Signaling: C29F4), mouse α-ubiquitin-HRP (1 µg/mL; Cytoskeleton: AUB01-HRP), α-rabbit-HRP (1:10,000; Thermo Fisher: 34102), and α-mouse-HRP (1:10,000; Sigma: SAB3701122). Antibodies for MNV nonstructural proteins, including mouse α-NS1, rabbit α-NS2, and rabbit α-NS6/7, were used as described previously (45).
## Coimmunoprecipitation
293T cells were seeded at 1 × 10 6 cells per well of a six-well plate and transfected with the indicated plasmids. At 24 hours post-transfection, cells were lysed in RIPA buffer (10 mM Tris [pH 7.5], 140 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 0.1% DOC, 0.1% SDS, 1% Triton X-100) containing HALT protease and phosphatase inhibitors. Clarified lysates were incubated with anti-FLAG-M2 (1:500; Sigma: F1804) with gentle rocking at 4°C. Two hours later, Protein A/G Agarose (Thermo Fisher: 20421) was added to lysates and incubated overnight with gentle rocking at 4°C. Samples were subjected to four washes with RIPA buffer, and proteins were eluted from beads with Laemmli buffer and subjected to SDS-PAGE and Western blotting as described above, with the addition of the TidyBlot Western Blot Detection Reagent (Bio-Rad: STAR209PT).
## Degradation assay
293T cells were seeded at 2 × 10 5 per well in a 24-well plate and subsequently transfec ted with 250 ng of the indicated plasmids. After 24 hours, cells were treated with either DMSO or 100 nM bortezomib (Sigma: 5043140001) for an additional 8 hours. Cells were lysed in Laemmli buffer and subjected to SDS-PAGE and Western blotting.
## Ubiquitination of NS1/2
293T cells were seeded at 1 × 10 6 per well of a six-well plate and transfected with the corresponding plasmid construct. At 24 hours post-transfection, cells were treated with 100 nM bortezomib for an additional 8 hours. Cells were then lysed in RIPA buffer containing HALT protease and phosphatase inhibitor cocktail, and lysates were incubated with anti-HA-agarose (Sigma: A2095) overnight on a nutator at 4°C. The next day, samples were subjected to four washes with RIPA buffer, and proteins were eluted from beads with Laemmli buffer and subjected to SDS-PAGE and Western blotting.
## Inhibition of NS1/2 processing
To verify that the caspase inhibitor z-VAD (OMe)-FMK (Cell Signaling Technology: 60332S) effectively prevents NS1/2 processing, HeLa-CD300lf cells were seeded at 2.5 × 10 5 per well in a 12-well plate and infected the next day with the indicated MNV strains at an MOI of 5. Eight hours post-infection, cells were treated with either DMSO or 50 µM z-VAD. Twenty-four hours post-infection, cells were lysed in Laemmli buffer and analyzed via SDS-PAGE and Western blotting. For MNV growth assays in the presence of the caspase inhibitor z-VAD, HeLa-CD300lf cells expressing either an empty vector or TRIM47 were infected with the indicated MNV strains at an MOI of 0.05. Eight hours post-infection, cells were treated with either DMSO or 50 µM z-VAD. Samples were frozen at -80°C 24 hours post-infection, and infectious virus was measured via plaque assay.
## References
1. Wikswo, Roberts, Marsh et al. (2009) "Enteric illness outbreaks reported through the national outbreak reporting system-United States" *Clin Infect Dis*
2. Chhabra, De Graaf, Parra et al. (2019) "Updated classification of norovirus genogroups and genotypes" *J Gen Virol*
3. Prasad, Atmar, Ramani et al. (2025) "Norovirus replication, host interactions and vaccine advances. Full-Length Text Journal of Virology December"
4. *Nat Rev Microbiol*
5. Karst, Wobus, Goodfellow et al. (2014) "Advances in norovirus biology" *Cell Host Microbe*
6. Baldridge, Nice, Mccune et al. (2015) "Commensal microbes and interferon-λ determine persistence of enteric murine norovirus infection" *Science*
7. Nice, Baldridge, Mccune et al. (2015) "Interferon-λ cures persistent murine norovirus infection in the absence of adaptive immunity" *Science*
8. Nice, Strong, Mccune et al. (2013) "A singleamino-acid change in murine norovirus NS1/2 is sufficient for colonic tropism and persistence" *J Virol*
9. Cadwell, Patel, Maloney et al. (2010) "Virus-plus-susceptibility gene interaction determines Crohn's disease gene Atg16L1 phenotypes in intestine" *Cell*
10. Kernbauer, Ding, Cadwell (2014) "An enteric virus can replace the beneficial function of commensal bacteria" *Nature*
11. Bouziat, Biering, Kouame et al. (2018) "Murine norovirus infection induces T H 1 inflammatory responses to dietary antigens" *Cell Host Microbe*
12. Jones, Watanabe, Zhu et al. (2014) "Enteric bacteria promote human and mouse norovirus infection of B cells" *Science*
13. Strine, Fagerberg, Darcy et al. (2024) "Intestinal tuft cell immune privilege enables norovirus persistence" *Sci Immunol*
14. Mcfadden, Bailey, Carrara et al. (2011) "Norovirus regulation of the innate immune response and apoptosis occurs via the product of the alternative open reading frame 4" *PLoS Pathog*
15. Wobus, Karst, Thackray et al. (2004) "Replication of norovirus in cell culture reveals a tropism for dendritic cells and macrophages" *PLoS Biol*
16. Karst, Wobus, Lay et al. (2003) "STAT1dependent innate immunity to a Norwalk-like virus" *Science*
17. Nelson, Wilen, Dai et al. (2018) "Structural basis for murine norovirus engagement of bile acids and the CD300lf receptor" *Proc Natl Acad Sci*
18. Biering, Choi, Halstrom et al. (2017) "Viral replication complexes are targeted by LC3-guided interferon-inducible GTPases" *Cell Host Microbe*
19. Lee, Liu, Wilen et al. (2019) "A secreted viral nonstructural protein determines intestinal norovirus pathogenesis" *Cell Host Microbe*
20. Lee, Wilen, Orvedahl et al. (2017) "Norovirus cell tropism is determined by combinatorial action of a viral non-structural protein and host cytokine" *Cell Host & Microbe*
21. Wilen, Lee, Hsieh et al. (2018) "Tropism for tuft cells determines immune promotion of norovirus pathogenesis" *Science*
22. Thackray, Wobus, Chachu et al. (2007) "Murine noroviruses comprising a single genogroup exhibit biological diversity despite limited sequence divergence" *J Virol*
23. Bouziat, Biering, Kouame et al. (2018) "Murine norovirus infection induces T H 1 inflammatory responses to dietary antigens" *Cell Host Microbe*
24. Grau, Roth, Zhu et al. (2017) "The major targets of acute norovirus infection are immune cells in the gutassociated lymphoid tissue" *Nat Microbiol*
26. Graziano, Alfajaro, Schmitz et al. (2021) "CD300lf conditional knockout mouse reveals strain-specific cellular tropism of murine norovirus" *J Virol*
27. Strine, Alfajaro, Graziano et al. (2022) "Tuftcell-intrinsic and -extrinsic mediators of norovirus tropism regulate viral immunity" *Cell Rep*
28. Zhu, Watanabe, Kirkpatrick et al. (2016) "Regulation of norovirus virulence by the VP1 protruding domain correlates with B cell infection efficiency" *J Virol*
29. Graziano, Walker, Kennedy et al. (2020) "CD300lf is the primary physiologic receptor of murine norovirus but not human norovirus" *PLoS Pathog*
30. Orchard, Sullender, Dunlap et al. (2019) "Identification of antinorovirus genes in human cells using genome-wide CRISPR activation screening" *J Virol*
31. Walker, Hassan, Peterson et al. (2021) "Norovirus evolution in immunodeficient mice reveals potentiated pathogenicity via a single nucleotide change in the viral capsid" *PLoS Pathog*
32. Sosnovtsev, Belliot, Chang et al. (2006) "Cleavage map and proteolytic processing of the murine norovirus nonstructural polyprotein in infected cells" *J Virol*
33. Chabot, Durantel, Lucifora (2025) "TRIM proteins: a "swiss army knife" of antiviral immunity" *PLoS Pathog*
35. Fan, Mar, Sari et al. (2021) "TRIM7 inhibits enterovirus replication and promotes emergence of a viral variant with increased pathogenicity" *Cell*
36. Wakeford, Werkmeister, Cayet et al. (2025) "Impaired K48-polyubiquitination downmodulates mouse norovirus propagation" *Front Cell Infect Microbiol*
37. Robinson, Van Winkle, Mccune et al. (2019) "Caspase-mediated cleavage of murine norovirus NS1/2 potentiates apoptosis and is required for persistent infection of intestinal epithelial cells" *PLoS Pathog*
38. Thornberry, Rano, Peterson et al. (1997) "A combinatorial approach defines specificities of Full-Length Text Journal of Virology December"
39. "members of the caspase family and granzyme B. Functional relation ships established for key mediators of apoptosis" *J Biol Chem*
40. Chu, Yang (2011) "SUMO E3 ligase activity of TRIM proteins" *Oncogene*
41. Hu, Liao, Yang et al. (2017) "Innate immunity to RNA virus is regulated by temporal and reversible sumoylation of RIG-I and MDA5" *J Exp Med*
42. Hu, Yang, Xie et al. (2016) "Sumoylation promotes the stability of the DNA sensor cGAS and the adaptor STING to regulate the kinetics of response to DNA virus" *Immunity*
43. Chang, Paul, Babu et al. (2008) "Deubiquitinating enzyme CYLD negatively regulates RANK signaling and osteoclastogenesis in mice" *J Clin Invest*
44. Azuma, Ikeda, Suzuki et al. (2021) "TRIM47 activates NF-κB signaling via PKC-ε/PKD3 stabilization and contributes to endocrine therapy resistance in breast cancer" *Proc Natl Acad Sci*
45. Dou, Li, Li et al. (2021) "Ubiquitination and degradation of NF90 by Tim-3 inhibits antiviral innate immunity"
46. Strong, Thackray, Smith et al. (2012) "Protruding domain of capsid protein is necessary and sufficient to determine murine norovirus replication and pathogenesis in vivo" *J Virol*
47. Orchard, Wilen, Doench et al. (2016) "Discovery of a proteinaceous cellular receptor for a norovirus" *Science*
48. Sullender, Pierce, Srinivas et al. (2022) "Selective polyprotein processing determines norovirus sensitivity to Trim7" *J Virol*
49. Song, Zhang, Moon et al. (2025) "Norovirus coopts NINJ1 for selective protein secretion" *Sci Adv*
50. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754525&blobtype=pdf
|
# Human bocavirus: As an emerging respiratory pathogen
Nawodhya Tharushi, Dilakshini Fernando, Rohitha Dayananda, Muthugala, Dayananda, Rohitha Muthugala
## Abstract
Acute respiratory infections (ARIs) are the main cause of morbidity and mortality worldwide, especially among children. The human bocavirus (HBoV) is a nonenveloped DNA virus that was recently identified as a respiratory pathogen associated with respiratory tract infections (RTIs), predominantly in infants and young children. It is also detected from the gastrointestinal tract in children. The prevalence of HBoV1 acute respiratory tract infection varies across age groups, ranging from 10.3% to 12.51% in individuals under 3 years of age. The spectrum of clinical presentation includes mild upper RTIs, acute exacerbation of asthma, bronchitis, bronchiolitis, pneumonia, and multi-organ failure. Although HBoV is often detected in patients with ARIs who have other respiratory viruses (17%-85%), recent studies have identified it as the sole aetiology for mild to severe ARIs. Children with pre-existing medical conditions infected with HBoV often have a risk of severe illness. HBoV infection is diagnosed primarily by detecting viral DNA in respiratory samples using molecular methods. Currently, there is no specific antiviral treatment for HBoV infections and the cases are managed symptomatically. General preventive measures used for the prevention of viral RTIs are applicable, as there is no effective vaccine against this virus. The HBoV has been implicated in RTIs, particularly in children, and has also been detected in cases of gastroenteritis. Despite its global prevalence, the exact pathogenic role of HBoV remains unclear due to frequent co-infections with other viruses. This minireview discusses the virology, epidemiology, clinical manifestations, diagnosis, and potential treatment approaches related to HBoV infections.
## INTRODUCTION
Acute respiratory infections (ARIs) are the main infectious cause of morbidity and mortality in the world[1]. Around 2.5 million of the global population lose their lives from ARIs each year, and most of these deaths are due to lower respiratory tract infections (LRTIs), which have become the fifth most common disease with a high mortality in the world [2,3].
ARIs are caused mainly by viruses: Human rhinoviruses, influenza A and influenza B viruses, respiratory syncytial virus (RSV), parainfluenza virus, human coronavirus, human metapneumovirus, human enterovirus and adenoviruses are the most common viruses detected from respiratory specimens in patients with ARIs [4,5]. Primary infections with viral pathogens can lead to secondary bacterial infections, which contribute to high mortality. The commonly reported bacteria include Staphylococcus aureus, Streptococcus pneumoniae, and Haemophilus influenzae [6,7].
Infectious respiratory viruses can cause outbreaks, epidemics and even pandemics. One such pandemic was coronavirus disease 2019, caused by the severe acute respiratory syndrome coronavirus 2, which has led to a global public health crisis. Initially, the disease progression was analyzed based on early clinical and diagnostic observations. For instance, Sabatino et al [8] reported findings from high-resolution computed tomography of the first 64 coronavirus disease 2019 patients, which were crucial for identifying diagnostic features and shaping early response strategies. Similarly, this pandemic underscored the importance of recognizing lesser-known viral agents capable of causing disease, particularly in children. One such pathogen is human bocavirus (HBoV), which is closely related to bovine parvovirus and canine minute virus [9]. The HBoV is recognized as a causative agent of respiratory infections in individuals of all ages, especially in children [10,11].
## DISCOVERY AND OVERVIEW OF HBOV
HBoV, a member of the Parvoviridae family and Bocaparvovirus genus, was first discovered in 2005 in Sweden from nasopharyngeal aspirates of seventeen children with LRTIs [9]. HBoV1, the first identified and the most studied genotype, is commonly associated with respiratory tract infections (RTIs) and affects both the upper and lower respiratory tracts [12]. Following this discovery, three further genotypes (HBoV2, HBoV3, and HBoV4) were subsequently detected in fecal specimens [13]. HBoV2, first reported in 2009, from fecal samples of children in Pakistan and the United Kingdom[14], has since been frequently associated with acute gastroenteritis (AGE) in children [15]. Additionally, co-infections with known human gastroenteritis viruses have frequently been observed, along with fecal discharge of asymptomatic children [16,17]. Therefore, the exact role of HBoV2 as an enteric pathogen remains uncertain. However, some studies show HBoV2 has been detected in fecal samples of children suffering from RTIs and is rarely present in respiratory tract samples [18,19]. HBoV3 was initially detected in a fecal sample in 2009 [15]. Several other studies have confirmed the presence of HBoV3 in fecal samples of patients with gastroenteritis. However, detection rates have been lower than those of HBoV2 [17,20,21]. HBoV4 was discovered recently in fecal samples of both children and adults, but the clinical significance of this virus remains unclear [21].
The HBoV1 has been proposed as a viral pathogen linked to respiratory infections, with global occurrences varying from 0.8% to 56.8% in children [10,11,22]. HBoV1 infection is commonly identified in patients experiencing pneumonia, acute wheezing, asthma, or bronchiolitis, and occasionally in severe lower respiratory infections that are life-threatening [23]. Additionally, HBoV1 has been detected in stool samples, indicating the potential for gastrointestinal involvement [24]. In recent years, this virus has become increasingly recognized as a causative agent of respiratory infections, affecting individuals of all ages [13,25,26]. Various studies have supported the presence of HBoV1 as a potential pathogen in LRTIs, mainly in children under the age of three. HBoV1 has also been detected in asymptomatic children, a finding initially disputed. However, research indicates that HBoV1 can persist in the nasopharynx for extended periods, leading to inaccurate polymerase chain reaction (PCR) diagnoses [27,28]. It is a co-infectious agent, often found with other respiratory viruses, and can also act as an isolated viral pathogen during RTIs [7]. Detection and treatment of co-infections depend on factors like illness severity and the presence of common pathogens [29].
Recent studies have shown that the most common Bocavirus in respiratory samples is HBoV1, compared to other genotypes, HBoV2, HBoV3, and HBoV4. While HBoV1 infections are often mild and self-limiting, severe cases can occur, as demonstrated in the study by Ziemele et al [26], especially in individuals with underlying health conditions. Some studies indicate the need for continued monitoring of HBoV in children with ARIs due to the seriousness of HBoV coinfections. Clinical observations suggest that HBoV infection might impact organ function in the liver, kidneys, heart, and blood acid-base balance. Accurate identification of the virus aids in providing specialized care and raises awareness of its risks in respiratory infections [26]. As demonstrated in the study conducted by Simon et al [30], the identification of HBoV in specimens obtained from the respiratory tracts of children did not necessitate a targeted antiviral treatment. However, it did prevent the need for additional intravenous antibiotics, thereby mitigating potential side effects and minimizing unnecessary costs associated with antibacterial medications [30]. Furthermore, raising awareness about the importance of hospital disinfection, sterilization and standardizing procedures is crucial. Further investigation is warranted to better understand the interactions between HBoV and other pathogens in the future [31]. Research on this virus is crucial for comprehending respiratory illness and developing potential treatments or vaccines. Access to updated and precise information is essential to alleviate strain on healthcare systems and enhance patient care for ARIs.
## VIROLOGY AND GENOMIC CHARACTERISTICS
HBoV has been classified as a member of the Bocaparvovirus genus in the Parvoviridae family [13]. It exhibits genetic similarity to bovine parvovirus and canine minute virus [13]. The HBoV is icosahedral in shape and is a non-enveloped virus. The viral genome comprises a single-stranded DNA of approximately 5.3-5.5 kb [9,32].
The genome contains three open-reading frames (ORF1, ORF2, and ORF3): (1) ORF1, located on the left half, encodes a series of non-structural protein (NS) 1-4; (2) ORF2, a smaller middle reading frame, encodes a unique nuclear protein (NP1), crucial for viral DNA replication and mRNA processing; and (3) ORF3, situated on the right half, encodes structural capsid proteins, viral structural protein (VP) 1, VP2, and VP3, with VP3 serving as the major capsid protein. The NS1 and NP1, which are relatively conserved proteins, play essential roles in viral DNA replication and are targeted for HBoV detection. In contrast, VP1 and VP2 exhibit greater mutational diversity. They are commonly used for phylogenetic analysis, contributing to the classification of HBoV into genotypes HBoV1, HBoV2, HBoV3, and HBoV4 based on nucleotide divergence in the VP1 capsid region. The VP1 is crucial for infectivity, aiding in the release of the virus from endocytic compartments into the nucleus of the host cell [32][33][34][35].
The non-enveloped capsid surface of the virus carries host determinants and plays various roles, including host tropism, cell recognition, pathogenicity, intracellular trafficking, genome packaging, assembly, and immune response [36].
## EPIDEMIOLOGY
Numerous clinical investigations have recognized HBoV1 as one of the most commonly detected respiratory viruses in infants and young children with RTIs. Most literature indicates that HBoV1 infection is prevalent among children aged 6 months to 2 years, with the highest detection rates occurring during the second year of life [31,37,38].
The prevalence of HBoV1 shows variation across age groups. Studies report a detection rate of 0.58 episodes per child by the age of two [25]. Prevalence rates range from 10.3% to 12.51% with the highest rates often found in children aged 1-3 years [39,40]. Some data show 63.1% of cases in infants under 12 months [40]. Other findings indicate a 28.1% prevalence with a median age of three, while the highest rate, 37.5% was seen in children aged ≥ 5 years [31].
Globally, the HBoV1 infection is reported in both low and high-income nations[25, [38][39][40][41]. For instance, detection rates in European countries such as Norway and Croatia have remained moderate. Whereas Asian regions, including China and Egypt, have reported a wider range [22,31,[42][43][44]. Additionally, some Asian regions, including Japan and Sri Lanka, have remained mid-range of detection rates [45,46]. A meta-analysis of 35 European studies found prevalence rates ranging from 2.0% to 45.69%, with a pooled rate of 9.57% [47]. In contrast, large-scale surveillance in the Middle East, such as in Saudi Arabia, found HBoV1 to be relatively uncommon among hospitalized children [48].
Regarding the seasonal pattern of HBoV1, several studies have indicated a greater prevalence during the late autumn and winter seasons [13,38]. Most diagnoses occurred during winter months, demonstrating that infections can happen throughout the year but are most common during the winter season [13,25,48]. However, another study in China showed a seasonal pattern, with a higher incidence of HBoV1 infections observed during summer [40]. Specifically, during the study duration, elevated detections of HBoV1 were primarily observed between May and August, as well as November and January [40].
## CLINICAL MANIFESTATIONS
HBoV infections present with a range of symptoms, including mild upper respiratory symptoms to respiratory distress. The range of illnesses associated with HBoV infection seems comparable to those reported for most of the other respiratory viruses, with a predominant impact on the upper respiratory tract [28,49].
The primary symptoms observed in both solitary infection and co-infection with HBoV were fever and cough. Respiratory distress ranked third in terms of frequency among both types of cases [26]. Rhinorrhea was notably more prevalent in cases of single infection compared to co-infection [26]. Conversely, respiratory failure was exclusively detected in cases of solitary HBoV infection [50]. Wheezing occurred more frequently in co-infection cases than in singleinfection cases, with a significant contrast. No substantial variation was observed, although pneumonia was more commonly associated with co-infection [38]. Previous studies have highlighted cough, rhinorrhea, and fever as the most prevalent symptoms in patients solely positive for HBoV, followed by severe respiratory distress, which required tracheal intubation [26]. Furthermore, patients infected with HBoV, whether as a solitary infection or co-infection with other viruses, exhibit a clinical presentation similar to those infected with other respiratory viruses such as RSV and human metapneumovirus [38,51]. Symptoms include upper RTIs, bronchitis, bronchiolitis, pneumonia, and acute exacerbation of asthma [26,52,53].
Although HBoV1 is primarily associated with respiratory symptoms, it has been detected in stool samples, indicating the potential for gastrointestinal involvement [24]. In contrast, HBoV genotypes 2-4 are linked with gastrointestinal issues, such as loss of appetite, vomiting, diarrhoea, and nausea [48]. Co-infections with intestinal pathogens, including human rotavirus, noroviruses, and certain strains of Escherichia coli or Salmonella are common, affecting up to 77.6% of HBoVpositive children [54]. However, the exact connection between the HBoV and gastroenteritis is still uncertain, despite these frequent associations [7].
A study conducted with the children who had symptoms of gastroenteritis, acute respiratory tract infection (ARTI) and symptoms of both shows that HBoV1 can be detected in patients with both ARTI and AGE and also patients with AGE alone, in addition to patients with ARTI symptoms. This shows HBoV1 can be found in stool samples during ARTI with or without gastroenteritis [55].
## CO-INFECTIONS
HBoV can be frequently detected in the first two years of life, and it is ranked as the fourth most prevalent respiratory virus, following RSV, rhinovirus, and adenovirus [25,56]. The HBoV1 can be detected with one or more respiratory viruses or bacteria (Table 1) [4,5,13,20,25,31,38,39,43,46,48,[57][58][59][60]. A notable feature of bocavirus infection is its frequent codetection with other pathogens, such as respiratory viruses and bacteria, likely due to the extended shedding of HBoV1 in the nasopharynx, which can last weeks or even months. Notably, some studies revealed that HBoV1 DNA shedding typically lasted up to four weeks in almost all instances [27]. Moreover, increasing age, winter seasons and attendance at childcare facilities are associated with higher rates of HBoV1 detection [25,27]. The initial release of HBoV1 is linked to mild respiratory symptoms, followed by an extended period of detecting HBoV1 DNA for about a year, which can lead to co-infection. Repeated infections with HBoV1 prolong this shedding process [61]. The prevalence of HBoV co-infections varies across different populations and geographical regions, highlighting the need for further investigation into its epidemiology and clinical significance [62].
## CO-INFECTIONS WITH VIRUSES AND BACTERIA
The prevalence of mixed viral infections in nasopharyngeal swab samples varies across different studies and regions. Coinfection rates range from around 17% to as high as 86% in twelve studies [4,5,13,25,38,39,43,46,48,[57][58][59].
Interestingly, HBoV1 can also be found in asymptomatic children, which can lead to high rates of co-infections, contributing to the high prevalence of the virus in the pediatric population. The role of HBoV in respiratory infections was questionable, with some studies suggesting it as a potential cause of severe respiratory illnesses in children [63]. In contrast, the other studies show it as a harmless virus [64]. However, recent studies increasingly support the fact that HBoV causes significant illness, even when it's the only infectious agent present. Therefore, timely diagnosis is crucial, especially for pediatricians, since this viral infection can potentially have severe and fatal consequences [26].
Several studies have highlighted the prevalence of bacterial co-infection among individuals who are positive for HBoV [31,38,39]. They have also reported complications such as pneumothorax, pneumomediastinum, and severe respiratory failure necessitating intensive care unit (ICU)/intermittent mandatory ventilation support in cases of HBoV coinfection with other pathogens [50]. Multiple co-infections in the respiratory tract were associated with more severe disease progression and the need for ICU admission [65].
## HBOV1 AS A SOLE PATHOGEN
Many studies across China demonstrate that HBoV1 frequently appears as a single infection in people with ARTIs, suggesting it can act as an independent respiratory pathogen. The mono infection rate ranges from 49.34% to 81.2% out of HBoV1 positive cases, reinforcing the virus's standalone pathogenic potential [5,45,66].
In Sri Lanka, out of 200 respiratory samples from suspected severe acute respiratory syndrome coronavirus 2 patients, only one sample tested positive for HBoV1 with common ARTI symptoms [67].
In the United Kingdom, HBoV1 mono-infected children who received oxygen, nebulizers, and mechanical ventilation; inappropriate antibiotic use was noted, raising concerns about resistance and microbiota disruption [13].
The literature shows that patients with ARTI had either high or low viral load, and HBoV1 single infections are not significantly different from co-infection with respect to clinical features; the virus can be as pathogenic by itself as other respiratory agents [68].
## RISK FACTORS AND COMPLICATIONS
Children infected with HBoV often have pre-existing medical conditions that elevate the risk of severe illness, including chronic lung diseases, congenital heart conditions, neuromuscular disorders, cancer, or immunological issues, with prevalence rates ranging from 14% to 77% [69,70]. HBoV1 infection in young children is linked to several factors, including prematurity, passive smoking, winter birth, and family history of asthma. There is a higher risk in children under 5, especially those attending childcare [25,71].
The main complications associated with HBoV infection include progressive respiratory distress leading to acute respiratory failure and dehydration [38]. Additionally, rare but life-threatening conditions such as pneumomediastinum and bilateral pneumothorax can develop in HBoV-infected children [48]. Apart from respiratory failure, the HBoV is also associated with acute heart failure at a significant rate of 10% [53].
Additionally, some children needed ICU admission due to serious events, such as heart failure, myocardial damage, and liver function damage. These studies show the importance of considering HBoV1 as a differential pathogen and diagnosis in pediatric patients presenting with respiratory symptoms [72]. Although most children recover with supportive care, severe and fatal cases have been documented, particularly among immunocompromised or chronically ill patients [73].
## DETECTION METHODS
Detection of HBoV infection primarily relies on identifying viral genomes present in human respiratory samples, though serum, blood, stool, and urine samples are also utilized. Various molecular assays, such as PCR, employing specific sets of primers targeting viral genes NP1, NS1, VP1, and VP2, are employed for this purpose [62]. The most prevalent techniques include quantitative PCR and real-time PCR, which quantify HBoV mRNA. Samples from different parts of the respiratory tract are examined in patients with RTIs, ranging from NP aspirates and swabs to bronchoalveolar lavage. Recent advances have introduced multiplex PCR panels and multiplex tandem PCR, allowing simultaneous analysis of samples where HBoV is detected alongside other respiratory viruses [59].
Moreover, serological assays are also employed to measure HBoV1-4 -immunoglobulin (Ig) G and IgM antibodies in plasma samples. These assays utilize in-house enzyme immunoassays, providing a complementary approach to molecular diagnostics for assessing HBoV infection. Such comprehensive diagnostic approaches enhance our understanding and management of HBoV-related illnesses, contributing to more effective patient care and public health strategies [74]. In addition to serological assays, advanced molecular surveillance tools, for instance, a metagenomic approach based on target-independent next-generation sequencing, have become recognized method for both known and novel viruses in clinical samples [75].
## CLINICAL MANAGEMENT
Currently, there is no approved specific antiviral treatment for HBoV infection, and comparative studies on antiviral drugs are lacking in the literature. While research on potential treatments is ongoing, there has been limited exploration of therapeutic options [7].
A study has been conducted to determine the effectiveness of steroids for acute HBoV1 infection. In this study, steroids were given to hospitalized children who suffered from wheezing episodes, who were confirmed to be positive for HBoV1 infection [76]. However, the results showed no effect in reducing hospitalization time, symptom duration, or relapse rates. Therefore, use of corticosteroids in managing acute HBoV1infection has not shown efficacy [76].
A case report showed that the use of cidofovir, used for herpesvirus infections, resulted in the elimination of HBoV1 infection. This report highlights the possibility of using antivirals for gastroenteritis caused by HBoV1, and may eliminate HBoV1, though evidence remains unclear [77].
Supportive measures like bronchoscopic intervention can be effective approaches for HBoV1-associated plastic bronchitis [78,79].
Case studies show that the supportive care is effective for HBoV1 infection, as no targeted antiviral therapy is currently available. Fever was controlled with paracetamol and tepid sponging. Early supportive interventions, particularly oxygen therapy, fever management, and hydration are effective in stabilizing patients with HBoV1 infection and can lead to favourable outcomes [80].
## PREVENTION
These findings highlight the urgent need for preventive actions such as vaccines. An experimental study using virus-like particles (VLPs) of HBoV1 and HBoV2 as vaccine candidates in mice demonstrates that HBoV1 VLPs successfully induced immune responses in mice. This has been characterized by the production of high-titre and high-avidity IgG antibodies. HBoV1 VLPs were capable of eliciting balanced T helper type 1/type 2 cellular immune responses. Additionally, cross-reactive antibody responses were observed between HBoV1 and HBoV2, with higher reactivity in HBoV1-immunized groups, suggesting that HBoV1 VP2 VLP-based vaccines represent a strong candidate for the prevention and management of HBoV1 infections. This can be used as a foundation for the development of a vaccine targeting HBoV1 [81].
Accurate identification of the virus aids specialized care and raises awareness of its risks in respiratory infections. As demonstrated by Simon et al [30], the identification of HBoV in specimens obtained from the respiratory tracts of children did not necessitate a targeted antiviral treatment. However, it prevented the need for additional intravenous antibiotics, thereby mitigating potential side effects and minimizing unnecessary costs associated with antibacterial medications [30]. Furthermore, raising awareness about the significance of hospital environment disinfection and sterilization and standardizing procedures is crucial. Further investigation is warranted to better understand the interactions between HBoV and other pathogens in the future [31]. Research on this virus is crucial for comprehending respiratory illness and developing potential treatments or vaccines. Access to updated and precise information is essential to alleviate strain on healthcare systems and enhance patient care for ARIs.
## CONCLUSION
HBoV1 is increasingly recognized as an emerging respiratory pathogen, commonly detected in young children and capable of causing illness ranging from mild respiratory symptoms to severe complications. Since no specific antiviral therapy exists, management relies mainly on supportive care.
Distinguishing true HBoV1 infection from co-infection or prolonged viral shedding remains essential to avoid misdiagnosis and unnecessary antibiotic use, emphasizing the importance of quantitative PCR, serological markers, and improved clinical interpretation.
Advances in molecular diagnostics, particularly multiplex PCR panels and metagenomic sequencing, have enhanced detection accuracy, while experimental virus-like particle vaccines have demonstrated strong immune responses in animal studies, offering a promising foundation for future preventive strategies against HBoV1.
## FOOTNOTES
Author contributions: Fernando TN conducted the literature search and drafted the manuscript; Dayananda D has conceptualized this manuscript and involved in revision and editing; Muthugala R has reviewed and edited the manuscript; all authors reviewed and approved the final version of the manuscript.
## Conflict-of-interest statement:
All authors declare no conflict of interest in publishing the manuscript.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers.
It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/ Country of origin: Sri Lanka ORCID number: Rohitha Muthugala 0000-0002-7546-1069.
## S-Editor:
Luo ML L-Editor: A P-Editor: Wang CH
## References
1. (2014) "Infection Prevention and Control of Epidemic-and Pandemic-Prone Acute Respiratory Infections in Health Care"
2. Shahbazi, Moslehi, Mirzaei et al. (2025) "The effect of addressing the top 10 global causes of death on life expectancy in 2019: a global and regional analysis" *Int Health*
3. Nair, Brooks, Katz et al. (2011) "Global burden of respiratory infections due to seasonal influenza in young children: a systematic review and meta-analysis" *Lancet*
4. Piñana, Vila, Andrés et al. (2023) "Molecular characterization and clinical impact of human bocavirus at a tertiary hospital in Barcelona (Catalonia, Spain) during the 2014-2017 seasons" *Infection*
5. De, Zhang, Wang et al. (2022) "Human bocavirus 1 is a genuine pathogen for acute respiratory tract infection in pediatric patients determined by nucleic acid, antigen, and serology tests" *Front Microbiol*
6. Morris, Cleary, Clarke (2017) "Secondary Bacterial Infections Associated with Influenza Pandemics" *Front Microbiol*
7. Trapani, Caporizzi, Ricci et al. (2023) "Human Bocavirus in Childhood: A True Respiratory Pathogen or a "Passenger" *Virus? A Comprehensive Review. Microorganisms*
8. Sabatino, Sergio, Muri et al. (2021) "COVID-19: high-resolution computed tomography findings in the first 64 patients admitted to the Hospital of Cremona, the epicentre of the pandemic in Europe" *Pol J Radiol*
9. Allander, Tammi, Eriksson et al. (2005) "Cloning of a human parvovirus by molecular screening of respiratory tract samples" *Proc Natl Acad Sci U S A*
10. Guido, Quattrocchi, Campa et al. (2011) "Human metapneumovirus and human bocavirus associated with respiratory infection in Apulian population" *Virology*
11. Longtin, Bastien, Gilca et al. (2008) "Human bocavirus infections in hospitalized children and adults" *Emerg Infect Dis*
12. Jartti, Hedman, Jartti et al. (2012) "Human bocavirus-the first 5 years" *Rev Med Virol*
13. Bagasi, Hc, Clark et al. (2020) "Human Bocavirus infection and respiratory tract disease identified in a UK patient cohort" *J Clin Virol*
14. Kapoor, Slikas, Simmonds et al. (2009) "A newly identified bocavirus species in human stool" *J Infect Dis*
15. Arthur, Higgins, Davidson et al. (2009) "A novel bocavirus associated with acute gastroenteritis in Australian children" *PLoS Pathog*
16. Han, Kim, Park et al. (2009) "Detection of human bocavirus-2 in children with acute gastroenteritis in South Korea" *Arch Virol*
17. Jin, Cheng, Xu et al. (2011) "High prevalence of human bocavirus 2 and its role in childhood acute gastroenteritis in China" *J Clin Virol*
18. Shan, Zhang, Guo et al. (2009) "The first detection of human bocavirus 2 infections in China" *J Clin Virol*
19. Kalaskar, Heresi, Wanger et al. (2009) "Severe necrotizing pneumonia in children" *Emerg Infect Dis*
20. Kantola, Sadeghi, Antikainen et al. (2010) "Real-time quantitative PCR detection of four human bocaviruses" *J Clin Microbiol*
21. Kapoor, Simmonds, Slikas et al. (2010) "Human bocaviruses are highly diverse, dispersed, recombination prone, and prevalent in enteric infections" *J Infect Dis*
22. Kamel, Hamed, Hassan et al. (2016) "A novel primer set for improved direct gene sequencing of human bocavirus genotype-1 from clinical samples" *J Virol Methods*
23. Zhao, Zhu, Qian et al. (2016) "Prevalence and Phylogenetic Analysis of Human Bocaviruses 1-4 in Pediatric Patients with Various Infectious Diseases" *PLoS One*
24. Cashman, Shea (2012) "Detection of human bocaviruses 1, 2 and 3 in Irish children presenting with gastroenteritis" *Arch Virol*
25. Saha, Fozzard, Lambert et al. (2023) "Human bocavirus-1 infections in Australian children aged < 2 years: a birth cohort study" *Eur J Clin Microbiol Infect Dis*
26. Ziemele, Xu, Vilmane et al. (2019) "Acute human bocavirus 1 infection in child with life-threatening bilateral bronchiolitis and right-sided pneumonia: a case report" *J Med Case Rep*
27. Martin, Fairchok, Kuypers et al. (2010) "Frequent and prolonged shedding of bocavirus in young children attending daycare" *J Infect Dis*
28. Allander, Jartti, Gupta et al. (2007) "Human bocavirus and acute wheezing in children" *Clin Infect Dis*
29. Lai, Wang, Hsueh (2020) "Co-infections among patients with COVID-19: The need for combination therapy with non-anti-SARS-CoV-2 agents?" *J Microbiol Immunol Infect*
30. Simon, Groneck, Kupfer et al. (2007) "Detection of bocavirus DNA in nasopharyngeal aspirates of a child with bronchiolitis" *J Infect*
31. Wang, Guan, Liu et al. (2022) "Epidemiologic and clinical characteristics of human bocavirus infection in children hospitalized for acute respiratory tract infection in Qingdao" *China. Front Microbiol*
32. Huang, Deng, Yan et al. (2012) "Establishment of a reverse genetics system for studying human bocavirus in human airway epithelia" *PLoS Pathog*
33. Gurda, Parent, Bladek et al. (2010) "Human bocavirus capsid structure: insights into the structural repertoire of the parvoviridae" *J Virol*
34. Zou, Cheng, Shen et al. (2016) "Nonstructural Protein NP1 of Human Bocavirus 1 Plays a Critical Role in the Expression of Viral Capsid Proteins" *J Virol*
35. Shao, Shen, Wang et al. (2021) "Recent Advances in Molecular Biology of Human Bocavirus 1 and Its Applications" *Front Microbiol*
36. Kailasan, Garrison, Ilyas et al. (2016) "Mapping Antigenic Epitopes on the Human Bocavirus Capsid" *J Virol*
37. Schildgen, Müller, Allander et al. (2008) "Human bocavirus: passenger or pathogen in acute respiratory tract infections?" *Clin Microbiol Rev*
38. Ji, Sun, Yan et al. (2021) "Epidemiologic and clinical characteristics of human bocavirus infection in infants and young children suffering with community acquired pneumonia in Ningxia, China" *Virol J*
39. Tang, Dai, Wang et al. (2022) "Comparison of the clinical features of human bocavirus and metapneumovirus lower respiratory tract infections in hospitalized children in Suzhou" *China. Front Pediatr*
40. Zhou, Zheng, Xiao et al. (2014) "Single detection of human bocavirus 1 with a high viral load in severe respiratory tract infections in previously healthy children" *BMC Infect Dis*
41. Ljubin-Sternak, Meštrović, Ivković-Jureković et al. (2019) "High Detection Rates of Human Bocavirus in Infants and Small Children with Lower Respiratory Tract Infection from Croatia" *Clin Lab*
42. Ljubin-Sternak, Slović, Mijač et al. (2021) "Prevalence and Molecular Characterization of Human Bocavirus Detected in Croatian Children with Respiratory Infection" *Viruses*
43. Christensen, Nordbø, Krokstad et al. (2008) "Human bocavirus commonly involved in multiple viral airway infections" *J Clin Virol*
44. Sun, Jiang, Chen et al. (2025) "Prevalence and molecular characterization of human bocavirus-1 in children and adults with influenza-like illness from Kunming, Southwest China" *Microbiol Spectr*
45. Murata, Shibata, Funakoshi et al. (2025) "Epidemiology and Diseases Burden of Human Bocavirus 1 Infection in a Children's Hospital in Japan" *Pediatr Infect Dis J*
46. Arunasalam, Pattiyakumbura, Shihab et al. (2023) "Demographic and clinical characteristics of human bocavirus-1 infection in patients with acute respiratory tract infections during the COVID-19 pandemic in the Central Province of Sri Lanka" *BMC Infect Dis*
47. Polo, Lema, Gándara et al. (2022) "Prevalence of human bocavirus infections in Europe. A systematic review and meta-analysis" *Transbound Emerg Dis*
48. Alkhalf, Almutairi, Almutairi et al. (2022) "Prevalence and Clinical Characterization of Bocavirus Infection in a Specialized Children's Hospital in Saudi Arabia" *Cureus*
49. Bastien, Chui, Robinson et al. (2007) "Detection of human bocavirus in Canadian children in a 1-year study" *J Clin Microbiol*
50. Ursic, Steyer, Kopriva et al. (2011) "Systematic review and meta-analysis of the prevalence of common respiratory viruses in children < 2 years with bronchiolitis in the pre-COVID-19 pandemic era" *J Clin Microbiol*
51. Madi, Al-Adwani (2020) "Human bocavirus (HBoV) in Kuwait: molecular epidemiology and clinical outcome of the virus among patients with respiratory diseases" *J Med Microbiol*
52. Zhang, Zheng, Zhu et al. (2021) "Human bocavirus-1 screening in infants with acute lower respiratory tract infection" *J Int Med Res*
53. Yu, Li, Xu et al. (2008) "Human bocaviruses are commonly found in stools of hospitalized children without causal association to acute gastroenteritis" *Eur J Pediatr*
54. Calvo, García-García, Pozo et al. (2008) "Clinical characteristics of human bocavirus infections compared with other respiratory viruses in Spanish children" *Pediatr Infect Dis J*
55. Esposito, Daleno, Prunotto et al. (2013) "Impact of viral infections in children with community-acquired pneumonia: results of a study of 17 respiratory viruses" *Influenza Other Respir Viruses*
56. Verbeke, Reynders, Floré et al. (2019) "Human bocavirus infection in Belgian children with respiratory tract disease" *Arch Virol*
57. Mijač, Ljubin-Sternak, Ivković-Jureković et al. (2023) "Comparison of MT-PCR with Quantitative PCR for Human Bocavirus in Respiratory Samples with Multiple Respiratory Viruses Detection" *Diagnostics (Basel)*
58. Sun, Sun, Hao et al. (2019) "Impact of RSV Coinfection on Human Bocavirus in Children with Acute Respiratory Infections" *J Trop Pediatr*
59. Martin, Kuypers, Mcroberts et al. (2015) "Human Bocavirus 1 Primary Infection and Shedding in Infants" *J Infect Dis*
60. Lindner, Karalar, Schimanski et al. (2008) "Clinical and epidemiological aspects of human bocavirus infection" *J Clin Virol*
61. Bastien, Brandt, Dust et al. (2006) "Human Bocavirus infection" *Canada. Emerg Infect Dis*
62. Schildgen (2013) "Human bocavirus: lessons learned to date" *Pathogens*
63. Eşki, Öztürk, Çiçek et al. (2021) "Is viral coinfection a risk factor for severe lower respiratory tract infection? A retrospective observational study" *Pediatr Pulmonol*
64. Wen, Yang, Luo et al. (2025) "Molecular epidemiological characterization of human bocavirus (HBoV) in acute respiratory infection (ARI) patients in Yucheng" *China. Front Public Health*
65. Fernando, Arachchige, Weerathunga et al. (2025) "Detection and genetic characterization of human bocavirus 1 infection: evolutionary insights in Western Province Sri Lanka" *Access Microbiol*
66. Ghietto, Majul, Soaje et al. (2015) "Comorbidity and high viral load linked to clinical presentation of respiratory human bocavirus infection" *Arch Virol*
67. Oldhoff, Bennet, Eriksson et al. (2023) "Human bocavirus 1 epidemiology in children in relation to virus load and codetection" *Acta Paediatr*
68. Tural, Yalcin, Emiralioglu et al. (2022) "Human bocavirus and human metapneumovirus in children with lower respiratory tract infections: Effects on clinical, microbiological features and disease severity" *Pediatr Int*
69. Campelo, De Aguiar Cordeiro, Moura (2022) "The role of human bocavirus as an agent of community-acquired pneumonia in children under 5 years of age in Fortaleza, Ceará (Northeast Brazil)" *Braz J Microbiol*
70. Tan, Huang, Tang et al. (2025) "Human bocavirus-1 infection in hospitalized pediatric patients with acute respiratory tract infections" *Microbiol Spectr*
71. Liao, Yang, He et al. (2022) "Respiratory tract infection of fatal severe human bocavirus 1 in a 13-month-old child: A case report and literature review" *Front Pediatr*
72. Nora-Krukle, Vilmane, Xu et al. (2018) "Human Bocavirus Infection Markers in Peripheral Blood and Stool Samples of Children with Acute Gastroenteritis" *Viruses*
73. Madi, Al-Nakib, Mustafa et al. (2018) "Metagenomic analysis of viral diversity in respiratory samples from patients with respiratory tract infections in Kuwait" *J Med Virol*
74. Jartti, Söderlund-Venermo, Allander et al. (2011) "No efficacy of prednisolone in acute wheezing associated with human bocavirus infection" *Pediatr Infect Dis J*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12844066&blobtype=pdf
|
# Editorial for the Special Issue "New Knowledge in the Study of Coronaviruses: Towards One Health and Whole Genome Sequencing Approaches, 2nd Edition"
Simone Peletto
## Abstract
The study of coronaviruses has undergone unprecedented acceleration over recent years, driven largely by the global impact of SARS-CoV-2 and growing recognition of the extraordinary diversity and zoonotic potential of coronaviruses across species. Advances in molecular biology, high-throughput sequencing, and bioinformatics have substantially improved our ability to detect, characterize, and monitor these viruses at the humananimal-environment interface. Nevertheless, major gaps in knowledge remain, particularly regarding the ecological drivers of spillover events, the role of intermediate hosts, and the integration of genomic data with epidemiological and ecological information within a One Health framework [1,2].Whole genome sequencing (WGS) has become a cornerstone of modern coronavirus research and surveillance, enabling high-resolution analyses of viral evolution, transmission dynamics, and the emergence of variants of concern [3,4]. Despite its transformative impact, the full potential of WGS is often limited by fragmented surveillance systems and insufficient integration with clinical and epidemiological datasets. As a result, the translation of genomic insights into actionable public health and veterinary interventions remains uneven across regions and host species [5,6].This Special Issue, "New Knowledge in the Study of Coronaviruses: Towards One Health and Whole Genome Sequencing Approaches, 2nd Edition", was conceived to contribute to addressing these challenges by bringing together studies that apply genomic and multidisciplinary approaches to the investigation of coronaviruses in both human and animal contexts. The collected contributions highlight how sequencing-based strategies can enhance variant detection and characterization, improve data quality, and provide insights into viral evolution and selective pressures. In this context, studies published in this Special Issue illustrate the complementary value of WGS and molecular diagnostics for variant tracking, the challenges posed by protein-level evolutionary drift, and the importance of sequence quality assessment for reliable downstream analyses [7][8][9][10].Taken together, the findings presented in this Special Issue reinforce the importance of embedding genomic surveillance within a broader One Health vision that integrates virology, veterinary science, ecology, and public health. Such an approach is essential not only for understanding ongoing coronavirus circulation, but also for strengthening preparedness and early warning systems for future emergence events. These perspectives align with broader international efforts emphasizing standardized sequencing, open data sharing, and interdisciplinary collaboration as key pillars of effective pathogen surveillance [5][6][7][8].Looking ahead, future research should prioritize the expansion of WGS-based surveillance to under-sampled geographic regions and host species, particularly wildlife and domestic animals that may act as reservoirs or intermediate hosts. Harmonization of
sequencing protocols, metadata standards, and analytical pipelines will be critical to ensure comparability and interoperability across studies. Moreover, closer integration between genomic surveillance and risk assessment frameworks will be necessary to translate molecular data into predictive models and evidence-based policy decisions. Sustained investment in One Health-oriented research and infrastructure will be crucial to improving global readiness for current and future coronavirus threats.
In conclusion, this Special Issue provides an updated and multifaceted overview of contemporary advances in coronavirus research, with particular emphasis on whole genome sequencing and One Health approaches. By bringing together complementary studies across disciplines and host systems, it highlights both the progress achieved and the challenges that remain. We hope that these contributions will inspire further research and collaboration aimed at strengthening integrated surveillance, preparedness, and response strategies for coronaviruses.
## References
1. Cui, Li, Shi (2019) "Origin and evolution of pathogenic coronaviruses" *Nat. Rev. Microbiol*
2. Anthony, Johnson, Greig et al. (2017) "Global patterns in coronavirus diversity" *Virus Evol*
3. Lu, Zhao, Li et al. (2020) "Genomic characterisation and epidemiology of 2019 novel coronavirus" *Lancet*
4. Oude Munnink, Nieuwenhuijse, Stein et al. (2020) "Rapid SARS-CoV-2 whole-genome sequencing for informed public health decision-making" *Nat. Med*
5. Gardy, Loman (2018) "Towards a genomics-informed, real-time global pathogen surveillance system" *Nat. Rev. Genet*
6. Hodcroft, De Maio, Lanfear et al. (2021) "Want to Track Pandemic Variants Faster? Fix the Bioinformatics Bottleneck" *Nature*
7. Carroll, Morzaria, Briand et al. (2018) "The Global Virome Project" *Science*
8. Na, Hong, Lee et al. (2025) "Tracing emergence of SARS-CoV-2 variants using RT-PCR and whole genome sequencing. Microorganisms"
9. Prokop, Alberta, Witteveen-Lane et al. (1863) "SARS-CoV-2 genotyping highlights challenges in spike protein drift. Microorganisms"
10. Xia (2187) "How trustworthy are the genomic sequences of SARS-CoV-2 in GenBank? Microorganisms 2024"
11. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645962&blobtype=pdf
|
# Highly efficient production of HIV-1 AD8 gp120 in mammalian cells
Tanvi Mathur, Shamim Ahmed, Durgadevi Parthasarathy, Alon Herschhorn, Alon R01ai167653, Herschhorn
## Abstract
Human immunodeficiency virus type 1 (HIV-1) envelope glycoproteins (Envs) interact with the CD4 receptor and CCR5/CXCR4 coreceptor expressed on target cells to mediate viral entry. Infection is initiated when the HIV-1 gp120 subunit of Envs binds to host receptors. gp120 can also be expressed as a soluble protein and is routinely used for different biophysical and immunization studies. Here, we compared the transient and stable expression of monomeric HIV-1 AD8 gp120 in two closely related 293 cell lines grown in suspension. We used an HIV-1 AD8 gp120-expressing plasmid to generate 293 cell lines that stably expressed HIV-1 AD8 gp120 under selection of puromycin. In parallel, we used polyethyleneimine (PEI) to transiently transfect FreeStyle 293 F cells with the identical HIV-1 AD8 gp120-expressing plasmid. Stable 293-AD8gp120 cells grew in media complemented with a feed supplement to a density of >1E7 cells/mL and produced an average of ~150 mg of gp120/liter of the culture, which was about 50-fold higher than the yield of gp120 produced by transient transfection of FreeStyle 293 F cells. SDS-PAGE analysis of HIV-1 AD8 gp120 produced by the two different methods showed similar patterns, but gp120 purified from the stable cell line exhibited higher homogeneity. Binding of anti-gp120 antibodies to gp120 that was produced by the two methods was comparable according to ELISA. Both gp120s could bind equally well to soluble CD4 and compete with infection of viruses pseudotyped with HIV-1 AD8 Envs. Our findings highlight multiple advantages of producing HIV-1 AD8 gp120 by stable expression for downstream applications. IMPORTANCE There are approximately 40.8 million people living with HIV-1 (PLWH) worldwide, with an estimate of about 1.3 million new HIV-1 infections in 2024, highlight ing the urgent need for an effective HIV-1 vaccine and cure strategies. Here, we describe a highly efficient method to produce soluble HIV-1 gp120, which is intensively used in viral assays and for vaccine development. The method may be helpful in research work that needs high amounts of proteins for diverse experiments.
KEYWORDS envelope glycoproteins, HIV-1, vaccineT he human immunodeficiency virus type 1 (HIV-1) pandemic continues to spread, with 1.3 million (1.0-1.7 million) people who acquired HIV-1 and 630,000 (490,000-820,000) people who died from HIV-1-related causes globally in 2024 (https:// www.who.int). FDA-approved antiretroviral therapy (ART) includes inhibitors of HIV-1 reverse transcriptase (1-3), integrase (4, 5), protease (6), envelope glycoproteins (Envs) (7-11), and drugs that target HIV-1 capsid (12-14) and the cellular host receptors CD4 and CCR5 (15,16). Although very effective, ART does not cure HIV-1 in people living with HIV-1 (PLWH), and upon ART discontinuation, HIV-1 replenishes the viral population from latent reservoirs and/or from low levels of replicating viruses (17-20). Thus, new approaches at different levels of development are and have been constantly tested for targeting HIV-1 persistence (21-28).
HIV-1 Envs mediate viral entry, have a key role in critical virus-host interactions, and are a target of intensive efforts to develop an HIV-1 vaccine (29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39). On the surface of HIV-1 virions, the Envs are assembled into a trimeric spike consisting of three gp120 exterior subunits and three gp41 transmembrane subunits (40)(41)(42). The gp120 subunit interacts with the host CD4 receptor and CCR5/CXCR4 coreceptors, while the gp41 subunit facilitates membrane fusion (43)(44)(45)(46)(47)(48). Receptor binding is associated with extensive structural rearrangements that result in Env conformational transitions from a closed conformation (State 1) to downstream states (State 2 and State 3) (49)(50)(51)(52). Env State 1 is a metastable conformation that contributes to low Env presentation on virions or infected cells and thus poses a significant challenge to study Envs in their native state. Notably, the combination of high Env robustness (53), which allows HIV-1 Envs to tolerate amino acid changes while maintaining entry competency; heavy glycan shield (54)(55)(56)(57)(58), which masks critical and vulnerable Env sites; Env conformational flexibility (59)(60)(61); and HIV-1 cell-to-cell transmission (62)(63)(64)(65) facilitates the exceptional ability of HIV-1 Envs to escape broadly neutralizing antibodies (66)(67)(68)(69)(70)(71)(72).
Studies of HIV-1 Env biology and designing HIV-1 vaccine candidates typically require large amounts of soluble HIV-1 gp120. In general, bacterial expression systems allow high protein yield, but production of active proteins can be challenging when the native proteins rely on post-translational modifications, i.e., glycosylation, or specific chaperones for function. Mammalian cells have become a popular choice for recombi nant protein production due to their ability to express proteins with native-like posttranslational modifications (PTMs) (73). Commonly used cells for recombinant protein production include human embryonic kidney (HEK293) and Chinese hamster ovary (CHO), along with their various derivatives. But other cell lines, such as Sf9, HKB11, and CAP-T, have been used as well (74). Recombinant proteins can be expressed by transient or stable transfection, and there are several advantages and disadvantages for each method. Transient transfection offers speed and flexibility but requires a relatively large amount of DNA. Due to the complexity of some proteins, it is sometimes difficult to produce significant amounts of specific proteins that could be effectively used for downstream applications. On the other hand, the development of stable cell lines is a long and time-consuming process but remains essential for applications requiring long-term, high-level protein expression and ease of upscaling. Protein expression using stable cell lines can significantly increase protein yields, and the process is reproducible, but long-term stability of the clones depends on multiple factors including the cells, specific gene of interest (GOI), protein expressed, and the sites of GOI integration into the cellular genome. Nevertheless, stable gene expression requires low amounts of plasmid DNA and can usually be scaled up for large production of proteins (75). CHO cells are genetically stable and can be transfected, enabling both transient and stable expression workflows (76)(77)(78). However, one of their limitations is nonhuman PTMs such as galactose-α1,3-galactose and N-glycolylneuraminic acid, which may be immunogenic in humans (79). Alternatively, HEK293 cells are often used; HEK293 cells can perform efficient γ-carboxylation, a critical modification for function of some proteins (79,80). Similar to CHO cells, HEK293 cells are well-suited for culturing in suspension and support both transient and stable expressions of recombinant proteins. Over the years, several derivatives of the HEK293 lineage have been developed to enhance specific character istics. One such variant, HEK293S GnTI -, has been adapted for growth in suspension (77). The optimal balance between growth and production is an important factor to consider when growing mammalian cells for either transient or stable protein expression. Limits on cell growth could significantly reduce the yield of the protein being produced. Here, we generated stable adherent HEK293 cells that express soluble HIV-1 gp120 and adapted the cells to grow in suspension. Growth in a BalanCD HEK293 medium enabled high-yield recombinant protein expression of ~150 mg/liter using a batch-feed scheme, which is designed to boost protein production by increasing viable cell density and prolonging culture duration. We compared the protein production, antigenicity, and expression patterns of gp120 purified from these cells to the gp120 purified after transient expression in the commonly used FreeStyle 293 F cells.
## MATERIALS AND METHODS
## Cells
Adherent HEK293 cells were obtained from the NIH HIV reagent program, and FreeStyle 293 F cells were purchased from Gibco (ThermoFisher Scientific). Adherent HEK293 cells were maintained at 37°C and 5% CO2 in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 100 µg/mL streptomycin, and 100 units/mL penicillin (Invitrogen, ThermoFisher Scientific). FreeStyle 293 F cells were maintained in suspension at 37°C and 8% CO2 in Freestyle expression medium (Gibco; ThermoFisher Scientific) with continuous shaking at 110-130 rpm.
## Plasmid construction
HIV-1 NL4-3 ASP Expression Vector (pcDNA3.4-based vector for expression of HIV-1 antisense protein) was obtained from the NIH HIV Reagent Program (cat# 13463). DNA sequence of the puromycin N-acetyl-transferase (puromycin resistance) gene was codon-optimized for Homo sapiens use, synthesized de novo, and cloned into the HIV-1 NL4-3 ASP vector replacing the neomycin-resistant gene. DNA sequence of gp120 followed by the DNA sequence of 3xGly and 8xHis tag was codon-optimized for Homo sapiens use, synthesized de novo, and cloned into the HIV-1 NL4-3 ASP/puromycin vector replacing the asp gene. The 1059-SOSIP expression plasmid was constructed by codon-optimizing the DNA sequence of 1059-SOSIP, de novo gene synthesizing, and cloning the gene into the pTwist CMV BetaGlobin WPRE Neo vector (Twist Bioscience). Gene synthesis and cloning were done by Gene Universal, Newark, DE.
## HEK293 transfection and selection
Exponentially growing HEK293 cells were used to generate stable AD8gp120-expressing cells. For transfection, 4 × 10 6 cells were seeded in a T75 flask and incubated overnight. The next day, chloroquine (25 µM final concentration; MilliporeSigma) was added to the medium dropwise, and 5 µg of pcDNA3.4.p-AD8gp120 plasmid was transfected using calcium phosphate as previously described (81). After 12-14 hours, the medium was replaced with 12 mL fresh DMEM to reduce chloroquine toxicity. Chloroquine has been shown to significantly improve transfection efficiency and is cytotoxic to cells if not removed after up to 14 hours (82,(82)(83)(84)(85)(86)(87). Cells were then selected with 0.5 µg/mL of puromycin (final concentration; MilliporeSigma), and the medium was replaced with a fresh medium containing the same puromycin concentration every few days to remove dead cells and allow selection. After ~4 weeks, cells grew with minimal cell death in the presence of 0.5 µg/mL puromycin. gp120 expression was assessed by Western blot or small-scale purification using Ni-NTA chromatography.
## HEK293 adaptation to growth in suspension
Stable HEK293-AD8gp120 cells were centrifuged, washed, and directly resuspended in BalanCD HEK293 medium (Fujifilm; Irvine Scientific) supplemented with 4 mM GlutaMAX at an initial cell density of 1 × 10 6 cells/mL and in a total volume of 20 mL. The culture was incubated in a 37°C, 100 RPM, 8% CO2 incubator for 4-5 days, and cell density and viability were monitored daily. Adapted cells were subsequently expanded to 4 × 10 6 cells/mL in 200 medium and then were subject to a feed schedule supplementing with 4% BalanCD HEK293 Feed (Fujifilm Irvine Scientific) every other day for ~10 days to maintain optimal growth and enhanced protein production.
## Transient transfection
FreeStyle 293 F cells or HEK293 were transfected with pcDNA3.4.p-AD8gp120 plasmid or co-transfected with pTwist 1059-SOSIP-expressing plasmid and a human furin-expressing plasmid at a ratio of 4:1 (total 1 mg DNA/1 L of cells) using PEI, as previously descri bed (88), or Turbo (Speed BioSystems, Cat # PXX1001) according to the manufacturer's instructions and grown for 5-10 days in a tissue culture incubator at 37°C, 8% CO 2 with continuous shaking.
## Expression optimization
Cells in suspension. Our previous optimization studies indicated that a 1:3 ratio of DNA:PEI and transfection of 2E6 cells/mL resulted in the most efficient transfection, and we used these conditions throughout the current studies.
Adherent cells. Adherent 0.5 × 10 6 HEK293 cells were transfected with 0.8, 2.5, or 5 µg of AD8gp120-expressing plasmid DNA and selected with 0.5 or 1 µg/mL puromycin. Expression of AD8 gp120 was assessed in HEK293 cells by small-scale Ni-NTA purification at 72 hours post-transfection and 96 hours post-transfection under puromycin selection; AD8 gp120 yields were evaluated and compared by SDS-PAGE. Based on these optimiza tions, 5 µg of AD8gp120 plasmid DNA was subsequently transfected into 4 × 10 6 HEK293 cells, followed by selection with 0.5 µg/mL puromycin.
## Protein purification
Culture supernatants containing soluble gp120 or 1059-SOSIP glycoproteins were harvested from the culture of transiently transfected FreeStyle 293 F cells, HEK293 cells, or from the culture of stable HEK293-AD8gp120 cells by centrifugation at 7,000 × g for 1 hour and filtered using a 0.45 µm vacuum filtering system (FOXX Life Sciences, Cat# FOX-1151-RLS). The supernatant was dialyzed against Ni-NTA buffer (50 mM NaH 2 PO 4 , 300 mM NaCl; pH 7.4) overnight using a 13,000 MWCO dialysis tube (VWR), or 10× of dialysis buffer was added to the supernatant to achieve a final 1× concentration. The equilibrated supernatant was then loaded on an Ni Sepharose High Performance column (Cytiva, Cat# 29051021) connected in an AKTA Go protein purification system (Cytiva) at 4°C-8°C. The column was washed with 500 mM NaCl in phosphate buffer pH 7.4, and the protein was eluted using a linear imidazole gradient containing 50-500 mM imidazole (Millipore Sigma) in Ni-NTA buffer. In some cases, 10 mL Econo-Column chromatography columns (BioRad, Cat# 7371512) were packed with Ni-NTA agarose beads (Qiagen, Cat# 30210), and proteins were purified by gravity flow. The supernatant was pre-dialyzed against Ni-NTA buffer (50 mM NaH 2 PO 4 , 300 mM NaCl; pH 8.0), which was recommended by the manufacturer and used for all steps. The equilibrated supernatant was loaded into the column, and the beads were then washed with the Ni-NTA buffer followed by stepwise elution using imidazole (50-250 mM). Eluted fractions were analyze by 8%-16% SDS-PAGE gel (mini-PROTEAN TGX protein gels; Bio-Rad), and fractions containing the gp120 glycoprotein were concentrated followed by buffer exchange to PBS using Vivaspin 6 centrifugal concentrators (30 kDa; Cytiva, Cat# GHC-28-9323-17). Purified gp120 glycoproteins were flash-frozen and stored in aliquots at -80°C. The concentration of the protein was determined by BCA assay (Pierce, Cat# A55864).
1059-SOSIP was purified by affinity chromatography using a Galanthus Nivalis Agglutinin (GNA) pre-packed column (Sterogene Bioseparations, Cat# 937614SF4). The clarified supernatant was equilibrated with five column volumes (CVs) of PBS, loaded on the column, and the column was washed with 10 CVs of PBS at a high flow rate, followed by an additional 5 CV wash with high salt (PBS contain ing 500 mM NaCl, pH 7.5). The 1059-SOSIP protein was eluted with 0.1-0.3 M of methyl-α-D-mannopyranoside/PBS solution and concentrated using Vivaspin 6 centrifugal concentrators (30 kDa; Cytiva). Purified 1059-SOSIP glycoproteins were then further separated according to protein size using a HiLoad 16/600 Superdex 200 pg size-exclusion chromatography column (Cytiva).
## Deglycosylation analysis
Four micrograms of AD8gp120, purified from 293 F cells or HEK293 stable cell line, were mixed with 1 µL Glycoprotein Denaturing Buffer (10×) in 10 µL reaction volume and incubated at 100°C for 10 minutes. The reaction was then cooled down by placing the tube on ice, the tube was centrifuged for 10 seconds, and 2 µL GlycoBuffer 2 (10×), 2 µL 10% NP-40, 6 µL water, and 1 µL PNGase F (New England Biolabs) were added to the reaction tube and then incubated at 37°C for 1 hour. Deglycosylation was evaluated by SDS-PAGE analysis. Untreated AD8gp120 samples were analyzed in parallel to compare side-by-side glycosylated and deglycosylated AD8 gp120 proteins.
## Enzyme-linked immunosorbent assay (ELISA)
We used ELISA to analyze antibody binding to gp120 monomers, as previously descri bed (89). Briefly, Immulon 2 HB 96-well plates (ThermoFisher, Cat # 3455) were coated by adding 100 µL of PBS containing 0.1 µg AD8 gp120 to each well and incubating overnight in 4°C. The next day, plates were washed three times with 0.1% Tween-20 (BioRad, Cat # 1706531) in PBS (TPBS) and blocked overnight at 4°C with 0.5% dry skim milk (BioRad, Cat # 1706404) and 0.1% Tween-20 in PBS (MTPBS). The next day, plates were washed three times with PBST and incubated with 100 µL of diluted antibodies for 1 hour at RT. All the antibodies were diluted in MTPBS. After 1 hour incubation, wells were washed again three times, and 100 µL of horseradish peroxidase (HRP)-conjugated donkey anti-human IgG (FC specific; Jackson ImmunoResearch Laboratories, West Grove, PA) prediluted 1:5,000 in MTPBS were added to each well. The plate was incubated for 1 hour at RT, wells were then washed three times with TPBS, and 100 µL of TMB substrate solution (300 µL of 4 mg/mL 3,3,5,5-tetramethylbenzidine (MilliporeSigma) in DMSO, 10 mL of 0.1 M sodium acetate, pH 5.0, and 5 µL of fresh 30% hydrogen peroxide) were added to each well. After ~1 hour incubation at RT, the HRP reaction was stopped by adding 50 µL of 0.5 M H 2 SO 4, and optical density at 450 nm was measured using Synergy|H1 microplate reader (BioTek).
## Western blot
The culture supernatant containing AD8 gp120 or purified protein was separated on 8%-16% SDS-PAGE gel (mini-PROTEAN TGX protein gels; Bio-Rad) and transferred to a 0.45 µm nitrocellulose membrane (cat # 1620115, Bio-Rad). The membrane was blocked with 5% blotting-grade dry skim milk (cat # 1706404, Bio-Rad) in PBS (5%MPBS) overnight at 4°C. The next day, the membrane was washed with PBS and incubated for 1 hour on a shaker with serum of a person living with HIV-1 (1:10,000 dilution) and sheep anti-gp120 IgG (1:4,000 dilution; cat # 288, NIH AIDS reagent program), both diluted in 5%MPBS. After three washes with 0.05% Tween 20 (cat # 1706531, Bio-Rad) in PBS (TPBS), the membrane was incubated with peroxidase-conjugated anti-human IgG (1:10,000 dilution) and anti-sheep IgG (1:10,000 dilution) (Jackson ImmunoResearch Laboratories) in 5%MPBS for 1 hour. The membrane was washed three times with TPBS, developed with SuperSignal West Pico PLUS Chemiluminescent Substrate (cat # 34580, ThermoFisher Scientific), and analyzed using the Odyssey imaging system (LI-COR Biosciences). In some cases, we used serum from a person living with HIV-1 (1:10,000 dilution) without sheep anti-gp120 IgG.
## Viral assay
Viral neutralization assay in the presence and absence of soluble purified AD8 gp120 was performed as previously described (81).
## RESULTS AND DISCUSSION
## Generation and selection of stable HEK 293-AD8gp120 cells
HEK293 cells, originally derived from human embryonic kidney tissue, are widely used for protein expression due to high transfection efficiency of DNA into these cells by most available methods. We transfected adherent HEK293 cells with a plasmid encoding for HIV-1 AD8 gp120 and the puromycin resistance gene by calcium phosphate (81). To generate a stable cell line expressing AD8gp120, transfected cells were selected with 0.5 µg/mL puromycin starting 24 hours post-transfection and selection continued for 3-4 weeks, with medium changes every 2-3 days to remove dead cells (Fig. 1). We transfected and selected adherent HEK293 cells rather than using HEK293 that have been already adapted for growth in suspension because adherent cells that die are readily removed with the medium, allowing us to maintain only cells that resist puromycin in culture. The cell culture supernatant was analyzed 72 hours post-transfection by Western blot, confirming the correct size of the expressed gp120 protein (Fig. 1c).
## Adaptation of HEK293-AD8gp120 to grow in suspension
Stable, adherent HEK293 cells expressing AD8 gp120 were next adapted to grow in suspension. In preliminary experiments, we tested media procured from different suppliers including Ex-Cell (MilliporeSigma), HEK GM (Sartorius), and BalanCD HEK293 (Irvine Scientific) and chose the latter. HEK293-AD8gp120 cells were then detached, washed, and suspended at an initial cell density of 1 × 10 6 cells/mL in a total volume of 20 mL BalanCD HEK293. Suspension cultures of HEK293 were grown in shaker flasks in the presence of 0.5 µg/mL puromycin, and cell density and viability of the stable HEK293-AD8gp120 cells were monitored daily. We observed a steady increase in viable cell density and successful adaptation to suspension. Notably, when supplemented with 4% feed every other day to maintain optimal growth and productivity, the cells grew to a density of >1 x 10 7 cells/ml with ~90% viability after 9 days in culture.
As reference/control, we used PEI to transiently transfect FreeStyle 293 F cells, with the same AD8gp120 expression plasmid used to generate the stable HEK293 cells. We used PEI for transient transfection and not calcium phosphate because, in our experi ence, 293 cells in suspension are more efficiently transfected with PEI, whereas adherent 293 cells are more efficiently transfected with calcium phosphate. Thus, we used optimized transfection reagent for each case. In addition, as the transfection efficiency increases, stable expression in cell populations may reflect more diverse integration sites (more integration events are potentially possible). However, once cells are selected, most cells are expected to express soluble AD8 gp120 with levels that depend on the specific integration site and cellular and external environments. FreeStyle 293 F transfected cells were grown in the recommended medium for 7 days. Supernatants of cells transiently or stably expressing AD8 gp120 were harvested by centrifugation and then equilibrated overnight against binding buffer. gp120 was purified from the supernatants by liquid chromatography system (AKTA Go; Cytiva) using 5 mL Ni-NTA column. We observed a few differences between the elution profiles of gp120 from the two sources of supernatants. The chromatogram of eluted gp120 from the stable HEK293-AD8gp120 supernatant showed sharp peaks with good separation and a dominant peak corresponding to gp120 according to subsequent SDS-PAGE and Western blot confirmation analysis (Fig. 2 and3). In contrast, the chromatogram of eluted gp120 from transient FreeStyle 293 F cell supernatants showed two peaks of comparable size, and the overall yield was ~50 fold lower compared to the stable expression (Fig. 2; Table 1). Purity of gp120 was evaluated by SDS-PAGE analysis, which showed a single band of the correct ~120 kDa size for both preparations (Fig. 3). However, gp120 purified from the stable HEK293 cells exhibited higher homogeneity compared to the gp120 purified from FreeStyle 293 F cells. Western blot analysis using either sheep polyclonal antibodies against gp120 or serum from a PLWH further confirmed the difference between the two preparations observed by the SDS-PAGE analysis. To test the contribution of glycans to gp120 homogeneity, we used PNGase F to actively remove the glycans from the purified gp120s. The PNGase F enzyme specifically recognized and removed N-linked glycans. Both PNGase-F-treated samples showed the same protein size and homogeneity after the removal of glycans, according to SDS-PAGE analysis. Thus, stable expression of AD8 gp120 was associated with more homogenous processing or modifications, most likely glycosylation, due to a more controlled environment during protein expression, lower copy number of the gp120 gene in each cell, and the slower rate of growth in comparison to the FreeStyle 293 F cells.
## HIV-1 AD8 gp120 antigenicity
We compared the antigenic profile of the monomeric gp120 purified from the 2 cell sources by ELISA. We used the following antibodies: VRC01, VRC03, 3BNC117, N6, 2G12, PGT121, 10-1074, PGT128, and 17b (with or without soluble CD4) and a control antibody against the SARS-CoV2 spike. These antibodies target the CD4-binding site (VRC01, VRC03, 3BNC117, and N6) (90)(91)(92)(93), the glycans at the base of gp120 V3 loop (2G12, PGT121, 10-1074, and PGT128) (94)(95)(96), and the CD4-induced site, which overlaps with the coreceptor binding site (17b) (97). The antibodies were tested at a wide range of concentrations between 0.0001 and 1 µg/mL, and all showed high binding to gp120. Binding of all antibodies showed a typical dose response and, except for 17b, reached saturation at 1 µg/mL antibody concentration. The addition of sCD4 increased the 17b binding to gp120 to similar levels of the binding of the other antibodies. For all antibodies, binding to the monomeric gp120 purified from two different cell sources was comparable, suggesting no significant difference in their antigenicity. As expected, binding of anti-SARS-CoV-2 spike to gp120 as well as binding of the anti-Env antibodies in the absence of gp120 was comparable to the background (Fig. 4).
## HIV-1 AD8 gp120 function
We next assessed the function of purified AD8 gp120 from the two cell sources. We first measured the binding of soluble gp120 to immobilized sCD4 by ELISA. The gp120 from both sources exhibited a typical dose-response curve that reached a saturation at 1 µg/mL gp120 concentration. No significant difference between the two preparations was observed, and controls without either sCD4 or gp120 showed very low nonspecific binding (Fig. 5). We then measured the ability of gp120 to compete with and block viral infection by viruses pseudotyped with HIV-1 AD8 Envs (gp160). The response to soluble gp120 from both cell sources exhibited similar patterns of neutralization, with an IC50 value of approximately 0.5 µg/mL. Thus, the gp120 function of the monomeric protein from two preparations was consistent with known biological activities of gp120 and exhibited the expected and very similar phenotype in both cases.
## Production of HIV-1 1059-SOSIP trimer
As 1059 gp140 has been used to measure antigen binding of serum from nonhuman primates vaccinated with DNA-based 1059 immunogen (99). Thus, the 1059-SOSIP trimer serves as a good example for trimer production. We transfected both 293 F in FreeStyle medium and HEK293 in BalanCD medium with the same plasmid supporting 1059-SOSIP expression. Both cells were grown in suspension and transfected with PEI. We purified the 1059-SOSIP trimer from the supernatant of cells using Galactose Nivalis lectin column, followed by size exclusion chromatography. Consistent with our results for the gp120 production, we detected a significantly higher yield of 1059-SOSIP that was produced in HEK293 grown in BalanCD medium (with feed) compared to the production in 293 F cells grown in FreeStyle medium. Notably, the higher fraction of trimeric protein was formed in HEK293 cells compared to the 293 F cells (Fig. 6) Here, we report the efficient production of HIV-1 AD8 gp120 from stable HEK293 cells grown in suspension. Establishment of the HEK293-AD8gp120 cells was straightforward and required only a few weeks of selection; even without further selection for single clones, we could purify hundreds of milligrams from a liter of cells grown in suspension, which was significantly higher (approximately 50-fold higher; Table 1) than the yield of b Yield of purified 1059-SOSIP was calculated from the FPLC chromatograms using the theoretical extinction coefficient 1.61 for 1059-SOSIP.
FreeStyle 293 F cells transiently transfected with the same gp120 expression plasmid. The stable HEK293-AD8 gp120 cells grew in suspension at a slower rate than the parental adherent 293 grown in DMEM that included serum; however, the stable HEK293-AD8 gp120 cells grew to very high density of >1 × 10 7 cells/mL in suspension. Thus, the high number of stable cells produced significantly higher amounts of proteins than transient transfection. Moreover, the slower and more controlled growth of the HEK293 in BalanCD medium in comparison with the 293 F cells led to more homogenous glycosylation patterns, which may be important for presentation to the immune system and for biological activity. We obtained similar results for production of HIV-1 gp120 of a second HIV-1 strain (1059) and about 10-fold higher yield for the stable HEK293-1059gp120 compared to the transient transfection of the same plasmid into FreeStyle 293 F cells (Table 1). Additionally, we obtained similar results with a different soluble format (SOSIP) of HIV-1 Envs. 1059 Envs have been used for DNA-based vaccination of humans (98), and 1059-SOSIP could be efficiently produced in HEK293 cells grown in suspension in BalanCD medium. Efficient 1059-SOSIP production was associated with increased yield and higher fraction of trimeric 1059-SOSIP protein compared to 1059-SOSIP produc tion in 293 F cells. Thus, at least for two different gp120s and 1 SOSIP Envs, we observed highly efficient production of soluble HIV-1 Envs. The method is appropriate for laboratories that may need high amounts of proteins for different experiments.
## References
1. Herschhorn, Hizi (2010) "Retroviral reverse transcriptases" *Cell Mol Life Sci*
2. Andries, Azijn, Thielemans et al. (2004) "TMC125, a novel next-generation nonnucleoside reverse transcriptase inhibitor active against nonnucleoside reverse transcriptase inhibitor-resistant human immunodeficiency virus type 1" *Antimicrob Agents Chemother*
3. Das, Clark, Jr et al. (2004) "Roles of conformational and positional adaptability in structure-based design of TMC125-R165335 (etravirine) and related non-nucleoside reverse transcriptase inhibitors that are highly potent and effective against wildtype and drug-resistant HIV-1 variants" *J Med Chem*
4. Hikichi, Grover, Schäfer et al. (2024) "Epistatic pathways can drive HIV-1 escape from integrase strand transfer inhibitors" *Sci Adv*
5. Steigbigel, Cooper, Kumar et al. (2008) "Raltegravir with optimized background therapy for resistant HIV-1 infection" *N Engl J Med*
6. Surleraux, Tahri, Verschueren et al. (2005) "A broad HIV-1 inhibitor blocks envelope glycoprotein transitions critical for entry" *Nat Chem Biol*
7. Pancera, Lai, Bylund et al. (2017) "Crystal structures of trimeric HIV envelope with entry inhibitors BMS-378806 and BMS-626529" *Nat Chem Biol*
8. Richard, Prévost, Bourassa et al. (2023) "Temsavir blocks the immunomodulatory activities of HIV-1 soluble gp120"
9. Gartland, Stewart, Zhou et al. (2024) "Characterization of clinical envelopes with lack of sensitivity to the HIV-1 inhibitors temsavir and ibalizumab" *Antiviral Res*
10. Prévost, Chen, Zhou et al. (2023) "Structure-function analyses reveal key molecular determinants of HIV-1 CRF01_AE resistance to the entry inhibitor temsavir" *Nat Commun*
11. (2025) *Full-Length Text Journal of Virology*
12. Paik (2022) "Lenacapavir: first approval" *Drugs (Abingdon Engl)*
13. Van Zyl, Prochazka, Schmidt et al. (2025) "Lenacapavir-associated drug resistance: implications for scaling up long-acting HIV pre-exposure prophylaxis" *Lancet HIV*
14. Phillips, Smith, Bansi-Matharu et al. (2025) "Potential impact and cost-effectiveness of longacting injectable lenacapavir plus cabotegravir as HIV treatment in Africa" *Nat Commun*
15. Abel, Back, Vourvahis (2009) "Maraviroc: pharmacokinetics and drug interactions" *Antivir Ther*
16. Iacob, Iacob (2017) "Ibalizumab targeting CD4 receptors, an emerging molecule in HIV therapy" *Front Microbiol*
17. Chun, Carruth, Finzi et al. (1997) "Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection" *Nature*
18. Wong, Hezareh, Günthard et al. (1997) "Recovery of replication-competent HIV despite prolonged suppression of plasma viremia" *Science*
19. Wietgrefe, Anderson, Duan et al. (2023) "Initial productive and latent HIV infections originate in vivo by infection of resting T cells" *J Clin Invest*
20. Deleage, Wietgrefe, Prete et al. (2016) "Defining HIV and SIV reservoirs in lymphoid tissues"
21. Herschhorn, Marasco, Hizi (2010) "Antibodies and lentiviruses that specifically recognize a T cell epitope derived from HIV-1 Nef protein and presented by HLA-C" *J Immunol*
22. Mao, Liao, Zhu et al. (2024) "Efficacy and safety of novel multifunctional M10 CAR-T cells in HIV-1-infected patients: a phase I, multicenter, single-arm, open-label study" *Cell Discov*
23. Caskey, Klein, Lorenzi et al. (2015) "Viraemia suppressed in HIV-1-infected humans by broadly neutralizing antibody 3BNC117" *Nature*
24. Bar-On, Gruell, Schoofs et al. (2018) "Safety and antiviral activity of combination HIV-1 broadly neutralizing antibodies in viremic individuals" *Nat Med*
25. Cale, Bai, Bose et al. (2020) "Neutralizing antibody VRC01 failed to select for HIV-1 mutations upon viral rebound" *J Clin Invest*
26. Darcis, Kula, Bouchat et al. (2015) "An in-depth comparison of latency-reversing agent combinations in various in vitro and ex vivo HIV-1 latency models identified bryosta tin-1+JQ1 and ingenol-B+JQ1 to potently reactivate viral gene expression" *PLoS Pathog*
27. Battivelli, Dahabieh, Abdel-Mohsen et al. (2018) "Distinct chromatin functional states correlate with HIV latency reactivation in infected primary CD4 + T cells" *Elife*
28. Herschhorn, Hizi (2008) "Virtual screening, identification, and biochemical characterization of novel inhibitors of the reverse transcriptase of human immunodeficiency virus type-1" *J Med Chem*
29. Ahmed, Herschhorn "2024. mRNA-based HIV-1 vaccines" *Clin Microbiol Rev*
30. Ahmed, Herschhorn (2024) "Insights from HIV-1 vaccine and passive immunization efficacy trials" *Trends Mol Med*
31. Ackerman, Das, Pittala et al. (2018) "Route of immunization defines multiple mechanisms of vaccine-mediated protection against SIV" *Nat Med*
32. Abbott, Lee, Menis et al. (2018) "Precursor frequency and affinity determine B cell competitive fitness in germinal centers, tested with germline-targeting HIV vaccine immunogens" *Immunity*
33. Gao, Wijewardhana, Mann (2018) "Virus-like particle, liposome, and polymeric particle-based vaccines against HIV-1" *Front Immunol*
34. Apostólico, De, Boscardin et al. (2016) "HIV Envelope trimer specific immune response is influenced by different adjuvant formulations and heterologous prime-boost" *PLoS One*
35. Ratnapriya, Perez-Greene, Schifanella et al. (2022) "Adjuvant-mediated enhancement of the immune response to HIV vaccines" *FEBS J*
36. Ahmed, Parthasarathy, Newhall et al. (2023) "Enhancing anti-viral neutralization response to immunization with HIV-1 envelope glycoprotein immunogens" *NPJ Vaccines*
37. Williams, Alam, Ofek et al. (2024) "Vaccine induction of heterologous HIV-1-neutralizing antibody B cell lineages in humans" *Cell*
38. Burton, Ahmed, Barouch et al. (2012) "A blueprint for HIV vaccine discovery" *Cell Host Microbe*
39. Caniels, Prabhakaran, Ozorowski et al. (2025) "Precise targeting of HIV broadly neutralizing antibody precursors in humans" *Science*
40. Kwong, Wyatt, Robinson et al. (1998) "Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody" *Nature*
41. Liu, Bartesaghi, Borgnia et al. (2008) "Molecular architecture of native HIV-1 gp120 trimers" *Nature*
42. Wyatt, Kwong, Desjardins et al. (1998) "The antigenic structure of the HIV gp120 envelope glycoprotein"
43. Choe, Farzan, Sun et al. (1996) "The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates" *Cell*
44. Trkola, Dragic, Arthos et al. (1996) "CD4-dependent, antibody-sensitive interactions between HIV-1 and its co-receptor CCR-5" *Nature*
45. (2025) *Full-Length Text Journal of Virology*
46. Wu, Gerard, Wyatt et al. (1996) "CD4induced interaction of primary HIV-1 gp120 glycoproteins with the chemokine receptor CCR-5" *Nature*
47. Dragic, Litwin, Allaway et al. (1996) "HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5" *Nature*
48. Alkhatib, Combadiere, Broder et al. (1996) "CC CKR5: a RANTES, MIP-1alpha, MIP-1beta receptor as a fusion cofactor for macrophage-tropic HIV-1" *Science*
49. Feng, Broder, Kennedy et al. (1996) "HIV-1 entry cofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor" *Science*
50. Herschhorn, Gu, Moraca et al. (2017) "The β20-β21 of gp120 is a regulatory switch for HIV-1 Env conformational transitions" *Nat Commun*
51. Herschhorn, Ma, Gu et al. (2016) "Release of gp120 restraints leads to an entrycompetent intermediate state of the HIV-1 envelope glycoproteins" *mBio*
52. Munro, Gorman, Ma et al. (2014) "Conforma tional dynamics of single HIV-1 envelope trimers on the surface of native virions" *Science*
53. Li, Qin, Nand et al. (2023) "HIV-1 Env trimers asymmetri cally engage CD4 receptors in membranes" *Nature*
54. Parthasarathy, Pickthorn, Ahmed et al. (2025) "Incompletely closed HIV-1 CH040 envelope glycoproteins resist broadly neutralizing antibodies while mediating efficient HIV-1 entry" *Npj Viruses*
55. Jeffy, Parthasarathy, Ahmed et al. (2023) "Alternative substitutions of N332 in HIV-1 AD8 gp120 differentially affect envelope glycoprotein function and viral sensitivity to broadly neutralizing antibodies targeting the V3-glycan" *bioRxiv*
56. Yen, Herschhorn, Haim et al. (2014) "Loss of a conserved N-linked glycosylation site in the simian immunodeficiency virus envelope glycoprotein V2 region enhances macrophage tropism by increasing CD4-independent cell-to-cell transmission" *J Virol*
57. Behrens, Crispin (2017) "Structural principles controlling HIV envelope glycosylation" *Curr Opin Struct Biol*
58. Koch, Pancera, Kwong et al. (2003) "Structure-based, targeted deglycosylation of HIV-1 gp120 and effects on neutralization sensitivity and antibody recognition" *Virology (Auckl)*
59. Lee, Ozorowski, Ward (2016) "Cryo-EM structure of a native, fully glycosylated, cleaved HIV-1 envelope trimer" *Science*
60. Parthasarathy, Pothula, Ratnapriya et al. (2024) "Conformational flexibility of HIV-1 envelope glycoproteins modulates transmitted/founder sensitivity to broadly neutralizing antibodies" *Nat Commun*
61. Han, Jones, Nicely et al. (2019) "Difficult-to-neutralize global HIV-1 isolates are neutralized by antibodies targeting open envelope conformations" *Nat Commun*
62. Flemming, Wiesen, Herschhorn (2018) "Conformation-dependent interactions between HIV-1 envelope glycoproteins and broadly neutralizing antibodies" *AIDS Res Hum Retroviruses*
63. Mazurov, Herschhorn (2024) "Ultrasensitive quantification of HIV-1 cell-to-cell transmission in primary human CD4 + T cells measures viral sensitivity to broadly neutralizing antibodies"
64. Abela, Berlinger, Schanz et al. (2012) "Cell-cell transmission enables HIV-1 to evade inhibition by potent CD4bs directed antibodies" *PLoS Pathog*
65. Kalinichenko, Komkov, Mazurov (2022) "HIV-1 and HTLV-1 transmission modes: mechanisms and importance for virus spread" *Viruses*
66. Malbec, Porrot, Rua et al. (2013) "Broadly neutralizing antibodies that inhibit HIV-1 cell to cell transmission" *J Exp Med*
67. Dingens, Haddox, Overbaugh et al. (2017) "Comprehensive mapping of HIV-1 escape from a broadly neutralizing antibody" *Cell Host & Microbe*
68. Bar, Tsao, Iyer et al. (2012) "Early low-titer neutralizing antibodies impede HIV-1 replication and select for virus escape" *PLoS Pathog*
69. Lynch, Boritz, Coates et al. (2015) "Virologic effects of broadly neutralizing antibody VRC01 administration during chronic HIV-1 infection" *Sci Transl Med*
70. Bar, Sneller, Harrison et al. (2016) "Effect of HIV antibody VRC01 on viral rebound after treatment interruption" *N Engl J Med*
71. Bouvin-Pley, Morgand, Meyer et al. (2014) "Drift of the HIV-1 envelope glycoprotein gp120 toward increased neutralization resistance over the course of the epidemic: a comprehensive study using the most potent and broadly neutralizing monoclonal antibodies" *J Virol*
72. Bouvin-Pley, Morgand, Moreau et al. (2013) "Evidence for a continuous drift of the HIV-1 species towards higher resistance to neutralizing antibodies over the course of the epidemic" *PLoS Pathog*
73. Herschhorn (2023) "Indirect mechanisms of HIV-1 evasion from broadly neutralizing antibodies in vivo" *ACS Infect Dis*
74. Wurm (2004) "Production of recombinant protein therapeutics in cultivated mammalian cells" *Nat Biotechnol*
75. L'abbé, Bisson, Gervais et al. (2018) "Transient gene expression in suspension HEK293-EBNA1 cells" *Methods Mol Biol Clifton NJ*
76. Hacker, Balasubramanian (2016) "Recombinant protein production from stable mammalian cell lines and pools" *Curr Opin Struct Biol*
77. Kim, Kim, Lee (2012) "CHO cells in biotechnology for production of recombinant proteins: current state and further potential" *Appl Microbiol Biotechnol*
78. Sun, Wang, Lu et al. (2023) "Protein production from HEK293 cell line-derived stable pools with high protein quality and quantity to support discovery research" *PLoS One*
79. Synoground, Gowtham, Lindquist et al. (2025) "Transcriptomic insights into serum-free medium adaptation and temperature reduction in chinese hamster ovary cell cultures" *Biotechnol J*
80. Dumont, Euwart, Mei et al. (2016) "Human cell lines for biopharmaceutical manufacturing: history, status, and future Full-Length Text Journal of Virology November"
81. *Crit Rev Biotechnol*
82. Berkner (1993) "Expression of recombinant vitamin K-dependent proteins in mammalian cells: factors IX and VII" *Methods Enzymol*
83. Ratnapriya, Chov, Herschhorn (2020) "A protocol for studying HIV-1 envelope glycoprotein function" *STAR Protoc*
84. Chang, Zhou, Liu et al. (2021) "Preparation of pseudo-typed H5 avian influenza viruses with calcium phosphate transfection method and measurement of antibody neutralizing activity" *J Vis Exp*
85. Oupický, Carlisle, Seymour (2001) "Triggered intracellular activation of disulfide crosslinked polyelectrolyte gene delivery complexes with extended systemic circulation in vivo" *Gene Ther*
86. Luthman, Magnusson (1983) "High efficiency polyoma DNA transfection of chloroquine treated cells" *Nucleic Acids Res*
87. Li, Trent, Eid (2023) "Optimized lentiviral vector transduction of adherent cells and analysis in sulforhodamine B proliferation and chromatin immunoprecipitation assays" *STAR Protocols*
88. Hasan, Subbaroyan, Chang (1991) "High-efficiency stable gene transfection using chloroquine-treated Chinese hamster ovary cells" *Somat Cell Mol Genet*
89. Kumar, Nagarajan, Uchil (2019) "Calcium phosphate-mediated transfection of eukaryotic cells with plasmid DNAs" *Cold Spring Harb Protoc*
90. Delafosse, Xu, Durocher (2016) "Comparative study of polyethyleni mines for transient gene expression in mammalian HEK293 and CHO cells" *J Biotechnol*
91. Rao, Onkar, Peachman et al. (2018) "Liposome-encapsulated human immunodeficiency virus-1 gp120 induces potent V1V2-specific antibodies in humans" *J Infect Dis*
92. Huang, Kang, Ishida et al. (2016) "Identification of a CD4-bindingsite antibody to HIV that evolved near-pan neutralization breadth" *Immunity*
93. Zhou, Georgiev, Wu et al. (2010) "Structural basis for broad and potent neutralization of HIV-1 by antibody VRC01" *Science*
94. Wu, Yang, Li et al. (2010) "Rational design of envelope identifies broadly neutralizing human monoclonal antibodies to HIV-1" *Science*
95. Scheid, Horwitz, Bar-On et al. (2016) "HIV-1 antibody 3BNC117 suppresses viral rebound in humans during treatment interruption" *Nature*
96. Julien, Sok, Khayat et al. (2013) "Broadly neutralizing antibody PGT121 allosterically modulates CD4 binding via recognition of the HIV-1 gp120 V3 base and multiple surrounding glycans" *PLoS Pathog*
97. Walker, Phogat, Wagner et al. (2009) "Broad and potent neutralizing antibodies from an African donor reveal a new HIV-1 vaccine target" *Science*
98. Walker, Huber, Doores et al. (2011) "Broad neutralization coverage of HIV by multiple highly potent antibodies" *Nature*
99. Alam, Paleos, Liao et al. (2004) "An inducible HIV type 1 gp41 HR-2 peptide-binding site on HIV type 1 envelope gp120"
100. Cohen, Fiore-Gartland, Walsh et al. (2023) "Trivalent mosaic or consensus HIV immunogens prime humoral and broader cellular immune responses in adults" *J Clin Invest*
102. Hulot, Korber, Giorgi et al. (2015) "Comparison of immunogenicity in rhesus macaques of transmitted-founder, HIV-1 group M consensus, and trivalent mosaic envelope vaccines formulated as a DNA prime, NYVAC, and envelope protein boost" *J Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645973&blobtype=pdf
|
# Human organoids for Risk Group 4 virus research: a new frontier in investigating Nipah virus infection of the central nervous system
Gabriella Worwa, Shuǐqìng Yú, Amanda Hischak, Julie Tran, Jeremy Bearss, John Bernbaum, Daniel Woodburn, Bapi Pahar, Jillian Geiger, Louis Huzella, Santiago Freire, Ian Crozier, César Muñoz-Fontela, Gustavo Palacios, Nicole Kleinstreuer, Lina Widerspick, Jens Kuhn
## Abstract
New approach methodologies, such as high-complexity in vitro systems, are increasingly prioritized in biomedical research as potential alternatives to animal experimentation. We show that cerebral organoids derived from human induced pluripotent stem cells can be leveraged to (i) investigate isolate-specific replication dynamics of Nipah virus and (ii) model key histopathological lesions found in the brain tissue of infected human patients. Furthermore, we discuss the importance of organoid models for the study of Risk Group 4 viruses. IMPORTANCE Advanced development of medical countermeasures against Risk Group 4 viruses, such as the Nipah virus, historically required testing in mammals under the FDA Animal Rule and translation of data to inform clinical trials in humans. Because the application of human organoids in research on viruses pathogenic for humans is conspecific, it bears the potential to reduce, refine, or replace animal studies where unnecessary. Human cerebral organoids are three-dimensional cell aggregates that resemble the developing human brain functionally and structurally. Brain organoids may be valuable in investigating the replication, neuroinvasion, pathogenesis, virulence, and persistence of neurotropic viruses and provide scientific discernment when developing medical countermeasures destined for the human end-user. KEYWORDS central nervous system infection, complex in vitro system, human cerebral organoid, maximum containment, NAMs, new approach methodologies, Nipah virus, NiV Bangladesh, NiV Malaysia, replication kinetics N ipah virus (NiV), a zoonotic and neurotropic henipavirus in the mononegaviral family Paramyxoviridae, is endemic to regions in Southeastern Asia, in particular Bangladesh, India, Malaysia, and the Philippines. Flying foxes of the Pteropus genus host NiV naturally and occasionally, during active shedding pulses, transmit the virus to other mammals, including humans, via bat excreta-contaminated food sources, contact with non-food fomites, or inhalation of aerosols (1-3).In humans, a hallmark of central nervous system (CNS) infection is vasculitis and endothelial cell necrosis (4), but NiV readily infects cells of the CNS parenchyma, including neurons and microglia (5). CNS involvement in NiV encephalitis frequently causes headaches, altered consciousness, and death (6, 7). Survivors may experience neurological sequelae, including headaches, seizures, static encephalopathy, oculomotor dysfunction, cognitive dysfunction, bradykinesia, cervical dystonia, and facial paralysis, and may even suffer from relapsing encephalitis (6, 7).
Significant differences in the transmission, clinical disease phenotypes, and lethal ity/case-fatality rates of Bangladeshi (NiV-B) and Malaysian (NiV-M) isolates have been described in both natural and experimental infections. Compared with NiV-M, NiV-B appears to transmit more efficiently (thereby contributing to increased person-to-per son spread) and has a shorter incubation period. Encephalitis occurs after infection with both isolates, but NiV-B infection is associated with a higher frequency of respira tory disease and higher overall case-fatality rates. Data from experimental exposures of nonhuman primates largely confirmed these differences (8,9). When grivets were exposed intranasally and intratracheally, NiV-B caused more severe clinical signs than NiV-M, resulting in 100% lethality, whereas NiV-M resulted in 50% lethality in this study and another that tested only intratracheal exposure (9). However, as with most nonhuman primate studies, these data were derived from small group sizes (n = 3 or 4), and the publications do not specify whether study personnel were blinded.
The host-virus exposure determinants of these between-isolate differences in disease severity and phenotype are understudied, including in the CNS. Indeed, "side-by-side" comparisons of cellular tropism, viral replication, and cellular/tissue damage have not been possible in human, animal model, or 2D in vitro systems.
Here, we describe the first effort to model these characteristics of NiV infection using a new approach methodology (NAM), that is, human cell-based organoids as a potential future alternative to animal experimentation.
## RESULTS
## Distinct Nipah virus isolates replicate with distinct kinetics in human cerebral organoids
NiV is a neurotropic virus. In a human biology-based in vitro system, we sought to determine whether NiV replication and neurovirulence, including after exposure to the two major isolates, NiV-B and NiV-M, can be investigated in human cerebral organoids. In a first proof-of-concept experiment, we differentiated human induced pluripotent stem cells (iPSCs) and subsequently matured matrix-embedded cerebral organoids for a total of 3 mo. All personnel examining endpoints as part of this experiment remained blinded to the type of exposure until the conclusion of the experiment and subsequent data analysis.
As proof of principle that spatial clustering and size of cell populations present in the organoids can be visualized, we stained an unexposed organoid with specific antibodies targeting astrocytes (cells positive for glial fibrillary acidic protein [GFAP + ]). A t-distrib uted stochastic neighbor embedding (t-SNE) plot revealed numerous macrophages positive for CD68 molecule (CD68 + ) and smaller areas of GFAP + astrocytes with inter spersed undifferentiated stem cells positive for SRY-box transcription factor 2 (SOX2 + ). An area positive for Ki-67 marker of proliferation (Ki67 + ), indicating proliferative activity, overlapped with an area positive for allograft inflammatory factor 1 (AIF1 + ) microglial cells and signs of neuronal differentiation as judged by detection of Ser133-phosphoryla ted cAMP responsive element binding protein (phosphoCREB + ) (Fig. 1A).
Next, we transferred cerebral organoids into maximum (biosafety level 4) contain ment for exposure to 10,000 PFU of NiV-B, NiV-M, or mock/media on Day 0. Organoid culture media were collected immediately after removal of inocula, and every 24 h until Day 12 post-exposure to generate growth curves. Plaque assays were performed using these media to quantify the infectious titers and define replication kinetics of both NiV isolates in the organoids over time (Fig. 1B). A small, but significantly higher NiV-B titer difference was noted (P = 0.01; n = 24) as residue following inoculation compared with NiV-M on Day 0, but we left titers unadjusted to track replication kinetics neatly. We observed an increase in NiV-B titers, which remained elevated until Day 4 to 3.96-5.26 log 10 PFU/mL, followed by a plateau, except for organoids #B22 and #B24 (Fig. 1B). In contrast, we observed a distinct peak of NiV-M replication on Day 4, with titers ranging between 4.20 and 5.56 log 10 PFU/mL, followed by a steady decline in titers until Day 12 (Fig. 1B). The final titers of organoids (n = 6 per group) remaining on Day 12 were 3.87- 5.15 and 2.86-4.04 log 10 PFU/mL for NiV-B and NiV-M, respectively (Fig. 1B). Despite the small increase in titers of NiV-B after inoculation (Day 0), NiV-M replicated to significantly higher titers at 1-4 d (Fig. 1C); titers were similar during subsequent time points (at 5-7 d) and, ultimately, higher for NiV-B at 11 and 12 d in accordance with the replication plateau described above (Fig. 1B andC). Overall, we observed slightly different replication dynamics in cerebral organoids: NiV-M replicated more efficiently early to a peak and subsequent small decline, whereas NiV-B replicated more slowly to a prolonged plateau over the course of the experiment.
Overall, we show that cerebral organoids usefully enable side-by-side investigation of viral replication dynamics, thus far not possible in human, animal model, or 2D tissue culture systems (8)(9)(10)(11).
## Distinct Nipah virus isolates cause characteristic histopathological lesions in human cerebral organoids
At 3, 6, 9, and 12 d post-exposure, three cerebral organoids per group were collected and bisected: one half was fixed in 10% neutral buffered formalin and assessed by histopathologic examination with immunohistochemical staining (by a board-certified veterinary pathologist who was blinded to the type of exposure) and the other half was fixed in 2.5% glutaraldehyde and 2.0% paraformaldehyde and assessed by transmis sion electron microscopy (TEM). All organoids examined by both light and electron microscopy had clearly identifiable cell types and a similar structural morphology that consistently included three characteristic regions. The outer, ependymal layer was thin and consisted of one to three ependymal cell layers. Beneath the ependymal layer(s), a cortical ("cortex") layer, consisting of neurons, astrocytes, microglia, and oligodendro cytes, was reliably present but varied in thickness, ranging from six to 20 cell layers among organoids. In all organoids, the cortical layer surrounded a large inner core region consisting of various cell types and materials, including fibroblasts, collagen, matrix proteins, and areas with lysed cells. Because we found consistent morphology among all organoids, we used only one batch of organoids for this experiment. After exposure to NiV, organoid morphology was altered as progressive cell lysis increasingly led to disruption of the structural, "multi-layered" organization of the organoid. The degree of staining for GFAP + and AIF1 + , markers of necrosis (cleaved caspase 3 [CASP3]), mature neurons (RNA binding fox-1 homolog 3 [RBFOX3]), and oligodendrocytes (oligodendro cyte transcription factor 2 [OLIG2]), was graded on a scale of 0-4 for each organoid and found to be similar between virus-exposed groups (Fig. 2A). Necrosis (a common finding, especially in the organoid centers) was present in all organoids to comparable degrees (Fig. 2A). Representative images for each stain are shown in Fig. 2B through G. Cyto pathic effects characteristic of NiV infection, including syncytia, nuclear and cytoplasmic viral inclusions, and nuclear atypia, were frequently observed in ependymal cells and cortical cells in all NiV-exposed organoids (Fig. 2H andI). We observed a progressive loss of ependymal cells and connective tissue in all NiV-exposed organoids but not in mock-exposed organoids. Unsurprisingly, NiV genomic nucleic acids were detected by in situ hybridization (ISH) in the outer/cortical regions of organoids at early time points; subsequently, increasing ISH signal with outer-to-inner progression was detected over the course of 3, 6, 9, and 12 d post-exposure (Fig. 2J andK). TEM images of NiV-exposed organoids revealed large quantities of NiV nucleocapsid inclusions (present in several cell types), widespread areas of syncytia, fused, or necrotic cells (Fig. 2L andM), and budding of mature NiV particles.
Together, these data demonstrate that NiV infection of cerebral organoids results in key pathological features, including syncytia formation and viral cytopathic effects noted in brain autopsies of humans who died with NiV encephalitis (2,12).
## DISCUSSION
Recognized as a Biodefense Pathogen by the U.S. National Institute for Allergy and Infectious Diseases (13) and other organizations (14), NiV is a priority focus of current research and medical countermeasure (MCM) development efforts globally. In the past, researchers approached candidate MCM assessment for Risk Group 4 pathogens such as NiV via the U.S. Food and Drug Administration (FDA) "Animal Rule, " that is, the replace ment of human clinical trials that cannot be performed due to ethical reasons with "wellcharacterized" animal models (15). Recent developments in organoid and other microphysiological system approaches, that is, NAMs may complement or possibly replace MCM evaluation in animal models. Indeed, the FDA recently announced its intention to strategically phase out reliance on the use of animal models for preclinical safety assessment of a broad range of therapeutics (16). Although the traditional use of nonhuman primates, for instance, served as a translational proxy for humans, the application of human-origin, three-dimensional organoids that mimic the anatomy and function of organs is attractive as a potential future substitute because such organoids eliminate species-related and therefore immune system-related variability, especially in the context of viral infection and MCM evaluation. Organoid creation based on the inclusion of stem cells from donors with varied and representative genetic backgrounds additionally enables organoid customization and forward-thinking diversification of experiments, thereby bringing precision medicine to the doorstep of Risk Group 4 pathogen research. Together with the ability to design blinded experiments that include high numbers of replicates (compared with the historically very low replicates of typically unblinded nonhuman primate studies), data sets derived from organoids and other NAMs provide the opportunity to robustly build confidence in experimental outcomes.
With NAMs as a focus in the FDA's "Modernization Act 2.0" (17) and the recently announced commitment by the National Institutes of Health to prioritize human-based technologies for biomedical research (18), it is expected that maximum (biosafety level 4) containment laboratories will start to pivot towards incorporating NAMs into their research portfolios. The application and handling of organoids within maximum containment are relatively straightforward: no specialized equipment or incubators are required; furthermore, because organoid tissue resembles animal tissue, organoids can easily be incorporated into established sample/pathogen inactivation protocols and Select Agent removal workflows for downstream analyses outside of containment.
Overall, we provide proof of principle that cerebral organoids can be used in the maximum containment setting to investigate NiV viral replication and character ize infection-associated tissue damage. To the best of our knowledge, side-by-side comparisons of NiV-B and NiV-M replication or tissue damage, that is, neurovirulence, in relevant CNS cell types have not been reported in humans, animal models, or 2D cell cultures.
We acknowledge the limitations of this first effort that reflect the considerable challenges associated with the use of any new technology. Ideally, experiments in the future should include at least two "batches" of organoids from the same iPSC donor (to account for potential batch-to-batch variability) as well as from multiple donors (e.g., that might vary in age or sex), depending on the goals of the experiment. Although progress has been made, optimization of histopathologic and immunologic assays for use in virus-exposed CNS organoid systems is needed (19). For instance, we demonstrated via a t-SNE plot that spatial clustering and size of cell populations can be visualized in a single organoid (Fig. 1A). Because this plot represented only the cells contained in one organoid (and cannot represent the heterogeneity likely present between organoids of the same batch), we intentionally did not provide quantitative characterization, which would of course be of future interest to characterize both unexposed and NiV-exposed organoids. We hope that this work (including our staining panel and methodology) will facilitate such analyses in the future. In addition, it will be of interest to characterize cytokine secretion from individual organoids, determine whether these data can be extrapolated to all organoids of a batch (given their morphological variabilities), and then compare those global cytokine responses to those known from animal model experimentation. Also, natural NiV infection causes endothelial cell necrosis, widespread microinfarction, and vasculitis-induced thrombosis (4)-features that are necessarily absent in the cerebral organoids used for this study, given the absence of blood vessels. These features may be better mimicked in microfluidic systems ("organs-on-chips") such as a brain-on-chip that houses microvascular endothelial cells and forms a blood-brain barrier. In organ-on-chips, experimental conditions (such as virus infection, therapeutic dosing, or addition of immune cells) can be controlled more precisely compared with organoids due to the existence of fluidic channels (including a vascular component) and a physically adaptable environment. The combination of NAMs will likely result in complementary and possibly synergistic data that may make the replacement of animal experimentation with more human-relevant models an achievable reality.
## MATERIALS AND METHODS
## Cerebral organoid generation
Human induced pluripotent stem cells (Healthy Control Human iPSC Line, Female, SCTi003-A, #200-0511, Stemcell Technologies, Cambridge, MA, USA) were used for the establishment and maturation of cerebral organoids following a commercial protocol using a STEMdiff Cerebral Organoid Kit (Stemcell Technologies). Briefly, for organoid formation (which took place from Day -10 to Day 5), one vial of frozen human iPSC was thawed on Day -10, and the cells were seeded into one well of a Matrigel-coated 6-well plate containing iPSC seeding medium. Cells were passaged three times prior to induction. On Day 0 (organoid embryoid body induction), high-quality iPSCs (distinct colonies, sharp edges, >60%) were harvested, counted, and seeded (9,000 cells per well) in a round-bottom, ultra-low-attachment 96-well plate containing organoid formation medium. On Days 2 and 4 post-seeding, 100 µL of fresh organoid formation medium were added. Prior to neuronal induction (5-7 d), the presence of organoid embryoid bodies with round and smooth edges was confirmed by light microscopy. On Day 5, the seeding/formation medium was removed and 250 µL of induction medium were added per well, and the plate was incubated for 48 h at 37°C and 5% carbon dioxide (CO 2 ). For organoid expansion (7-10 d), each organoid embryoid body was drawn up in 25 µL of medium using a wide-bore pipette tip and transferred to a mold on the embedding surface. Excess medium was removed from each mold, and 15 µL of Matrigel were added dropwise onto each embryoid body. Matrigel-embedded embryoid bodies were placed in a container and incubated at 37°C and 5% CO 2 for 30 min to achieve polymerization. Next, the embedding surface containing the Matrigel droplets was grasped, and expansion medium was used to carefully wash off each Matrigel droplet from the embedding surface into a well of an ultra-low-attachment 6-well plate. Embedded organoids were subsequently incubated in 3 mL of expansion medium at 37°C and 5% CO 2 for 3 d. Budding of the organoids' surfaces was monitored under the light microscope to document the development of neuroepithelia. On Day 11, wide-bore pipette tips were used to transfer expanded organoids to ultra-low-attachment 24-well plates containing 1 mL of maturation medium per well (one organoid per well to avoid fusion of multiple organoids). For organoid maturation (11-89 d), the medium was replaced with 1 mL of maturation medium per well without disturbing the embedded organoids. Plates containing organoids were then transferred onto an orbital shaker at 37°C for incubation and maturation. Maturation medium was exchanged every 3-4 d and growth was monitored weekly using light microscopy until virus exposure on Day 89. Prior to virus exposure, cerebral organoids were randomly assigned to experimental groups.
## Antibody staining
A single-cell suspension of each cerebral organoid was achieved using the Papain Dissociation System protocol (#LLK003150, Worthington Biochemical Corp, Lakewood, NJ, USA). Whole-organoid staining was performed using the following antibodies: Alexa Fluor 594 Anti-GFAP Antibody (Clone 2E1.E9, #644708, BioLegend, San Diego, CA, USA), Brilliant Violet 785 Anti-Human CD68 Antibody (Clone Y1/82A, #333826, BioLegend), PerCP/Cyanine5.5 Anti-Human CX3CR1 Antibody (Clone 2A9-1, #341614, BioLegend), Phospho-CREB (Ser133) (4D11) Rabbit mAb APC Conjugate (Clone CREBS133-4D11, #2129, AbwizBio, San Diego, CA, USA), BD Horizon BV605 Mouse Anti-Ki-67 (Clone B56, #567122, BD Biosciences, Franklin Lakes, NJ, USA), BD Horizon BUV395 Mouse Anti-EOMES (Clone X4-83, #567171, BD Biosciences), Pacific Blue Anti-SOX2 Antibody (Clone 14A6A34, #656112, BioLegend), Alexa Fluor 750 MxA/Mx1 Antibody OTI2G12 (Clone OTI2G12, #NBP2-72838AF750, Novus Biologicals), and anti-NiV antibody (in-house). The AIF-1/Iba1 Antibody, goat, polyclonal (#NB100-2833, Novus Biologicals, Centennial, CO, USA) was used in an in-house custom conjugation process utilizing a PE-Cy7 Conjugation Kit (#ab102903, Abcam, Cambridge, UK). Live/dead cells were identified with a Live/Dead Fixable Aqua Dead Cell Stain Kit, at 405 nm excitation (#L34966, Thermo Fisher Scientific, Waltham, MA, USA).
Initially, the organoid samples were stained for surface markers. Subsequently, the cells were permeabilized using Cytofix/Cytoperm (# 554714, BD Biosciences), followed by intracellular staining. The cells were subsequently washed and fixed with 500 µL of Cytofix/Cytoperm, resuspended in phosphate-buffered saline (PBS), and data acquisition was performed using a five-laser Cytek Aurora Flow Cytometer (Cytek Biosciences, Fremont, CA, USA). The t-SNE analysis was performed using FlowJo software version 10.8.1.
## Virus exposure
On Day 0, 89-day-old cerebral organoids were exposed in maximum (biosafety level 4) containment to either mock/media or 10,000 PFU of a laboratory stock of NiV (Para myxoviridae: Henipavirus nipahense), isolate Bangladesh 2004 (GenBank accession no. PV892943), or isolate Malaysia 1998 (GenBank accession no. PQ463988) diluted in 500 µL of STEMdiff Cerebral Organoid Basal Medium with Supplement E (#08571, Stemcell Technologies; henceforth: organoid media). After incubation for 1 h at 37°C, inocula were removed, and organoids were washed once in 1 mL of Gibco Dulbecco's PBS (DPBS; #14190094, Thermo Fisher Scientific), followed by the addition of 1.5 mL of fresh organoid media. Daily harvests of organoid media were used for plaque assay titration as described previously (20).
## Histopathology
Cerebral organoids were carefully bisected, and halves were fixed in either 10% neutralbuffered formalin (for histology) or in 2.5% glutaraldehyde/2.0% paraformaldehyde (for transmission electron microscopy) for at least 72 h prior to removal from maximum containment.
Formalin-fixed organoid halves were routinely processed with an optimized abbreviated protocol in a Tissue-Tek VIP-6 vacuum infiltration automated tissue processor (Sakura Finetek USA, Torrance, CA, USA), paraffin-embedded using a Tissue-Tek TEC-6 embedding console system (Sakura Finetek), sectioned at 4 µm using a stand ard semiautomated rotary microtome (Leica RM2255, Leica Biosystems), mounted on positively charged uncoated glass slides, and then air-dried for routine hematoxylin and eosin (H&E) staining.
Organoid slides were immunohistochemically stained for rabbit anti-GFAP (#Z0334, Dako, 1:3,000 dilution), Anti-NeuN antibody EPR12763 Neuronal Marker (RBFOX3, #ab177487, Abcam, 1:3,000 dilution), Anti-OLIG2 Antibody (#AB9610, Millipore, 1:100 dilution), Anti-Iba1 antibody (AIF1, #ab107159, Abcam, 1:1,300 dilution), and Cleaved Caspase-3 (CASP3) Antibody 9661 (#9661L, Cell Signaling, 1:150 dilution). All tissues were visualized with either 3,3'-diaminobenzidine brown chromogen (#BDB2004L, Betazoid DAB Chromogen Kit, Biocare Medical) or red chromogen (#WR806S, Warp Red Chromo gen Kit, Biocare Medical) hematoxylin (blue) counterstain. Sections were examined by light microscopy and scored with a subjective percentage of positive cells.
RNAscope in situ hybridization was performed to detect NiV genomic nucleic acids in sections of each organoid. Routinely processed 4 µm formalin-fixed paraffin-embedded tissue sections were mounted on positively charged uncoated glass slides and stained using a manual RNAscope 2.5 HD Red kit (#322360, Advanced Cell Diagnostics, Newark, CA, USA), following the manufacturer's protocol, including modifications for optimization validated by appropriate controls. A synthesized 20-ZZ pair probe targeting the NiV nucleoprotein (N) gene was used (V-Nipah-StrM.B.-N, #439251, Advanced Cell Diagnos tics), and the sections were counterstained with hematoxylin and glass-coverslipped with a non-alcohol-based mounting medium to prevent chromogen fading. Sections were examined by light microscopy and scored with a subjective percentage of positive cells.
## Transmission electron microscopy
For conventional thin-section microscopic evaluation, organoids were preserved and inactivated in 2.5% glutaraldehyde/2.0% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) in Millonig's Sodium Phosphate Buffer (Tousimis Research, Rockville, MD, USA). After 72 h of fixation, organoids were washed in the same buffer and incubated for 2 h in 1.0% osmium tetroxide (Electron Microscopy Sciences). After rinsing in water and en bloc staining with 2.0% uranyl acetate (Ted Pella, Redding, CA, USA), samples were dehydrated in a series of graded ethanols and then infiltrated and embedded using a SPURR Low-Viscosity Embedding Kit (Electron Microscopy Sciences). After polymerization, embedded blocks were cut into 70-80-nm sections using an EM UC7 ultramicrotome (Leica Microsystems, Deerfield, IL, USA). Sections were collected on 150-mesh copper grids, stained with lead citrate, and examined using a FEI Tecnai Spirit twin-transmission electron microscope (model G2 F20, Thermo Fisher Scientific) operating at 80 kV.
## References
1. Chua, Bellini, Rota et al. (2000) "Nipah virus: a recently emergent deadly paramyxovi rus" *Science*
2. Chua, Goh, Wong et al. (1999) "Fatal encephalitis due to Nipah virus among pig-farmers in Malaysia" *The Lancet*
3. Spengler, Lo, Welch et al. (2025) "Henipaviruses: epidemiology, ecology, disease, and the development of vaccines and therapeutics" *Clin Microbiol Rev*
4. Erbar, Maisner (2010) "Nipah virus infection and glycoprotein targeting in endothelial cells" *Virol J*
5. Wong, Shieh, Kumar et al. (2002) "Nipah virus infection: pathology and pathogenesis of an emerging paramyxoviral zoonosis" *Am J Pathol*
6. Sejvar, Hossain, Saha et al. (2007) "Long-term neurological and functional outcome in Nipah virus infection" *Ann Neurol*
7. Tan, Goh, Wong et al. (2002) "Relapsed and late-onset Nipah encephalitis" *Ann Neurol*
8. Mire, Satterfield, Geisbert et al. (2016) "Pathogenic differences between Nipah virus Bangladesh and Malaysia strains in primates: implications for antibody therapy" *Sci Rep*
9. Johnston, Briese, Bell et al. (2015) "Detailed analysis of the African green monkey model of Nipah virus disease" *PLoS One*
10. Geisbert, Daddario-Dicaprio, Hickey et al. (2010) "Development of an acute and highly pathogenic nonhuman primate model of Nipah virus infection" *PLoS One*
11. Goldin, Liu, Rosenke et al. (2025) "Nipah virus-associated neuropathology in African green monkeys during acute disease and convalescence" *J Infect Dis*
12. Clayton, Middleton, Bergfeld et al. (2012) "Transmission routes for Nipah virus from Malaysia and Bangladesh" *Emerg Infect Dis*
13. (2022) "Prioritizing diseases for research and development in emergency contexts"
14. (2024) "Food and Drug Administration"
15. (2025) "FDA announces plan to phase out animal testing requirement for monoclonal antibodies and other drugs"
16. (2022) "S.5002 -FDA Modernization Act"
17. (2025) "NIH to prioritize human-based research technologies. New initiative aims to reduce use of animals in NIH-funded research"
18. Thomas, Sirois, Li et al. (2024) "CelltypeR: a flow cytometry pipeline to characterize single cells from brain organoids"
19. Cong, Lentz, Lara et al. (2017) "Loss in lung volume and changes in the immune response demonstrate disease progression in African green monkeys infected by small-particle aerosol and intratracheal exposure to Nipah virus" *PLoS Negl Trop Dis*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12826980&blobtype=pdf
|
# Structural polymorphism of two-dimensional lattices assembled from baculoviral capsid proteins
Kexing Tian, Heya Na, Yan Fu, Tingting Chong, Chao Leng, Fanxing Meng, Yaozhou Liang, Manli Wang, Zhihong Hu, Xi Wang, Guibo Rao, Sheng Cao, Virologica Sinica
## Abstract
Protein nanotubes (PNTs) can be regarded as two-dimensional (2D) lattices with p1 or p2 symmetry rolled into tubes. However, attempts to re-assemble their building blocks into stable 2D nanomaterials often fail. Here, starting from two baculoviral capsid proteins, we screened protein variants for the in vitro assembly of various nanotubes and nanosheets. These high-order assemblies were structurally characterized by cryo-electron microscopy techniques. Interfacial analysis of three groups of PNTs revealed that helical heterogeneity is largely the result of the redundancy of p2 symmetry-related contacting interfaces. The assembled nanosheets showed similar interfacial networks to their nanotubular counterparts. In addition, foreign macromolecules could be efficiently displayed on the sizecontrollable double-layered nanosheets. This study sheds light on the rational design of flexible nanosheets, and it also provides novel 2D protein scaffolds for developing biocompatible materials.
## INTRODUCTION
Owing to the distinct geometry-related properties of one-(1D) or three-dimensional (3D) protein nanomaterials, such as nanowires and nanocages, two-dimensional (2D) protein nanosheets have attracted attention for developing biocompatible applications (Hamley, 2019;Zhang et al., 2020). Naturally occurring 2D protein lattices are found in the purple membrane of Halobacterium species and the surface (S-) layer of bacteria and archaea (Henderson and Unwin, 1975;Baneyx and Matthaei, 2014;Fagan and Fairweather, 2014). To provide versatile platforms for nanosheets, computer-aided protein engineering focusing on interfacial interactions has been used to interconnect small building blocks to form continuous 2D lattices based on 17 known planar symmetries (Gonen et al., 2015;Suzuki et al., 2016;Ben-Sasson et al., 2021). Geometrically, protein nanotubes (PNTs) can be viewed as rolled planar lattices, and only lattices with p1 or p2 symmetry can roll into tubular structures (Yeates, 2017). Thus, it is feasible to develop novel nanosheets using the same building blocks as those of PNTs.
Viral capsid structures serve as a rich natural source for the development of protein-based nanomaterials (Liu et al., 2012;Wen and Steinmetz, 2016). The most extensively studied PNTs are based on tobacco mosaic virus (TMV), which possesses a rigid helical capsid encapsulating a genomic RNA strand (Klug, 1999). TMV nanotubes have been explored for different applications (Koch et al., 2016). Many other rod-shaped viruses show some degree of polymorphism, either because of local external forces, such as bending or flattening of soft nanotubes, or because of different symmetrical parameters, such as the helical twist and rise or the start number (He and Scheres, 2017). By using sophisticated sorting strategies for different viral populations, the tubular regions of vesicular stomatitis virus and baculoviruses have been structurally studied at increasedresolution by cryo-electron microscopy (cryo-EM) (Jenni et al., 2022;Zhou et al., 2022;Benning et al., 2024). However, the molecular mechanism underlying structural heterogeneity is not well understood, mainly because of the difficulty of obtaining a sufficient number of particles with less prominent variations for structural analysis.
Cryo-EM studies of the cylindrical trunk of nucleocapsids from Autographa californica multiple nucleopolyhedrovirus (AcMNPV, a group I alphabaculovirus) have revealed the structural arrangement of the major capsid protein (VP39) on virions at near-atomic resolution (Jia et al., 2023;Benning et al., 2024). The tubular VP39 trunk with a diameter of ~50 nm is formed by helical wrapping of 14 protein strands, which comprise tandemly arranged VP39 homodimers. In early work, we reported a controllable approach for preparing PNTs derived from another baculoviral capsid protein (CP) of Helicoverpa armigera nucleopolyhedrovirus (HearNPV, a group II alphabaculovirus) (Rao et al., 2018). When the purified HearNPV nucleocapsids were stored for an extended period of time and then analysed by transmission electron microscopy (TEM), 2D lattices (after 2 weeks) or lattice-like aggregations (after 8 weeks) were identified along with distorted nucleocapsids (Supplementary Fig. S1A-B), showing that 2D lattices of HaCP (CP of HearNPV) were conformationally flexible during the transition from "curved" states to "flattened" states.
In this study, we analysed the molecular interaction networks formed by neighbouring subunits in a variety of single-walled PNTs (SWPNTs) and multi-walled PNTs (MWPNTs) based on HaCP or VP39 mutants by cryo-EM, allowing us to explore the structural basis for the nanotubular flexibility at near-atomic resolution. The single-layered planar materials were also in vitro assembled, and their structural relationships with the nanotubular assemblies were investigated. In addition, size-controllable nanosheets were assembled as protein bilayers for displaying large foreign molecules through the SpyTag-SpyCatcher system (Hatlem et al., 2019). The results demonstrated that using the same building blocks as those of naturally occurring flexible nanotubes is a promising strategy for obtaining protein nanosheets.
## RESULTS
## Native helical arrangement of HaCP in HearNPV nucleocapsids
Cryo-EM analysis of the purified HearNPV nucleocapsid (Fig. 1A) revealed that HaCP in the trunk region had a similar helical arrangement (twist = 18.7 • , rise = 44.3 Å and C14 rotational symmetry) (Supplementary Table S1) to VP39 in the AcMNPV nucleocapsid (twist = 18.5 • , rise = 44.1 Å and C14). A HaCP homodimer (hereafter called a primary dimer, p-dimer) was the helical asymmetric unit (ASU) (Fig. 1B). For simplicity, we call the right-handed strands consisting of helically arranged p-dimers p-filaments, and we define the tube based on the number of p-filaments. The HaCP tube in the nucleocapsid is called H14, and the tube derived from the AcMNPV nucleocapsid is called A14, as both contain 14 p-filaments. We defined the mean radius (R m ) of the tube as the distance from the centre of mass of the p-dimer to the main axis of the tube. H14 (R m = 216.5 Å) was slightly narrower than A14 (R m = 226.6 Å) (Supplementary Table S2). HaCP shared high amino acid sequence identity (~45%) with VP39 (Supplementary Fig. S1C). Local reconstruction focusing on four neighbouring HaCP p-dimers at 3.4 Å resolution (Supplementary Fig. S2) verified that the HaCP protomer was also structurally similar to VP39 with a Cα root mean square deviation (RMSD) of 1.1 Å (233 of the total of 293 atoms).
According to the nomenclature of the VP39 structure (Benning et al., 2024), HaCP contains Zn-finger, antenna, claw, glider and lasso domains (Supplementary Fig. S1D andE). Similar to VP39 (Jia et al., 2023;Benning et al., 2024), the p-dimer is mainly stabilized by multiple hydrophobic contacts between the antenna domain and wrapping-around lasso domain. The Zn-finger motif, CCCH (containing a cluster of four residues, C17, C35, C48 and H51), the β-hairpin in the antenna domain and most of the lasso domain largely face the central channel. The thickness of the nanotube is ~40 Å. A loop (called a CG-loop) connecting the claw domain to the glider domain is crucial for maintaining the continuity of the p-filaments in H14 (Fig. 1C-E). Previous structural studies have suggested that a disulfide bond (Cys132-Cys169) in A14 plays a role in the interaction of the CG-loop with neighbouring p-dimers (Jia et al., 2023;Benning et al., 2024). Our H14 density map revealed that in addition to this conserved interaction (Cys131-Cys165), Y164 inserts into a hydrophobic pocket sandwiched by two helices (α4 and α5) from neighbouring HaCP subunits, which are also responsible for maintaining the connectivity of the p-filaments. In contrast to the structural definition based on the 14 p-filaments, the H14 tube can also be viewed as stacking of C14 ring-like structure comprising laterally arranged 14 p-dimers (Fig. 1F). This ring can be subdivided into 14 different HaCP dimers (called secondary dimers, s-dimers) consisting of two HaCP molecules from neighbouring p-filaments. Hydrophobic interactions are important for stabilizing the s-dimers (Fig. 1G), and they thus also contribute to the lateral connectivity between the p-filaments.
To investigate the molecular interactions, we hereafter focused our structural analysis on four local dyad axes (Fig. 1H): type I (intra-pdimer), type II (inter-p-dimer), type III (lasso-lasso) and type IV (intra-sdimer), corresponding to the symmetry axes featured in the p2 group, as described in S-layer lattices (Sleytr et al., 2014). In H14, one HaCP molecule within a p-dimer interacts with three neighbouring HaCP molecules (related by type II, III or IV dyad axes) in a relatively balanced way, and the buried surface area for each of these interfaces is larger than 400 Å 2 (Fig. 1I).
## Cryo-EM analysis of the in vitro assembled HaCP nanotubes
The second group of PNTs for structural characterization was HaCP nanotubes assembled in vitro. Recombinant expression of HaCP fused with a glutathione S-transferase (GST) tag and proteolytic cleavage of the GST tag drive simultaneous orderly assembly of HaCP into two types of nanotubes, and the relatively narrow nanotubes significantly outnumber wider nanotubes (Rao et al., 2018). However, the structural bending of these PNTs hampered our efforts in high-resolution cryo-EM reconstruction. Considering that HaCP (molecular weight ~33 kDa) is smaller than VP39 (~39 kDa), largely because HaCP does not contain a C-terminal 39-residue tail (Supplementary Fig. S1C), we designed and tested over 60 HaCP mutants of interfacial residues (Supplementary Table S3), including the C-terminal ones which are also at the intermolecular interface (Fig. 1C), to obtain long straight nanotubes (Fig. 2A-C, and Supplementary Fig. S3). We collected two cryo-EM data sets, HaCP-R293C (consisting of 11 p-filaments) and HaCP-C218A (consisting of 12 p-filaments), for reconstruction of H11 (twist = 66.6 • , rise = 6.8 Å and C1) and H12 (twist = -23.0 • , rise = 36.0 Å and C6) (Fig. 2D), respectively. The cryo-EM images of HaCP-R293C showed three types of assemblies: H11, H12 (<2% PNTs) and some 2D lattices (see below, Fig. 3A). The cryo-EM images of the HaCP-C218A mutants exhibited a higher portion of H12 (25% PNTs), but no 2D lattices were detected (Fig. 2A, and Supplementary Fig. S2). In contrast to H14, the helical ASU for the H11 and H12 reconstructions was the s-dimer (Fig. 2E).
Local reconstruction of four neighbouring p-dimers produced density maps with improved resolution (2.8 Å for H11 and 3.3 Å for H12), which allowed us to model most of the HaCP structure, except for twelve residues at the N-terminus, nine residues (D162-T170) in the CG-loop and two residues at the C-terminus (including R293). Although H11 (R m = 98.3 Å) is narrower than H12 (R m = 107.3 Å), the basic assembly units (p-dimers) and the contacting interface between these units are highly conserved (Fig. 2F andG), with a Cα RMSD of 1.1 Å (2081 of the total of 2160 Cα atoms). C218A (Fig. 2F) and R293C mutations did not cause apparent variations in the nearby density distribution, suggesting that these residues may be involved in more dynamic processing of the nanotubular assembly that cannot be captured in cryo-EM reconstructions.
Compared with H14, H11 showed significant structural variations at three contacting interfaces (II, III and IV) (Fig. 2G). At the type II interface, the disordered CG-loops indicated that the connectivity along the H11 p-filaments is relatively weak. When the CG-loop (residues 159-172) was replaced by a flexible loop containing SpyTag (HaCP-159S172 in Supplementary Fig. S3), single-walled nanotubes with diameters of 30-75 nm formed, suggesting that inter-p-dimer interactions are largely dispensable during in vitro assembly of HaCP PNTs. At the type III interface, the lasso loops are responsible for the interactions between adjacent p-filaments by forming two anti-parallel β-strands (Fig. 2H), and these interactions are well preserved in all of the HaCP tubes. However, compared with the H14 lasso loops, the H11 (or H12) lasso loops are dislocated in the HaCP protomer, resulting in an ~16 Å shift of the relative positions of the two p-dimers involved in the lasso-lasso interactions (Supplementary Fig. S4A-B). Replacement of the lasso region with the homologous sequence from VP39 (HaCP-Exclasso in Supplementary Fig. S3) still produced nanotubes, but they were less stable. Deletion of the lasso region ) or the interacting β-strand (HaCP-Δ[252-262]) caused complete loss of the assembly of the nanotubes, although soluble protein could be expressed in E. coli (Supplementary Fig. S4C-D). At the type IV interface, hydrophobic interactions are crucial for the intra-s-dimer interactions of H11 (Fig. 2I-K). Fragment insertion or replacement within a loop near this interface caused the formation of less stable SWPNTs or even MWPNTs (HaCP-Sd1, Sd2 and Sd3, Supplementary Fig. S3).
## Structural variations between nanotubes and real single-layered 2D nanosheets
Interestingly, 2D protein sheets were detected in ~10% of the cryo-EM images of the HaCP-R293C mutant, in addition to two types of nanotubes (H11 and H12) (Fig. 3A). These sheets appeared as black--white striped patterns, and some of the sheets captured in the cryo-EM images partially overlapped. Because these sheets were substantially larger than any unrolled lattices from nucleocapsids (Supplementary Fig. S1A-B), we conclude that they are not simply misassembled or unrolled H11/H12 tubes, but instead represent an alternative assembly pathway under the same in vitro conditions. The boundaries of the sheets were difficult to identify due to the low signal-to-noise ratio. This limitation was overcome by masking the diffraction points contributed by periodical stripe signals in the power spectrum and applying inverse fast Fourier transformation, which improved the signal-to-noise ratio and clarified the sheet edges (Fig. 3A-C, and Supplementary Fig. S5). Direct measurement of the line intensity profiles revealed that the inter-line spacing was ~33 Å. The extraction and averaging of local regions at randomly selected sites in the cryo-EM images resulted in only one 2D class average, which exhibited periodically arranged density for the p-dimers and s-dimers (Fig. 3D andE). The s-filaments composed of tandemly arranged s-dimers also had a width of ~33 Å (Fig. 3E), indicating that they were the "stripes" identified in the raw cryo-EM images.
The four types of interactions of the p2 symmetry were analysed similarly to those in the nanotubes (Fig. 3E). The p-dimer density on the 2D average was well aligned with the 2D projection of the H11 p-dimer, suggesting that the p-dimer was also the basic building block of the unrolled 2D lattice. The density of the CG-loop was largely absent near the type II dyad axis, similar to the H11 or H12 PNTs. The density of the lasso-lasso (type III) interactions was detectable between neighbouring s-dimers along the s-filament. Although some local features of the 2D projection of the H11 s-dimer could be identified on the 2D average, separation and rotation of HaCP protomers near the type IV axis are expected to occur during "unwrapping" of the nanotubes. Given that these 2D lattices and nanotubes formed under the same physical and chemical conditions, the p2-symmetrical interactive network is highly conserved among these related nanomaterials.
## In vitro assembled multi-walled VP39 nanotubes
The building blocks for the third group of PNTs were derived from AcMNPV VP39. In vitro assembly of the VP39 nanotubes produced unstable nanotubes (Supplementary Fig. S6). A site-specific alanine mutation at R174, which is in close proximity to the solvable last C-terminal residue P321 [based on 8TAF.pdb (Benning et al., 2024)], yielded MWPNTs with various diameters (Fig. 4A). We further modified the C-terminal end of VP39 by appending a short peptide sequence (His 6 Tag-SpyTag), which was expected to emanate outwardly from the tube. The resulting PNTs mainly showed two-layer walls, with some discontinuous triple-walled regions (Fig. 4B-D).
A symmetry search failed to identify the specific helical symmetry of the double-walled PNT, and we therefore solved the structures of the inner tube (twist = 25.0 • , rise = 27.8 Å and C7) and outer tube (twist = 19.4 • , rise = 30.8 Å and C1) (Fig. 4E) using two different tubular masks during 3D refinement (Supplementary Fig. S7D). The inner tube (called A7, containing 7 p-filaments) at 6.6 Å resolution and the outer tube (called A10, containing 10 p-filaments) at 9.3 Å resolution both showed clear density boundaries for the p-filaments (Fig. 4E). The two concentric VP39 layers (R m(A7) = 160.9 Å and R m(A10) = 211.5 Å) were separated by a gap of ~10 Å (Fig. 4F). Because A7 and A10 are characterized by different helical parameters, we expected no specific radial inter-wall interactions in the double-walled PNTs, and the possible driving forces for lateral interactions may arise from charged patches on the inner or outer surface of the nanotubes (Supplementary Fig. S8).
Local refinement of the A7 nanotubes produced an improved density map at 3.4 Å resolution, which is sufficient for tracking the residues along the main chain. Compared with the A14 tubes, the most dramatic variations were located at the type III interface (Fig. 4G). The lasso (S256-L284) density was not visible, and the last C-terminal residues (E307-V347), which are stabilized by the lasso in A14, were also missing. Although local refinement of the A10 nanotubes at 3.8 Å was insufficient for accurate model building, A7 p-dimers could be well docked into the A10 density (Fig. 4H). There was also no density near the A10 type III interface (Supplementary Fig. S4E), suggesting that A7 and A10 shared highly conserved interactive networks for neighbouring VP39 subunits.
## Wrapping vectors of the rolled 2D lattices
Geometrically, given a planar 2D lattice, a tubular assembly can be achieved by rolling the lattice along a wrapping (or circumferential) vector (Tsai and Nussinov, 2011). The unit cell of the 2D lattice is defined by two edges (a and b) and the angle between the two edges (γ) (Fig. 5A). The integers n 1 and n 2 are the numbers of unit cells along edges a and b of the 2D lattice, respectively. The wrapping vector is a vector connecting two lattice points, and the two ends of the vector coincide when the tubular assembly successfully forms. Structural determination of the closely related PNTs enabled us to re-examine this conventional model using real biological assemblies. Between the two types of ASUs (p-dimers and s-dimers) used in 3D reconstructions, p-dimers are stable building blocks with a larger dimeric interface (Fig. 5B). For comparison, the 2D unit cell on the "unrolled" lattice approximates the space occupied by a p-dimer (Supplementary Fig. S9). The length of the wrapping vector is nearly equal to the real circumference of each nanotube (Supplementary Table S2), verifying that this "wrapping" strategy is mathematically reasonable.
According to the above interfacial analysis near the p2-related dyad axes, the PNTs can be categorized into three groups: groups A-C. Group A contains H14 and A14, which are derived from baculoviral nucleocapsids, showing balanced interactive networks at all four interfaces. Group B contains the H11 and H12 tubes, showing almost no type II interactions (less than 20 Å 2 is buried). Group C contains A7 and A10, showing no lasso-lasso (type III) interactions. The unrolled 2D lattices were calculated and mapped for the three groups (Fig. 5C-E). The unit cells of the group A PNTs, with the shortest |a|, the longest |b| and an angle γ of ~90 • were significantly different from those of the in vitro assembled group B and C tubes. Pairwise comparison between the two members of each group revealed that the accumulative subtle variations in the unit cells resulted in large variations in the final 2D lattices of groups A and C, while the lattices of H11 and H12 in group B had almost identical repeating unit cells. It is important to note that groups A, B and C also appeared to be distinguishable by the direction of the wrapping vector (Fig. 5D), indicating that the different connectivity of the subunits at the molecular level is correlated with the wrapping direction of the rollable 2D lattice.
## Development of novel double-layered nanosheets
To obtain a stable 2D platform for displaying protein molecules, we appended SpyTag to the C-terminus of HaCP with a flexible linker including His 6 Tag, based on the observation that C-terminal extension of the HaCP sequence produced only 2D lattices (HaCP-C-3G, HaCP-C-6G, and HaCP-C-9G mutants, Supplementary Fig. S3). Following a similar protocol to that for preparing the nanotubes, the resulting HaCP-CHS protein self-assembled into flexible sheets with dimensions of up to 100 μm, which could serve as a platform for displaying SpyCatchertagged enhanced green fluorescent protein (EGFP) (Fig. 6A). Further analysis using cryo-electron tomography (cryo-ET) revealed that the sheets were protein bilayers, with an inter-layer distance of ~5 nm (Fig. 6B-D).
Two strategies were explored to load foreign proteins onto the surface of the 2D nanosheets. The first strategy was to take advantage of the SpyTag-SpyCatcher system. As a proof of concept, Gc glycoprotein, a viral spike protein from Crimean-Congo haemorrhagic fever virus, fused to SpyCatcher could be readily covalently linked to the HaCP-CHS sheets (Fig. 6E). The second strategy was to prepare hybrid sheets by mixing genetically modified HaCP with HaCP-CHS during self-assembly of the 2D lattices. mCherry-HaCP, as a fusion protein, could be efficiently coassembled into the HaCP-CHS lattice, which could also serve as a scaffold for loading SpyCatcher-tagged EGFP (Fig. 6F). Using confocal fluorescence microscopy, we found that the size of the hybrid sheets was influenced by the ratio of the two building components: when more mCherry-HaCP molecules were introduced into the hybrid sheets, smaller 2D sheets were produced (Fig. 6F and Supplementary Fig. S4F). This suggested that the disruption of the inter-layer interactions by filling more mCherry between the two layers could block the expansion of the protein bilayer. By combining these two loading strategies, we can exploit size-controllable HaCP sheets as attractive building scaffolds for the in vitro presentation of foreign proteins, which would have broad applications.
## DISCUSSION
The structural comparison of the nanotubes and nanosheets in this study suggested that the basic building block (in this case, p-dimers) is structurally stable and the connectivity of the blocks in the nanotubes and nanosheets is closely related, which provides us with the opportunity to understand the organization of nanotube-derived 2D materials through structural studies of helical assemblies. Theoretically, at least three distinct contact regions are required for planar connectivity for the p2 symmetry (Wukovitz and Yeates, 1995). Consequently, the four contact regions at the local dyad axes play important roles in maintaining the connectivity of the rolled 2D lattice of protein nanotubes in a redundant way. Three contact regions are sufficient for assembling H11/H12 and A7/A10 nanotubes, but all four interfaces are fully used in H14/A14 assemblies. Our mutagenesis studies showed that the C-terminal region of HaCP or VP39 molecules at the intermolecular interface provided mutational hotspots (e.g., HaCP-R293C, HaCP-CHS and VP39-R174A-CHS) for generating various nanotubes or nanosheets.
Formation of a new nanotube can be generally divided into two steps: expansion of the 2D lattice consisting of the basic unit cells and wrapping of this 2D lattice in a specific direction. These two steps are inseparable in time and space. It is a simplified strategy for creating all possible initial tubular models by rolling a 2D lattice along wrapping vectors defined by two variable integers n 1 and n 2 , assuming that all of the unit cells are identical for the PNTs formed under similar conditions ( Jenni et al., 2022;Benning et al., 2024). This strategy is suitable for some cases, such as for H11 and H12 (Fig. 5D), in cases where no other molecular forces influence the growth of the nanotubes. However, in MWPNTs, structural interference from the proteins in other layers would slightly change the unit cell (Fig. 5E). The wrapping of 2D lattices made of multiple p-filaments (large n 1 values) would exaggerate the overall difference in the distribution of lattice points by accumulative local variations in the unit cells.
When the two ends of the wrapping vector are not superimposed, the 2D lattice (e.g., HaCP-R293C and HaCP-CHS mutants) further expands into a spreading sheet instead of forming a cylindrical tube. The intrinsic structural flexibility of HaCP lattices, arising from p2-symmetry related redundancy or variable unit cells as discussed above, may help to maintain the connectivity of unrolled 2D lattices. In addition, doublelayered HaCP-CHS nanosheets are probably stabilized by electrostatic interactions between adjacent layers, similar to the stabilization in in vitro assembled multiple-walled VP39 nanotubes (Supplementary Fig. S8). In contrast to the length control of TMV particles, which is achieved by taking advantage of the specific length of RNA (Eber et al., 2015), size control of 2D nanosheets is usually unachievable. As an unexpected approach to solve this problem, double-layered HaCP nanosheets can be size-controllably assembled by filling the layer-to-layer space with protein fragments.
The high-level assembly of small structural modules through specific assembly mechanisms is a common way to obtain high-molecularweight materials with different properties. Such structural modules can be atoms or molecules. One of the most widely used types of singleatom assemblies is carbon-atom-based nanomaterials, such as singlelayered sheets (graphene), multi-layered sheets (graphite), singlewalled carbon nanotubes and multi-walled carbon nanotubes (Speranza, 2021). Compared with carbon-based nanomaterials, protein sheets or nanotubes described in this study are based on molecule-scale building blocks with complex loading surfaces. The easier biodegradation of biomaterials means that these protein-based materials are much less toxic to higher organisms (Audette et al., 2019). We expect that more biological modules derived from tubular capsids with intrinsic flexibility could be developed as versatile platforms for different biocompatible applications in the future.
## CONCLUSIONS
Our high-resolution cryo-EM study demonstrated an important relationship between the polymorphism of protein nanotubes and the redundant p2-symmetrical interactive network. From an application perspective, molecules derived from flexible viral nanotubes have great potential to serve as a starting material for the development of novel functionalized 2D protein lattices.
## MATERIALS AND METHODS
## Gene cloning
The full-length HaCP (GenBank AAG53821) and VP39 (GenBank NP_054119) genes were codon-optimized for the E. coli expression system (Sangon). The sequences were subcloned into pGEX-6P-1 (Cytiva), and the resulting fusion protein products contained a GST tag at the N-terminus, a 3C cleavage site, a flexible glycine-serine (GS) linker, and an optimized HaCP/VP39 sequence at the C-terminus. Based on these two expression plasmids, more than 200 other expression vectors containing deletions, insertions or substitutions were then constructed (Supplementary Table S3). The expression plasmid of SpyCatchertagged Gc (Gc-Catcher) was constructed by inserting the ectodomain of Gc from Crimean-Congo haemorrhagic fever virus (Li, N. et al., 2022) and SpyCatcher (Li, L. et al., 2014) at either end of the V5-tag in the pMT/BiP/V5-His vector (Invitrogen). The expression plasmid of SpyCatcher-tagged EGFP (EGFP-Catcher) was constructed by subcloning the fusion sequence of EGFP and SpyCatcher into pGEX-6P-1.
## Preparation of the HearNPV nucleocapsids
Fourth instar larvae of Helicoverpa armigera were fed with 5 μL of the occlusion bodies (OBs) of HearNPV (1 × 10 8 OBs/mL). After 5-6 days of infection, the liquefied larval cadavers were ground with 0.01% (w/v) sodium dodecyl sulphate (SDS). After centrifugation at 3000×g for 5 min, the pellet was resuspended with 0.01% SDS and incubated at 37 • C for 30 min. To remove the larval debris, the suspension was filtered with cotton gauze, and the filtrate was centrifuged at 20×g for 30 min. The OBs in the supernatant were pelleted at 3000×g for 30 min and dissolved in an alkaline buffer containing 100 mM Na 2 CO 3 , 150 mM NaCl, and 10 mM EDTA (pH 11.0) at 25 • C for 5 min, followed by neutralization with 1/10 the volume of 500 mM Tris-HCl (pH 7.5). After removing the non-dissolved debris by centrifugation at 3000×g for 5 min, the occlusion-derived viruses (ODVs) were purified by 30%-65% (w/w) sucrose density gradient ultracentrifugation at 20,000×g for 1 h at 4 • C. To release the nucleocapsids, the band containing the ODVs was collected and then treated with 1% (v/v) NP-40 for 30 min on ice. The HearNPV nucleocapsids were pelleted by centrifugation at 20,000×g for 30 min at 4 • C and resuspended with 1 × phosphate-buffered saline (PBS, pH 7.4).
## Preparation of the HaCP and VP39 nanomaterials
The in vitro assembled nanotubes and nanosheets were prepared following a previously described protocol for HaCP PNTs with modification (Rao et al., 2018). The E. coli BL21 (DE3) cells were transformed with recombinant expression plasmids and grown at 37 • C in Luria-Bertani medium until the optical density at 600 nm reached ~0.8. The cells were induced by 0.5 mM isopropyl β-D-thiogalactopyranoside (IPTG) for 20 h at 16 • C and then harvested by centrifugation at 4000×g for 30 min at 4 • C. The cell pellets were washed once with 1 × PBS and then resuspended in an optimized assembly buffer containing 1 × PBS (pH 7.4), 5% (v/v) glycerol, 0.01% Triton X-100, and 5 mM dithiothreitol. The cells were disrupted by sonication, and then the lysates were centrifuged at 12,000×g for 30 min at 4 • C. The supernatant was collected and filtered through a 0.45-μm filtration unit (Millipore). The soluble fraction was applied a gravity column containing glutathione agarose resin (GE Healthcare), and the resin was washed with the assembly buffer. On-column cleavage of the fusion protein with GST-3C protease (a fusion protein with the GST tag fused to the C-terminal end of human rhinovirus B14 strain 3C protease, prepared in our laboratory) yielded HaCP or VP39 derivatives, which were self-assembled into nanotubes or nanosheets at 4 • C overnight. The nanomaterials were eluted by 2 bed-volumes of the assembly buffer and then concentrated to 6 mL/ml using Amicon filter units (MWCO 100 kDa). The nanosheets and their stacks were pelleted by centrifugation at 30-130×g for 1-2 min at 4 • C, while the nanotubes were pelleted by centrifugation at 2000-4500×g for 1-2 min at 4 • C, before being resuspended in 1 × PBS. The nanomaterials were washed another three times with 1 × PBS, and the final concentration was ~5 mL/mL.
## Loading foreign molecular modules onto the hybrid nanosheets
The hybrid nanosheets were prepared with different molar ratios of mixed HaCP-CHS and mCherry-HaCP before 3C digestion. Gc-Catcher was purified from the supernatant derived from the Drosophila expression system by a Ni affinity column. EGFP-Catcher was expressed in E. coli and purified by a GST affinity column (GST tag was removed by oncolumn cleavage). The HaCP-CHS nanosheets or hybrid nanosheets were mixed with excess SpyCatcher-labelled molecules (EGFP-Catcher or Gc-Catcher) for 30 min, followed by the pelleting process, as described above for purifying the nanosheets.
## Negative staining and transmission electron microscopy (TEM)
For each HaCP-or VP39-derived nanomaterial, 3 μL of the purified sample at a concentration of 0.5 mL/mL was loaded on a glowdischarged copper grid and allowed to adsorb on the continuous carbon film for 30 s. Excess solution was removed by blotting using dustfree paper, and the grid was rinsed with a few drops of doubledistilled H 2 O. The grid was then stained with 10 μL of 2% (w/v) uranyl acetate solution for 10 s, and the excess stain was removed by blotting. This staining process was repeated three times. Each grid was air-dried for 1 h before imaging by TEM using a Talos L120C electron microscope equipped with a 4k × 4k Ceta CMOS camera (Thermo Fisher Scientific). size of 1.728 Å. The acquisition covered the tilt range from -60 • to +60 • with a tilt increment of 3 • , using a dose-symmetrical tilt scheme facilitated by the PACEtomo.py (Eisenstein et al., 2023) script in SerialEM.
The defocus was set between 3 and 5 μm. A sequence of ten frames was recorded at each tilt angle, resulting in a total cumulative dosage of ~120 e -/Å 2 for each tilt series.
## Cryo-ET data processing and subtomogram averaging
The frame alignment and CTF estimation of the raw movies were performed by Warp (Tegunov and Cramer, 2019). The assembled tilt series were aligned based on the gold fiducial markers using autoa-lign_dynamo (Burt et al., 2021), and the full tomograms were reconstructed at a pixel size of 10 Å by Warp. The tomographs were imported into Dynamo (Scaramuzza and Castano-Diez, 2021) for particle selection. A total of 3794 particles were selected for the HaCP-CHS lattices. The sub-volumes were extracted with a 32-pixel box size in Warp, and they were then imported into RELION for sub-tomogram averaging with a final resolution of ~20 Å. The signals of the double-layered nanosheets in the tomograms were enhanced using the Morphological Gradient module in Amira 3D (Thermo Fisher Scientific).
## Structural analysis and figure preparation
The unwrapping of the nanotubular surface into a planar 2D lattice was performed using the program e2unwrap3d.py (Ludtke, 2016). The stripe pattern in the cryo-EM micrographs was extracted and enhanced using ImageJ (Schneider et al., 2012), as outlined in Supplementary Fig. S5. The 2D projection of the p-dimer or s-dimer was performed in cryoSPARC, after the C2 symmetry axis of the dimeric models was aligned along the z axis by relion_align_symmetry. The interfacial analysis was performed using PDBePISA (Krissinel and Henrick, 2007). The figures were prepared using ChimeraX (Goddard et al., 2018) and ImageJ.
## References
1. Afonine, Poon, Read et al. (2018) "Real-space refinement in PHENIX for cryo-EM and crystallography" *Acta Crystallogr D Struct Biol*
2. Audette, Yaseen, Bragagnolo et al. (2019) "Protein nanotubes: from bionanotech towards medical applications" *Biomedicines*
3. Baneyx, Matthaei (2014) "Self-assembled two-dimensional protein arrays in bionanotechnology: from S-layers to designed lattices" *Curr. Opin. Biotechnol*
4. Ben-Sasson, Watson, Sheffler et al. (2021) "Design of biologically active binary protein 2D materials" *Nature*
5. Benning, Jenni, Garcia et al. (2024) "Helical reconstruction of VP39 reveals principles for baculovirus nucleocapsid assembly" *Nat. Commun*
6. Burt, Gaifas, Dendooven et al. (2021) "A flexible framework for multiparticle refinement in cryo-electron tomography" *PLoS Biol*
7. Chen, Arendall 3rd, Headd et al. (2010) "MolProbity: all-atom structure validation for macromolecular crystallography" *Acta Crystallogr D Biol Crystallogr*
8. Eber, Eiben, Jeske et al. (2015) "RNA-controlled assembly of tobacco mosaic virus-derived complex structures: from nanoboomerangs to tetrapods" *Nanoscale*
9. Eisenstein, Yanagisawa, Kashihara et al. (2023) "Parallel cryo electron tomography on in situ lamellae" *Nat. Methods*
10. Emsley, Lohkamp, Scott et al. (2010) "Features and development of coot" *Acta Crystallogr D Biol Crystallogr*
11. Fagan, Fairweather (2014) "Biogenesis and functions of bacterial S-layers" *Nat. Rev. Microbiol*
12. Goddard, Huang, Meng et al. (2018) "UCSF ChimeraX: meeting modern challenges in visualization and analysis" *Protein Sci*
13. Gonen, Dimaio, Gonen et al. (2015) "Design of ordered two-dimensional arrays mediated by noncovalent protein-protein interfaces" *Science*
14. Hamley (2019) "Protein assemblies: nature-inspired and designed nanostructures" *Biomacromolecules*
15. Hatlem, Trunk, Linke et al. (2019) "Catching a SPY: using the SpyCatcher-SpyTag and related systems for labeling and localizing bacterial proteins" *Int. J. Mol. Sci*
16. He, Scheres (2017) "Helical reconstruction in RELION" *J. Struct. Biol*
17. Henderson, Unwin (1975) "Three-dimensional model of purple membrane obtained by electron microscopy" *Nature*
18. Jenni, Horwitz, Bloyet et al. (2022) "Visualizing molecular interactions that determine assembly of a bullet-shaped vesicular stomatitis virus particle" *Nat. Commun*
19. Jia, Gao, Huang et al. (2023) "Architecture of the baculovirus nucleocapsid revealed by cryo-EM" *Nat. Commun*
20. Klug (1999) "The tobacco mosaic virus particle: structure and assembly" *Philos. Trans. R. Soc. Lond. B Biol. Sci*
21. Koch, Eber, Azucena et al. (2016) "Novel roles for wellknown players: from tobacco mosaic virus pests to enzymatically active assemblies" *Beilstein J. Nanotechnol*
22. Krissinel, Henrick (2007) "Inference of macromolecular assemblies from crystalline state" *J. Mol. Biol*
23. Tian (2025) *Virologica Sinica*
24. Li, Fierer, Rapoport et al. (2014) "Structural analysis and optimization of the covalent association between SpyCatcher and a peptide tag" *J. Mol. Biol*
25. Li, Rao, Li et al. (2022) "Cryo-EM structure of glycoprotein C from Crimean-Congo hemorrhagic fever virus" *Virol. Sin*
26. Liu, Qiao, Niu et al. (2012) "Natural supramolecular building blocks: from virus coat proteins to viral nanoparticles" *Chem. Soc. Rev*
27. Ludtke (2016) "Single-particle refinement and variability analysis in EMAN2.1" *Methods Enzymol*
28. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J. Struct. Biol*
29. Pettersen, Goddard, Huang et al. (2004) "UCSF chimera -a visualization system for exploratory research and analysis" *J. Comput. Chem*
30. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat. Methods*
31. Rao, Fu, Li et al. (2018) "Controllable assembly of flexible protein nanotubes for loading multifunctional modules" *ACS Appl. Mater. Interfaces*
32. Scaramuzza, Castano-Diez (2021) "Step-by-step guide to efficient subtomogram averaging of virus-like particles with dynamo" *PLoS Biol*
33. Schneider, Rasband, Eliceiri (2012) "NIH image to ImageJ: 25 years of image analysis" *Nat. Methods*
34. Sleytr, Schuster, Egelseer et al. (2014) "S-layers: principles and applications" *FEMS Microbiol. Rev*
35. Speranza (2021) "Carbon nanomaterials: synthesis, functionalization and sensing applications" *Nanomaterials*
36. Suzuki, Cardone, Restrepo et al. (2016) "Self-assembly of coherently dynamic, auxetic, two-dimensional protein crystals" *Nature*
37. Tegunov, Cramer (2019) "Real-time cryo-electron microscopy data preprocessing with warp" *Nat. Methods*
38. Tsai, Nussinov (2011) "A unified convention for biological assemblies with helical symmetry" *Acta Crystallogr D Biol Crystallogr*
39. Wen, Steinmetz (2016) "Design of virus-based nanomaterials for medicine, biotechnology, and energy" *Chem. Soc. Rev*
40. Wukovitz, Yeates (1995) "Why protein crystals favour some space-groups over others" *Nat. Struct. Biol*
41. Yeates (2017) "Geometric principles for designing highly symmetric self-assembling protein nanomaterials" *Annu. Rev. Biophys*
42. Zhang, Alberstein, De Yoreo et al. (2020) "Assembly of a patchy protein into variable 2D lattices via tunable multiscale interactions" *Nat. Commun*
43. Zhou, Si, Ge et al. (2022) "Atomic model of vesicular stomatitis virus and mechanism of assembly" *Nat. Commun*
44. Zivanov, Nakane, Forsberg et al. (2018) "New tools for automated high-resolution cryo-EM structure"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12656724&blobtype=pdf
|
# Leveraging Classical Virology and High Throughput Sequencing for Viral Discovery Using a Historical Viral Collection
Mark Sistrom, Matthew Neave, Ancy Joseph, Kim Newberry, Hannah Andrews, Cathy Shilton, Vidya Bhardwaj, Richard Weir
## Abstract
Northern Australia has long been a hotbed of arboviral discovery, and collections of viral isolates from Northern Australia represent an invaluable resource for both our knowledge of viral diversity and for disease preparedness and treatment. While discovery of novel viruses via classical virology methods is on the decline, next generation sequencing offers the possibility to speed up viral discovery, albeit at the expense of the collection of valuable life history data. By sequencing unknown isolates from historical viral collections, we may leverage both the rich data collected through classical virology and the power of identification using contemporary sequencing technologies. In the present study, we sequenced 76 historical viral isolates held at the Berrimah Veterinary Laboratory in Darwin, northern Australia, for which serological typing had yielded ambiguous results. We determined that 43 of these isolates belong to the genera Hapavirus, Orbivirus, and Orthobunyavirus. Several of these isolates are putatively novel genotypes or potential taxa, which has significant potential implications for human and animal health. This study demonstrates the utility of historical collections for viral discovery and characterisation and how considerable past efforts to isolate and characterise viruses can be enhanced using next generation sequencing approaches.
## 1. Introduction
Arbovirus discovery has a rich and storied history [1] and has played a critical role in the documentation and development of knowledge of viral diversity broadly speaking. While there has been a decline in the discovery of novel viruses via classical methods of virology [1], the development of high throughput sequencing has facilitated a suite of new methods for the identification and characterisation of novel viruses. While the pace of viral discovery offered by next generation sequencing techniques is unequivocally more rapid than classical techniques [2] and often allows for the ascertainment of genomic data for novel viruses, it often negates the ability to collect important biological parameters facilitated by the isolation of novel viruses that classical virological techniques provide.
One resource that has the potential to leverage the power of high throughput sequencing and classical virology is historical virus isolate collections [3]. Sequencing of virus isolates held in historical collections can allow for the linkage of critical serological and biological data with genome sequences, allow for the identification of type isolates for viruses discovered in sequencing surveys, prevent redundancy and duplicated effort in the systematic characterisation of viruses, and allow for the identification of unknown viruses isolated in the past-potentially shedding light on historical epidemiological and disease events and allowing for improved prediction of, and responses to, future outbreaks.
A growing body of research is demonstrating a link between climate change, changes in trading patterns of animals and animal products, and the expansion in geographic ranges of several pathogenic disease threats [4]. In particular, the recent spread of formerly obscure arboviral pathogens such as Chikungunya and Zika virus highlights the importance of viral discovery and taxonomy in order to rapidly identify the etiological agents and mount responses to emergent disease outbreaks [5]. This is especially relevant to Australia, where a number of undifferentiated febrile illnesses go undiagnosed [6] and a number of poorly understood arboviruses (e.g., Edge Hill, Kokobera, Alfuy, Sinbis, and Stratford viruses) are known to cause human infection, but diagnostic tests are not readily available [7]. It stands to reason that: (1) a number of the undiagnosed febrile illnesses are caused by undiscovered arboviruses and (2) a considerable diversity of potentially human infectious arboviruses are yet to be discovered in the Australasian region.
Northern Australia has been a historical hotbed of virus discovery, with over 80 novel RNA viruses from 9 families and 16 genera reported from Australia and Papua New Guinea, largely driven by research efforts in Northern Australia [1]. A particular focus of viral diversity for Northern Australia has been arboviral taxa-often driven by outbreaks of both human and zoonotic encephalitis [1]. Unlike other arboviral hotspots such as the Amazon Basin and Southeast Asia, which tend to be endemic areas for widely distributed arboviruses of high clinical health and veterinary importance such as Zika, West Nile, Dengue, Chikungunya, and Yellow Fever [8,9], Northern Australia tends towards a higher number of unique arboviruses, thought to be due to the monsoonal wet-dry tropical climate that supports a range of distinct host and vector species [10,11]. A considerable proportion of these viruses remain insufficiently characterised with respect to their potential for human infectivity and subsequent risk of zoonotic spill-over [6].
Two of the most diverse arboviral genera in Northern Australia are Orthobunyavirus and Orbivirus [11]. Orthobunyavirus is a large genus (~170 species) of the family Peribunyaviridae, a family of negative sense, single stranded RNA viruses characterised by tripartite genomes and spherical, enveloped capsids [12]. They are transmitted by arthropod vectors, with amplification cycles in a diversity of vertebrate hosts. Orthobunyaviruses are largely transmitted by mosquitoes; however, some have been isolated from tabanoids [13], phlebotomines [14], ticks [15], and bedbugs [16]. A number of Orthobunyaviruses are known to cause human infections, including La Crosse virus in the United States [17], members of the Bunyamwera serogroup in sub-Saharan Africa and the Americas [18], and Oropuche orthobunyavirus in South America. Akabane serogroup orthobunyaviruses are a significant cause of livestock disease in Australasia, the Middle East and sub-Saharan Africa [19], and are vectored by biting midges of the genus Culicoides [20]. Orthobunyaviruses generally cause acute febrile illness and encephalitis when presenting in human infections [21]. The taxonomy of the Orthobunyavirises is complex and remains fluid as serotyping has led to incomplete characterisation of viruses within the genus [12].
The genus Orbivirus is within the Reoviridae family, comprising 22 serogroups and several unclassified isolates [22]. Orbiviruses have a double stranded RNA genome comprising 10 segments encapsulated by a non-enveloped, architecturally complex icosahedral capsid [23]. Orbiviruses are transmitted by ticks [24] and various haematophagus insect vectors (e.g., Culicoides midges [25], mosquitoes [26], and sandflies) and have a wide host range, including bovids, equids, camelids, marsupials, sloths, bats, birds, canines, felines, and humans [27]. There are three economically important animal diseases in the Orbivirus genus-Bluetongue virus, African horse sickness virus, and epizootic haemorrhagic disease virus [28]. They produce a spectrum of disease ranging from subclinical infection to high morbidity and mortality haemorrhagic fevers [29].
In the present study, we undertook high throughput sequencing of 76 virus isolates isolated between 1982 and 2004 for which serological testing at the time of isolation yielded uncertain results. The 76 isolates were selected based on serological ambiguity in conjunction with considerations of host range and geographic distribution, to ensure broad representation across potential sources of viral diversity. Isolates were held in the viral isolate collection at the Northern Territory government's Berrimah Veterinary Laboratory and the Australian Centre for Disease Preparedness (ACDP).
## 2. Materials and Methods
## 2.1. Isolation of Viruses from Mosquitoes
Mosquitoes were trapped using encephalitis virus surveillance CO 2 baited light traps [30]. Collections were transferred to the entomology laboratory for speciation and separation of gravid specimens. Mosquitoes were manually sorted into monospecific pools of not more than 50 individuals, and stored in liquid nitrogen prior to transfer to the virology laboratory for virus isolation and identification.
Isolation methods between 1981 and 1991 were as described [31]. Briefly, monospecific pools of mosquitoes were removed from an ultracold (-70 • C) freezer and homogenised in a chilled mortar and pestle containing a small quantity of sterile fine grain sand and 5 mL of brain heart infusion broth containing antibiotics penicillin (6 mg/mL), streptomycin (20 mg/mL) and amphotericin B (2.5 mg/mL). The resulting suspension was transferred to a 10 mL polypropylene centrifuge tube and held overnight at 4 • C. The following day the homogenates were clarified by centrifugation (2000× g, 4 • C, 15 min) and inoculated to two baby hamster kidney clone 21 (BHK-21) [32] cell cultures. This method was followed from 1981 to 1984. Between 1985 and 1992, two Aedes albopictus (C6/36) [33] stationary cell cultures growing in tubes were used as the first passage amplification step. First passage tubes were incubated for seven days before pooling and passage to two BHK-21 and two hamster lung (HmLu-1) [34] cell culture tubes. Second passage tubes were incubated for seven days at 37 • C before pooling and final passage to one BHK-21 and one HmLu-1. Cell culture tubes were inspected for the presence of cytopathic effect (CPE) from day three to day seven during the second and third passages.
In 1992, Kontes pellet pestles replaced mortars and pestles. A small quantity of lapidary grinding powder was added to each microcentrifuge tube, followed by 200 µL of brain heart infusion broth containing penicillin G (6 mg/mL), streptomycin sulphate (20 mg/mL), and amphotericin B (2.5 mg/mL). The tubes were returned to 4 • C and held prior to use. From one to 50 monospecific mosquitoes were added to each 2 mL tube containing brain heart infusion broth and lapidary grinding powder. The mosquitoes were homogenised using a plastic disposable pestle attached to either a Kontes grinding motor or a laboratory homogeniser. The homogenised mosquitoes were transferred to a 10 mL centrifuge tube containing 5 mL of brain heart infusion broth and antibiotics and held overnight at 4 • C (to restrict contamination of cell cultures) prior to clarification by centrifugation (2000× g for 15 min at 4 • C). Two confluent C6/36 cell culture tubes containing 2 mL of minimum essential medium and 10% foetal bovine serum and mosquito homogenate were incubated stationary at 25 • C for seven days. On day seven, tubes were inspected for fungal or bacterial contamination, and placed in an ultrasonic cleaning bath for 60 min. Each pair of C6/36 tubes was pooled together before passage to two BHK-21 and two HmLu-1 cell culture tubes. Contaminated cultures were processed last and filtered through a 0.22 µm sterile membrane filter (Sartorius Minisart Plus Cat. 178 23K or Millipore Millex-GP (Burlington, MA, USA)) before inoculation. The BHK-21 and HmLu-1 cultures were inspected for CPE from day three to day seven. At passage two on day seven, CPE negative cultures were pooled and inoculated to single BHK-21 and HmLu-1 cell culture tubes. Cultures were inspected for the presence of CPE for seven days. CPE positive cultures were removed, and CPE negative cultures were discarded. Seed virus stocks were prepared from all CPE positive cultures by inoculation of six tubes of BHK-21 or HmLu-1, or if both cell lines were CPE-positive, BHK-21 was used to prepare the seed virus stock. Seed virus stock CPE was allowed to progress to 80-90% before harvesting, pooling, stabilisation with 1% bovine serum albumin, and storage at 4 • C and -70 • C. Storage of some unidentified viruses, particularly Orbiviruses, proved superior at 4 • C.
## 2.2. Virus Isolation from Cattle Blood
During 1979, permanent sentinel cattle herds were established at Coastal Plains Research Station (lat. 12 • 39 ′ S, long. 131 • 20 ′ E), Tortilla Flats Research Farm (lat. 13 • 05 ′ S long. 131 • 14 ′ E) and Berrimah Research Farm (lat. 12 • 26 ′ S, long. 130 • 55 ′ E). The cattle were bled weekly for virus isolation and monthly for retrospective serology. Each year, the herds were replaced with animals that were seronegative to bluetongue, epizootic haemorrhagic disease, Palyam virus, bovine ephemeral fever, and Simbu group viruses. Total herd size and composition have varied considerably. Over the years, there are generally forty to fifty cattle comprising an equal number of cows and steers.
Each heparinised blood sample was inoculated to each of three embryonated chicken eggs for virus isolation as previously described [35,36]. Briefly, heparinised blood samples were chilled immediately upon collection and transported to the laboratory and held at 4 • C overnight. A 50 µL sample of blood was aspirated from the sealed tube and lysed with 450 µL of sterile distilled water. The lysate was used as inoculum from three-, nine-, to eleven-day-old embryonated chicken eggs. Each egg received a 100 µL intravenous inoculation of blood lysate using a 1 mL Tuberculin syringe and 29 ga needle. Embryos dying during the first 24 h were recorded and discarded. Embryos dying from 24 to 120 h post inoculation were recorded and held at 4 • C prior to harvesting. At 120 h post inoculation, all remaining embryos were euthanised by placing them at -20 • C for 30 min (eggs were not frozen). Embryos were aseptically harvested, and those receiving the same inoculum and dying at about the same time were pooled.
Embryos, minus the head, were then homogenised in brain heart infusion broth containing penicillin (5 mg), streptomycin (3 mg) and amphotericin B (12 µg). The resulting homogenate was transferred to a 10 mL polypropylene centrifuge tube and centrifuged at 2000× g for 10 min. All homogenates were held at 4 • C overnight prior to inoculation to cell culture tubes.
In 1993, a less labour-intensive method was derived from a method used by the University of Western Australia [37]. Duplicate wells of a 96 well microtitre plate containing C6/36 mosquito cell cultures were used as a first passage, followed by two mammalian cell culture passages each comprising six cell lines. In the first passage, 15 µL of each sample (egg or mosquito homogenate) was inoculated into duplicate wells containing C6/36 mosquito cell cultures (seeding rate 2 × 10 5 /mL) and 150 µL of minimum essential medium growth medium in a 96 well flat bottomed microtitre plate. Plates were incubated at ambient room temperature (25 • C) in a humidified container for seven days. In the second passage, 15 µL of first passage supernatant was inoculated to identical wells of seven plates, each containing a different cell type. The wells of column A1-H1 were mixed, and 15 µL was transferred to the duplicate column A2-H2, mixed, and the procedure reversed. On completion of mixing, 15 µL was transferred to identical rows in each of seven cell culture plates, each containing a single cell type as follows: (1) BSR clone of BHK-21 cells [38] grown in basal medium Eagle's growth medium with Earle's salts, supplemented with 10% foetal bovine serum. (2) BHK 21 cells grown in basal Eagle's growth medium.
(3) Hamster lung cells grown in minimum essential medium growth medium. (4) Vero cells derived from African green monkey kidney cells grown in Medium 199 supplemented with 10% foetal bovine serum (Medium 199 growth medium). ( 5) Porcine stable equine kidney cells grown in minimum essential medium growth medium. ( 6) Calf pulmonary artery endothelial cells [39] grown in minimum essential medium growth medium. (7) C6/36 cells grown in minimum essential medium growth medium. At 80-100% CPE, the supernatant was aseptically removed from the CPE positive wells and inoculated to 25 cm 2 tissue culture flasks for the production of seed stock virus. Inoculation of the third cell culture passage used the C6/36 s passage plates.
Each supernatant (150 or 300 µL) from CPE positive well(s) was aseptically removed from the infected wells and diluted 1:100. The diluted supernatant (virus) was then adsorbed in a BSR confluent 25 cm 2 tissue culture flask containing 3 mL of basal medium Eagle's supplemented with 5% heat inactivated foetal bovine serum for one hour. A further 8 mL of basal medium Eagle's maintenance medium was added to the flask following adsorption. Generally, CPE was present in BSR cells, and this was the cell line of choice for the production of seed stock virus. Tissue culture supernatant was removed at 90% CPE, stabilised using 1% bovine serum albumin, aliquoted, and stored at -70 • C.
## 2.3. Virus Identification: Serogrouping of Virus Isolates
Mosquito viruses isolated from 1982 to 1992 were initially screened in a plaque reduction neutralisation test [40]. Briefly, 24 well cluster plates containing BHK-21 or BSR monolayers were overlayed with 250 µL of a homologous antibody of known titer in 4 wells down the plate. Each column contained a different antibody, and 4 viruses could be screened in a single plate. A single virus was inoculated per column and allowed to incubate with the antibody for 1-2 h. After the incubation period, a 2% agarose solution was added and mixed thoroughly and allowed to cool. Polyvalent hyperimmune antibodies prepared in rabbits to members of the bovine ephemeral fever, bluetongue, epizootic haemorrhagic disease, Palyam, Simbu, alphavirus, and flavivirus virus groups were used in a grouping plaque reduction neutralisation test. Isolates not neutralised were then screened against a panel of antibodies using a modification of the indirect fluorescent antibody method described by [41,42].
Further serology to group isolates using a comprehensive IFA panel-a panel of antisera provided by Charlie Calisher from CDC Fort Collins-allowed us to test for ~400 agents. Preliminary identification of virus isolates was conducted using serological grouping assays applied to cultured virus, rather than to host sera. All virus isolates were grown as adherent monolayers in tissue culture flasks; cells were only briefly placed in suspension during trypsinisation prior to preparation of cell suspensions for the IFA. Isolates were screened by plaque reduction neutralisation tests (PRNT) against panels of polyclonal hyperimmune antisera [40] targeting major arbovirus groups (e.g., flaviviruses, alphaviruses, orbiviruses). Isolates not neutralised were further screened using an indirect immunofluorescent antibody assay (IFA) with a comprehensive reference antisera panel [43].
In this context, serological assays were used to assess antigenic relationships between isolates and reference viruses, providing a first level of taxonomic classification prior to sequencing.
## 2.4. Virus Identification: Indirect Immunofluorescent Assay
A 25 cm 2 tissue culture flask was used for each unidentified virus and infected with 100 TCID50. Cytopathic effect was allowed to progress to 50%. At 50% (2+) CPE, supernatant was removed, and the cells remaining in the flask were trypsinized from the flask and combined with the decanted supernatant. Trypsinising the remaining cells from the flask had no significant effect on the outcome of the indirect fluorescent antibody test. It did, however, save a considerable amount of time (days) waiting for cells to slough off the flask. The suspension was centrifuged at 2000× g for 15 min at 4 • C, and the pellet was resuspended in 4 mL of phosphate buffered saline containing 5% foetal bovine serum. Spot slides were prepared essentially as described by [42]. Briefly, 30 µL of cell suspension was added to each spot of ten, twelve spot-slides. The slides were air-dried and then immersed and fixed in cold acetone for 15 min. The slides were removed from acetone and air-dried prior to rinsing in phosphate buffered saline (to remove crystalline deposits) and acetone (to aid drying) before storage at -20 • C. It was realised that trypsinisation can alter or remove antigens of some viruses. Alternatively, if CPE was allowed to continue until the majority of cells were dislodged, identification of some viruses becomes difficult as the antigen concentration falls rapidly once CPE is advanced. Comparisons were made, and this method was determined to be a good compromise.
Group-or type-specific antibody (List 1) was diluted 1:25 and added to a prerecorded spot on each slide for each unidentified virus. All slides were incubated in a humidified container (150 mm disposable Petri dishes containing damp paper towels) at 37 • C for 1 h. After incubation, the slides were rinsed in phosphate buffered saline and then washed for 15 min in phosphate buffered saline. The slides were air dried and 30 µL of reconstituted anti-species fluorescein isothiocyanate conjugate (goat-anti-mouse fluorescein isothiocyanate, Cappel Cat. 55496) containing 0.4% trypan blue was added to each spot. The slides were returned to the humidified container and placed at 37 • C for 1 h. After incubation the slides were washed as previously described, and were not allowed to dry. A bead (~100 µL) of Aquamount mountant (BDH Cat. 36086) was laid in a strip down the centre of each wet slide, and a large coverslip (60 × 24 mm) was placed over the slide and air bubbles removed.
Fluorescence was visualised using an Olympus fluorescent microscope and subjectively estimated from 0 (negative) to 4 + (strong positive) and recorded for each spot. Spots recorded as 2+ or greater were considered positive.
## 2.5. Sequencing and Bioinformatics Analysis
RNA was extracted using an RNeasy Plus Mini Kit (Qiagen, Gernamy). Library preparation was undertaken using the Illumina Stranded Total RNA Prep, Ligation with Ribo-Zero Plus kit as per the manufacturer's recommendations. Isolates were sequenced at the Australian Centre for Disease Preparedness (ACDP) using an Illumina NextSeq2000 platform. Sequences were adapter trimmed and quality filtered using Trimmomatic v0.4.0 [44] and assembled using SPAdes v3.15.5 [45] at k = 31, 55, 75, 95, and 127 and otherwise default settings. Contigs were compared with the Rdrp-scan database [46] using the Blastx function of Diamond v2.1.8.162 [47]. Genomes with significant hits (e > 10 -5 ) were used as references for alignment using BWA [48] on trimmed reads for confirmation of sample identity.
Alignments of the Sedoreoviriade RdRp gene [22] and Peribunyaviridae L segment sequence [12] were downloaded from the International Committee on the Taxonomy of Viruses (ICTV) website. Consensus fasta files of aligned viral isolates were produced using Samtools [49], and an alignment of samples generated in this study was aligned with the datasets downloaded from ICTV using MAFFT [50] using the E-INS-I algorithm. The most appropriate substitution model for the Sederoreoviridae and Peribunyaviridae alignments was determined using jModelTest [51], and phylogenies were generated using MrBayes v3.2.7 [52] using 4 heated chains and 1,100,000 generations sampled every 200 generations. The first 10% of trees were discarded as burn in and run parameters were evaluated using Tracer v1.7.1 [53].
## 3. Results
For many of the isolates, inadequate nucleic acid could be recovered to perform effective sequencing. Resultantly, we obtained de novo assemblies and RdRp blast matches for 43 of the 76 isolates initially selected for sequencing (Table 1). In these, we had provisional serotype identifications for 30 isolates. Viruses were isolated between 1982 and 2004, with two samples isolated from Aedes lineatopennis, three from Aedes (Och) normanensis, one from Anopheles annulipes, two from Anopheles farauti, two from Anopheles amictus, six from Anopheles meraukensis, and nine from Culex annulirostris. Two samples were isolated from pooled mosquitoes and two from unnamed insects. The remainder were isolated from mammalian hosts, including ten from Bos indicus, one from Bubalus bubalis, two from Equus caballus, and one from Osphranter rufus. Samples were collected primarily from regions within the Northern Territory of Australia, including Darwin, Katherine, Kakadu National Park, Jabiru, Nganmarriyanga (Palumpa), Larrimah, Mataranka, and Beatrice Hill Farm (Coastal Plains Research Centre). The two unnamed insect samples were collected in Western Australia. All viruses were matched to at least 91% similarity with RdRp genes from known viruses (Table 1). Isolates included one > 99 match to Hapavirus Holmes (Rhabdovirivdae/Alpharhabdoviridae), 23 isolates were matched to eleven species in the genus Orbivirus (Reovirales/Sedoreoviridae), and 19 to six species in the genus Orthobunyavirus (Bunyavi-rales/Peribunyaviridae). Phylogenetic analysis of both the Sedoreoviridae (Figure 1) and Peribunyaviridae (Figure 2) confirmed Blastx results both with respect to species identification and the percentage match with respect to topology and branch lengths.
## 4. Discussion
All the 43 virus isolates that we were able to generate sequence data that was adequate for identification via RdRp Blast comparison were closely related (between 91 and >99% sequence identity) to known viral taxa (Table 1). Most of these identifications were confirmatory with provisional serological identifications, with the exception of an isolate serotypically identified as Bluetongue virus and sequenced as Epizootic haemorrhagic disease virus. It is not necessarily surprising that the taxa identified in this study are not completely novel, as the process of isolation and amplification is likely to select for a specific subset of viruses already characterised. However, the data does add considerably to the body of genomic data for several relatively understudied species, and potentially sheds light on novel serotypes and genotypes within species groups. The relatively low recovery rate (43/76 isolates) reflects variability in nucleic acid integrity across samples. Future studies may benefit from unbiased enrichment approaches (e.g., rRNA depletion, host subtraction, random-primed amplification) alongside optimised extraction and preservation methods to increase recovery from poorly preserved or low-titre material.
Species delimitation is a field fraught with challenges, as taxonomy places discrete boundaries on evolutionary processes that operate on continuums [54]. While some describe viral taxonomy as a categorisation of convenience, there is practical utility to taxonomic nomenclature for researchers in the field [54,55]. There are varying levels of genetic divergence between lineages considered distinct species, and frequently significant divergences within species. In this study, most isolates are likely members of existing species based on genetic evidence; however, some are significantly divergent from their closest match. However, due to incomplete genome sequencing, the provided virus classifications are tentative. For example, V1664, V197, V3265, V3289, and V6250 are all ~5% divergent from Wongorr virus, and V409 is 9% divergent from Warrego virus based on Blastx matches, but recovery of the RdRp gene was not of sufficient coverage for phylogenetic placement. This could be driven by inadequate sequencing depth but is also possibly due to poor mapping to the reference sequence due to divergence from it. As the identity of the isolates was unknown prior to sequencing, specific sequence-targeted amplification could not be performed. Although additional rounds of enrichment could have increased genome coverage per isolate, our study focused on maximising virus discovery across the collection rather than achieving complete genome sequences for each isolate. V6013 was a 95% match to Sango virus based on Blastx results, but alignment of the RdRp L segment gene using MAFFT reveals a pairwise identity of only 80.7%. By comparison, pairwise identity of Sango virus to its nearest relative-Peaton virus-is 84.7%. Based purely on genetic identity, it would prima facie appear that V6013 is a novel taxon; however, follow up evidence from other life history criteria is necessary to support this. This study shows, at least putatively, that multiple novel species of virus may be preliminarily identified by sequencing historical virus isolate collections, considerably narrowing down the isolates worthy of further investigation for the discovery of novel taxa.
Hapaviruses form a monophyletic group within the Alpharhabdovirinae subfamily of the Rhaboviridae, within a larger clade of arthropod-borne rhabdovirsues. They have been primarily isolated from culicine mosquitoes and passerine birds [56], although there is evidence for Hapavirus antibodies in marsupials [56], cattle [57], and a hospitalised human patient [58]. Pairwise alignment of the whole genome sequence recovered from Sample V1163 has a 99.9% pairwise nucleotide identity with the Hapavirus holmes genome, and a 76.4% pairwise identity with the nearest relative-Wongabel virus [58], strongly indicating that the virus isolate sequenced in this study is Hapavirus holmes. The Hapavirus genus was recently established in 2017 [59], and relatively little is known about the life history and gene function of the group [58]. Another complete genome from this enigmatic group of arboviruses presents a resource in generating a more complete understanding of the group.
Orbiviruses are highly prevalent in Northern Australia [60], and a number are emergent zoonotic pathogens (e.g., Corriparta virus [61], Lebombo virus [62], Orungo virus [63]) or have recently spilled over into commercially important livestock species, e.g., Peruvian horse sickness virus and Yunnan virus [64]. Similarly, orthobunyaviruses are especially diverse in Northern Australia, and responsible for a wide spectrum of zoonotic disease-especially species from the Bunyamwera [65], California [66], and Simbu [67] serocomplexes. The impacts of human orthobunyaviral infections can be significant; e.g., 10% of paediatric infections of La Crosse virus result in long-term cognitive sequelae [68], and Oropouche virus causes symptomology similar to Zika and Dengue viruses in South America [69]. Given the high proportion of undiagnosed febrile illnesses in northern Australia [6], it is likely that many of these cases are caused by either novel or poorly characterised Orbivirus and Orthobunyavirus taxa. The role of novel arboviral etiological agents in infections and their role in febrile illnesses in non-specific fevers in humans is of significant public health interest [6,70], and due to the increase in factors associated with novel epizootic arboviral pathogens [71,72], will be of continued importance. As this study demonstrates, historical viral collections can allow for the pre-emptive characterisation of novel orbivurses, providing an invaluable resource for disease outbreak preparedness and can provide a critical tool for the development of both diagnostic and therapeutic agents for the treatments of these diseases.
## References
1. Vasilakis, Tesh, Popov et al. (2019) "Exploiting the Legacy of the Arbovirus Hunters" *Viruses*
2. (2025) "Next Generation Sequencing Technologies for Insect Virus Discovery-PubMed"
3. Jones, Boonham, Adams et al. (2021) "Historical Virus Isolate Collections: An Invaluable Resource Connecting Plant Virology's Pre-Sequencing and Post-Sequencing Eras" *Plant Pathol*
4. Petersen, Holcomb, Beard (2022) "Climate Change and Vector-Borne Disease in North America and Europe" *J. Health Monit*
5. Huang, Higgs, Vanlandingham (2019) "Emergence and Re-Emergence of Mosquito-Borne Arboviruses" *Curr. Opin. Virol*
6. Gyawali, Bradbury, Aaskov et al. (2017) "Neglected Australian Arboviruses and Undifferentiated Febrile Illness: Addressing Public Health Challenges Arising From the" *Developing Northern Australia' Government Policy. Front. Microbiol*
7. Gyawali, Bradbury, Aaskov et al. (2017) "Neglected Australian Arboviruses: Quam Gravis? Microbes Infect"
8. Gould, Solomon, Flaviviruses (2008) *Lancet*
9. Messina, Brady, Scott et al. (2014) "Global Spread of Dengue Virus Types: Mapping the 70 Year History" *Trends Microbiol*
10. Taylor-Robinson (2024) "Complex Transmission Epidemiology of Neglected Australian Arboviruses: Diverse Non-Human Vertebrate Hosts and Competent Arthropod Invertebrate Vectors" *Front. Microbiol*
11. Huang, Allcock, Warrilow (2016) "Newly Characterized Arboviruses of Northern Australia" *Virol. Rep*
12. Hughes, Adkins, Alkhovskiy et al. (2020) *Virus Taxonomy Profile: Peribunyaviridae. J. Gen. Virol*
13. (2025) "Isolations of Jamestown Canyon Virus (Bunyaviridae: Orthobunyavirus) from Field-Collected Mosquitoes (Diptera: Culicidae) in Connecticut, USA: A Ten-Year Analysis, 1997-2006-PubMed"
14. Labuda (1991) "Arthropod Vectors in the Evolution of Bunyaviruses" *Acta Virol*
15. Lasecka, Baron (2014) "The Molecular Biology of Nairoviruses, an Emerging Group of Tick-Borne Arboviruses" *Arch. Virol*
16. Williams, Imlarp, Top et al. (1976) "Kaeng Khoi Virus from Naturally Infected Bedbugs (Cimicidae) and Immature Free-Tailed Bats" *Bull. World Health Organ*
17. Henderson, Coleman (1971) "The Growing Importance of California Arboviruses in the Etiology of Human Disease" *Prog. Med. Virol*
18. Dutuze, Nzayirambaho, Mores et al. (2018) "A Review of Bunyamwera, Batai, and Ngari Viruses: Understudied Orthobunyaviruses with Potential One Health Implications" *Front. Vet. Sci*
19. Hayama, Moriguchi, Yanase et al. (2016) "Spatial Epidemiological Analysis of Bovine Encephalomyelitis Outbreaks Caused by Akabane Virus Infection in Western Japan in 2011" *Trop. Anim. Health Prod*
20. Jennings, Mellor, Culicoides (1989) "Biological Vectors of Akabane Virus" *Vet. Microbiol*
21. Weidmann, Rudaz, Nunes et al. (2003) "Rapid Detection of Human Pathogenic Orthobunyaviruses" *J. Clin. Microbiol*
22. Matthijnssens, Attoui, Bányai et al. (1782) "ICTV Virus Taxonomy Profile: Sedoreoviridae 2022" *J. Gen. Virol*
23. Roy (1996) "Orbivirus Structure and Assembly" *Virology*
24. Belaganahalli, Maan, Maan et al. (2015) "Genetic Characterization of the Tick-Borne Orbiviruses" *Viruses*
25. Calisher, Mertens, Mellor et al. (1998) "Taxonomy of African Horse Sickness Viruses"
26. Liehne, Anderson, Stanley et al. (1981) "Isolation of Murray Valley Encephalitis Virus and Other Arboviruses in the Ord River Valley 1972-1976" *Aust. J. Exp. Biol. Med. Sci*
27. Maan, Belaganahalli, Maan et al. (2020) "Emerging and Transboundary Animal Viruses"
28. Maclachlan, Guthrie (2010) "Re-Emergence of Bluetongue, African Horse Sickness, and Other Orbivirus Diseases" *Vet. Res*
29. Debiasi, Tyler (2015) "Orthoreoviruses and Orbiviruses"
30. Rohe, Fall (1979) "A Miniature Battery Powered CO2 Baited Light Trap for Mosquito Borne Encephalitis Surveillance" *Bull. Soc. Vector Ecol*
31. Whelan, Weir (1982) "The Isolation of Alpha and Flavi Viruses from Mosquitoes in the Northern Territory"
32. Macpherson, Stoker (1962) "Polyoma Transformation of Hamster Cell Clones-An Investigation of Genetic Factors Affecting Cell Competence" *Virology*
33. Igarashi (1978) "Isolation of a Singh's Aedes Albopictus Cell Clone Sensitive to Dengue and Chikungunya Viruses" *J. Gen. Virol*
34. Hsu, Zenzes (1964) "Mammalian Chromosomes in Vitro" *XVII. Idiogram Chin. Hamster. J. Natl. Cancer Inst*
35. Gard, Shorthose, Weir et al. (1987) "The Isolation of a Bluetongue Serotype New to Austrlia"
36. Goldsmit, Barzilai (1968) "An Improved Method for the Isolation and Identification of Bluetongue Virus by Intravenous Inoculation of Embryonating Chicken Eggs" *J. Comp. Pathol*
37. Lindsay, Broom, Wright et al. (1993) "Ross River Virus Isolations from Mosquitoes in Arid Regions of Western Australia: Implication of Vertical Transmission as a Means of Persistence of the Virus" *Am. J. Trop. Med. Hyg*
38. Sato, Maeda, Yoshida et al. (1977) "Plaque Formation of Herpes Virus Hominis Type 2 and Rubella Virus in Variants Isolated from the Colonies of BHK21/WI-2 Cells Formed in Soft Agar" *Arch. Virol*
39. Del Vecchio, Smith (1981) "Expression of Angiotensin-Converting Enzyme Activity in Cultured Pulmonary Artery Endothelial Cells" *J. Cell. Physiol*
40. Gard, Kirkland, Bluetongue (1993) "Virology and Serology"
41. Zeller, Karabatsos, Calisher et al. (1989) "Electron microscopic and antigenic studies of uncharacterized viruses. II. Evidence suggesting the placement of viruses in the familyBunyaviridae" *Arch. Virol*
42. Zeller, Karabatsos, Calisher et al. (1989) "Electron mi-croscopic and antigenic studies of uncharacterized viruses. III. Evidence suggesting the placement of viruses in the family Reoviridae" *Arch. Virol*
43. Calisher (1994) "Medically important arboviruses of the United States and Canada" *Clin. Microbiol. Rev*
44. Bolger, Lohse, Usadel (2014) "Trimmomatic: A Flexible Trimmer for Illumina Sequence Data" *Bioinformatics*
45. Bankevich, Nurk, Antipov et al. (2012) "SPAdes: A New Genome Assembly Algorithm and Its Applications to Single-Cell Sequencing" *J. Comput. Biol*
46. Charon, Buchmann, Sadiq et al. (2022) "RdRp-Scan: A Bioinformatic Resource to Identify and Annotate Divergent RNA Viruses in Metagenomic Sequence Data" *Virus Evol*
47. Buchfink, Reuter, Drost (2021) "Sensitive Protein Alignments at Tree-of-Life Scale Using DIAMOND" *Nat. Methods*
48. Li "Aligning Sequence Reads"
49. Danecek, Bonfield, Liddle et al. "Twelve Years of SAMtools and BCFtools"
50. Katoh, Misawa, Kuma et al. (2002) "MAFFT: A Novel Method for Rapid Multiple Sequence Alignment Based on Fast Fourier Transform" *Nucleic Acids Res*
51. Darriba, Taboada, Doallo et al. (2012) "jModelTest 2: More Models, New Heuristics and Parallel Computing" *Nat. Methods*
52. Ronquist, Teslenko, Van Der Mark et al. (2012) "MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice across a Large Model Space" *Syst. Biol*
53. Rambaut, Drummond, Xie et al. (2018) "Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7" *Syst. Biol*
54. Peterson (2014) "Defining Viral Species: Making Taxonomy Useful" *Virol. J*
55. Bobay, Ochman "Biological Species in the Viral World"
56. Gubala, Davis, Weir et al. (2010) "Ngaingan Virus, a Macropod-Associated Rhabdovirus, Contains a Second Glycoprotein Gene and Seven Novel Open Reading Frames" *Virology*
57. Doherty, Carley, Standfast et al. (1969) "Isolation of Arboviruses from Mosquitoes, Biting Midges, Sandflies and Vertebrates Collected in Queensland" *Trans. R. Soc. Trop. Med. Hyg*
58. Gubala, Walsh, Mcallister et al. (2017) "Identification of Very Small Open Reading Frames in the Genomes of Holmes Jungle Virus, Ord River Virus, and Wongabel Virus of the Genus Hapavirus, Family Rhabdoviridae" *Evol. Bioinform. Online*
59. Amarasinghe, Bào, Basler et al. (2017) "Taxonomy of the Order Mononegavirales: Update" *Arch. Virol*
60. Cowled, Melville, Weir et al. (2007) "Genetic and Epidemiological Characterization of Middle Point Orbivirus, a Novel Virus Isolated from Sentinel Cattle in Northern Australia" *J. Gen. Virol*
61. Boughton, Hawkes, Naim (1990) "Arbovirus Infection in Humans in NSW: Seroprevalence and Pathogenicity of Certain Australian Bunyaviruses"
62. Moore, Causey, Carey et al. (1975) "Arthropod-Borne Viral Infections of Man in Nigeria, 1964-1970" *Ann. Trop. Med. Parasitol*
63. Tomori, Fabiyi (1976) "Neutralizing Antibodies to Orungo Virus in Man and Animals in Nigeria" *Trop. Geogr. Med*
64. Attoui, Mendez-Lopez, Rao et al. (2009) "Peruvian Horse Sickness Virus and Yunnan Orbivirus, Isolated from Vertebrates and Mosquitoes in Peru and Australia" *Virology*
65. Venter (2018) "Assessing the Zoonotic Potential of Arboviruses of African Origin" *Curr. Opin. Virol*
66. Kosoy, Rabe, Geissler et al. (2016) "Serological Survey for Antibodies to Mosquito"
67. Reusken, Van Den Wijngaard, Van Beek et al. (2012) "Lack of Evidence for Zoonotic Transmission of Schmallenberg Virus" *Emerg. Infect. Dis*
68. Boutzoukas, Freedman, Koterba et al. "La Crosse Virus Neuroinvasive Disease in Children: A Contemporary Analysis of Clinical/Neurobehavioral Outcomes and Predictors of Disease Severity" *Clin*
69. Da Rosa, De Souza, De Pinheiro et al. (2017) "Oropouche Virus: Clinical, Epidemiological, and Molecular Aspects of a Neglected Orthobunyavirus" *Am. J. Trop. Med. Hyg*
70. Endy, Ryan, Hill et al. (2020) "In Hunter's Tropical Medicine and Emerging Infectious Diseases"
71. Cao-Lormeau, Musso (2014) "Emerging Arboviruses in the Pacific" *Lancet*
72. Gould, Pettersson, Higgs et al. (2017) "Emerging Arboviruses: Why Today? One Health"
73. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12592443&blobtype=pdf
|
# Correction: A new biomarker combination differentiates viral from bacterial infections and helps monitoring response to antibiotics in hospitalized children
Federica Pagano, Stefano Brusa, Giusy Arrichiello, Valentina Cioffi, Marco Poeta, Dario Bruzzese, Giuseppe Portella, Alfredo Guarino, Eugenia Bruzzese
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12391543&blobtype=pdf
|
# Corrections & amendments Author Correction: EGFR core fucosylation, induced by hepatitis C virus, promotes TRIM40-mediated-RIG-I ubiquitination and suppresses interferon-I antiviral defenses
Qiu Pan, Yan Xie, Ying Zhang, Xinqi Guo, Jing Wang, Min Liu, Xiao-Lian Zhang, Nature Communications
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724372&blobtype=pdf
|
# Annexin A2 stabilizes the endoplasmic reticulum and actin cytoskeleton and influences the formation of reovirus factories
Raquel Tenorio, Gwen Taylor, Ana Cayuela, C Sorzano, Isabel Fernández De Castro, Sara Fernández-Sánchez, Alexa Roth, Xayathed Somoulay, Gavin Treadaway, Terence Dermody, Cristina Risco
## Abstract
Early in infection, mammalian orthoreoviruses (reoviruses) build neo-organ elles called viral factories (VFs). These structures incorporate fragments of endoplasmic reticulum (ER) and serve as sites of viral genome replication and particle assembly. Two reovirus nonstructural proteins, σNS and µNS, remodel the ER to produce the vesicles and tubules that form the VF matrix. Using co-immunoprecipitation assays followed by mass spectrometry, we identified annexin A2 (ANXA2), which binds actin and cellular membranes, as a host factor that interacts with reovirus nonstructural proteins. In the absence of ANXA2, VF formation was accelerated in the early phases of reovirus infection, and yields of reovirus were increased. Moreover, organization of the ER network and structure of the actin cytoskeleton were altered in the absence of ANXA2, suggesting that ANXA2 binding to actin is required to maintain ER network morphology. These findings provide evidence that interactions of reovirus nonstructural proteins directly or indirectly with ANXA2 are required for ER remodeling and formation of the membranous fragments that serve as the matrix to assemble reovirus factories.IMPORTANCE Reovirus uses ER fragments to build the membranous scaffold of viral factories (VFs). Host proteins that participate in the ER remodeling that precedes factory biogenesis are not known. We identified actin-binding protein ANXA2 as a cellular factor required for maintenance of ER morphology. The absence of ANXA2 destabilizes the actin cytoskeleton and consequently the ER, which accelerates VF biogenesis and enhances reovirus replication. Uncovering the cellular factors used by viruses to form VFs deepens an understanding of viral cell biology and highlights new targets for antiviral drug development.
KEYWORDS annexin A2, reovirus, viral factories, σNS, µNS, endoplasmic reticulum, actinV iral factories (VFs) are formed during infection and concentrate viral and host components required for viral replication. These specialized intracellular compart ments establish an ideal environment for essential steps in viral infection, including genome replication and packaging and assembly of progeny particles. VFs can also shield viral genomes and proteins from detection by innate immune response sensing mechanisms (1, 2). VFs can be membrane-enclosed (3), devoid of membranes (4), or nonmembrane-enclosed with internal membranes (5, 6). Viruses often transform cellular membrane organelles, such as the endoplasmic reticulum (ER), Golgi, lysosomes, or mitochondria, to build VFs (7-9). The ER is most commonly subverted by viruses to build membrane-enclosed and nonmembrane-enclosed factories (10). A variety of viruses, including mammalian orthoreovirus (reovirus) (5), nidoviruses (11,12), potyviruses (13), tombusviruses ( 14), vaccinia virus (15), and Zika virus ( 16), use the ER to form VFs.
Reoviruses are nonenveloped, segmented, double-stranded RNA (dsRNA) viruses (17) that remodel the ER to build nonmembrane-enclosed VFs (5,6,18). Reovirus has a broad host range and infects many mammalian species (17). Reovirus infections in humans are mild or asymptomatic and have been implicated in the development of celiac disease (19). Reovirus factories contain a combination of viral and cellular components (20) and are nucleated by viral nonstructural proteins σNS and µNS (21). In infected cells, the nonstructural proteins induce an unusual remodeling of the ER. σNS binds ER cister nae and transforms these structures into thin tubules, whereas µNS binds the tubules, eliminates their branches, and fragments these structures into small membranous pieces that contribute to the VF scaffold (6,22). Cellular proteins required for the transformation of the ER, a key step in VF biogenesis, are not known.
In this study, we used co-immunoprecipitation assays followed by mass spectrometry (MS) to identify cellular proteins that interact with reovirus σNS and µNS. We found that both nonstructural proteins precipitated annexin A2 (ANXA2), an actin-binding protein. ANXA2 binds membranes using type II calcium-binding domains (23). In the absence of ANXA2, the actin cytoskeleton and ER network are disrupted, and reovirus factory formation and infection progress more rapidly than in cells expressing ANXA2. These results demonstrate that interactions of ANXA2 with actin stabilize the ER network and suggest that interactions of reovirus nonstructural proteins with ANXA2 function in ER remodeling and VF biogenesis during reovirus infection.
## RESULTS
## ANXA2 associates with reovirus nonstructural proteins
To identify host factors that interact with reovirus nonstructural proteins σNS and µNS, we used co-immunoprecipitation assays followed by MS. HeLa cells were either infected with reovirus strain type 1 Lang (T1L) M1 P208S, which forms large, globular VFs (24), or transfected with plasmids encoding σNS, µNS, or both. Cells were lysed and incuba ted with antibodies specific for σNS or µNS to capture complexes containing reovirus nonstructural and host proteins. Immunoprecipitated proteins were identified by MS at 24 h post-infection or 48 h post-transfection (Fig. 1A).
Proteins identified by MS were distributed in five groups: viral proteins, transcription factors, cytoskeletal proteins, chaperones, and energetic metabolism proteins (Fig. 1B). Only two proteins, ANXA2 and myosin-9 (MYH9), were identified in infected and transfected cells as putative interacting partners of both σNS and µNS (Fig. 1A andC). In reovirus-infected cells, ANXA2 and MYH9 were co-immunoprecipitated by a µNS-specific antibody but not by a σNS-specific antibody. However, ANXA2 and MYH9 were coimmunoprecipitated by the σNS-specific antibody in cells transfected with σNS alone or in combination with µNS. ANXA2 and MYH9 were also co-immunoprecipitated by the anti-µNS antibody, but only in cells expressing both σNS and µNS. ANXA2 is an actinbinding protein involved in endosomal trafficking and cholesterol homeostasis (25). It was also identified as a proviral gene candidate for reovirus replication in siRNA screens (26,27). Based on the known functions of ANXA2 and its identification in unrelated genetic screens, we evaluated a potential role for ANXA2 in reovirus replication.
ANXA2 distributes to intracellular vesicles and the plasma membrane and is involved in vesicular trafficking, actin remodeling, and regulation of translation (28,29). To determine the distribution of ANXA2 during reovirus infection, we imaged reovirusinfected cells using confocal immunofluorescence microscopy (Fig. 1D through I). ANXA2 was diffusely distributed in the cytoplasm of mock-infected cells (Fig. 1D). However, in reovirus-infected cells, ANXA2 redistributed and became more intense surrounding the periphery of VFs, where reovirus nonstructural proteins concentrate (Fig. 1G through I). and late stages of infection. We adsorbed WT and KO cells with reovirus and quantified the number of infected cells at 14 and 24 h post-adsorption using immunofluorescence staining for µNS. Following adsorption with reovirus, there was an approximate twofold increase in the number of KO cells containing VFs relative to VF-containing WT cells at both 14 and 24 h post-adsorption (Fig. 2A), suggesting that ANXA2 limits the formation of reovirus VFs. To determine whether levels of µNS, which is expressed early in infection, are altered in the absence of ANXA2, we quantified µNS levels in reovirus-infected WT and KO cells using immunoblotting. At 14 and 24 h post-adsorption, µNS levels were comparable in WT and KO cells (Fig. 2B), suggesting that ANXA2 is not required for the expression of reovirus proteins.
To quantify viral titer by plaque assay, we adsorbed WT and KO cells with reovirus and collected culture supernatants and cell lysates at 0, 14, and 24 h post-adsorption.
Comparison of viral titers in culture supernatants and cell lysates allows us to determine whether there are differences in viral egress. Significantly more infectious virus was detected in supernatants of KO cells than in WT cells at 24 h post-adsorption (Fig. 2C). However, viral titers in lysates of WT and KO cells were comparable at 14 and 24 h postadsorption (Fig. 2D). These findings suggest that ANXA2 influences steps in reovirus infection that lead to viral release.
To determine whether ANXA2 regulates the formation of VFs early in infection, we quantified cells containing VFs in WT and KO cells at 7, 8, 9, and 10 h post-adsorption in a synchronized infection in which cells were incubated for 30 min on ice during adsorp tion. In WT cells, VFs were small punctate structures first detectable by immunofluorescence microscopy at approximately 7 h post-adsorption. At all times post-adsorption, the number of VF-containing KO cells was significantly greater than VF-containing WT cells (Fig. 2E; Fig. S1 andS2), and significantly more advanced-infected cells, defined as those with five or more VFs/cell, were observed in KO than WT cells (Fig. 2E). Large VFs (≥1 µm) were visible starting at 10 h post-adsorption in both WT and KO cells (Fig. S3). However, there were no significant differences in σNS protein levels at these early times postadsorption (Fig. 2F). These data suggest that the absence of ANXA2 accelerates VF formation early in infection, independent of effects on protein expression.
During reovirus infection, σNS and µNS colocalize within VFs and at the factory periphery (21,30). To determine whether the absence of ANXA2 alters the distribution of σNS and µNS, WT and KO cells were adsorbed with reovirus, and the distribution of σNS and µNS was defined using confocal immunofluorescence microscopy. In WT cells, σNS and µNS colocalize within VFs and concentrate at the VF periphery (Fig. 2G). However, in KO cells, while σNS and µNS distribute to VFs, their colocalization is significantly reduced. (Fig. 2; Fig. S4). Instead, σNS and µNS appear to be in close proximity. These data suggest that ANXA2 influences the distribution of σNS and µNS inside VFs.
## ANXA2 depletion alters the morphology of the ER network
VFs contain membranous fragments that are derived from extensive remodeling of the ER by σNS and µNS during reovirus infection (6). To determine whether depletion of ANXA2 alters ER morphology, we transfected WT and KO cells with a plasmid expressing mCherry-ER-3 to label the ER, visualized ER morphology using confocal microscopy followed by deconvolution image processing (Fig. 3A through D), and quantified the ER morphologies observed (Fig. 3E andF). In WT cells, the mCherry-ER-3 signal revealed an ER network with a typical reticular pattern (Fig. 3A). However, in KO cells, the ER network was less well defined with an 81.6% increase in cells with fragmented ER and a 44% increase in cells with collapsed ER structures (Fig. 3B, E, andF), demonstrating that the ER network is altered in the absence of ANXA2.
We conducted similar experiments to evaluate ER morphology in WT and KO cells during reovirus infection. Following infection of WT cells, we observed the characteristic ER remodeling induced by reovirus (6), in which the ER becomes thin and fragmented (50% increase) and forms small collapsed structures (39% increase) (Fig. 3C, E, andF), similar to the ER morphology of mock-infected cells in the absence of ANXA2. In Full-Length Text reovirus-infected KO cells, the collapsed ER aggregates were more prominent than in mock-infected KO cells (8% increase), and we observed more unbranched linear ER tubules (61% increase) (Fig. 3D, E andF). Representative images of the observed ER morphologies are shown in Fig. S5.
To further characterize these morphological changes, we used transmission electron microscopy (TEM) to image mock-infected and reovirus-infected WT and KO cells. The ER in mock-infected WT cells formed a regular branched tubular network (Fig. 3G). However, in mock-infected KO cells, the ER network was not well organized, and ER tubules were fragmented (Fig. 3H). In reovirus-infected WT cells, ER tubules, including branched tubules, surround the VFs (Fig. 3I). However, in reovirus-infected KO cells, the ER tubules surrounding VFs were long, thin, and unbranched (Fig. 3J). These data demonstrate that the ER network is disrupted in the absence of ANXA2 and during reovirus infection. However, while ANXA2 localizes at the periphery of VFs at 14 and 24 h post-adsorption, it does not colocalize with ER membranes within VFs (Fig. S6).
## ANXA2 depletion disrupts the actin cytoskeleton
ANXA2 is an actin-binding protein with multiple functions, including stabilizing the actin cytoskeleton (31,32). The actin cytoskeleton also interacts with components of the ER to maintain ER morphology (33,34). To determine whether ANXA2 depletion disrupts the actin cytoskeleton, we used fluorescently labeled phalloidin and confocal microscopy to visualize actin filaments in WT and KO cells. In WT cells, we observed peripheral cortical actin and groups of long actin filaments and stress fibers (Fig. 4A). In KO cells, we observed cortical actin networks only at the cell periphery, and we did not detect long actin filaments (Fig. 4B), suggesting that ANXA2 functions in stress fiber formation or maintenance (35). We also imaged actin filaments in WT and KO cells infected with reovirus at 14 h post-adsorption. Fewer stress fibers were observed in infected WT cells, while cortical actin appeared intact (Fig. 4C). The organization of the actin cytoskeleton in infected KO cells appeared similar to that observed in uninfected KO cells, with only cortical actin networks visible (Fig. 4D). Together, these data demonstrate that the absence of ANXA2 and reovirus infection alters the actin cytoskeleton.
To assess interactions between the actin cytoskeleton and ER network during reovirus infection, we imaged infected WT and KO cells using confocal microscopy and assembled 3D reconstructions using LAS X software. Fluorescent phalloidin was used to detect actin filaments, transfection with mCherry-ER-3 was used to detect the ER network, and an antibody specific for σNS was used to detect VFs. Videos were recorded from the 3D reconstructions of mock-infected (Videos S1, S3, S5 and S7) and reovirus-infected cells (Videos S2, S4, S6 and S8). In frames from videos of mock-infected cells, actin was associated with the ER in both WT and KO cells (Fig. 4E andF). Actin and ER were also in close proximity in VFs of reovirus-infected WT and KO cells (Fig. 4G andH). Together, these data demonstrate that the actin cytoskeleton and ER network are remodeled during reovirus infection and are disrupted in the absence of ANXA2.
## µNS and σNS differentially modify the ER network in WT and KO cells
To determine how ANXA2 depletion affects ER remodeling by σNS and µNS, we transfec ted WT and KO cells with mCherry-KDEL together with σNS, µNS, or both proteins and monitored ER morphology using confocal microscopy. In WT cells expressing σNS, ER tubules were stretched at the cell periphery (Fig. 5A through C), as observed previously (6). However, in the absence of ANXA2, the ER remained as clusters of small collapsed cisternae throughout the cell (Fig. 5D through F). In both the presence and absence of ANXA2, σNS was distributed diffusely in the cytoplasm. In WT cells expressing µNS, ER tubules were unbranched and fragmented, with some long unbranched tubules remaining (Fig. 5G through I), as observed previously (6,36). In the absence of ANXA2, the ER was mostly collapsed at the periphery of the nucleus (Fig. 5J through L). In both WT and KO cells, µNS was distributed in the cytoplasm and formed factory-like struc tures. In WT cells expressing both σNS and µNS, ER remodeling was similar to that observed during reovirus infection, and σNS and µNS co-localized in factory-like structures (Fig. 5M through P). In ANXA2 KO cells transfected with both nonstructural proteins, the ER was highly unstructured, and some small foci of collapsed ER were observed adjacent to factory-like structures (Fig. 5Q through T). The most common ER morphologies observed were quantified (Fig. S7). Viral nonstructural proteins also were concentrated in these factory-like structures, but they did not colocalize, similar to that observed in reovirus-infected KO cells (Fig. 2H). Collectively, these data suggest that the absence of ANXA2 modifies the effects of reovirus nonstructural proteins on the morphology of the ER network, increasing the portions of collapsed ER and uncoupling σNS and µNS interactions.
## The distance between reovirus nonstructural proteins, ER, and actin is altered in the absence of ANXA2
To better understand the dynamics of reovirus nonstructural proteins and ANXA2 in relation to the ER and actin cytoskeleton, we quantified minimum distances between the fluorescence signals of σNS, µNS, ANXA2, the ER, and the actin cytoskeleton in mockinfected and reovirus-infected WT and KO cells using the z-stack analyzer plug-in for FIJI software. This quantification method is based on the minimum distances separating the signals of two different fluorescent channels. More than 1 million measurements of minimum distances were obtained and ordered from shortest (0 nm) to longest (several µm). Adjacent signal distances were obtained by subtracting measurements between 0 and 150 nm from all measurements obtained. Thus, the adjacent signal measurements correspond to all measurements between the first channel and second channel with a maximum separation of three pixels (with a pixel size of 52 nm) (Fig. S8). The percentage of adjacent signal measurements was calculated as the ratio of all measurements within the adjacent signal interval to the total minimum measurements multiplied by 100 (Fig. 6).
We first determined the distance between ANXA2 and the ER and ANXA2 and the actin cytoskeleton in mock-infected and reovirus-infected WT cells. The distance between ANXA2 and the ER (Fig. 6A) and ANXA2 and actin (Fig. 6B) increased in infected cells relative to mock-infected cells, suggesting that the association of ANXA2 with both the ER and actin is altered during reovirus infection. We also analyzed the distance between ANXA2 and the nonstructural proteins in infected WT cells. The distance between ANXA2 and σNS and ANXA2 and µNS was comparable (Fig. 6C), suggesting that the nonstructural proteins form a complex during infection that is in close proximity to ANXA2.
We next determined the distance between the ER and actin cytoskeleton in mockinfected and reovirus-infected WT and KO cells. The distance between the ER and actin decreased in both mock-infected and reovirus-infected KO cells relative to mock-infected WT cells (Fig. 6D). Additionally, the distance between the ER and actin decreased in infected WT cells relative to mock-infected WT cells. Together, these data suggest that the disruption of the ER network and actin cytoskeleton that occurs during reovirus infection or in the absence of ANXA2 increases an association of actin with the ER. We also analyzed the distance between the reovirus nonstructural proteins and the ER and actin in reovirus-infected WT and KO cells. The proximity of the ER to either µNS (Fig. 6E) or σNS (Fig. 6F) was significantly greater in KO cells than in WT cells, suggesting that in the absence of ANXA2, the nonstructural proteins are more closely associated with the ER than they are in WT cells. However, the distance between µNS and actin was decreased in KO cells (Fig. 6G), while the distance between σNS and actin was increased (Fig. 6H). These data suggest that ANXA2 influences the distribution of σNS, µNS, and actin in VFs and are consistent with our morphological observations.
## Complementation of ANXA2 KO cells with ANXA2 recovers ER network morphology
To ensure that potential off-target effects of ANXA2 gene disruption do not contribute to the alteration of the ER network observed in KO cells relative to WT cells, we transfected KO cells with a plasmid encoding WT ANXA2 and compared ER morphology in mockinfected and reovirus-infected cells (Fig. 7). Mock-infected and reovirus-infected WT and ANXA2 KO cells were used as controls for ER morphology (Fig. 7A through L). In KO cells transfected with ANXA2, the ER morphology in mock-infected cells (Fig. 7M) was similar to that in mock-infected WT cells (Fig. 7A), with restoration of the tubular ER network. However, some collapsed ER cisternae were observed in the complemented cells (Fig. 7M). The relative ER network recovery was proportional to the level of ANXA2 in the transfected cells (Fig. 7P andQ). ER remodeling in reovirus-infected KO cells complemen ted with ANXA2 (Fig. 7N) was comparable to ER remodeling in reovirus-infected WT cells (Fig. 7B), with an increase in thinned and fragmented ER tubules and areas of collapsed ER (Fig. 7B andN). These data demonstrate that transient expression of ANXA2 in KO cells restores ER morphology.
To test whether ANXA2 maintains ER structure by binding ER membranes, we transfected KO cells with a plasmid encoding ANXA2 cmamxA2+, which lacks the calcium-binding domains and cannot interact with membranes (37), and compared ER morphology in mock-infected and reovirus-infected cells. In mock-infected KO cells expressing ANXA2 cmamxA2+, we observed a tubular ER network, similar to the ER morphology in WT cells and KO cells expressing ANXA2 (Fig. 7S). In reovirus-infected KO cells complemented with cmamxA2+ (Fig. 7T), ER remodeling was similar to ER remodeling in reovirus-infected WT cells and KO cells expressing ANXA2, with an increase in fragmented ER tubules and areas of collapsed ER. As observed previously, levels of ANXA2 correlate with the extent of ER network restoration (Fig. 7V andW). Collectively, these data suggest that the capacity of ANXA2 to bind ER membranes is not required to maintain ER structure.
To confirm a function for ANXA2 in reovirus infection, we used CRISPR/Cas9 gene editing to engineer a new clonal HeLa cell line with a nonfunctional ANXA2 gene. As a control, the rederived ANXA2 KO cells were complemented with wild-type ANXA2 (KO+) (Fig. S9). The rederived WT, KO, and KO+ cells were adsorbed with reovirus, and virus replication was quantified. The percentage of cells with VFs was significantly increased in KO cells at 14 and 24 h post-infection relative to WT and KO+ cells, and significantly higher titers of infectious virus were detected in culture supernatants of KO cells relative to WT or KO+ cells at 14 h post-adsorption (Fig. S10). Thus, these data suggest that ANXA2 expression maintains ER morphology and influences the rate of VF formation.
## DISCUSSION
During the early stages of reovirus infection, nonstructural proteins σNS and µNS function cooperatively to build functional VFs. These proteins remodel the ER to form small membranous fragments that contribute to the VF matrix. How σNS and µNS fragment and cleave the ER is not known. In this study, we identified ANXA2 as a host factor that interacts with nonstructural proteins σNS and µNS. We discovered that ANXA2, which binds actin, calcium, and lipids and functions in endocytosis, exocytosis, and the formation of lipid microdomains, also maintains the morphology of the ER network and influences VF formation. In the absence of ANXA2, the ER network is fragmented, similar to that observed during reovirus infection. In reovirus-infected cells lacking ANXA2, VF formation is accelerated relative to WT cells. Based on these data, we propose a model in which σNS and µNS interact directly or indirectly with ANXA2 to interrupt engagement of ANXA2 with actin fibers associated with the ER (Fig. 8). Disruption of ANXA2-actin interactions facilitates dismantling of the actin cytoskeleton and ER network, allowing the formation of ER membranous fragments that coalesce to form the reovirus factory matrix. In the absence of ANXA2, the actin cytoskeleton and ER are dismantled prior to infection, allowing for rapid recruitment of ER fragments to VFs and accelerated reovirus factory formation.
Early in reovirus infection, when VFs are forming, all viral and host factors required for VF assembly must be available. We hypothesize that alterations in ANXA2 distribution during infection destabilize the ER and facilitate the acquisition of ER fragments that nucleate VFs. In the absence of ANXA2, ER fragments are available at the initiation of infection and, therefore, VFs can form more rapidly in the presence of unchanged levels of nonstructural proteins. Mutagenesis experiments with the reovirus nonstructural proteins are required to test this hypothesis.
Several viruses use ANXA2 at different steps of replication, including entry, assembly, and egress. ANXA2 is a putative receptor for respiratory syncytial virus (38), mediates membrane fusion and internalization of porcine epidemic diarrhea virus (39), regulates hepatitis C virus RNA synthesis and assembly (40,41), and contributes to assembly of human immunodeficiency virus (42). ANXA2 also functions in replication of members of the Reovirales order, including avian reovirus (43) and bluetongue virus (BTV) (44). ANXA2 is required for caveolin-dependent endocytosis of avian reovirus (45,46) and functions in BTV egress by interacting with viral nonstructural protein NS3 (47). In each of these cases, ANXA2 interactions with membranes are required for its activities in viral replication. We found that ANXA2 stabilizes the ER network and slows VF biogenesis using a mechanism that is not dependent on membrane binding. Complementation of ANXA2 KO cells with an ANXA2 mutant lacking the membrane-binding domain (cmamxA2+) restored ER morphology in a manner comparable to WT cells and ANXA2 KO cells complemented with WT ANXA2 (Fig. 7). These data suggest that the function of ANXA2 in maintaining ER morphology is independent of binding ER membranes. We identified a new function for ANXA2 in viral infection that is dependent on the interaction of ANXA2 with the actin cytoskeleton. The actin cytoskeleton scaffolds the ER network using TPM4 (34), which binds actin and forms copolymers with short actin filaments that interact with the ER. In the absence of TPM4, the ER network is destabilized (34,48), similar to that observed in the absence of ANXA2 (Fig. 3). However, a function for ANXA2 in maintaining ER morphology has not been previously reported. Our data demonstrate that ANXA2 is required to maintain ER network morphology (Fig. 3). In the absence of ANXA2, fragmented and collapsed ER increased by 81.6% and 44%, respectively (Fig. 3I andJ). A function for ANXA2 in maintaining ER morphology is further supported by our finding that the distance between ANXA2 and the ER and ANXA2 and actin increases in reovirus-infected WT cells in which the ER is fragmented (Fig. 6A andB). The increased association of actin with the ER in reovirus-infected WT and KO cells relative to mock-infected WT and KO cells may be due to the distribution of actin and ER fragments within VFs (Fig. 6D). These data suggest that ANXA2 stabilizes the ER by interactions with the actin cytoskeleton. We also observed that destabilization of the ER network by ANXA2 gene editing (Fig. 3; Fig. S1) results in accelerated VF biogenesis, further supporting our model in which preemptive dismantling of the ER allows for rapid recruitment of ER fragments to VFs and accelerated reovirus factory formation.
Dynamic changes in the actin cytoskeleton are required for many cellular functions, including endocytosis, intracellular transport of macromolecular cargo, and exocytosis. Viruses interact with elements of the cytoskeleton at several steps of infection, including cell entry, protein synthesis, genome replication, and egress. Cytoskeletal elements, including microtubules (49) and actin filaments (Fernandez de Castro et al., unpublished results), are required for reovirus replication. Reovirus binding to PirB, which serves as a receptor for reovirus in some types of cells (50,51), activates receptor signaling and triggers endocytosis (51). Binding of some ligands to PirB induces depolymerization of actin filaments (52), which may be required for reovirus entry. We used fluorescently labeled phalloidin to demonstrate that reovirus infection alters the actin cytoskeleton. Reovirus-infected cells have fewer stress fibers than mock-infected cells, while cortical actin appears intact (Fig. 4C). Thus, actin serves a key function in reovirus replication.
Nonmembrane-enclosed VFs have properties of biomolecular condensates, including liquid-like properties (53)(54)(55), and can be either devoid of membranes or contain membranes within the factory. Reovirus factories are nonmembrane-enclosed but contain fragments of ER membranes that contribute to the VF matrix and may serve as a scaffold for RNA synthesis and viral particle assembly (6). Other members of the Reovirales, such as rotaviruses, also assemble nonmembrane-enclosed VFs. However, membranes have not been observed within these factories, suggesting that the composition of the factories formed by reovirus and rotavirus differs.
Biomolecular condensates were once thought to be devoid of membranes (56). However, membrane-associated condensates form at the plasma membrane, ER, nuclear envelope, peroxisomes, and autophagosomes (57). The thermodynamics of condensate formation depend on the local concentration of phase-separating macromolecules, such as proteins or RNA. A lower threshold concentration is required for membrane-asso ciated condensate formation (58). Our data suggest that reovirus factories are mem brane-associated condensates. Reovirus nonstructural proteins induce fragmentation and vesiculation of ER membranes (Fig. 3C) (6), and 3D electron tomography shows ER membranes surrounding VFs (6) as well as membrane fragments within VFs (6,18). When ER membrane fragments are present prior to reovirus infection, as occurs in the absence of ANXA2, VFs are visible by immunofluorescence staining of nonstructural proteins as early as 7 h post-infection (Fig. 2), which is approximately 2 h earlier than observed in cells expressing ANXA2. These data suggest that ER fragments serve as nucleation sites for functional VF biogenesis.
Within the order Reovirales, several cellular factors required for VF formation or function have been described. For example, casein kinase 2 and phosphatase 2A are required for BTV (59), ADP ribosylation factor 1 (ARF1) is required for grass carp reovirus (60), and casein kinase 1α (61), casein kinase 2 (62), small ubiquitin-like modifier (63), and perilipin1 (64) are required for rotavirus. In the case of reovirus, the TRiC chaperonin (65), hsc70 (66), and ER fragments and vesicles (6) localize to factories and are required for VF function but not formation. We found that ANXA2 localizes to the periphery of VFs (Fig. 1) and may regulate VF biogenesis by direct or indirect interactions with nonstructural proteins σNS and µNS.
In this study, we discovered that ANXA2 stabilizes the ER. Disrupting the binding of ANXA2 to actin perturbs ER membranes, leading to ER dismantling. Reovirus infection of cells lacking ANXA2 leads to accelerated VF formation and enhances the kinetics of reovirus replication. We found that reovirus dismantles actin fibers, which is essential for the formation of reovirus factories. This work establishes the foundation for future research on how ER membranes, the cytoskeleton, and the cellular factors identified in our proteomic screens coordinate with reovirus proteins to assemble functional VFs. As these mechanisms may be conserved across virus families, this work could illuminate new targets for broadly applicable antiviral therapeutics.
## MATERIALS AND METHODS
## Cells
HeLa cells (WT) and ANXA2 knockout (KO) HeLa cells were obtained from Dr. Martin Kast at the University of Southern California (36). Both cell lines were propagated in Dulbecco's modified Eagle's medium (DMEM; D6429; Sigma) supplemented to contain 10% fetal bovine serum (FBS), 100 U/mL penicillin G, 100 µg/mL streptomycin (Gibco), 0.25 µg/mL amphotericin B, nonessential amino acids, 2 mM L-glutamine, and 1 mM sodium pyruvate (Sigma). L929 fibroblast cells were propagated in DMEM (D6429; Sigma) supplemented to contain 10% FBS, 100 U/mL penicillin G, 100 µg/mL streptomycin (Gibco), nonessential amino acids (Sigma), and 2 mM L-glutamine.
New clonal HeLa ANXA2 knockout cells (ANXA2 KO cells) and ANXA2 complemented cells (ANXA2 KO+) (Fig. S9 andS10) were engineered by transfecting HeLa S3 cells with either empty CRISPR-KO transfer vector (lentiCRISPRv2-blast) or lentiCRISPRv2-blast encoding ANXA2-specific guide sequence (5′-GGTCCTTCTCTGGTAGGCGA-3′ from the human Brunello CRISPR knockout pooled library) using LipfectAMINE 3000 according to the manufacturer's instructions (67). At 48 h post-transfection, cells were selected with medium containing 10 µg/mL blasticidin for 5 days. Single-cell clones were selected from surviving cells and screened for ANXA2 expression. To confirm the specificity of Anxa2 knockout, the rederived ANXA2 KO cells were transduced with lentivirus vector pCSIB-expressing ANXA2 (engineered by Gibson assembly). Two days post-transduction, cells were selected with medium containing 10 µg/mL blasticidin for 5 days, and ANXA2 expression in the surviving cell population was assessed (ANXA2 KO+).
## Viruses
Reovirus strain T1L M1-P208S is identical to reovirus strain T1L, with the exception of a proline-to-serine substitution at position 208 of the µ2 protein (M1 gene). This substitution results in a change in VF morphology from filamentous to globular (24). Site-directed mutagenesis was used to engineer the P208S substitution in the M1 gene with the following primers: forward, 5′ CATTTCGGGGTAGCAATTGATGAAAATGTGCCAAC ATTAAATCTAG 3′; reverse, 5′ CTAGATTTAATGTTGGCACATTTTCATCAATTGCTACCCCGAAAT G 3′. Both strains were recovered by plasmid-based reverse genetics (68), purified using cesium gradient centrifugation (69), and propagated at a multiplicity of infection (MOI) of 5 PFU/cell at 33°C for 65 h to yield working stocks. Viral titers were determined by plaque assay using L929 cells (70).
## Viral titration by plaque assay
Titers of infectious reovirus were determined by plaque assay using L929 cells (70). Cells were adsorbed with tenfold serial dilutions of the infected sample or viral stock at 37°C for 1 h. Following viral adsorption, cells were overlaid with DMEM supplemented to contain 0.5% agarose, 1% penicillin/streptomycin, 1% nonessential amino acids, 0.1% gentamicin, and 2% FBS. Cells were incubated for 7 d, fixed with 10% formaldehyde for 1 h, and stained with 0.1% crystal violet for 5 min to visualize viral plaques. Viral titer is expressed as plaque-forming units per mL of culture supernatant or cell lysate (PFU/mL).
## Co-immunoprecipitation
HeLa cells cultivated in two P150 plates per condition were infected with reovirus T1L M1 P208S at an MOI 20 PFU/cell or transfected with T3D σNS, µNS, or both expression plasmids (71) using the calcium phosphate method (72). At 24 h post-infection (hpi) or 48 h post-transfection, cells were collected and lysed using co-immunoprecipitation (co-IP) buffer (20 mM Tris [pH 8], 137 mM NaCl, 2 mM EDTA, and 1% NP-40 substitute) supplemented with protease inhibitors (Roche, 11873617001) at 4°C for 30 min with rotation. Lysates were cleared by centrifugation, and supernatants were incubated with σNS-specific or µNS-specific antibodies at 4°C overnight and incubated with A Dyna beads (Thermo Fisher, 10001D) at room temperature (RT) for 1.5 h with rotation. Beads were collected from the supernatant using a magnet and washed six times with cold lysis buffer. Bound proteins were eluted by boiling in Laemmli sample buffer (Bio-Rad) with 10% β-mercaptoethanol for 15 min. Proteins were identified by MS (matrix-assisted laser desorption/ionization with time-of-flight [MALDI-ToF) and analyzed using MASCOT software (Matrix Science) to obtain qualitative identification of the resultant peptides.
## Immunoblotting
Cells harvested for protein extraction were lysed in 2× Laemmli sample buffer (Bio-Rad) containing 10% β-mercaptoethanol and incubated at 95°C for 15 min. A volume of 10 µL of each sample was electrophoresed in 4-20% Mini-Protean TGX gels (Bio-Rad). Following electrophoresis, proteins were transferred from gels to PVDF membranes using a Transblot turbo transfer pack (Bio-Rad). Membranes were blocked at RT for 1 h with 3% nonfat milk in phosphate-buffered saline (PBS) containing 0.05% Tween 20 and incubated overnight at 4°C with 1:1,000 rabbit σNS-specific polyclonal antibody VU82 (73), 1:12,000 chicken µNS-specific antiserum provided by John Parker (Cornell University) (20), and 1:2,000 mouse ANXA2-specific monoclonal antibody (Proteintech) in blocking buffer. Mouse tubulin-specific, rabbit Tomm22-specific, or mouse GAPDHspecific antibodies in blocking buffer were used for controls. After washing with 0.1% Tween 20 in PBS, membranes were incubated with secondary antibodies conjugated with horseradish peroxidase (HRP, POD) at RT for 1 h. Target proteins were detected using ECL solutions (SuperSignal West Dura Extended Duration Substrate, Thermo Scientific) and a ChemiDoc Imager (BioRad). Band intensity was calculated using Quantity One software (BioRad) and normalized to the respective control proteins. Three independent replicates were analyzed for each experiment.
## Confocal microscopy, immunofluorescence, quantification of ER morphology, and nonstructural proteins colocalization
HeLa cells cultivated on glass coverslips in 6-well plates were adsorbed at 37°C with reovirus at an MOI of 1 PFU/cell and incubated at 37°C for 1 h. For experiments probing events at early times post-adsorption (7, 8, 9, and 10 h post-adsorption), cells were adsorbed on ice for 30 min to synchronize the onset of infection. Following incubation, cells were fixed with 4% PFA in PBS (pH 7.4) at RT for 20 min, permeabilized with 0.25% saponin, and blocked in saturation buffer (0.25% saponin, 2% FBS in PBS) for 30 min. Antibodies and probes were diluted in saturation buffer as follows: 1:1,000 for rabbit σNS-specific antibody VU82, 1:200 for guinea pig σNS-specific antiserum (21), chicken µNS-specific antiserum, rabbit giantin-specific polyclonal antiserum (BioLegend), 4′,6-diamidino-2-phenylindole (DAPI), and phalloidin (Thermo Fisher Scientific), 1:50 for mouse ANXA2-specific monoclonal antibody (Abcam), and 1:500 for Alexa Fluor-conju gated secondary antibodies (Invitrogen). Images were obtained using a Leica Stellaris 5 confocal microscope, followed by a deconvolution processing using the integrated module Lightning in LAS X software (Leica Microsystems).
Confocal images obtained for 3D reconstruction were processed and built using the 3D integrated module in LAS X software. Quantification of ER morphology changes was conducted using confocal images of mock-infected WT cells (n = 46), reovirus-infected WT cells (n = 46), mock-infected ANXA2 KO cells (n = 49), and reovirus-infected ANXA2 KO cells (n = 41). ER structures were classified in four categories: (i) regular ER, (ii) fragmented ER, (iii) unbranched ER, and (iv) collapsed ER. Images were obtained using a Leica Stellaris 5 confocal microscope at a magnification of x63 or x100 using LAS X software.
Confocal images used for colocalization studies were processed using the JACoP plugin (74) in FIJI software (75). We studied σNS and µNS colocalization in 36 VFs per condition. Pearson's and Mander's coefficients were obtained for each VF. Mander's coefficients were calculated as Mander's 1 (fraction of red signal [σNS] overlapping green signal [µNS]) and Mander's 2 (fraction of green signal [µNS] overlapping red signal [σNS]).
## Quantification of reovirus infection
Cells were adsorbed with reovirus at an MOI of 1 PFU/cell at 37°C for 1 h. The inoculum was removed, and cells were incubated at 37°C for various intervals. Cells were fixed with 4% PFA in PBS (pH 7.4) at RT for 20 min, permeabilized with 0.25% saponin, and blocked with 0.25% saponin and 2% FBS in PBS. Fixed and permeabilized cells were incubated sequentially with chicken µNS-specific antiserum and Alexa Fluor 488-conju gated chicken-specific antibody in PBS containing 0.25% saponin and 2% FBS. Stained cells were imaged using a Leica DMI6000B fluorescence microscope equipped with a 40× air objective. Ten fields of view per condition were imaged by an observer blinded to the conditions of the experiment. In each field of view, 25-75 cells were quantified. Cells with at least one VF were considered infected. Cells with VFs were quantified using LAS X software.
## Transmission electron microscopy
Cells were propagated in six-well plates, infected with reovirus, and processed for embedding in resin as described (1,6). Cells were incubated for 24 h and fixed with a mixture of 4% PFA and 1% glutaraldehyde in 0.4 M HEPES buffer, pH 7.4, at RT for 1 h. Cells were postfixed at 4°C for 1 h with a mixture of 1% osmium tetroxide and 0.8% potassium ferricyanide in water and dehydrated in 5-minute steps with increas ing concentrations of acetone (50%, 70%, 90%, and twice in 100%) at 4°C. Cells were processed for embedding in the epoxy resin EML-812 (TAAB Laboratories) by incubation overnight with a 1:1 mixture of acetone and resin at RT. After embedding in 100% resin for 8 h, samples were polymerized at 60°C for 48 h. Serial sections of 50-70 nm thickness were obtained using a UC6 ultramicrotome (Leica Microsystems) and collected on uncoated 300-mesh copper grids (TAAB Laboratories). Sections were stained with 4% uranyl acetate and Reynold's lead citrate prior to TEM imaging. Images were acquired using a 100 kV JEOL JEM 1011 TEM, equipped with a Gatan ES1000WW digital camera.
## Reovirus nonstructural protein transfections and ANXA2 complementation
WT and KO cells were transfected with σNS, µNS, or both expression plasmids in combination with the mCherry-ER-3 plasmid expressing mCherry fused with calreticulin, ER signal peptide, and KDEL (Addgene) using Fugene HD (Promega) according to the manufacturer's instructions. At 24 h post-transfection, cells were fixed with 4% parafor maldehyde (PFA) in PBS at RT for 20 min. The σNS and µNS proteins were imaged using confocal immunofluorescence microscopy.
To complement ANXA2 in ANXA2 KO cells, we used the ANXA2-GFP (Addgene #107196) and ANXA2 cmamxA2-GFP (Addgene # 107197) plasmids encoding WT ANXA2 and an ANXA2 mutant in which all type II calcium-binding sites are deleted, respectively. The GFP tag was deleted in both plasmids prior to use. Linear fragments of ANXA2 and ANXA2-cmamxA2 flanked by NheI and XhoI restriction enzyme sites were ampli fied using PCR and primers ANXA2-NheI-F (5′-TAAGCAGCTAGCACCATGTCTACTGTTCAC GAAATC-3′) and ANXA2-XhoI-R (5′-TGCTTACTCGAGTCAGTCATCTCCACCACACAGG-3′) to engineer untagged ANXA2 plasmids. PCR fragments were cloned into pcDNA3.1 using NheI (NEB, R3131S) and XhoI (NEB, R0146S).
ANXA2 KO cells were either mock-transfected or transfected with engineered plasmids in combination with the mCherry-ER-3 plasmid using Fugene HD (Promega). At 24 h post-transfection, cells were fixed with 4% PFA in PBS at RT for 20 min and imaged using confocal microscopy.
## Minimum distance analysis
Minimum distance analyses were conducted using the Z-stack analyzer plug-in for FIJI software (75). This method was developed to evaluate the spatial distribution of two fluorescence channels within a specified region of interest. It determines whether these channels are independent or follow the same spatial distribution, going beyond colocalization analysis by identifying dependencies between channels that do not necessarily colocalize. It operates by computing the minimum distance from a particle in one channel (Marker A) to the flipped particles in the other channel (Marker B), simulating a scenario where two channels are independent. The minimum distances calculated in this scenario serve as a baseline for comparison (Fig. S8A). Then the Kolmogorov-Smirnov test is used to compare the minimum distance distributions in the actual and simulated scenarios. The test computes a P-value, which denotes the probability that the two distributions are the same. A small P-value indicates significant distribution differences, suggesting a spatial dependency among the channels, even when one channel's particles are reversed.
WT and KO cells were mock-infected or reovirus-infected, at 14 h post-adsorption, fixed with 4% PFA in PBS at RT for 20 min, and processed for immunofluorescence. From the collected confocal images, the distances between the ER, NS proteins, actin, and ANXA2 fluorescent channels (in groups of two by two) were measured and quantified. The ratio of adjacent signals was calculated by taking all measurements of distances between channels from 0 to 150 nm and dividing by the total measurements (all lengths) (Fig. S8B).
To verify the robustness of the method, minimal measurements were made between the fluorescence channel marking the nucleus and the Golgi in all conditions, as well as measurements between the µNS channel and the Golgi channel in infected cells. In all cases, no significant differences were observed (Fig. S8C andD).
## Quantification and statistical analysis
All experiments were conducted with a minimum of three independent replicates. The data are presented as the mean ± s.e.m. Statistical significance was determined using a two-sample unequal variance t-test with a two-tailed distribution (α = 0.05). Graphs and statistical analyses were conducted using Microsoft Excel.
## References
1. Fernandez De Castro, Tr (2020) "Virus factories"
2. Su, Wilson, Samuel et al. (2021) "Formation and function of liquid-like viral factories in negative-sense single-stranded RNA virus infections" *Viruses*
3. Netherton, Wileman (2011) "Virus factories, double membrane vesicles and viroplasm generated in animal cells" *Curr Opin Virol*
4. Nevers, Albertini, Lagaudrière-Gesbert et al. (2020) "Negri bodies and other virus membrane-less replication compartments" *Biochim Biophys Acta Mol Cell Res*
5. De Castro, Zamora, Ooms et al. (2014) "Reovirus forms neo-organelles for progeny particle assembly within reorganized cell membranes" *mBio*
6. Tenorio, De Castro, Knowlton et al. (2018) "Reovirus σNS and μNS proteins remodel the endoplasmic reticulum to build replication neo-organelles" *mBio*
7. Dolnik, Gerresheim, Biedenkopf (2021) "New perspectives on the biogenesis of viral inclusion bodies in negative-sense RNA virus infections" *Cells*
8. De Castro, Tenorio, Risco (2016) "Virus assembly factories in a lipid world" *Curr Opin Virol*
9. Nguyen-Dinh, Herker (2021) "Ultrastructural features of membranous replication organelles induced by positive-stranded RNA viruses" *Cells*
10. Romero-Brey, Bartenschlager (2016) "Endoplasmic reticulum: the favorite intracellular niche for viral replication and assembly" *Viruses*
11. Wang, Chen, Qi et al. (2024) "Mechanism, structural and functional insights into nidovirus-induced double-membrane vesicles" *Front Immunol*
12. Zhang, Lan, Sanyal (2020) "Membrane heist: coronavirus host membrane remodeling during replication" *Biochimie*
13. He, Li, Bernards (2023) "Manipulation of the cellular membrane-cytoskeleton network for RNA virus replication and movement in plants" *Viruses*
14. Kovalev, De Castro Martín, Pogany et al. (2016) "Role of viral RNA and co-opted cellular ESCRT-I and ESCRT-III factors in formation of tombusvirus spherules harboring the tombusvirus replicase" *J Virol*
15. Weisberg, Maruri-Avidal, Bisht et al. (2017) "Enigmatic origin of the poxvirus membrane from the endoplasmic reticulum shown by 3D imaging of vaccinia virus assembly mutants" *Proc Natl Acad Sci*
16. Cortese, Goellner, Acosta et al. (2017) "Ultrastructural characterization of zika virus replication factories" *Cell Rep*
17. Dermody, Parker, Sherry (2023) "Fields virology"
18. Liu, Xia, Martynowycz et al. (2024) "Molecular sociology of virus-induced cellular condensates supporting reovirus assembly and replication" *Nat Commun*
19. Bouziat, Hinterleitner, Brown et al. (2017) "Reovirus infection triggers inflammatory responses to dietary antigens and development of celiac disease" *Science*
20. Desmet, Anguish, Parker (2014) "Virus-mediated compartmen talization of the host translational machinery" *mBio*
21. Becker, Peters, Dermody (2003) "Reovirus sigma NS and mu NS proteins form cytoplasmic inclusion structures in the absence of viral infection" *J Virol*
22. Tenorio, De Castro, Knowlton et al. (2019) "Function, architecture, and biogenesis of reovirus replication neoorganelles" *Viruses*
23. Bharadwaj, Bydoun, Holloway et al. (2013) "Annexin A2 heterotetramer: structure and function" *Int J Mol Sci*
24. Parker, Broering, Kim et al. (2002) "Reovirus core protein mu2 determines the filamentous morphology of viral inclusion bodies by interacting with and stabilizing microtubules" *J Virol*
25. Rentero, Blanco-Muñoz, Meneses-Salas et al. (2018) "Annexins-coordinators of cholesterol homeostasis in endocytic pathways" *Int J Mol Sci*
26. Konopka-Anstadt, Mainou, Sutherland et al. (2014) "The Nogo receptor NgR1 mediates infection by mammalian reovirus" *Cell Host Microbe*
27. Ortega-Gonzalez, Taylor, Jangra et al. (2025) *Full-Length Text Journal of Virology*
28. Sachse, Risco, Dermody (2022) "Reovirus infection is regulated by NPC1 and endosomal cholesterol homeostasis" *PLoS Pathog*
29. Grieve, Moss, Hayes (2012) "Annexin A2 at the interface of actin and membrane dynamics: a focus on its roles in endocytosis and cell polarization" *Int J Cell Biol*
30. Strand, Hollås, Sakya et al. (2021) "Annexin A2 binds the internal ribosomal entry site of c-myc mRNA and regulates its translation" *RNA Biol*
31. Miller, Broering, Parker et al. (2003) "Reovirus sigma NS protein localizes to inclusions through an association requiring the mu NS amino terminus" *J Virol*
32. Hayes, Rescher, Gerke et al. (2004) *Annexin-actin interactions. Traffic*
33. Morel, Parton, Gruenberg (2009) "Annexin A2-dependent polymerization of actin mediates endosome biogenesis" *Dev Cell*
34. Joensuu, Jokitalo (2015) "ER sheet-tubule balance is regulated by an array of actin filaments and microtubules" *Exp Cell Res*
35. Kee, Bryce, Yang et al. (2017) "ER/Golgi trafficking is facilitated by unbranched actin filaments containing Tpm4" *Cytoskeleton (Hoboken)*
36. Rescher, Ludwig, Konietzko et al. (2008) "Tyrosine phosphorylation of annexin A2 regulates Rho-mediated actin rearrangement and cell adhesion" *J Cell Sci*
37. Taylor, Fernandez, Thornton et al. (2018) "Heterotetrameric annexin A2/S100A10 (A2t) is essential for oncogenic human papillomavirus trafficking and capsid disassembly, and protects virions from lysosomal degradation" *Sci Rep*
38. Rescher, Zobiack, Gerke (2000) "Intact Ca(2+)-binding sites are required for targeting of annexin 1 to endosomal membranes in living HeLa cells" *J Cell Sci*
39. Malhotra, Ward, Bright et al. (2003) "Isolation and characterisation of potential respiratory syncytial virus receptor(s) on epithelial cells" *Microbes Infect*
40. Song, Li, Han et al. (2024) "Host restriction factor Rab11a limits porcine epidemic diarrhea virus invasion of cells via fusion peptide-mediated membrane fusion" *Int J Biol Macromol*
41. Backes, Quinkert, Reiss et al. (2010) "Role of annexin A2 in the production of infectious hepatitis C virus particles" *J Virol*
42. Saxena, Lai, Chao et al. (2012) "Annexin A2 is involved in the formation of hepatitis C virus replication complex on the lipid raft" *J Virol*
43. Ryzhova, Vos, Albright et al. (2006) "Annexin 2: a novel human immunodeficiency virus type 1 Gag binding protein involved in replication in monocyte-derived macro phages" *J Virol*
44. (2006)
45. Huang, Chi, Chiu et al. (2017) "Avian reovirus p17 and σA act cooperatively to downregulate Akt by suppressing mTORC2 and CDK2/cyclin A2 and upregulating proteasome PSMB6" *Sci Rep*
46. Bhattacharya, Roy (2013) "Cellular phosphoinositides and the maturation of bluetongue virus, a non-enveloped capsid virus" *Virol J*
47. Huang, Wu, Liao et al. (2023) "Cell entry of avian reovirus modulated by cell-surface annexin A2 and adhesion G proteincoupled receptor latrophilin-2 triggers Src and p38 MAPK signaling enhancing caveolin-1-and dynamin 2-dependent endocytosis" *Microbiol Spectr*
48. Wang, Zhao, Zhang et al. (2024) "Annexin A2: a double-edged sword in pathogen infection" *Pathogens*
49. Beaton, Rodriguez, Reddy et al. (2002) "The membrane trafficking protein calpactin forms a complex with bluetongue virus protein NS3 and mediates virus release" *Proc Natl Acad Sci*
50. Meiring, Bryce, Wang et al. (2018) "Co-polymers of actin and tropomyo sin account for a major fraction of the human actin cytoskeleton" *Curr Biol*
51. Mainou, Zamora, Ashbrook et al. (2013) *Reovirus cell entry requires functional microtubules. mBio*
52. Fiske, Brigleb, Sanchez et al. (2024) "Strain-specific differences in reovirus infection of murine macrophages segregate with polymorphisms in viral outer-capsid protein σ3" *J Virol*
53. Shang, Simpson, Taylor et al. (2023) "Paired immunoglobulin-like receptor B is an entry receptor for mammalian orthoreovirus" *Nat Commun*
54. Yuan, Yang, Lan et al. (2020) "Paired immunoglobulinlike receptor B inhibition in muller cells promotes neurite regeneration after retinal ganglion cell injury in vitro" *Neurosci Bull*
55. Nunzio (2023) "Stress-induced condensate switch awakens sleeping viruses" *Cell Host Microbe*
56. Glon, Léonardon, Guillemot et al. (2024) "Biomolecular condensates with liquid properties formed during viral infections" *Microbes Infect*
57. Lopez, Camporeale, Salgueiro et al. (2021) "Deconstructing virus condensation" *PLoS Pathog*
58. Vazquez, Toledo, Gianotti et al. (2022) "Protein conformation and biomolecular condensates" *Curr Res Struct Biol*
59. Kim, Yun, Lee et al. (2024) "Interplay between membranes and biomolecular condensates in the regulation of membrane-associated cellular processes" *Exp Mol Med*
60. Snead, Gladfelter (2019) "The control centers of biomolecular phase separation: how membrane surfaces, PTMs, and active processes regulate condensation" *Mol Cell*
61. Mohl, Roy (2016) "Cellular casein kinase 2 and protein phosphatase 2A modulate replication site assembly of bluetongue virus" *J Biol Chem*
62. Zhang, Li, Lu et al. (2022) "Structural and functional analysis of the small GTPase ARF1 reveals a pivotal role of its GTPbinding domain in controlling of the generation of viral inclusion bodies and replication of grass carp reovirus" *Front Immunol*
63. Campagna, Budini, Arnoldi et al. (2007) "Impaired hyperphosphorylation of rotavirus NSP5 in cells depleted of casein kinase 1alpha is associated with the formation of viroplasms with altered morphology and a moderate decrease in virus replication" *J Gen Virol*
64. Papa, Venditti, Arnoldi et al. (2019) "Recombinant rotaviruses rescued by reverse genetics reveal the role of NSP5 hyperphosphorylation in the assembly of viral factories" *J Virol*
65. Campagna, Villar, Arnoldi et al. (2013) "Rotavirus viroplasm proteins interact with the cellular SUMOylation system: implications for viroplasm-like structure formation" *J Virol*
66. (2025) *Full-Length Text Journal of Virology*
67. Criglar, Crawford, Zhao et al. (2020) "A genetically engineered rotavirus NSP2 phosphorylation mutant impaired in viroplasm formation and replication shows an early interaction between vNSP2 and cellular lipid droplets" *J Virol*
68. Knowlton, De Castro, Ashbrook et al. (2018) "The TRiC chaperonin controls reovirus replication through outer-capsid folding" *Nat Microbiol*
69. Kaufer, Coffey, Parker (2012) "The cellular chaperone hsc70 is specifically recruited to reovirus viral factories independently of its chaperone function" *J Virol*
70. Doench, Fusi, Sullender et al. (2016) "Optimized sgRNA design to maximize activity and minimize offtarget effects of CRISPR-Cas9" *Nat Biotechnol*
71. Kobayashi, Antar, Boehme et al. (2007) "A plasmid-based reverse genetics system for animal double-stranded RNA viruses" *Cell Host Microbe*
72. Furlong, Nibert, Fields (1988) "Sigma 1 protein of mammalian reoviruses extends from the surfaces of viral particles" *J Virol*
73. Virgin Hw 4th, Bassel-Duby, Fields et al. (1988) "Antibody protects against lethal infection with the neurally spreading reovirus type 3 (Dearing)" *J Virol*
74. Kobayashi, Chappell, Danthi et al. (2006) "Gene-specific inhibition of reovirus replication by RNA interference" *J Virol*
75. Kwon, Firestein (2013) "DNA transfection: calcium phosphate method" *Methods Mol Biol*
76. Becker, Goral, Hazelton et al. (2001) "Reovirus sigmaNS protein is required for nucleation of viral assembly complexes and formation of viral inclusions" *J Virol*
77. Bolte, Cordelières (2006) "A guided tour into subcellular colocaliza tion analysis in light microscopy" *J Microsc*
78. Schindelin, Arganda-Carreras, Frise et al. (2012) "Fiji: an opensource platform for biological-image analysis" *Nat Methods*
79. Perez-Riverol, Bandla, Kundu et al. (2025) "The PRIDE database at 20 years: 2025 update" *Nucleic Acids Res*
80. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793369&blobtype=pdf
|
Tianyan Hu, Alon Yehoshua, Joseph Cappelleri, Meghan Gavaghan, Manuela Fusco, Xiaowu Sun
## Abstract
Background. This study characterized Health-Related Quality of Life (HRQoL) and Work Productivity and Impairment (WPAI) among outpatient symptomatic adults during the first week of a test-confirmed influenza infection in the US in the 2024/25 respiratory season.Methods. Symptomatic adults with test-confirmed influenza infection were enrolled at CVS Health between 10/24/2024-4/15/2025 (CT.gov: NCT05160636). Questionnaires on socio-demographics and clinical characteristics were administered to participants via an online survey platform at enrollment. Validated instruments (EQ-5D-5L, WPAI-GH) were used to assess HRQoL and WPAI for the pre-infection baseline (through recall at enrollment) and at Week 1. Outcomes were summarized using descriptive statistics for each time point and compared between time points using paired t-tests.Results. Of 720 participants, mean age was 42.0 (SD: 13.0), 73.8% were female, 47.1% had at least one comorbidity, and 588 (81.7%) were employed at baseline. The mean (SD) of EQ-Visual Analogue Scale (EQ-VAS) and Utility Index (UI) scores were 89.7 (SD: 7.8) and 0.96 (SD: 0.07), respectively, at the pre-infection baseline. Both scores statistically significantly decreased to 84.9 (SD: 11.0) for EQ-VAS and 0.94 (SD: 0.10) for UI scores at Week 1, with a mean change of -4.9 (p< 0.001) for EQ-VAS, and -0.02 (p< 0.001) for UI scores. Relative to baseline, the mean changes at Week 1 were all statistically significant, with 55.0% (p< 0.001) in work productivity time loss, 33.2% (p< 0.001) in absenteeism, 42.2% (p< 0.001) in presenteeism, and 45.3% (p< 0.001) in activity impairment.Conclusion. Influenza infections negatively impacted the health-related quality of life and especially work productivity among US adults during the 2024/25 season up to the first week of infection. These findings underscore the broad and lasting consequences of influenza infections and help raise awareness of the value of preventive measures such as influenza vaccination.Disclosures.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645913&blobtype=pdf
|
# Development of a cross-protective common cold coronavirus vaccine
Tanushree Dangi, Shiyi Li, Pablo Penaloza-Macmaster, Pablo Penaloza- Macmaster
## Abstract
Common cold coronaviruses, such as OC43 and HKU1, typically cause mild respiratory infections in healthy people. However, they can lead to severe illness in high-risk groups, including immunocompromised individuals and older adults. Currently, there is no clinically approved vaccine to prevent infection by common cold corona viruses. Here, we developed an mRNA vaccine expressing a stabilized spike protein derived from OC43 coronavirus and tested its efficacy in different challenge models in C57BL/6 mice. This novel OC43 vaccine elicited OC43-specific immune responses, as well as cross-reactive immune response against other embecoviruses, including HKU1 and mouse hepatitis virus (MHV-A59). Interestingly, this OC43 vaccine protected mice not only against a lethal OC43 infection but also against a distant embecovirus, MHV-A59. These findings provide insights for the development of common cold coronavirus vaccines, demonstrating their potential to protect against various coronaviruses. IMPORTANCE Human coronaviruses like OC43 cause disease in vulnerable populations, yet no approved vaccines exist. We developed an mRNA vaccine targeting the OC43 spike protein that protects mice not only against homologous OC43 challenges but also against the distantly related embecovirus MHV-A59. These findings demonstrate the feasibility of a single vaccine conferring broad protection across multiple coronaviruses within the same subgenus, advancing strategies for pan-coronavirus vaccine develop ment.KEYWORDS coronavirus, HCoV-OC43, vaccines, cross-protective immunity S evere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in over 7 million deaths globally. Despite this high number of deaths, the rapid develop ment of SARS-CoV-2 vaccines reduced the death toll of the pandemic. In addition to SARS-CoV-2, several endemic coronaviruses like OC43, 229E, HKU1, and NL63 circulate within the human population, causing frequent re-infections. While these endemic coronaviruses typically cause mild to moderate upper respiratory infections, they can also lead to severe conditions, pneumonia, and bronchitis, particularly in the elderly and immunocompromised individuals (1-6). Among these, the human common cold coronavirus, OC43, causes frequent respiratory infections worldwide, but no effective vaccine is currently available.The widespread circulation of OC43 poses public health concerns due to its propensity for mutation and potential recombination with other coronaviruses. OC43 can cause a burden on healthcare systems, resulting in economic losses. The high incidence of OC43 re-infections further highlights the need for effective vaccines (7-10). Addressing these challenges with an effective vaccine would not only improve overall health in the population but also have the potential to alleviate the economic burden associated with recurrent coronavirus infections. Here, we develop an mRNA vaccine encoding a stabilized OC43 spike protein. Our results demonstrate that this vaccine is
immunogenic and highly protective not only against OC43, but also a distant embecovi rus.
## RESULTS
## Design of a novel mRNA-LNP vaccine encoding HCoV-OC43 spike glycopro tein (mRNA-OC43)
We developed an mRNA vaccine encoding the full-length spike glycoprotein of the OC43 coronavirus. We utilized the spike protein sequence from NCBI (AAA03055.1) and introduced two proline mutations at 1070 and 1071 amino acid positions (replaced with A and L amino acids) in the S2 domain to stabilize the spike protein in its pre-fusion state. A pcDNA3.1 (+) plasmid construct was designed by incorporating the codon-optimized spike gene sequence flanked with untranslated regions (UTRs) at 3′ and 5′ ends, and a T7 RNA polymerase site before the 3′ end of coding sequence. Expression of OC43 spike glycoprotein was confirmed by transfecting HEK293T cells with the corresponding mRNA followed by Western Blot analysis to validate protein expression (Fig. 1). mRNA molecules were then encapsulated into a lipid nanoparticle using a Nanoassemblr.
## Immunogenicity and protective efficacy of mRNA-OC43 following homolo gous OC43 challenge
To evaluate immunogenicity, we first immunized C57BL/6 mice intramuscularly with 3 µg of mRNA-OC43 (Fig. 2A). After vaccination, we measured antibody and T cell responses in plasma, using enzyme-linked immunosorbent assay (ELISA) with OC43 spike as coating antigen, and intracellular cytokine assay (ICS) using overlapping OC43 spike peptide pools. The mRNA-OC43 vaccine elicited potent antibody responses to the OC43 spike protein (Fig. 2B). Moreover, the vaccine elicited CD8 and CD4 T-cell responses by ICS (Fig. 2C through G).
To examine vaccine efficacy, mice were challenged intranasally with 50 µL of OC43 neurovirulent (NV) strain (1 × 10 10 genome copies) at week 2 post-vaccination. Mice were monitored daily for any changes in clinical signs or symptoms, body mass, and mortality (Fig. 3A). Upon OC43 challenge, unvaccinated mice exhibited severe weight loss, severe clinical pathology, and showed only 20% survival (Fig. 3B through E). In contrast, mice that received the OC43 vaccine showed no weight loss and 100% survival, with no clinical signs of disease (Fig. 3F).
## Protective efficacy of mRNA-OC43 following a heterologous coronavirus challenge
Further, we interrogated whether the mRNA-OC43 vaccine could cross-protect against another embecovirus (MHV-A59). MHV-A59 is a well-studied mouse virus (11)(12)(13). OC43 and MHV-A59 share only ~65% sequence identity in their spike proteins, rendering MHV-A59 a stringent challenge model to examine cross-protection by our mRNA-OC43 vaccine (Fig. 4). While MHV-A59 is not considered a significant threat to humans, it serves as a useful proof-of-principle model to evaluate the protective breadth of our OC43 vaccine.
To interrogate the cross-protective efficacy of the OC43 vaccine, mice were immu nized intramuscularly with 3 µg of the mRNA-OC43 vaccine. After 2 weeks post-vaccina tion, mice were challenged intraperitoneally (i.p.) with 2 × 10 6 plaque forming units (PFU) of MHV-A59 (Fig. 5A). All mice experienced weight loss following infection, but the mice that were vaccinated with the mRNA-OC43 vaccine exhibited significantly less weight loss compared with control (Fig. 5B). Further, the clinical signs were significantly milder in mRNA-OC43 vaccinated mice (Fig. 5C). MHV infection is very transient in C57BL/6 mice. In our hands, all mice resolve acute MHV infection after 5-7 days following intraperitoneal challenge without succumbing to infection, regardless of immunization status. There fore, we selected day 3 post-challenge as the optimal time for evaluating viral loads in vaccinated mice. All mice were sacrificed at day 3 post-infection for assessing viral load in tissues. Importantly, the mRNA-OC43 vaccinated mice showed enhanced viral control in lung, brain, and liver (Fig. 5D through F). These data suggest that the mRNA-OC43 vaccine provides cross-protection against a heterologous coronavirus with only a 65% antigen match.
## Humoral responses elicited by mRNA-OC43 confer cross-protection against MHV-A59
Vaccine protection is typically mediated by humoral and cellular responses. To specifically assess the role of humoral protection, we performed a passive immunization study (Fig. 5G). First, we immunized C57BL/6 mice with the mRNA-OC43 vaccine on days 0 and 21 and then collected immune plasma on day 35. Prior to adoptive transfer, antibody titers specific to the OC43 spike antigen were confirmed in immune plasma using ELISA (Fig. 5H). These mRNA-OC43-immune plasma also exhibited cross-reactivity against an MHV-spike antigen encoded by HEK293T cell lysate (Fig. 5I). Each recipient mouse received 800 µL of pooled plasma via i.p. injection. Control mice received plasma from naïve animals. On the following day, all recipient mice were challenged i.p. with 2 × 10 6 PFU of MHV-A59. Mice were monitored for three consecutive days for weight loss and Next, we investigated whether T cells elicited by the mRNA-OC43 vaccine contribute to the control of MHV-A59 (Fig. 5N). To assess the role of CD8 + T cells, we depleted them at the time of infection, using depleting antibodies. CD8 + T cell depletion had no significant effect on viral control (Fig. 5O). Similarly, in separate experiments, we depleted CD4 + T cells to determine whether they impaired vaccine protection. CD4 + T cell depletion also did not impair vaccine protection (Fig. 5P). These data suggest that T cells are dispensable for mRNA-OC43 vaccine cross-protection against MHV, although it is important to clarify that depleting antibodies may not fully deplete all T cells in tissues (14). Moreover, functional compensation between CD4 and CD8 T cells remains a possible explanation.
## mRNA-MHV vaccine confers heterologous protection against OC43
We have shown that an mRNA-OC43 vaccine confers heterologous protections against MHV. We also performed the "inverse" vaccination challenge study. Mice were immu nized intramuscularly with an mRNA-MHV vaccine followed by a lethal challenge with OC43 (Fig. 6A). On day 15 after vaccination, antibody and T cell responses were measured. As expected, the mRNA-MHV vaccine induced antibody responses against its matched antigen, the MHV spike protein (Fig. 6B). Interestingly, this mRNA-MHV vaccine also elicited cross-reactive antibody responses against other coronaviruses, including OC43, HKU1, and SARS-CoV-2 (Fig. 6C through E). This vaccine elicited MHV-specific CD8 + T cell responses (K b S598) (Fig. 6F andG), and also cross-reactive OC43-specific CD8 + T cell responses by ELISpot assays (Fig. 6H).
To evaluate cross-protective efficacy by the MHV vaccine, mice were intranasally challenged with OC43 (NV strain) 3 weeks after vaccination and then monitored for weight loss and clinical score. Following an OC43 challenge, unvaccinated mice exhibited more significant weight loss and worse disease compared with vaccinated mice (Fig. 6I andJ). There was also a pattern of improved survival with the mRNA-MHV vaccine relative to control, but the difference was not statistically significant (Fig. 6K). These results suggest that an mRNA-MHV vaccine confers partial protection against a distant OC43 coronavirus.
## DISCUSSION
There are four endemic human coronaviruses that typically cause mild respiratory infections. These include two alphacoronaviruses (HCoV-229E and HCoV-NL63) and two betacoronaviruses (HCoV-OC43 and HKU1), which are both part of the embecovirus sublineage. In this study, we focused on OC43, given the availability of a mouse model and the fact that it accounts for a great fraction of common cold coronavirus infections in humans (15,16). OC43 belongs to the betacoronavirus genus, alongside SARS-CoV-2, SARS-CoV, and MERS-CoV, which were responsible for outbreaks in 2019, 2003, and 2012, respectively.
While previous studies have shown that coronavirus vaccines can generate crossreactive antibodies against endemic coronaviruses, it remains unclear whether these antibodies are cross-protective in vivo (17)(18)(19)(20). Building on this, we hypothesized that an mRNA vaccine targeting the human common cold coronavirus OC43, for which no effective vaccine currently exists, could also confer cross-protection against other coronaviruses. To test this hypothesis, we developed a novel mRNA vaccine encoding a stabilized OC43 spike protein and assessed its immunogenicity and protective efficacy in vivo. This mRNA-OC43 vaccine elicited adaptive immune responses to OC43 and conferred protection against OC43 infection, which was expected given that the vaccine antigen was matched to the challenge antigen. However, an interesting finding was that the mRNA-OC43 vaccine also provided cross-protection against MHV-A59, despite OC43 and MHV-A59 having only 65% identity in their spike proteins. Notably, antibodies induced by the OC43 vaccine exhibited cross-reactivity with MHV-spike. This antibodymediated cross-reactivity was further supported by adoptive plasma transfer experi ments, which showed that antibodies elicited by the mRNA-OC43 vaccine conferred protection to mice against MHV infection, underscoring the importance of humoral immunity in cross-protection. Interestingly, unlike previous reports that highlight a key role for virus-specific CD8 + and CD4 + T cells in cross-protection (21)(22)(23)(24), our model did not reveal a cross-protective role for these T cell subsets in the context of MHV infection. "This could be because T cell-depleting antibodies do not reach all tissues, rendering some tissue-resident memory T cells (Trms) undepleted" (14). In addition, it is possible that depletion of CD8 T cells results in functional compensation by CD4 T cells, and vice versa.
Thus, the depleting antibodies used in our study may have effectively eliminated T cells in the blood, but not in tissues, potentially obscuring their contribution to crossprotection.
To further substantiate our findings, we immunized mice with an mRNA vaccine encoding the MHV spike protein and evaluated its protective efficacy against OC43 infection. This also conferred cross-protection, as demonstrated by reduced weight loss, milder clinical symptoms, and increased survival rates in vaccinated animals. Interest ingly, this mRNA-MHV vaccine elicited cross-binding antibody responses to multiple betacoronaviruses, including OC43, HKU1, and SARS-CoV-2, as well as cross-reactive T-cell responses targeting the OC43 spike protein. These results suggest that vaccines targeting a single coronavirus strain can confer broad protection against other coronavi ruses from the same embecovirus subgenus. These data may be important for improving vaccine preparedness against circulating and emerging coronaviruses, for example, by pre-emptively developing vaccines to representative coronaviruses from each subgenus.
## MATERIALS AND METHODS
## Mice and Immunizations
Six -to 8-week-old C57BL/6 mice were used. Mice were purchased from Jackson laboratories (approximately half males and half females). Mice were immunized intramuscularly with mRNA-LNPs (made in-house) diluted in sterile PBS. Mice were housed at Northwestern University's Center for Comparative Medicine (CCM).
## Synthesis of modified mRNA
We synthesized mRNA vaccines encoding for the codon-optimized OC43 spike protein from HCoV-OC43 (accession number AAA03055.1) and codon-optimized MHV-spike protein from MHV-JHM strain (accession number YP_209233.1). For in vitro transcription of mRNA (IVT-mRNA), plasmid constructs were designed by incorporating codon-opti mized immunogens (OC43-spike or MHV-spike), UTRs, and phase T7 RNA polymerase promoter and purchased from Genscript. The sequences of the 5′-and -3′-UTRs were identical to those used in a previous publication (20). Modified nucleotide pseudouridine-5′-triphosphate (ΨTP), along with canonical nucleotides ATP, CTP, and GTP (CellScript, Cat. No. ICTY110510), was used to synthesize nucleoside-modified IVT-mRNA from the plasmid construct. To enhance mRNA stability, an N7-methylguanosine cap (Cap 1, m⁷G) was added to the 5′ end, and a ~ 150-nucleotide poly(A) tail was incor porated at the 3′ end using CellScript Capping and Tailing Kits (Cat. Nos. SCCS1710 and PAP5104H). The IVT-mRNA was purified via ammonium acetate precipitation and quantified using a NanoDrop ONE spectrophotometer (Thermo Scientific). To evaluate protein expression, purified mRNA was transfected into female human embryonic kidney (HEK) 293T cells using the TransIT-mRNA Transfection Kit (Mirus, Cat. No. MIR2250). Cell lysates from transfected HEK293T cells were analyzed by Western blot to confirm spike protein expression. Following confirmation, the mRNA was encapsulated in lipid nanoparticles as described below.
## mRNA-LNP formulation
All purified mRNAs generated above were encapsulated into lipid nanoparticles using the NanoAssemblr Benchtop system (Precision NanoSystems) and confirmed to have similar encapsulation efficiency (∼95%). In brief, mRNA was diluted in 50 mM sodium acetate buffer, pH 5.0 to achieve a working concentration of 0.096 mg/mL (Cayman Chemical, Cat. No. 35425). An ethanolic lipid mixture was prepared using four lipids -SM-102, 1,2-distearoyl-sn-glycero-3-PC, cholesterol, and DMG-PEG (Cayman, Cat. No. 35425) in a molar ratio of 50:10:38.5:0.38. Subsequently, diluted mRNA in an aqueous phase and lipid mixture was run through a microfluidic laminar flow cartridge (Nano Assemblr Ignite NxGen, Cat. No. NIN0061). This was done by maintaining a nitrogen-tophosphate (N/P) ratio of 4.0 (Lipid mix to mRNA ratio of 4), an RNA-to-lipid flow ratio of 3:1, and a total flow rate of 12 mL/min to generate mRNA-lipid nanoparticles (mRNA-LNPs). The resulting mRNA-LNPs were concentrated and purified using an Amicon Ultra-15 filtration unit and a 0.2-µm Acrodisc filter. Encapsulation efficiency and the concentration of encapsulated mRNA were determined using the Quant-iT RiboGreen RNA Assay Kit (Invitrogen, Cat. No. R11490).
## Reagents, flow cytometry, and equipment
To determine the T-cell responses in the blood and spleen, single-cell suspensions of PBMCs and spleen were prepared as described previously (25). Dead cells were gated out using Live/Dead fixable dead cell stain (Invitrogen). The CD8 and CD4 responses specific to the OC43 spike were measured by stimulating splenocytes with the OC43 spike peptide pools (NR-53728, BEI) in intracellular cytokine staining (ICS). Biotinylated MHC class I monomer (K b S598, sequence RCQIFANI) was used for detecting MHV spike-specific CD8 T cells and was obtained from the NIH tetramer facility at Emory University. Cells were stained with fluorescently labeled antibodies against anti-mouse CD8α (53-6.7 on PerCP-Cy5.5), anti-mouse CD4 (RM4-5 FITC), anti-mouse CD44 (IM7 on Pacific Blue), anti-mouse IFNγ (XMG1.2 on APC), anti-mouse IL-2 (JES6-5H4 on PE), and anti-mouse TNFα (MP6-XT22 on PE/Cyanine7). Fluorescently labeled antibodies were purchased from BD Pharmingen, except for anti-CD44 (which was from Biolegend). Dead cells were gated out using LIVE/ DEAD fixable dead cell stain (Invitrogen). Flow cytometry samples were acquired with a Becton Dickinson Canto II or an LSRII and analyzed using FlowJo v10 (Treestar). The following reagent was obtained through BEI Resources, NIAID, NIH: Peptide Array, Human Coronavirus OC43 Spike (S) Glycoprotein, NR-53728.
## OC43 spike, HKU1 spike, SARS-CoV-2 spike, and MHV spike-specific ELISA
Binding antibody titers were measured using ELISA as described previously (26,27). In brief, 96-well flat bottom plates MaxiSorp (Thermo Scientific) were coated with 0.1 mg/ well of the respective spike protein for 24 h at 4°C. For detection of MHV spike-specific antibody responses, we utilized a lysate of HEK293T cells transfected with a plasmid encoding MHV spike, as coating antigen (incubated for 48 h at room temperature). Plates were washed with PBS + 0.05% Tween-20. Blocking was performed for 4 h at room temperature with 200 μL of PBS + 0.05% Tween-20 + bovine serum albumin. Then, 6 µL of sera was added to 144 μL of blocking solution in the first column of the plate, 1:3 serial dilutions were performed until row 12 for each sample, and plates were incubated for 60 min at room temperature. Plates were washed three times followed by the addition of goat anti-mouse IgG horseradish peroxidase-conjugated (Southern Biotech) diluted in blocking solution (1:5,000) at 100 µL/well and incubated for 60 min at room temperature.
Plates were washed three times, and 100 µL/well of Sure Blue substrate (Sera Care) was added for approximately 8 min. The reaction was stopped using 100 µL/well of KPL TMB stop solution (Sera Care). Absorbance was measured at 450 nm using Spectramax Plus 384 (Molecular Devices). OC43 spike protein was produced in-house using a mammalian expression vector obtained from Addgene (Cat. No. 166015).
## Propagation and determination of OC43-NV titers
OC43-NV stocks were propagated in 1-2 days old suckling neonates from C57BL/6 mice using a protocol from a prior paper (28). In brief, ten 1-2-day-old mice were inocula ted intracerebrally with 10 µL of brain homogenates infected with OC43-NV (kind gift from Dr. Stanley Perlman's laboratory). After 2 days, whole brains were collected from the neonates and homogenized in 2 mL of sterile PBS. The lysates were clarified by centrifugation at 2,000 rpm for 10 min, as previously described (29), and small aliquots of the supernatant were stored in a -80°C. For adult mouse challenges, the above viral stock was diluted 10-fold, and 50 µL was administered intranasally. Viral load in brain lysates was quantified by quantitative real-time RT-PCR targeting the OC43 nucleocapsid gene, using TaqMan chemistry as previously described (20).
## OC43-NV challenge and disease severity score
On day 15 post-vaccination, mice were challenged intranasally with 50 µL of OC43-NV stock (1 × 10 10 genome copies), administered as 25 µL per nostril. All mice were monitored for weight loss and clinical severity over the course of 3 weeks. Disease severity was measured in terms of clinical scores ranging from 0 to 3, defining the body posture, appearance of fur, eye secretions or closure, animal activity, lethargy, body temperature, and neurological symptoms (30,31). The highest score was represented by severe disease status counting piloerection, puffy appearance, non-responsive and stationary even when provoked, severely hunched posture, completely closed eyes, stopped eating/drinking, rapid or labored breathing with gasps, cold body temperature, shivering, showing no response upon stimuli, and neurological symptoms. Score 2 was defined as moderate disease including moderately hunched posture, majority of fur on back is ruffled, active only when provoked, stationary, no response to auditory or slowed response to touch, eyes half-closed, potential eye secretions, lethargy, less active, and consistently labored breathing. The mild disease was scored as one showing mildly hunched, slightly ruffled fur, active, avoiding standing upright, and slowed response to auditory/touch stimuli. The normal active animal with a smooth coat was scored as zero.
## MHV propagation and quantification
Seed stock of MHV-A59 was obtained from ATCC. The virus was propagated in 17 CL-1 cells and tittered on L2 cells (kind gift from Dr. Susan R. Weiss). The 17 CL-1 and L2 cells were passaged in DMEM supplemented with 10% fetal bovine serum (FBS), 1% L-glutamine and 1% penicillin/streptomycin. For virus propagation, 17 CL-1 cells were inoculated at a low multiplicity of infection (MOI) of 0.1 in 1% DMEM. After 72 h of incubation, the supernatant was collected and clarified by centrifugation at 2,000 rpm for 10 min. The titer of the viral stock was determined by plaque assay using L2 cell monolayers. To determine the viral titer in infected tissues, the lung, liver, and brain were collected on day 3 after the challenge and stored immediately in a -80°C until processing. For plaque assay, 1 × 10 6 L2 cells per well were seeded in 6-well plates in 10% DMEM medium. After 24 h, when the monolayer was 90%-00% confluent, tissue samples were thawed in a water bath at 37°C and processed to assess viral titer. The tissues were homogenized using a standard TissueRuptor homogenizer, and 10-fold serial dilutions of the homogenized samples were prepared in 1% DMEM and applied dropwise onto the cell monolayer. The six-well plates were placed in a 37°C, 5% CO 2 incubator for 1 h and manually rocked every 10 min. After 1 h of incubation, a 1:1 agarose and 2 × 199 overlay was added to the monolayer, and plates were incubated at 37°C, 5% CO for 48 h. After 48 h, the overlay was removed, the cells were stained with 0.1% crystal violet, and plaques were counted.
## Adoptive plasma transfers
C57BL/6 donor mice were immunized with two doses of the mRNA-OC43 vaccine at a 3-week interval. Seven days after the final dose, OC43 spike-specific antibody responses were confirmed by ELISA, and plasma from the immunized mice was pooled. These pooled plasmas were adoptively transferred into C57BL/6 recipient mice via the i.p. route. Control mice received plasma from naïve donors. The following day, all mice were infected i.p. with 2 × 10⁶ PFU of MHV-A59. On day 3 post-infection, tissues were harvested and homogenized using a TissueRuptor Homogenizer (QIAGEN). Viral loads were quantified by plaque assay, as described above.
## Antibody treatments for CD4 and CD8 depletion
All antibodies used for in vivo treatments were purchased from BioXCell or Leinco, diluted in sterile PBS, and administered via intraperitoneal (i.p.) injection. CD4 + T cell-depleting antibody (GK1.5) and CD8 + T cell-depleting antibody (2.43) were given at a dose of 200 µg daily, starting 1 day before infection and continuing through day three postinfection. IgG isotype controls were included in all experiments as controls.
## Detection of IFN-γ-producing T cell responses via ELISpot
To detect IFN-γ-producing antigen-specific T cells, 96-well ELISpot plates (Millipore, Burlington, MA) were coated with anti-mouse IFN-γ monoclonal antibody (clone AN-18, BioLegend 517902) at 5 µg/mL and incubated overnight at 4°C. The following day, plates were washed twice with 200 µL/well of sterile 1× PBS and blocked with 200 µL/well of RPMI medium supplemented with 10% FBS, 1% L-glutamine, and 1% Pen/Strep for 2 h at 37°C in a CO ₂ incubator. Single-cell suspensions from spleen or lymph nodes were prepared at 2.5 × 10⁶ cells/mL in supplemented RPMI. After blocking, media were discarded, and plates were seeded with 2.5 × 10⁵ cells/well and stimulated using 4 µg/mL of an OC43-spike peptide pools. Cells were incubated for 18-20 h at 37°C in a CO ₂ incubator. Cells from naïve mice served as negative controls. For positive controls, cells were treated with either 10 ng/mL phorbol 12-myristate 13-acetate (PMA, Sigma) plus 500 ng/mL ionomycin (Sigma), or with anti-mouse CD3 and CD28 antibodies (1 µg/well). An equimolar concentration of DMSO was included as a negative control. Unstimulated controls included cells incubated with media alone. After 18-20 h of incubation, cells were discarded, and plates were washed five times with wash buffer (1× PBS + 0.05% Tween-20). Plates were then incubated for 90 min with biotinylated anti-IFN-γ antibody (clone R4-6A2, BioLegend 505704) at 0.5 µg/mL diluted in PBS with 10% FBS. After washing, streptavidin-alkaline phosphatase (Bio-Rad 170-6432), diluted 1:1000 in 10% PBS, was added and incubated for 45 min. Plates were washed with wash buffer and developed using substrate (Bio-Rad 170-6432) for 8 min. The reaction was stopped by rinsing plates with running water. Spot-forming cells (SFCs) were analyzed using an ImmunoSpot Image Analyzer (Cleveland, USA). The number of spot-forming units (SFU) per million cells was calculated as the mean of duplicate wells after subtrac tion of negative control wells (no antigen).
## Statistical analysis
## References
1. El-Sahly, Atmar, Glezen et al. (2000) "Spectrum of clinical illness in hospitalized patients with "common cold" virus infections" *Clin Infect Dis*
2. Falsey, Walsh, Hayden (2002) "Rhinovirus and coronavirus infection-associated hospitalizations among older adults" *J Infect Dis*
3. Mcintosh, Chao, Krause et al. (1974) "Coronavirus infection in acute lower respiratory tract disease of infants" *J Infect Dis*
4. Pene, Merlat, Vabret et al. (2003) "Coronavirus 229E-related pneumonia in immunocompromised patients" *Clin Infect Dis*
5. Vabret, Mourez, Gouarin et al. (2003) "An outbreak of coronavirus OC43 respiratory infection in Normandy, France" *Clin Infect Dis*
6. Van Der Hoek (2007) "Human coronaviruses: what do they cause?" *Antivir Ther*
7. Walsh, Shin, Falsey (2013) "Clinical impact of human coronavi ruses 229E and OC43 infection in diverse adult populations" *J Infect Dis*
8. Gaunt, Hardie, Claas et al. (2010) "Epidemiology and clinical presentations of the four human coronavi ruses 229E, HKU1, NL63, and OC43 detected over 3 years using a novel multiplex real-time PCR method" *J Clin Microbiol*
9. Paules, Marston, Fauci (2020) "Coronavirus infections-more than just the common cold" *JAMA*
10. Choi, Kim, Park et al. (2021) "Comparison of the clinical characteristics and mortality of adults infected with human coronavi ruses 229E and OC43" *Sci Rep*
11. Lavi, Gilden, Wroblewska et al. (1984) "Experimental demyelination produced by the A59 strain of mouse hepatitis virus" *Neurology (ECronicon)*
12. Yang, Du, Chen et al. (2014) "Coronavirus MHV-A59 infects the lung and causes severe pneumonia in C57BL/6 mice" *Virol Sin*
13. Boscarino, Logan, Lacny et al. (2008) "Envelope protein palmitoylations are crucial for murine coronavirus assembly" *J Virol*
14. Szabo, Miron, Farber (2019) "Location, location, location: tissue resident memory T cells in mice and humans" *Sci Immunol*
15. Zeng, Chen, Tan et al. (2018) "Epidemiology and clinical characteristics of human coronaviruses OC43, 229E, NL63, and HKU1: a study of hospitalized children with acute respiratory tract infection in Guangz hou, China" *Eur J Clin Microbiol Infect Dis*
16. Zhou, Qiu, Ge (2021) "The taxonomy, host range and pathogenicity of coronaviruses and other viruses in the Nidovirales order"
17. Anderson, Goodwin, Verma et al. (2021) "Seasonal human coronavirus antibodies are boosted upon SARS-CoV-2 infection but not associated with protection" *Cell*
18. Saunders, Lee, Parks et al. (2021) "Neutralizing antibody vaccine for pandemic and pre-emergent coronaviruses" *Nature*
19. Jacob-Dolan, Feldman, Mcmahan et al. (2021) "Coronavirus-specific antibody cross reactivity in rhesus macaques following SARS-CoV-2 vaccination and infection" *J Virol*
20. Dangi, Palacio, Sanchez et al. (2021) "Cross-protective immunity following coronavirus vaccination and coronavirus infection" *J Clin Invest*
21. Dos, Alves, Timis et al. (2024) "Human coronavirus OC43-elicited CD4 + T cells protect against SARS-CoV-2 in HLA transgenic mice" *Nat Commun*
22. Humbert, Olofsson, Wullimann et al. (2023)
23. "cells established in early childhood decline with age" *Proc Natl Acad Sci*
24. Mallajosyula, Ganjavi, Chakraborty et al. (2021)
25. "+ T cells specific for conserved coronavirus epitopes correlate with milder disease in COVID-19 patients" *Sci Immunol*
26. Sagar, Reifler, Rossi et al. (2021) "Recent endemic coronavirus infection is associated with less-severe COVID-19" *J Clin Invest*
27. Sanchez, Palacio, Dangi et al. (2021) "Fractionating a COVID-19 Ad5-vectored vaccine improves virus-specific immunity" *Sci Immunol*
28. Dangi, Sanchez, Lew et al. (2022) "Pre-existing immunity modulates responses to mRNA boosters"
29. Sanchez, Dangi, Awakoaiye et al. (2024) "Delayed reinforcement of costimulation improves the efficacy of mRNA vaccines in mice" *J Clin Invest*
30. Butler, Pewe, Trandem et al. (2006) "Murine encephalitis caused by HCoV-OC43, a human coronavirus with broad species specificity, is partly immune-mediated" *Virology (Auckl)*
31. (2025) *Full-Length Text Journal of Virology*
32. Xie, Fang, Baloch et al. (2022) "A mouse-adapted model of HCoV-OC43 and its usage to the evaluation of antiviral drugs" *Front Microbiol*
33. Gonzalez, Ijezie, Balemba et al. (2018) "Attenuation of influenza A virus disease severity by viral coinfection in a mouse model" *J Virol*
34. Couto, Gonçalves, Lamas et al. (2023) "Protocol for infecting and monitoring susceptible k18-hACE2 mice with SARS-CoV-2" *STAR Protoc*
35. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793530&blobtype=pdf
|
Said Rachida, Henok Tafesse, S Biology, ; Hannah, Cayla Dakanay, Alaa Ahmed, Anne Piantadosi, Disclosures Piantadosi, Advisor Consultant|delve, Bio Board Member
## Abstract
Background. Human respiratory syncytial virus (RSV) is a leading cause of lower respiratory tract infections in high-risk individuals. The recent introduction of vaccines and prophylactic monoclonal antibodies is expected to reduce the clinical and public health impact of RSV but may also create selective pressure for immune escape among circulating viruses.Methods. We evaluated clinical and demographic data from 183 adult patients who presented to Emory University Hospital between September 21, 2024, and March 26, 2025. We collected residual upper respiratory swab samples and performed subgroup-specific RT-PCR followed by viral whole-genome sequencing using the xGen™ Respiratory Virus Amplicon Panel and phylogenetic analyses using Nextstrain.Results. Among 183 adult patients, the median age was 61 (range: 18-99), and 69.5% were female. Most patients (97.8%) were symptomatic; 12.9% were hospitalized, 2.2% required ICU care, and 2.8% died. Of 83 vaccine-eligible patients, only 17 (20.5%) had been vaccinated. RT-PCR identified 84 RSV-A and 91 RSV-B infections. Fever was more common in RSV-B (76% vs. 61%, p = 0.05); other outcomes were similar. Viral whole-genome sequencing achieved >95% coverage in 108 samples (60%). RSV-A sequences exhibited high clade and lineage diversity. The majority (35%) belonged to lineage A.D.3.1, a descendant of clade A.D.3, while other sequences were distributed across clades and lineages including A.D.1, A.D.1.5, A.D.1.6, A.D.3, and A.D.5.2, each representing 5-25% of cases. RSV-B sequences were predominantly from lineage B.D.E.1 (55%), a major lineage within clade B.D.E, with additional lineages including B.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793533&blobtype=pdf
|
# P-2198. Effect of IV Fluids on Diuretic Need in Pediatric and Adult Dengue Patients: A Network Meta Analysis and Systematic Review
Sophia Costa, Leticia Campos, Gisella Carpi, Jose Luis Boene, Thiago Netto, ; Oscar, Hernández Rios, Taniela Bes
## Abstract
for each ventilation condition and viral shedding potential. Bright red line denotes R 0 of 1. 0 ) of influenza under different ventilation conditions and viral shedding potentials 0 for each ventilation condition and viral shedding potential. Bright red line denotes R 0 of 1. 0 of SARS-CoV-2 and influenza in these settings, as well as under different ventilation conditions and viral shedding levels. 0 above 1 among high virus shedders.
Background. Dengue fever (DF), transmitted by Aedes aegypti and Aedes albopictus, mostly in tropical and subtropical regions, causes high fever, muscle cramps, and joint pain. Fluid resuscitation is the mainstay of treatment, with Ringer's Lactate (RL) being the first line recommended by the World Health Organization (WHO). Capillary leakage during resuscitation may happen and require diuretics, so the best IV fluid option for managing dengue remains in discussion. Methods. A literature search was performed in PubMed, Embase, and the Cochrane Library databases for Randomized Controlled Trials (RCTs) comparing RL with alternative IV fluids in patients with DF. The primary outcome was the use of furosemide after IV infusion. Data were analyzed using R, with heterogeneity assessed. Reference management and deduplication were done via PROSPERO, and risk of bias was evaluated using ROB-2.
Results. Five RCTs were included, and direct/indirect comparisons were made (Figure 1). In this network meta-analysis of fluid therapies, Dextran vs Gelatin showed a non-significant trend favoring Dextran (Direct OR: 0.51; 95% CI: [0.10, 2.54]; Network OR: 0.78; [0.18, 3.34]). Dextran vs Ringer's Lactate showed no clear difference (Direct and Network OR: 0.88), though the indirect estimate was highly imprecise. Dextran vs Saline also showed no meaningful difference (Network OR: 0.99; [0.22, 4.35]). Gelatin vs Ringer's Lactate and Saline vs Ringer's Lactate both showed no significant differences (Network ORs: 1.13 and 0.89, respectively), with wide confidence intervals. Heterogeneity was generally low (I² = 0%), except for Dextran vs Ringer's Lactate (I² = 62.4%), indicating moderate inconsistency in that comparison (Figure 2).
Conclusion. There was no significant association between the type of IV fluid administered and the need for furosemide due to plasma leakage. Further robust, direct comparisons are needed to strengthen the current evidence base Disclosures. All Authors: No reported disclosures
Poster Abstracts • OFID 2026:13 (Suppl 1) • S1333
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12871658&blobtype=pdf
|
# Conserved assembly architecture of the essential herpesvirus packaging accessory factor
Elizabeth Bailey, Swapnil Devarkar, Renata Szczepaniak, Laura Meißner, Xinyu Chen, Chunxiang Wu, Sandra Weller, Yong Xiong, Allison Didychuk
## Abstract
To create a new wave of infectious virions, all herpesviruses require an accessory factor of unknown function to package their viral genomes into nascent capsids. Here, we present cryo-EM structures of the packaging accessory factor from the α-herpesvirus herpes simplex virus type 1 (HSV-1, UL32) and the β-herpesvirus human cytomegalovirus (HCMV, UL52). Unlike homologs from the γ-herpesviruses, neither UL32 nor UL52 form stable homopentameric rings. UL52 forms incomplete pentameric rings lacking one or two protomers. UL32 does not form stable higher-order species, but stabilization through chemical crosslinking revealed a novel quaternary structure where three pentameric rings assemble into a "tripentamer." Our results reveal that herpesvirus packaging accessory factors adopt distinct oligomeric states but are constrained to pentameric symmetry. Assembly of protomers into a ring creates a positively charged central channel that we show is critical for infectious virus production in HSV-1. Taken together, our study points to a structurally conserved, essential function of packaging accessory factors across the Herpesviridae.
Nearly every human is infected with at least one herpesvirus by adulthood. Nine herpesviruses from three subfamilies (α, β, γ) infect humans and cause clinically significant disease. Herpes simplex virus type 1 (HSV-1) is an α-herpesvirus that causes oral and genital lesions and can lead to disseminated disease in neonates 1 . The β-herpesvirus cytomegalovirus (HCMV) is the leading non-genetic cause of congenital birth defects 2 and causes complications in transplant patients. Kaposi's Sarcoma-associated herpesvirus (KSHV) is a γ-herpesvirus that causes cancer and other malignancies, particularly in immunocompromised individuals 3,4 .
Late in their replication cycle, herpesviruses compress their double-stranded DNA genome into capsids that mature to form infectious progeny virions 5 . Empty icosahedral capsids, with a unique portal vertex through which DNA is inserted, are assembled in the nucleus of infected cells [6][7][8][9][10] . Genome packaging is driven by the virus-encoded terminase motor, which recognizes newly synthesized viral genomes, translocates the viral genome into the nascent capsid, and finally cleaves the packaged, unit-length viral genome [11][12][13] . The genome-filled capsid is stabilized by binding of the portal cap, followed by egress of the capsid through the nuclear membrane to continue maturation [14][15][16][17] .
While this packaging process is largely conserved with tailed bacteriophages 18,19 , herpesviruses require an additional packaging accessory factor of unknown function. This factor is unique to herpesviruses and essential for viral genome packaging across the Herpesviridae [20][21][22][23][24][25] . The viral genome is replicated in the absence of the packaging accessory factor, but the concatenated genome is not cleaved, consistent with a defect early in packaging 20,22,23 . Loss of the packaging accessory factor results in immature "B" capsids that retain the cleaved scaffold and lack packaged DNA 20,22- 24 . The packaging accessory factor is not incorporated into capsids or virions 20,26 , nor is it thought to be a constitutive component of the terminase, although a recent study in HCMV suggests that the packaging accessory factor and terminase may interact 27 .
The KSHV packaging accessory factor, ORF68, adopts a novel fold and oligomerizes into a stable homopentameric ring with a positively charged central channel that is essential for production of infectious virions 28 . While homologs from related γ-herpesviruses can partially complement for loss of ORF68 during KSHV infection, packaging accessory factors from the more distantly related α-and β-herpesviruses cannot 28 . Thus, we sought to determine if the packaging accessory factor is structurally conserved across the herpesvirus family. Here, we report cryo-electron microscopy (cryo-EM) structures of the packaging accessory factor from HSV-1 (UL32) and HCMV (UL52). We find that individual protomers adopt a highly conserved core tertiary structure, but their propensity to form stable oligomers varies across the Herpesviridae. HCMV UL52 assembles into incomplete pentameric rings with three or four protomers and a slight helical twist. HSV-1 UL32 formed an unstable pentamer in solution, leading us to pursue chemical crosslinking that stabilized a novel quaternary assembly composed of three pentameric rings. Assembly into a ring concentrates positively charged residues into a central channel, and we demonstrate that these positively charged residues are essential to productive infection in HSV-1. Our structures of UL52 and UL32 form the basis for further mechanistic dissection of this essential yet enigmatic viral protein.
## RESULTS
## The conserved herpesvirus packaging accessory factor adopts different oligomeric states across the Herpesviridae
HSV-1 UL32 and HCMV UL52 have insertions and amino (N)-terminal extensions relative to KSHV ORF68, resulting in protomers that are 12 and 22 kDa larger than ORF68 (Fig. 1a, Supplementary Fig. 1a). Recombinantly expressed ORF68, UL32, and UL52 were purified to homogeneity from baculovirus-infected insect cells (Fig. 1b). During size exclusion purification, we noticed that ORF68 eluted as a single peak at a volume consistent with a pentameric assembly, while UL32 and UL52 eluted later than expected for pentameric assemblies. To monitor oligomeric states in solution, we performed size exclusion chromatography coupled with multi-angle light scattering (SEC-MALS) on purified samples of the three packaging accessory factors. ORF68 eluted in a single peak with a calculated molecular weight of ~280 kDa, consistent with a pentamer (Fig. 1c, d) 28 . In contrast, UL52 exhibited a broad peak with a molecular weight corresponding to between 1-2 protomers and UL32 had two well-separated peaks consistent with the molecular weights of a pentamer and a monomer (Fig. 1c,d). Thus, unlike the γ-herpesvirus accessory factors ORF68 and the Epstein-Barr virus (EBV) homolog BFLF1 28 , UL32 and UL52 do not form stable pentameric rings in solution. We next sought to structurally characterize UL32 and UL52 to illuminate the tertiary structure and quaternary assemblies of these essential proteins.
## UL52, the HCMV packaging accessory factor, forms an incomplete pentameric ring
We carried out structural studies of UL52 using single-particle cryo-EM and identified two classes of incomplete pentameric rings with three or four well-resolved protomers (Fig. 2a, Supplementary Fig. 1). We were unable to identify any particles with complete pentameric rings. The map for the 3-membered incomplete ring had a global resolution of 3.28 Å and the map for the 4membered incomplete ring had a global resolution of 3.32 Å (hereafter referred to as 3-mer and 4-mer, respectively). The UL52 model was first built for the protomer with the highest local resolution (the central protomer in the 3-mer map), then used as a starting model to build the other protomers and create the final 4-mer model (Fig. 2b, Supplementary Fig. 2e). In the 3-mer map, a poorly-resolved fourth protomer is visible at low contour levels, suggesting that an additional protomer is present but highly flexible in this class (Supplementary Fig. 3a). The best-resolved portions of the cryo-EM density of the weak fourth protomer in the 3-mer map are at the interface between protomers, including density for a helix-loop-helix motif (residues 478-502) and another short helix (residues 657-666) (Supplementary Fig. 3b). Density for a helix spanning the central channel (residues 633-652) is also visible in the weak fourth protomer (Supplementary Fig. 3b). Thus, UL52 primarily forms a 4-membered incomplete ring with pentamer-like symmetry where the fourth protomer can be flexibly engaged.
## 2, Supplementary Movie 1, Supplementary Table
For the purposes of our discussion, we define the orientation of packaging accessory factors based on the crown-like ring in Didychuk et al. 28 , where the "top" surface of the ring is crenellated by short semi-structured loops and the "bottom" surface of the ring is relatively flat. The core fold of UL52 is highly similar to that of ORF68 (Fig. 2c). The outer surface is more variable, with long insertions that form surfaceexposed helices and loops and an N-terminal extension. The 51-amino acid N-terminal extension is almost entirely disordered, with residues 1-48 unresolved (Fig. 2c; Supplementary Fig. 3c,d). Also unresolved are two long, negatively charged insertions on the top surface of the incomplete ring (residues 82-181, 401-430) (Fig. 2c; Supplementary Fig. 3c,d). Both the 3-mer and 4-mer cryo-EM classes have additional unidentified low-resolution density in the central channel (Fig. 2a), which could be attributed to co-purified nucleic acid or part of the unresolved, negatively charged loop (residues 82-181) interacting with the positively charged central channel.
UL52 has two regions that are remodeled relative to ORF68, altering the size and shape of the top and side of the UL52 incomplete ring. A long α-helix insertion and subsequent 30 amino acid disordered loop (remodeled region 1: residues 372-432) reshape the top peripheral surface of the incomplete ring (Fig. 2c; Supplementary Fig. 3c,d). A short α-helical insertion and a 16 amino acid disordered loop (remodeled region 2: residues 502-547) reshape the side, peripheral surface (Fig. 2c; Supplementary Fig. 3c,d). A sequence alignment of packaging accessory factors from the nine human herpesviruses revealed that the insertion forming UL52 remodeled region 2 is unique to β-herpesvirus factors (Supplementary Fig. 1a). These differences in the exterior of the protein alter the available surface for binding partners, potentially enabling virus-specific interactions. The UL52 3-mer and 4-mer both form incomplete rings with pentamer-like symmetry wherein the protomers are arranged ~72º from each other. However, the incomplete ring twists out of plane, creating a right-handed screw axis with a helical rise of ~5 Å per protomer (Fig. 2d). This slight rise is insufficiently steep for UL52 to form a helical filament. In this helical arrangement UL52 protomers adopt a different relative conformation than they would as a planar ring. Examination of the protomer interface reveals that the twist arises from a ~12º rotation from the helix-loop-helix that extends into a pocket on the adjacent protomer (residues 487-497, hereafter referred to as the protomer interface loop) (Fig. 2e). The protomer interface loop has several residues buried deep in the pocket of the adjacent protomer, including N489 and K491, as well as polar-charged interactions between loop-side R495 and pocket-side Q286 (Fig. 2f). Several residues in the loop-pocket regions of the protomer interface are universally-or well-conserved across herpesviruses (Fig. 2f; Supplementary Fig. 1b,c). A previous study showed that an insertion in the protomer interface loop of the HSV-1 homolog UL32 was not tolerated 29 . Thus, this loop is the conserved "key" that fits into the "lock" of the adjacent protomer, controlling oligomerization of the packaging accessory factor.
## UL32, the HSV-1 packaging accessory factor, forms an unstable pentameric ring that can be stabilized in a novel quaternary assembly
Like HCMV UL52, SEC-MALS of the HSV-1 packaging accessory factor UL32 revealed multiple oligomeric states. In contrast to the poorly defined UL52 oligomer, we observed two well-separated peaks for UL32 consistent with monomer and pentamer species (Fig. 1c,d). We first attempted to visualize the pentameric species by negative stain electron microscopy but were unable to identify particles of the expected dimensions. We suspect that UL32 forms an unstable pentamer at higher concentrations that is disrupted at lower concentrations. We previously reported that ORF68 binds double-stranded DNA (dsDNA) in vitro 28 and hypothesized that if UL32 similarly bound dsDNA, we could use this approach to stabilize an oligomeric complex. We used an electrophoretic mobility shift assay (EMSA) to monitor binding of UL32 to fluorescently labeled dsDNA. UL32 bound to the shortest probe tested (10 bp, Supplementary Fig. 4a). The migration of this complex was comparable to ORF68-probe complexes 28 , suggestive of UL32 binding as a pentamer. As the probe length increased to 30 bp, a larger discrete complex formed (Supplementary Fig. 4a). UL32-DNA complexes were unstable on cryo-EM grids and thus we pursued chemical crosslinking with ethylene glycol bis(succinimidyl succinate) (EGS), which yielded distinct higher molecular weight bands by SDS-PAGE (Supplementary Fig. 4b,c). By negative stain electron microscopy, crosslinked samples of UL32 lacking dsDNA or with probes of 10, 20, 30, or 67 bp showed well-dispersed particles, including monomers, pentamers, and a larger species (Supplementary Fig. 4d). We anticipated that the larger species would be a dimer of homopentamers (D5 symmetry) as previously observed for the EBV packaging accessory factor BFLF1 28 .
We imaged samples of UL32 crosslinked in the presence of 30 bp dsDNA by single-particle cryo-EM and generated a map with a global resolution of 2.90 Å (Fig. 3a, Supplementary Fig. 5, Supplementary Movie 2, Supplementary Table 1). Our cryo-EM reconstruction of UL32 revealed a novel quaternary assembly wherein three pentameric UL32 rings form a trimer of homopentamers, hereafter referred to as a "tripentamer". Within the tripentamer, UL32 forms a pentameric planar ring with similar dimensions to those of the ORF68 pentamer, where the top of each pentameric ring faces outward (Fig. 3b). Like UL52, the core fold of UL32 is similar to that of ORF68 (Fig. 3c). Compared to ORF68, the surface of UL32 is altered by structured insertions (residues 315-322, 353-390), disordered loops (residues 64-106, 227-234, 265-274), and a partially disordered N-terminal extension (residues 1-31) (Fig. 3c, Supplementary Fig. 5g). A long α-helical insertion (residues 353-390) and two disordered loops (residues 64-106, 265-274) are present in UL32 at similar locations to these features in UL52. Residues 315-322 extend an α-helix on the bottom surface of the ring (residues 308-323) that is displaced laterally by 12 Å relative to the corresponding helices in UL52 and ORF68 (Supplementary Fig. 5h).
The tripentamer consists of three pentameric rings arranged around a 3-fold symmetry axis (Fig. 3d). The interface between two pentameric rings involves four protomers (two protomers from each pentamer) engaged in a local 2-fold symmetric interaction (Fig. 3d). The tripentamer architecture lacks global 2-fold symmetry because the three pentamers are splayed outward by ~18° relative to a regular prism (Fig. 3d). This arrangement forms a triangular frustrum of interior space, where the smaller equilateral triangle has a length of ~70 Å, and the larger widens to ~100 Fig. 2 caption continued. Additional unmodeled density (grey) is resolved in the central channel of both classes. (b) Model of the UL52 4-mer shown from top and side views, with each protomer colored a shade of blue. (c) UL52 protomer colored by the Cα RMSD to ORF68 (PDB ID: 6XF9), blue is 1 Å, white is 3 Å, and red is 5 Å, yellow are residues without RMSD values where there are insertions in UL52 relative to ORF68. (d) UL52 adopts pentamer-like symmetry with a twist out of plane. Central channel helices for ORF68 (grey) and UL52 are aligned to a UL52 protomer (sage green). Models are shown in tube helices inside transparent UL52 4-mer model surface at 10 Å resolution. Insert: (Left) shows a top view of the 5-fold axis (72°, helical twist (Δϕ)). (Right) shows a side view of the right-handed twist of the UL52 protomers relative to the planar arrangement of ORF68 and the ~5 Å helical rise (Δz) between protomers. (e) UL52 twists at the protomer interface loop. UL52 protomer models (sage, blue) shown relative to a UL52 protomer docked to a hypothetical C5 planar arrangement (salmon). Models are shown within a transparent UL52 4-mer model surface at 10 Å resolution. Inset shows details of the protomer interface pocket/loop region, where the pocket protomer model is shown as a sage surface and the loop protomer models are shown as blue or salmon cartoons. (f) Residues in the protomer interface, colored as in e but with loop residues shown as sticks with heteroatom coloring. Å (Fig. 3d). For the purposes of our discussion, we define the wider opening as the "top" of the tripentamer.
At the top of the tripentamer, one protomer from each pentameric ring does not contact neighboring pentamers (Fig. 3d). In the UL32 consensus map, one of the three top protomers had weaker density and lower local resolution (Supplementary Fig. 5e). Further 3D classification revealed that the majority of UL32 particles form complete tripentamers (15-mers, ~60%), while the remainder form incomplete tripentamers containing two 5-membered rings and an incomplete ring missing one or more protomers (Supplementary Fig. 6a). Over 40% of incomplete tripentameric particles (17% of total particles) formed a 14mer, with a twisted, 4-membered incomplete ring. This UL32 4-membered incomplete ring is reminiscent of the UL52 4membered incomplete ring, but close inspection shows that while the UL52 4-mer twist has a consistent rise, the UL32 4-mer twist is asymmetric, arising largely from the rotation of one protomer (Supplementary Fig. 6b,c). As in UL52, the twist arises from hinging at the loop-pocket motif at the protomer interface (Supplementary Fig. 6d).
Since the UL32 cryo-EM sample was prepared in the presence of 30 bp dsDNA, we carried out further classification to resolve UL32 interactions with dsDNA. Symmetry expansion followed by 3D classification with a mask focused on the potential dsDNA region revealed two classes of particular interest (Supplementary Fig. 6e). In one class (30% of all particles), a rod-shaped extra density spans the open interior space of the tripentamer between the central channels of two pentamers. A 30 bp dsDNA model, displayed as a molecular surface at 10 Å resolution, approximates the length and shape of this additional density; the absence of clear helical character in this density suggests that the potential dsDNA density is rotationally averaged (Supplementary Fig. 6f). In another class, the extra density near the central channel has a distinct righthanded twist that fits well with the minor groove of a B-form DNA model (Supplementary Fig. 6g). In both classes, the extra density inserts into, but does not entirely span, the central channel of the UL32 pentameric ring.
## Tripentamer assembly interfaces may play a role in infectious virion production in HSV-1
The calculated electrostatic surface of ORF68 The tripentameric architecture creates a distinct interface between the protomers of adjacent pentameric rings. Each of the three locally 2-fold symmetric inter-pentamer interfaces has two contact points between the same α-helix from protomers of adjacent pentamers (residues 523-536) (Fig. 4a). Two copies of residue C535 are separated by ~8 Å (Cα-Cα distance) with connecting map density visible at low to moderate contour levels (Fig. 4a, Supplementary Fig. 7a). A lysine residue (K532) on this helix could contribute to stabilization of the tripentamer in our crosslinked sample, as EGS targets primary amines 30 . HCMV UL52 and KSHV ORF68 have an α-helix in the equivalent position to the UL32 tripentamer interface (UL32 residues 523-536, Supplementary Fig. 7b). An equivalent Fig. 3 caption continued. Center and right views show the pentamer model inside the UL32 tripentamer map blurred to B-factor 2000 and shown as a transparent surface. (c) UL32 protomer colored by the Cα RMSD to ORF68 (PDB ID: 6XF9), blue is 1 Å, white is 3 Å, red is 5 Å, and yellow are residues without RMSD values where there are insertions relative to ORF68. (d) Views of the UL32 tripentamer model as tube helices, with the three pentamers shown in medium green, light green, and grey. Top row shows the rotation around the 3-fold symmetry axis, while the bottom row shows views from along the 3-fold symmetry axis and a view along the local 2-fold symmetry axis at the tripentamer interface. cysteine to UL32 C535 is conserved in nearly all α/βherpesvirus homologs but is absent in the γ-herpesvirus homologs (Supplementary Fig. 1d). Nearby in the tripentamer interface, a segment of non-continuous protein density can be modeled as a short α-helix from the Nterminus of UL32 (residues 5-11) that is sufficiently long to reach from the last modeled residue (P19) to the tripentamer interface (Supplementary Fig. 7a). While a subset of human herpesvirus packaging accessory factors have Nterminal extensions, they are poorly conserved, with the UL32 N-terminal extension sharing sequence similarity only with that of the closely related α-herpesvirus HSV-2 (Supplementary Fig. 1e).
We wanted to understand if the tripentamer interface is required for UL32 function in HSV-1 infection. Loss of UL32 prevents viral genome packaging and infectious virion production 20 . We used a previously described transient complementation assay 21 to test the effect of mutations at the UL32 tripentamer interface (Supplementary Fig. 7c). Loss of the N-terminal α-helix (Δ11) or a charge swap mutation at the interface (R440E) (Supplementary Fig. 7a,d) had minimal effect on infectious virion production (Fig. 4b; Supplementary Fig. 7e,f). However, mutation of K532A/C535A reduced infectious virion production by half (Fig. 4b), suggesting that the tripentamer interface may play a role in the viral life cycle.
## All herpesvirus packaging accessory factors contain a conserved core fold stabilized by zinc fingers
The structure of ORF68, the KSHV packaging accessory factor, revealed that each protomer contained three CCCH-type zinc fingers (ZnF) 28 . The first two zinc fingers are present in our structures of HSV-1 UL32 and HCMV UL52, with identical coordination and highly similar local structures conserved across all homologs (Fig. 5a). ZnF1 forms part of the pocket of the pocket-loop motif of the protomer interface. Unlike the α-and γ-herpesvirus accessory factors, HCMV UL52 lacks a third zinc finger (Fig. 5b), consistent with sequence alignments indicating that this motif is absent in the β-herpesviruses 28 . In the position where ORF68 and UL32 have ZnF3, UL52 has two residues (C653, H333) that could coordinate zinc, but it lacks two additional residues to fulfill tetrahedral coordination (Fig. 5b; Supplementary Fig. 1d,f). The third zinc finger in ORF68 consists of residues C415, H452, and the unusual use of adjacent cysteines C191 and C192 (Fig. 5b). These four metal-binding residues are not conserved across herpesvirus homologs (Supplementary Fig. 1d,f), but two adjacent cysteines were identified in HSV-1 UL32 (C308 and C309) and proposed to be the third and fourth metalbinding residues 28 . Our structure shows that while C308 does bind Zn3, C309 is not involved in coordination (Fig. 5b). Instead, C297 -present on the same loop as C308 -is the fourth coordinating residue of ZnF3 (Fig. 5b, Supplementary Fig. 5h).
The zinc finger motifs in ORF68 and UL32 were previously found to be required for the structural stability of these proteins in vivo, and mutations in ZnF1 (C128A), ZnF2 (C502A) or ZnF3 (C544A, H581A, and the double mutant C308A/C309A) greatly reduced UL32 levels in transfected HEK293T cells 28 . We show that mutation to alanine of the four metal-binding residues of ZnF3 observed in our structure (UL32 C297, C308, C544, and H581) reduced UL32 expression, while C309A alone had no effect (Fig. 5c). These data support our structural identification of the residues comprising the essential zinc fingers required for stability of packaging accessory factors across the Herpesviridae.
## A conserved positively charged central channel is required across the Herpesviridae
Our structures show that packaging accessory factors from all three herpesvirus subfamilies can assemble into rings with pentameric symmetry, creating a channel in the center of the ring. Each protomer contributes a wellstructured α-helix and a semi-structured loop that form the surface of the central channel (Fig. 6a). These central channel regions are studded with lysine and arginine residues, creating a positively charged, structurally conserved channel (Fig. 6a; Supplementary Fig. 1d,f). The central channel in ORF68 was previously shown to be essential for viral genome packaging, as mutation of positive charges in the central channel (the single mutation K435A and triple mutation K174A/R179A/K182A) phenocopied total loss of ORF68 28 .
We tested if the positively charged central channel residues of UL32 were required to produce infectious virion in HSV-1 using the transient complementation assay (Supplementary Fig. 7c) 21 . We generated the UL32 mutants K563A (positionally equivalent to ORF68 K435, at the top of the channel), R572A (in the middle of the channel), R579A/R580A (at the bottom of the channel), and K289A/R293A/R302A (extending throughout the central channel, positionally equivalent to ORF68 K174A/R179A/K182A) (Fig. 6a,b). All mutations drastically decreased production of infectious virus particles (Fig. 6c; Supplementary Fig. 7e,f). Mutating a single positively charged residue (K563A or R572A) reduced viral yield by an order of magnitude, and the double-and triple-point mutants were equivalent to total loss of UL32 (Fig. 6c, Fig. Supplementary Fig. 7f). This effect was specific to mutation of positively charged residues in the central channel, as mutation of two positively charged residues on the outer surface of UL32 (R366A/R370A) did not significantly impact virus production (Fig. 6b,c; Supplementary Fig. 7e,f). Thus, UL32 requires a positively charged central channel, and the structural conservation of this feature is suggestive of a shared function across the Herpesviridae (Fig. 6d).
## DISCUSSION
Our structures of packaging accessory factor homologs from across the herpesvirus family illuminate the conserved architecture of this essential viral factor. The herpesvirus packaging accessory factor homologs share a core fold whose surface is punctuated with insertions and variations. Some sites of remodeling are common to α/βherpesvirus homologs, while others are subfamily-specific (Supplementary Fig. 1a). For example, both UL32 and UL52 have an additional α-helix on the top peripheral surface of the ring compared to ORF68, as well as an Nterminal extension and two disordered loops on the top surface. β-herpesvirus homologs furthermore have a remodeled loop on the side of the ring, while α-herpesvirus homologs have an additional disordered loop and a shifted α-helix on the bottom of the ring. The variability of the outer surface of the ring suggests that these regions are involved in virus-or host-specific interactions. Indeed, factors from different subfamilies are not interchangeable: HSV-1 UL32 and HCMV UL52 cannot complement for loss of ORF68 during KSHV infection 28 . Even homologs from within the γherpesvirus subfamily (EBV BFLF1, MHV68 muORF68) are unable to fully replace ORF68 in KSHV 28 . Despite a common core fold, the packaging accessory factor homologs displayed a striking difference in their propensity to form stable assemblies, with a notable shared aspiration towards pentameric symmetry. The γ-herpesvirus homologs ORF68 and BFLF1 form stable pentamers 28 . In contrast, UL52 prefers to exist as an incomplete ring with pentamerlike symmetry, while UL32 can assemble into pentameric rings when stabilized by crosslinking. A structurally conserved loop-pocket motif acts as a hinge to modify the orientation of interacting protomers, but the degree of rotation is constricted such that the ring ultimately maintains near-5-fold symmetry.
Although the packaging accessory factor homologs have poor surface conservation overall, they likely serve the same function in viral genome packaging given that viruses lacking these factors have identical phenotypes 20,[22][23][24] . The shared ability to form a pentameric ring points to a common, essential function of this oligomeric state. Indeed, ring formation concentrates positive charges that are required for successful herpesvirus packaging within a central channel. Could this central channel be used to bind the viral dsDNA genome? While the central channel, 25 Å at its narrowest point, could technically accommodate dsDNA, it is unlikely to allow for sliding along dsDNA. In contrast, the sliding clamp involved in DNA replication (i.e., proliferating cell nuclear antigen (PCNA)) forms a symmetric ring with a positively charged central channel ~35 Å in diameter, easily accommodating duplex DNA 31,32 . In an interesting analogy, herpesvirus encode their own sliding clamp processivity factor (HSV-1 UL42/HCMV UL44/KSHV ORF59) that shares a common PCNA-like fold but whose oligomeric state varies across the herpesviruses and forms an incomplete ring [33][34][35][36] .
A surprising finding in our study was the novel tripentameric quaternary assembly of UL32. Cysteine residues at the tripentamer interface could form disulfide bonds, and we find that mutation of C535 reduces UL32 function during HSV-1 infection. Indeed, UL32 has been implicated in redox regulation during HSV-1 infection 21 . Interestingly, ORF68 has also been shown to form discrete higher molecular weight bands in EMSA on dsDNA pieces 30 bp and longer 23,28 , suggestive of supra-pentameric assemblies in other herpesviruses. We speculate that the formation of higher-order assemblies is crucial for the function of the packaging accessory factor.
## MATERIALS AND METHODS
## Plasmids
ORF68 and UL32 with N-terminal Twin-Strep tags and HRV 3C protease cleavage sites ("TSP") were subcloned from their respective pUE1-TSP plasmids (Addgene #162650, 162658) into pLIB (Addgene #80610) digested with BamHI/HindIII to generate pLIB-TSP-ORF68 and pLIB-TSP-UL32 (Addgene #250565, 250566). The coding region of UL52 was subcloned from a pcDNA4/TO-2xStrep vector (Addgene #162628) into a pHEK293 UltraExpression I vector (Clontech) that encodes an Nterminal Twin-Strep tag and HRV 3C protease cleavage site (pUE1-TSP) to generate pUE1-TSP-UL52. Then, TSP-UL52 was subcloned into pLIB as described above to generate pLIB-TSP-UL52 (Addgene #250567). UL32 mutants used for transient expression experiments (Addgene 250568-250578) were generated by inverse PCR of pcDNA4/TO-2xStrep-UL32 (Addgene #162629). Plasmids expressing UL32 C308A/C309A, C544A, and H581A were previously described (Addgene #162646-162648).
## Expression and purification of recombinant ORF68, UL52, UL32
TSP-ORF68 and TSP-UL52 were expressed in Sf9 insect cells using the Bac-to-Bac Baculovirus Expression System (Thermo Fisher Scientific). pLIB-TSP-ORF68 or UL52 was transformed into E. coli DH10Bac competent cells, from which bacmid was isolated and transfected into Sf9 cells with Cellfectin II (Thermo Fisher Scientific) to generate baculovirus. Sf9 cells were infected with baculovirus for protein expression and purification. Three days later, cells were pelleted at 1000 × g for 5 min at 4°C and resuspended in 30 mL lysis buffer (100 mM Tris-HCl pH 8.0, 300 mM NaCl, 5% glycerol, 0.5% CHAPS, 1 μg/mL avidin, 1 mM dithiothreitol (DTT), cOmplete™ EDTA-free Protease Inhibitor Cocktail (Millipore Sigma)) for 30 min at 4°C. The cell suspension was sonicated and the lysate clarified by ultracentrifugation at 50,000 × g for 30 min. The lysate was filtered through a 0.45 μm filter before purification by a gravity column with Strep-Tactin XT 4Flow resin (IBA Lifesciences). The column was washed with wash buffer (100 mM Tris pH 8.0, 300 mM NaCl, 0.1% CHAPS, 1 mM DTT), and the protein was eluted in wash buffer containing 50 mM biotin. The protein was concentrated using a 30 K Amicon Ultra-15 concentrator (Millipore) before injection onto a Superose 6 Increase 10/300 GL (24 mL, 5kDa-5000KDa, 500 μL load) equilibrated in SEC buffer (20 mM HEPES pH 7.6, 100 mM NaCl, 5% glycerol, 1 mM DTT). The peak fraction was pooled, concentrated, and exchanged into running buffer containing 10% glycerol for freezing using a 30 kDa MWCO Amicon Ultra-15 centrifugal filter (Millipore).
TSP-UL32 was expressed as above, with the exception that Strep-Tactin Sepharose resin was used for purification and the elution buffer contained 400 mM NaCl and 2.5 mM desthiobiotin. Additionally, the SEC buffer contained 500 mM NaCl. Protein used in SEC-MALS experiments were purified as above, but recombinant expression occurred in High Five insect cells for 52 h (Thermo Fisher Scientific), lysis buffer contained 20 μg/mL DNase I (Gold Biotechnology), buffers contained 1 mM TCEP instead of DTT, and proteins were eluted from the Strep-Tactin column by cleavage with HRV-3C protease.
## SEC-MALS
The oligomeric state of purified packaging accessory factors was determined using Size Exclusion Chromatography (SEC) coupled with Multi-Angle Light Scattering (MALS). SEC was performed with a Superdex 200 Increase column. Each sample injection consisted of 100 µL of 0.5-1 mg/mL purified protein in running buffer (50 mM HEPES pH 7.6, 100 mM NaCl, 5% glycerol, 1 mM TCEP for UL52 and ORF68; for UL32 the NaCl concentration was 500 mM) at a flow speed of 0.4 mL/min. MALS was monitored with a DAWN Heleos II light scattering detector and an Optilab T-rEX refractive index detector (both from Wyatt Technology). Light scattering data were collected at 1 second interval and analyzed with ASTRA6 software (Wyatt Technology).
## Electrophoretic mobility shift assay
Fluorescein-labeled dsDNA probes (Integrated DNA Technologies, previously described in 28 , except for the 67 bp probe: /56-FAM/cgccgccgggcctgcggcgcctcccgcccggg Catggggccgcgcgccgcctcagggcccggcgcgg + ccgcgccgggccc tgaggcggcgcgcggccccatgcccgggcgggaggcgccgcaggcccggc ggcg), were prepared as 2x stocks (20 nM) in binding buffer (100 mM NaCl, 20 mM HEPES pH 7.5, 5% glycerol, 0.05% CHAPS, 1 mM DTT). UL32 was diluted in binding buffer containing 0.2 mg/mL Bovine Serum Albumin (BSA). Binding reactions were prepared by mixing equal volumes of probe DNA and protein solutions, resulting in final concentrations of 10 nM DNA probe, 100 mM NaCl, 20 mM HEPES pH 7.5, 5% glycerol, 0.05% CHAPS, 0.1 mg/mL BSA, and variable concentrations of UL32. Concentrations are indicated as calculated for a UL32 protomer. Samples were incubated at room temperature for 30 min prior to electrophoresis on a 5% polyacrylamide (29:1 acrylamide:bis-acrylamide)/ 1x Tris borate gel at 2W at 4˚C. Gels were imaged using an Amersham Typhoon imaging system (Cytiva).
## Crosslinking of UL32 and negative stain EM
UL32 (4 µM) was incubated with 0.5-4 mM ethylene glycol bis(succinimidyl succinate) (EGS) for 30 min on ice. For samples with DNA, UL32 (4 µM) was incubated with 5 µM dsDNA probe for 30 min at room temperature prior to crosslinking. DNA probes were the same as described for EMSA. Crosslinking products were resolved and visualized by Coomassie stained 4-15% TGX SDS-PAGE (BioRad) and by negative stain microscopy. Samples for negative staining were diluted to a concentration of 250 nM, then applied to glow-discharged, continuous carbon EM grids (Electron Microscopy Science (EMS), CF400-Cu-50) and stained with 2% uranyl acetate (EMS, 22400-2). EM grids were imaged in a Talos L120C transmission electron microscope operated at 120 kV. Micrographs were collected manually at magnifications of 36,000×, 45,000x, or 73,000x
## Cryo-EM sample preparation and data collection
Cryo-EM samples were prepared with purified UL52 (7 µM final concentration) and purified UL32 (15 µM final concentration). Prior to freezing UL32, 4 µM UL32 was incubated with 5 µM 30 bp dsDNA (as used for EMSA) for 30 min at room temperature, chemically crosslinked with 2 mM EGS for 30 min on ice, quenched with 50 mM Tris-HCl, and concentrated to 15 µM in a 30 kDa MWCO Amicon Ultra-15 centrifugal filter. Samples (4 µL) were applied to C-flat R2/1, 300 mesh, holey carbon copper grids (Electron Microscopy Sciences) after glow discharging the grids at 11 mA for 30 sec (Pelco). CHAPSO detergent was added to a final concentration of 0.05% just before grid freezing to ensure uniform ice thickness and particle distribution. Samples were vitrified with a Vitrobot Mark IV system (Thermo Fisher Scientific) maintained at 10°C and 100% humidity. Grids were blotted at blot force 1 for 6 sec and plunge frozen in liquid ethane.
Grids were screened for optimal ice thickness and particle concentration. Data collection was carried out from a single screened grid using a 300 kV Titan Krios cryo-transmission electron microscope (Thermo Fisher Scientific) equipped with a K3 camera (Gatan) and an imaging energy filter (Gatan) operated at a slit width of 15 eV. The dataset was collected in counting super-resolution mode with a nominal magnification of 81,000x leading to a physical pixel size of 1.07 Å (super-resolution pixel size is 0.535 Å). The data were collected at a dose rate of 15 e-/pixel/sec with a total electron dose of 50 e-/Å2 applied over 40 frames and a targeted defocus range of -1.0 µm to -2.0 µm.
## Cryo-EM data processing and model building
The cryo-EM data processing workflow was carried out in CryoSPARC v4 37 . Movie frames were motion corrected using Patch motion correction and binned two-fold to yield a motion-corrected micrograph stack with a pixel size of 1.07 Å. Micrographs were manually curated and an initial round of particle picking was performed using blob picker on a subset of 1,000 micrographs. A 2D classification job on the initial particle stack yielded 2D templates that were used for template-based particle picking on the entire dataset. The template-picked particle stack was curated using 2D classification and used for ab initio reconstruction and heterogenous refinement of the partial stack. For UL52, classes for the 3-mer and 4-mer were separately refined using additional 2D classification and heterogenous refinement. No symmetry restrains were applied to maps (C1) though in the case of UL32 C3 symmetry could be applied and increased the resolution of the map (C1 2.90 Å vs. C3 2.74 Å). Final particle stacks were polished using Reference-based motion correction and used for NUrefinement to create the final reconstructions. Final reconstructions were uploaded to the Electron Microscopy Data Bank (EMDB).
Further analysis of the UL32 consensus particle stack to resolve incomplete tripentamers (i.e., 14-vs. 15-mers) was done with iterative 3D classification with focus masks, homogenous, and non-uniform refinement. The UL32 consensus particle stack was further analyzed for the presence of dsDNA probe included in the cryo-EM sample. Classes with additional density consistent with the DNA were generated by C3 symmetry expansion followed by iterative 3D classification with a focus mask and homogenous refinement.
Model building was done in Coot 38,39 using starting templates generated with AlphaFold3 40 . After initial rigidbody docking, residues were manually fit into the highresolution map using real-space refinement modules. Rounds of real-space refinement in PHENIX 41 were carried out and the outliers were manually corrected in Coot before deposition in the Protein Data Bank (PDB).
## Structure analysis and sequence alignment
Structures were visualized and figures generated with ChimeraX 42 . Electrostatic surfaces are shown with coulombic coloring. Structural analysis included two additional models generated by Alphafold3 40 : in Fig. 6a an ORF68 model was used to represent regions that were unmodeled in the ORF68 crystal structure (6XF9); and in Fig. S6f,g a 30 bp dsDNA model was docked into densities that could likely be attributed to dsDNA. The sequence of packaging accessory factors from the nine human herpesviruses were obtained from UniProt 43 (HSV-1, P10216; HSV-2, P89455; VZV, P09282; HCMV, P16793; HHV-6A, P52463; HHV-6B, Q9QJ33; HHV-7, P52464; EBV, P03184; KSHV, F5HF47) and aligned using Clustal Omega.
## Cell lines and viruses
HEK293T cells (ATCC CRL-3216) were maintained at 37°C with 5% CO2 in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (LifeTech). Vero cells (ATCC CCL-81) and the UL32complementing cell line 158 20 were maintained at 37°C with 5% CO2 in DMEM supplemented with 5% FBS and 1% penicillin-streptomycin. The HSV-1 UL32-null mutant virus hr64FS was previously described 21 .
## Transient transfection and western blot analysis
HEK293T cells were plated in 6-well plates and transfected after 24 hours at 70% confluency with polyethylenimine (PEI). Cells were harvested 24 hours later by washing in DPBS, then resuspending cells in DPBS prior to centrifugation at 500 x g for 5 minutes at 4°C. Cell pellets were stored at -80°C. Samples were prepared for western blot analysis by resuspension in lysis buffer (150 mM NaCl, 50 mM Tris pH 8.0, 1 mM EDTA, 0.5% NP-40, cOmplete protease inhibitor [Roche]) and rotating at 4°C for 1 hour. Insoluble material was removed through centrifugation at 15,000 x g for 10 minutes at 4°C, then the total protein concentration in the cell lysate was determined by a Bradford protein assay. Lysate was mixed with Laemelli loading buffer and 30 μg of total protein was separated on a 4-15% TGX SDS-PAGE gel (Bio-Rad) prior to transfer onto PVDF membrane. Membranes were blocked with 5% milk in TBST buffer (Tris-buffered saline, 0.2% Tween 20) and incubated overnight at 4°C with mouse monoclonal anti-GAPDH (1:5000, cat. no. AM4300 clone 6C5, ThermoFisher) or mouse monoclonal anti-Strep Tag II (1:500, cat. no. PIMA537747, Invitrogen) primary antibodies. The next day, the membrane was washed in TBST and incubated for 1 hour at room temperature with secondary antibody (polyclonal goat anti-mouse HRP, 1:5000, Southern Biotech). The blot was washed with TBST then incubated with Clarity Western ECl substrate (Bio-Rad) and imaged on an Azure Biosystems 600 imager.
## Transient complementation assay and western blot analysis
The UL32 transient complementation assay using hr64FS was performed as previously described 21 . Briefly, Vero cells grown to ∼75% confluence in a 12-well plate were transfected with 0.5 μg of empty, wild-type, or mutant plasmid using the Lipofectamine 2000 reagent (Thermo Fisher) according to the manufacturer's protocol. At 14-16 h posttransfection, cells were superinfected with hr64FS at an MOI of 5 PFU. At 24 h post infection, media and cells were harvested, centrifuged to separate virus-containing supernatant from cells. Cell pellets were washed with 1x PBS, reconstituted in 1x SDS loading buffer and subjected to western blot analysis for protein expression as described above. A plaque assay to quantify total viral yield was performed by infecting 158 (UL32-complementing) cells with serial dilutions of the viral supernatants. Infected cells were overlayed with 2% w/v methylcellulose prepared in 2% FBS DMEM and incubated for 72 h. Cells were fixed with final concentration of 2% formaldehyde and stained with 1% crystal violet solution. Plaques were counted and viral titers were determined. The percent complementation was calculated by dividing the titer obtained for the mutant plasmid by the titer obtained for the wild-type plasmid and multiplying by 100. The background titers from the empty plasmid samples were subtracted. The following primary antibody were used for western blot analysis: rabbit polyclonal anti-UL32 (1:500; antibody to synthetic antigenic peptide 21 and mouse monoclonal anti-gamma tubulin (1:5,000, Sigma, T6557) used as a loading control.
## Supplementary
## References
1. Samies, James, Kimberlin (2021) "Neonatal Herpes Simplex Virus Disease: Updates and Continued Challenges" *Clin Perinatol*
2. Lawrence (2024) "Human cytomegalovirus and neonatal infection" *Curr Res Microb Sci*
3. Cesarman, Damania, Krown et al. (2019) "Kaposi sarcoma" *Nat Rev Dis Primers*
4. Stern, Withers, Avdic et al. (1186) "Human Cytomegalovirus Latency and Reactivation in Allogeneic Hematopoietic Stem" *Cell Transplant Recipients. Front Microbiol*
5. Heming, Conway, Homa (2017) "Herpesvirus Capsid Assembly and DNA Packaging" *Adv Anat Embryol Cell Biol*
6. Thomsen, Newcomb, Brown et al. (1995) "Assembly of the herpes simplex virus capsid: requirement for the carboxyl-terminal twenty-five amino acids of the proteins encoded by the UL26 and UL26.5 genes" *Journal of Virology*
7. Patel, Rixon, Cunningham et al. (1996) "Isolation and characterization of herpes simplex virus type 1 mutants defective in the UL6 gene" *Virology*
8. Newcomb, Juhas, Thomsen et al. (2001) "The UL6 gene product forms the portal for entry of DNA into the herpes simplex virus capsid" *J Virol*
9. Draganova, Valentin, Heldwein (1913) "The Ins and Outs of Herpesviral Capsids: Divergent Structures and Assembly Mechanisms across the Three Subfamilies" *Viruses*
10. Heymann (2025) "The Triplex-Centric Assembly and Maturation of the Herpesvirus Procapsid" *Viruses*
11. Preston, Coates, Rixon (1983) "Identification and Characterization of a Herpes Simplex Virus Gene Product Required for Encapsidation of Virus DNA" *Journal of Virology*
12. Gao, Matusick-Kumar, Hurlburt et al. (1994) "The protease of herpes simplex virus type 1 is essential for functional capsid formation and viral growth" *Journal of Virology*
13. Nadal, Mas, Blanco et al. (2010) "Structure and inhibition of herpesvirus DNA packaging terminase nuclease domain" *Proceedings of the National Academy of Sciences*
14. Gong, Dai, Jih et al. (2019) "DNA-Packing Portal and Capsid-Associated Tegument Complexes in the Tumor Herpesvirus KSHV" *Cell*
15. Liu, Jih, Dai et al. (2019) "Cryo-EM structures of herpes simplex virus type 1 portal vertex and packaged genome" *Nature*
16. Bigalke, Heuser, Nicastro et al. (2014) "Membrane deformation and scission by the HSV-1 nuclear egress complex" *Nat Commun*
17. Draganova, Zhang, Zhou et al. (2020) "Structural basis for capsid recruitment and coat formation during HSV-1 nuclear egress"
18. Rixon, Schmid (2014) "Structural similarities in DNA packaging and delivery apparatuses in Herpesvirus and dsDNA bacteriophages" *Current Opinion in Virology*
19. Tandon, Mocarski, Conway (2015) "The A, B, Cs of Herpesvirus Capsids" *Viruses*
20. Lamberti, Weller (1998) "The Herpes Simplex Virus Type 1 Cleavage/Packaging Protein, UL32, Is Involved in Efficient Localization of Capsids to Replication Compartments" *Journal of Virology*
21. Albright, Kosinski, Szczepaniak et al. (2015) "The Putative Herpes Simplex Virus 1 Chaperone Protein UL32 Modulates Disulfide Bond Formation during Infection" *Journal of Virology*
22. Borst, Wagner, Binz et al. (2008) "The Essential Human Cytomegalovirus Gene UL52 Is Required for Cleavage-Packaging of the Viral Genome" *Journal of Virology*
23. Gardner, Glaunsinger (2018) "Kaposi's Sarcoma-Associated Herpesvirus ORF68 Is a DNA Binding Protein Required for Viral Genome Cleavage and Packaging" *Journal of Virology*
24. Fuchs, Klupp, Granzow et al. (2009) "Characterization of Pseudorabies Virus (PrV) Cleavage-Encapsidation Proteins and Functional Complementation of PrV pUL32 by the Homologous Protein of Herpes Simplex Virus Type 1" *J Virol*
25. Pavlova, Feederle, Gärtner et al. (2013) "An Epstein-Barr Virus Mutant Produces Immunogenic Defective Particles Devoid of Viral DNA" *Journal of Virology*
26. Couté, Kraut, Zimmermann et al. (2020) "Mass Spectrometry-Based Characterization of the Virion Proteome" *Phosphoproteome, and Associated Kinase Activity of Human Cytomegalovirus. Microorganisms*
27. Harmening, Bogdanow, Wagner et al. (2025) "Interaction of human cytomegalovirus pUL52 with major components of the viral DNA encapsidation network underlines its essential role in genome cleavage-packaging" *J Virol*
28. Didychuk, Gates, Gardner et al. (2021) "A pentameric protein ring with novel architecture is required for herpesviral packaging"
29. Palmer (2010) "Studies on the herpes simplex virus type 1 UL32 DNA packaging protein" *Thesis*
30. Mattson, Conklin, Desai et al. (1993) "A practical approach to crosslinking" *Mol Biol Rep*
31. Krishna, Kong, Gary et al. (1994) "Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA" *Cell*
32. De March, Merino, Barrera-Vilarmau et al. (2017) "Structural basis of human PCNA sliding on DNA" *Nat Commun*
33. Zuccola, Filman, Coen et al. (2000) "Bound to the C Terminus of Its Cognate Polymerase"
34. Appleton, Loregian, Filman et al. (2004) "The Cytomegalovirus DNA Polymerase Subunit UL44 Forms a C Clamp-Shaped Dimer" *Molecular Cell*
35. Baltz, Filman, Ciustea et al. (2009) "The Crystal Structure of PF-8, the DNA Polymerase Accessory Subunit from Kaposi's Sarcoma-Associated Herpesvirus" *Journal of Virology*
36. Murayama, Nakayama, Kato-Murayama et al. (2009) "Crystal Structure of Epstein-Barr Virus DNA Polymerase Processivity Factor BMRF1 *" *Journal of Biological Chemistry*
37. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
38. Emsley, Lohkamp, Scott et al. (2010) "Features and development of Coot" *Acta Cryst D*
39. Afonine, Poon, Read et al. (2018) "Real-space refinement in PHENIX for cryo-EM and crystallography" *Acta Crystallogr D Struct Biol*
40. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
41. Liebschner, Afonine, Baker et al. (2019) "Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix" *Acta Crystallogr D Struct Biol*
42. Goddard, Huang, Meng et al. (2018) "UCSF ChimeraX: Meeting modern challenges in visualization and analysis"
43. (2025) "UniProt: the Universal Protein Knowledgebase in 2025" *Nucleic Acids Res*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793477&blobtype=pdf
|
# P-2189. Performance of Dried Blood Spots Cards for Serologic Detection of HPV16 Antibodies in Oropharyngeal Squamous Cell Carcinoma Patients
Maisha Rahman, Soma Bose, ; Chen, Jessica Burris, Rony Aouad, Susanne Arnold, Melvyn Yeoh, Birgitta Michels, Tim Waterboer, Krystle Kuhs
## Abstract
other known risk factors, much of the risk remains unexplained. Latent cytomegalovirus (CMV) has been hypothesized to affect lung function through effects on natural killer cells, systemic inflammation, and direct effects on lung tissue. Prior analyses have been performed in populations with limited racial/ethnic, biologic sex, and/or geographic representation. The National Health and Nutritional Examination Survey (NHANES III) was a population-based sample from the United States.1 /forced vital capacity (FVC) < 0.7]; 2) FEV 1 /FVC (continuous), and FEV 1 (continuous). Binary outcomes were analyzed using logistic regression and continuous outcomes were analyzed using linear regression. All analyses used robust variance estimation for confidence intervals and P-values and were adjusted for important covariates listed in the tables below.1 /FVC ratio and FEV 1 (Table 1). Results did not significantly differ in never smokers vs ever smokers (Table 2).
Background. Antibodies against the human papillomavirus type 16 (HPV16) E6 oncoprotein are a promising biomarker for the early detection of HPV-driven oropharyngeal squamous cell carcinoma (HPV+OPSCC). However, standard serologic testing poses logistical barriers in underserved regions where HPV+OPSCC incidence is high. Dried blood spot (DBS) cards offer a low-resource alternative, but have not been evaluated for HPV antibody detection.
Methods. 25 OPSCC patients were recruited from the University of Kentucky and provided paired serum (venipuncture) and DBS (finger prick) samples. HPV16 antibodies (L1, E1, E2, E4, E6, E7) were measured using multiplex serology; levels were quantified as median fluorescence intensity (MFI) and dichotomized using established cutoffs. Correlation between serum (gold standard) and DBS MFI values was assessed using linear regression and Bland-Altman plots. Sensitivity, specificity, and Cohen's kappa were calculated to evaluate agreement.
Results. Mean MFI levels were lower in DBS than serum, but values were strongly correlated with R² ranging from 0.54 to 0.93. The R-squared for HPV16 E6 antibodies was 0.68. Among the 20 HPV16 E6 seropositive participants (serum), 18 were seropositive on DBS (Sensitivity: 90%); all five HPV16 E6 seronegative participants (serum) also tested seronegative on DBS (Specificity: 100%; kappa: 0.787). Specificity was 100% across all markers, while sensitivity varied from 0% (HPV16 L1) to 100% (HPV16 E2).
Conclusion. DBS cards are an accurate, inexpensive, and scalable alternative to serum-based testing for HPV16 E6 antibodies, particularly in medically underserved regions. Further studies are needed to validate these findings and support broader implementation.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12513077&blobtype=pdf
|
# Arthropod-borne disease challenges from planetary warming, urbanization, and migration
Maggie Bartlett, Kristi Miley, Sten Vermund, Scott Weaver
## Abstract
As the world confronts simultaneous climate and health emergencies, the spread of emerging infectious diseases, particularly arboviruses, underscores the intersection of planetary health, global mobility, and disease risk. Viral pathogens like Oropouche, dengue, and chikungunya are extending their reach, with expanding vector habitats (ticks, mosquitoes, and others) driven by global warming and changing ambient humidity. Arboviral risks due to these unfavorable vector dynamics are exacerbated by voluntary and involuntary migration of people, urbanization with attendant crowding, and suboptimal water, sanitation, and garbage disposal capacities. The poor surveillance and infectious disease control capacities in low-income settings are now exacerbated by public health infrastructure retrenchments in high income nations like the United States. We emphasize the need for urgent, transdisciplinary integration of climate science, epidemiology, human and animal research, and global health security, suggesting bold strategies to prepare for a new era of cross-border microbial threats.
trypanosomiasis, bartonellosis, and others.) Dengue has been documented for centuries, with major outbreaks becoming more frequent after World War II, spreading globally with the proliferation of Aedes mosquitoes. Chikungunya was first described in Tanzania in 1952 and had caused episodic outbreaks across Africa and Asia before its explosive emergence in the Indian Ocean region in the mid-2000s and the Americas by 2013. Zika virus, initially isolated in Uganda in 1947, was obscure until its rapid spread across the Pacific and into the Americas in 2015, where it caused a global health emergency due to its link with congenital birth defects. Yellow fever, one of the oldest known arboviral diseases, has caused deadly urban outbreaks in Africa and the Americas since the 17th century, with recent surges highlighting gaps in vaccine coverage. Oropouche virus has caused recurring outbreaks in the Amazon basin since the 1960 s and is increasingly recognized as an increasing global threat due to
## The expanding climate-pathogen nexus
Arbovirus dynamics depend on vectorial capacity, human habitats, migration of humans and vectors, and public health capacities. Viral infections like dengue, chikungunya, Zika, yellow fever, and Oropouche have emerged and re-emerged in complex patterns driven by ecological, social, and environmental change [1]. (Vector dynamics are relevant for a host of parasitic and a few bacterial diseases as well, e.g., malaria, filariasis, leishmaniasis, urbanization and deforestation. These viruses and many others illustrate how changing landscapes, global movement, and vector ecology have shaped past transmission dynamics, and continue to threaten human health through arboviral disease emergence and expansion.
The accelerating spread of arboviral diseases is a consequence of extreme weather patterns, fuelled by greenhouse gas emissions that warm the planet and increase humidity in many venues. Record-breaking heat waves, now frequent around the globe, are no longer environmental anomalies and they are driving forces behind expanding mosquito populations and heightened transmission of viruses like dengue, chikungunya, and Zika {Meher, 2025 #332}. Warmer temperatures accelerate mosquito reproduction, reduce viral incubation time inside vectors, and expand the geographic range of species like Aedes aegypti and Ae. albopictus into previously temperate zones. Simultaneously, extreme rainfall and flooding create ideal larval habitats, while drought can push communities to store water in containers that become additional Aedes spp. or Anopheles spp. habitats. These climatic shifts have coincided with surges in arboviral outbreaks from South America to South Asia, underscoring the urgent need to address vector-borne disease preparedness as an integral part of climate resilience planning.
Rising temperatures, intensified rainfall and drought cycles, water storage practices, and loss of biodiversity (e.g., deforestation, declining bird populations) optimize conditions for arbovirus vectors to thrive, even in previously inhospitable regions in temperate latitudes and higher altitudes [2]. In 2024 following outbreaks in Brazil, Peru, and Cuba, the World Health Organization flagged Oropouche virus as a disease of global concern [1]. Enviromental imbalance is a critical factor contributing to spread of Oropouche virus and other arboviruses {Barreto, 2024 #333}. Dengue incidence has exploded in Southeast Asia, Latin America, and parts of Africa where it was hitherto unknown; dengue is now an emerging threat to the southeastern United States and southern Europe where competent mosquito vectors, especially Ae. aegypti, are highly prevalent [3][4][5][6].
Climate change is acting not only as a threat multiplier, but as an architect for new emerging transmission zones [6]. Scientists use a number of terms to describe the geographic elements of changing disease ecology driving emerging and reemerging infectious diseases, including geographic suitability and expansion, spatial epidemiology, and risk mapping. Following heavy rains, the temporal clustering of outbreaks typically follows 2-4 weeks later, attributed to the increased water accumulation from storms that create ideal larval habitats for Aedes spp. mosquitoes. (Paradoxically, extreme weather with very high winds (such as hurricanes with North Atlantic Ocean, central and eastern North Pacific Ocean, or Caribbean Sea origins, cyclones with South Pacific or Indian Ocean origins, and typhoons with northwest Pacific Ocean origins, typically affected East/ Southeast Asia, may diminish mosquito populations, decreased short-term arbovirus and malaria risks.) Rising dengue cases and annual virus detection in mosquitoes in Florida, Texas, Arizona, and southern France underscore growing risks in high income nations, with autochthonous (i.e., local, not travel related) arboviral cases in recent summers in Florida and France are likely due to climate shifts. It now is incumbent in newly geographically exposed areas for public health agencies to implement preparedness measures ahead of widespread outbreaks and perhaps prevent endemicity.
## Migration, displacement, and public health blind spots
As of April 2025, the United Nations High Commissioner for Refugees (UNHCR) estimated over 122.1 million people worldwide to have been forcibly displaced from their homes due to civil strife, war, or other violence or persecution. Climate-driven migration adds to this human suffering, placing enormous strain on already fragile health systems [7]. Crowded refugee settlements, usually lacking vector control or sanitation, allow for vector-host interactions that can be flashpoints for viral amplification. Typically, migrants and displaced people are not included in national surveillance, diagnostics, and vaccine access that focuses on the indigenous citizenry. Varying border policies hamper the effective tracking of pathogen spread. These blind spots are both inequitable and dangerous to the worldwide community, given the ability of arboviruses to amplify locally and then spread globally. A new epidemic strain of chikungunya virus that arose in East Africa in 2004, associated with drought, spread to Indian Ocean islands, India, Southeast Asia, and, by 2013, to the Caribbean and the Americas, causing millions of cases of severe, often chronic arthralgia [8].
## Surveillance systems are not fit for purpose
Despite decades of discussion about "One Health, " global infectious disease surveillance remains fragmented, often neglecting non-human animal surveillance [9]. Hence, zoonotic spillover events are poorly monitored and may go undetected. Syndromic surveillance detects diseases late, leaving vulnerable populations at higher risk of infection transmission. Many nations lack both real-time diagnostic capacities and robust genomic sequencing, tools to enable prompt detection that would allow timely outbreak responses to prevent spread. Wastewater surveillance is a key surveillance tool since it was used for poliovirus in the 1940 s and more widely using molecular diagnostics in the 1970 s; it remains underutilized in regions where it could be most transformative for early outbreak detection.
Global threats demand global visibility. Investments made by high income nations in global community-based surveillance, real-time data sharing, mobile diagnostic platforms, and cross-border early warning networks can help with public health responses to protect local and international populations. Highly vulnerable ecosystems typically have surveillance gaps that can results in amplification of transmission and increased disease. Artificial intelligence (AI) is a powerful tool in how we understand and model threats, but will depend on accurate and adequate data that feeds the models. Social media are a new source of such information, as persons close to disease emergence typically make social media queries that can be harvested and interpreted. The growing use of AI underscores the need to fund collection of surveillance data to create the most robust and reliable AI tools. These tools are powerful, but without proper input lack the capacity to be truly impactful in preventing outbreaks. These efforts were already underfunded and have been curtailed significantly by funding changes in the USA to global programs. And while several outbreak detection platforms have been developed globally, universal usage is lagging for cross-country comparisons and sustained funding to continue operations is under threat {Zhang, 2024 #7;Zhang, 2014 #336}. There exist frameworks for detection of outbreaks from foodborne illnesses that could be adapted to aid in detection, mitigation, and response efforts to arboviruses as well as expansion of novel point-of-care testing strategies that were pivotal during the SARS-CoV-2 pandemic {Kennedy, 2025 #337}. Numerous rapid tests have been devised in academic labs, but few are available for commercial use {Zamil, 2025 #339}{Liu, 2025 #338}{Limothai, 2025 #340}. These rapid tests, while beneficial for diagnosing even in austere environments, limit the public health infrastructure's surveillance capacity and implementation should take into consideration how those data can be collected as with platforms that leverage geo-tagged results {Coopersmith, 2025 #341}.
## Mitigation measures are suboptimal
The current landscape of arbovirus vaccines and therapeutics remains uneven, with substantial progress for some viruses and critical gaps for others. For dengue virus, licensed vaccines like Dengvaxia® and Qdenga® offer partial protection but require careful consideration due to limitations such as prior exposure requirements and varying efficacy by serotype and age group. A chikungunya virus vaccine (Ixchiq®) was approved by the FDA in 2023, marking a milestone in proactive vaccine development for emerging threats. However, no licensed vaccines exist yet for Zika, Mayaro, or Oropouche viruses, despite ongoing clinical trials and/or preclinical efforts. Therapeutic options for arboviral infections are largely limited to supportive care, as no specific antiviral treatments have been approved. Research on small-molecule inhibitors and monoclonal antibodies is advancing, but challenges remain in scalability, cost, and efficacy across virus strains. The sporadic and unpredictable nature of outbreaks complicates both vaccine uptake and sustained investment in therapeutic development, underscoring the urgent need for global strategies that prioritize flexible, broad-spectrum countermeasures. This is further compromised by public health blindspots such as the gap between presentation at a hospital to diagnosis of cause. West nile virus was largely ignored by the public health community prior to it's widespresd emergence in the Americas in the 2000 s, as was Zika virus prior to the 2015-2016 epidemic, and Chikungunya in 2013-2014 {Musso, 2018 #331}. }. Novel vector control strategies have shown promise, such as Wolbachiabased interventions and gene drive technologies but have been underutilized and applied to control the threat {Minwuy{Macias, 2017 #335}elet, 2023 #334}.
## Training the next generation
There is a critical need to train the next generation of researchers to address the growing threat posed by arboviruses. Climate change, urbanization, and increased global travel have expanded the geographic range of mosquito vectors, leading to more frequent and severe outbreaks in both endemic and previously unaffected regions. Despite this, there remains a significant shortage of trained scientists with expertise in vector biology, virus evolution, transmission dynamics, field surveillance, diagnostic development, and outbreak response. Building a robust pipeline of well-trained arbovirologists, particularly from low-and middle-income countries where these diseases are most prevalent, is essential for advancing research, improving early detection, strengthening public health infrastructure, and supporting global pandemic preparedness.
Multidisciplinary training programs must integrate virology, entomology, genomics, ecology, and public health to equip emerging scientists with the knowledge and tools to confront the evolving challenges of arboviral threats with the context of new AI assets. Since 2010, global funding for virologists, especially for those working on emerging infectious diseases, has experienced a troubling pattern of fluctuation and overall decline in sustained investment. Elimination of the Fogarty International Center of the U.S. National Institutes of Health is a declared priority of the current U.S. Administration in 2025.
Following the 2009 H1N1 influenza pandemic, interest in virology surged, but funding levels quickly plateaued.
Major funding increases typically occur during crises, as with the 2014-2016 Ebola outbreak and the 2019-2022 COVID-19 pandemic, but these were followed by an eventual plummeting of resources once disease frequencies declined. For example, several national research agencies, including in the U.S. and Europe, reallocated pandemic-era funds by 2023, and virologists reported difficulties renewing grants or sustaining lab infrastructures. Virologists in low-and middle-income countries have been hit especially hard, facing reduced international collaborations and cuts in donor funding, particularly from the U.S. This decline is compounded by reduced investment in basic virology, with more emphasis on translational or pandemic-specific outputs, making long-term viral ecology, vector-virus interaction studies, and vaccine pipeline research harder to maintain. Without consistent support, the global virology workforce is shrinking, aging, and losing expertise, just as planetary warming and globalization are accelerating viral emergence. New funding routes to ensure that training continues is critical to meet a growing unmet need. Enhanced support from philanthropists and Foundations can aid in sustained research in arbovirology.
## The ongoing impacts of post-arbovirus chronic conditions
A post-arboviral "pandemic" of arthralgia and myalgia, persistent joint and muscle pain following infection with arboviruses such as chikungunya, Zika, and dengue, represents a growing yet underrecognized global health challenge, alongside long-COVID and other post-viral conditions [10]. As millions of individuals recover from the acute phases of these infections, an alarming number continue to suffer from debilitating, chronic musculoskeletal symptoms that can last for months or even years. This post-acute arthralgia disproportionately affects communities in low-and middle-income countries, where arboviruses are endemic in the face of suboptimal Aedes spp. vector control, and where access to long-term care is limited. Insecticide resistance and adaptive mosquito behaviors demand new approaches. Arboviruses can be lethal, can reduce quality-of-life for survivors, and place a significant economic burden on health systems and households due to lost productivity and increased healthcare needs. The pathogenesis of post-arboviral arthralgias remains poorly understood; research into biomarkers, diagnostics, integrated healthcare models, treatment options, and enhanced surveillance systems are needed to address the long-term impacts of arboviral infections on human health.
## Conclusion
Pandemics are not just biomedical crises; they are events with planetary significance. To prevent the next pandemic, we must act now to understand and address the drivers of pathogen emergence in a changing world, especially as climate factors and urbanization continue to increase the ability of viruses, especially arboviruses, to shift their range, increasing risk to vulnerable populations. It is paramount that we invest in research to monitor, understand, and develop medical countermeasures to arboviruses as they continue to be widespread endemic threats, cause chronic diseases, and emerge in new locations exposing new populations. Collaborative partnerships nurtured by international agencies and the private sector (as with the Global Virus Network in which the authors are embedded) are essential to building regional response capacity that benefit the global community, knowing that a viral threat anywhere is a viral threat everywhere. We must advance knowledge of viruses through (i) data-driven research and solutions, (ii) fostering the next generation of virology leaders, and (iii) enhancing global resources for readiness and response to emerging viral threats through integrated One Health and global health security contexts.
## References
1. Scachetti, Forato, Claro (2025) "Re-emergence of Oropouche virus between 2023 and 2024 in Brazil: an observational epidemiological study" *Lancet Infect Dis*
2. Souza, Weaver (2024) "Effects of climate change and human activities on vector-borne diseases" *Nat Rev Microbiol*
3. Tian, Zheng, Guo (2019) "Dengue incidence trends and its burden in major endemic regions from 1990 to" *Trop Med Infect Dis*
4. Boehmler, Pruszynski (2023) "Response to an outbreak of locally transmitted dengue in Key Largo, FL, by the Florida Keys Mosquito Control District" *J Am Mosq Control Assoc*
5. Rodriguez, Levitt, Khamisani et al. (2024) "Local transmission of dengue in South Florida: a case report" *Cureus*
6. Stephenson, Coatsworth, Waits (2021) "Geographic partitioning of dengue virus transmission risk in Florida" *Viruses*
7. Benson, Brand, Christianson et al. (2023) "Localisation of digital health tools used by displaced populations in low and middle-income settings: a scoping review and critical analysis of the participation revolution" *Confl Health*
8. Chretien, Anyamba, Bedno (2007) "Drought-associated chikungunya emergence along coastal East Africa" *Am J Trop Med Hyg*
9. Begeman, Van Riel, Koopmans et al. (2023) "The pathogenesis of zoonotic viral infections: lessons learned by studying reservoir hosts" *Front Microbiol*
10. Lozano-Parra, Herrera, Calderón (2024) "Chronic rheumatologic disease in Chikungunya virus fever: results from a cohort study conducted in Piedecuesta, Colombia" *Trop Med Infect Dis*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793294&blobtype=pdf
|
# P-1799. Targeting the Endocannabinoid System: Cannabinoid Receptor 2 Activation Modulates Outcomes in Influenza A Virus Infection -Insights from a Murine Model
Ahmed Alsuwaidi, Junu George, Ifrah Ali, Shreesh Ojha
Background. Emerging evidence suggests that Cannabinoid Receptor 2 (CB2) plays a critical role in regulating inflammation during viral infections. This study investigates the therapeutic potential of CB2 modulation in influenza A virus (IAV) infection using a murine model.
Methods. BALB/c mice were infected intranasally with IAV (A/PR/8/34) and treated with either JWH-133 (CB2 agonist) or AM-630 (CB2 inverse agonist). Four groups were studied: 1) IAV+PBS, 2) IAV+AM-630, 3) IAV+JWH-133, and 4) uninfected PBS controls. AM-630 (2 mg/kg) was administered intraperitoneally (i.p) twice daily starting on day -1 (one day before virus inoculation), while JWH-133 (5 mg/kg) was administered i.p from day 0 through day 5. Mice were monitored daily for survival, clinical scores (ruffled fur, lethargy, labored breathing, and hunched posture), and body weight. CB2 expression in lung tissue was quantified via qRT-PCR.
Results. Mice treated with the CB2 agonist (JWH-133) or inverse agonist (AM-630) survived the 15-day study period, whereas IAV-infected mice without treatment experienced 100% mortality by day 13. Clinical signs appeared later in the CB2 agonist group (day 6) compared to the CB2 inverse agonist (day 4) and IAV-only groups, correlating with delayed weight loss and recovery in the agonist group. CB2 agonist-treated mice regained weight after day 8, similar to the control group, while other infected groups did not recover. Lung CB2 mRNA expression was significantly upregulated in the agonist group but reduced in the antagonist group, confirming receptor activation and blockade.
Conclusion. Pharmacologic activation of CB2 reduces disease severity and enhances survival in influenza-infected mice. These findings highlight CB2 as a promising immunomodulatory target for future antiviral therapies, especially in the context of severe respiratory viral infections like influenza.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12497800&blobtype=pdf
|
# Co-infection of canine parvovirus and circovirus in fatal gastroenteritis outbreak among service dogs in Kazakhstan, 2023
Temirlan Sabyrzhan, Marat Kumar, Aidyn Kydyrmanov, Yermukhammet Kassymbekov, Nailya Klivleyeva, Baiken Baimakhanova, Kobey Karamendin, Iryna Goraichuk, April Davis, Amin Tahoun
## Abstract
Introduction: Between November 2023 and January 2024, a severe gastroenteritis outbreak with high mortality occurred among working dogs based in the Almaty region of Kazakhstan. The epidemic was characterized by an acute onset, rapid progression, and resulted in the death of over 100 juveniles (under 12-month-old) and several vaccinated adult dogs. In this study, we investigated the co-occurrence of canine circovirus and canine parvovirus DNAs in clinical samples from affected dogs, performed genetic characterization of the identified viruses, and evaluated their role in the outbreak. Methods: Polymerase Chain Reaction and Massive Parallel Sequencing methods were used in this research. Results: Polymerase chain reaction analysis of clinical samples revealed the presence of canine parvovirus in eight of the ten samples examined. Further, high-throughput sequencing of pooled oral, rectal, and blood swabs revealed that the majority of viral sequences corresponded to viruses in the Circovirus genus (Circoviridae, 42.3%), followed by Protoparvovirus genus (Parvoviridae, 38%), together accounting for over 80% of all viral reads. Discussion: Viral co-infections are a leading cause of mortality in dogs, with canine parvovirus enteritis often complicated by other pathogens such as canine distemper virus, canine coronavirus, and rotavirus. The presence of multiple pathogens can obscure the primary etiology, highlighting the need for comprehensive molecular diagnostics. Our findings underscore the critical importance of advanced molecular diagnostics in resolving complex infectious disease outbreaks in canine populations and inform future strategies for outbreak prevention and control.
## Introduction
Canine infectious diseases represent a persistent and urgent threat to both domestic and working dog populations worldwide, with outbreaks of highly contagious pathogens such as canine parvovirus (CPV), canine distemper virus (CDV), and canine circovirus (CanineCV) frequently resulting in significant morbidity and mortality, especially among young or immunologically naïve animals. Co-infections and viral recombination commonly occur in high-risk canine populations, such as those in shelters and among working dogs, posing additional challenges for accurate diagnosis and effective prevention (Decaro and Buonavoglia, 2012). The rapid transmission dynamics of these viruses, exacerbated by highdensity environments such as kennels and shelters, can lead to explosive outbreaks with case fatality rates exceeding 90% in unvaccinated puppies (Godsall et al., 2010;Miranda and Thompson, 2016). Moreover, studies have shown that parvovirus infections can persist subclinically in partially immunized or maternally protected dogs, allowing silent viral shedding and subsequent spread (Elia et al., 2015). Additionally, the emergence of novel viral strains and co-infections further complicates diagnosis and control, as seen in recent reports of CPV and CanineCV co-circulation, which have been associated with increased disease severity and vaccine breakthrough cases (Faraji et al., 2022). The continued genetic diversification of these pathogens, coupled with factors such as maternal antibody interference and incomplete vaccination coverage, underscores the critical need for enhanced surveillance, rapid molecular diagnostics, and optimized vaccination strategies to mitigate the impact of canine infectious diseases on animal health and welfare (Decaro and Buonavoglia, 2012;Day et al., 2016).
CPV belongs to the genus Protoparvovirus in the Parvoviridae family and is considered to be one of the smallest viruses known (Chung et al., 2020;Singh et al., 2021). Members of the Parvoviridae family can infect various species of mammals, including domestic dogs, raccoons, cats, coyotes, wolves, and marine mammals such as seals. The virus is widespread in the environment and can remain viable for more than a year under suitable conditions (Mylonakis et al., 2016;Karamendin et al., 2024). Dogs with CPV are more susceptible to co-infection with various pathogens (Anderson et al., 2017). In turn, circovirus replication is enhanced in tissues undergoing cellular regeneration, such as those damaged by CPVinduced necrosis (Thaiwong et al., 2016). Circoviruses are usually detected in association with parvoviruses, which are often associated with the occurrence of enteritis. Dual infections, mainly attributed to CPV-2, suggest their synergy in disease development (Hsu et al., 2016).
CanineCV is a non-enveloped, icosahedral, single-stranded, covalently closed circular DNA virus of ≈2 kb in size. It is a species in the family Circoviridae (Li et al., 2013;Giraldo-Ramirez et al., 2020). CanineCV was first identified in the United States in 2012 (Kapoor et al., 2012). In dogs, infection is associated with vasculitis, hemorrhaging, hemorrhagic enteritis, and diarrhea (Niu et al., 2020). Recent studies have suggested that CanineCV may contribute to the severity of disease in coinfections by modulating host immunity. The Rep Canine CV protein suppresses the host immune response, thereby facilitating the replication of CPV-2. This may exacerbate the clinical manifestations of coinfection in dogs (Hao et al., 2022).
This study documents a fatal outbreak of canine gastroenteritis in Kazakhstan, characterized by co-infection with CPV and CanineCV, and highlights the importance of advanced molecular diagnostics in outbreak investigations. To date, no studies have reported on the molecular characteristics of CPV and CanineCV co-infections in Central Asia. Therefore, this study also provides new regional insights into the evolution and circulation of these viruses in previously uncharacterized dog populations.
## Materials and methods
Laboratory and field studies were carried out in compliance with all bioethical standards in accordance with the instructions of the local ethics commission №02-09-130 from 20.10.2022.
## Collection of animal samples
Clinical samples, including oropharyngeal and rectal swabs and blood serum, were collected from a total of 10 dogs (both symptomatic and asymptomatic). Due to late reporting, clinical material was only obtained from symptomatic dogs housed in the facility's quarantine unit. The facility's remaining animals were either asymptomatic or had already died, precluding further sampling. All these factors, unfortunately, did not allow us to collect more samples from dogs. Selection included three juveniles under 6 months of age, three neutered males, and four unsterilized females. Each animal was double-swabbed to enable molecular diagnostic screening. For molecular diagnostics, swabs were placed in DNA/RNA Shield reagent (Zymo Research, USA) and stored at ambient temperature until processing. Duplicate swabs were preserved in viral transport media and archived at -80 °C for future analyses.
## Dog breed representation
## Vaccination history
Until 2023, the kennel used the combined live attenuated vaccine Nobivac ® DP (MSD Animal Health) for prophylaxis against CDV and CPV. Subsequently, the vaccination protocol was updated to the Biocan Novel DHPPi (Bioveta, CZ), a polyvalent live vaccine against parvovirus, CDV, infectious hepatitis, infectious laryngotracheitis, and parainfluenza.
While adult dogs were reportedly vaccinated in accordance with the manufacturer's standard protocols, juveniles were vaccinated earlier than recommended due to the urgent spread of the disease. This crisis-driven decision was made by facility personnel in an attempt to control viral transmission.
## Reverse transcription and polymerase chain reaction
PCR/RT-PCR screening was performed using the OneTaq One-
Step PCR kit (New England Biolabs, USA) with diagnostic primers targeting CDV, Influenza A virus (IAV), canine coronaviruses (CCoV), rotavirus, and canine parvovirus. Primer sequences and cycling conditions followed published protocols (Senda et al., 1995;Jensen et al., 2002;Payungporn et al., 2004;Moës et al., 2005;Ortega et al., 2017). Briefly: DNA/RNA extraction was performed using the QIAamp Viral RNA Mini Kit (Qiagen, Germany) (QIAGEN, 2015) following the manufacturer's instructions. Amplification of the targeted genes of viruses with a concentration of 0.5 mM each primer in a 25 mL final volume of PCR mix. The reactions were performed in an Eppendorf Gradient thermocycler.
## Genome library preparation and sequencing
Libraries were prepared using the QIAseq FX Single Cell RNA Library Kit (Qiagen, Germany) and NEBNext Ultra DNA Library Prep Kit for Illumina (NEB, USA). cDNA fragmentation (450-500 bp) was performed using the NEB Fragmentase kit (NEB, USA). Adenine overhangs and Illumina-compatible adapters were ligated, and libraries were purified and amplified per the manufacturer's protocols. Illumina v.3 chemistry kit (2 x 300 cycles) was used for sequencing on the MiSeq sequencer (Illumina, USA).
## Bioinformatic analyses
Raw reads were processed using the LAZYPIPE pipeline (Plyusnin et al., 2020) that included initial quality control and adapter/low-quality base trimming with Trimmomatic v0.39. Hostderived reads were filtered using BWA-MEM v0.7.17. The remaining reads were de novo assembled with MEGAHIT v1.2.9 (-presets meta-sensitive), and analyzed on a high-performance computer with Geneious Prime software (Biomatters, New Zealand). Homology searches were performed using BLASTn and BLASTx against GenBank's non-redundant and viral reference databases (E-value <10 -25 ).
Alignment and phylogenetic analyses were performed in MEGA X (neighbor-joining, 500 bootstrap replicates, Tamura-Nei model) (Kumar et al., 2018) based on complete genome sequences. Donut charts visualizing viral family distributions were generated in R (v4.2.0) with ggplot2 (Wickham, 2016). Comparative ORF analyses were performed in Geneious (v2025.1). All trees were constructed based on the complete genomic sequences presented in the NCBI database to assess the phylogenetic relationships of the canine circovirus KZ_2024 and canine parvovirus KZ_2024 strains obtained in this study.
## Results
## Outbreak description, clinical presentation, and PCR screening
In September 2023, an outbreak of acute gastroenteritis occurred among high-breed service dogs in suburban Almaty, Kazakhstan. Clinical signs included lethargy, seizures, anorexia, vomiting, and hemorrhagic diarrhea, with a high case fatality rate.
By the end of November 2023, a total of 110 dogs were affected, resulting in significant mortality, particularly among juveniles.
Table 2 summarizes the PCR screening results. All samples tested negative for IAV, CDV, and CCoV. Parvovirus was detected in 8 swabs, while two dogs tested positive for rotavirus. Seven of eight PCR-positive for CPV dogs were vaccinated (Table 1).
## Next-generation sequencing and virome composition
Illumina MiSeq sequencing yielded 3,274,832 raw reads, with 2,968,286 retained after quality filtering. De novo assembly produced 20,158 contigs (mean length: 502 nt; median: 360 nt), which were aligned to local viral protein databases.
The final consensus sequence for KZ_2024 circovirus was 2,063 nt in length, supported by 14,453 mapped reads (42.3%), while KZ_2024 parvovirus consensus was 5,058 nt in length, supported by 12,961 mapped reads (38%), together accounting for over 80% of all viral reads (Figure 1A). This distribution is consistent with previous studies of the sick canine enteric virome (Moreno et al., 2017;Esposito et al., 2022).
The visualization (Figure 1B) highlights that the Circoviridae and Parvoviridae families have the strongest signals (dark blue) in most samples, consistent with the co-infection pattern. Other viruses, such as betapapillomavirus and lentivirus, were less represented in the analysis.
## Phylogenetic analyses
Despite the geographical distance, the KZ_2024 parvovirus strain's complete genome exhibited high genetic similarity to the 2016CPV strains' genome from China (MF805796), suggesting either a stable phylogenetic lineage or a recent introduction from neighboring countries. The data support CPV-2c as the dominant variant in the region. The KZ_2024 circovirus strain formed a distinct clade with Southeast Asian strains (Thailand: MZ826142; Vietnam: MT740195, MT740196) and was clearly separated from most circoviruses (e.g., MT063074) from China (Figures 2A,B).
## Genome organization of detected viruses
## Genomic analysis of canine circovirus KZ
Sequence analysis revealed a high degree of similarity (>99% nucleotide identity) between the Canine circovirus KZ strain and previously described CCV genomes. The complete genome consists of 2,063 nucleotides with a GC content of 50.6%. It contains two predicted open reading frames (ORFs) transcribed from opposite DNA strands, as well as two intergenic regions. These ORFs encode the viral replicase (Rep) and capsid protein (Cap), comprising 269 and 303 amino acids, respectively. Four unique amino acid substitutions were identified in the Cap gene and one in the Rep gene that were not observed in other available CCV sequences, suggesting potential strain-specific adaptations.
The Cap gene shows a balanced nucleotide composition, with a GC content of 53.2% and AT content of 46.8%. Codon position analysis indicates a preference for guanine in the first position (33.6%) and adenine in the second (32.6%). The Rep gene spans 809 nucleotides and displays a moderate AT bias (52.5%), with adenine (31.3%) being the most prevalent nucleotide. This codon usage pattern is consistent with that observed in other global CCV isolates and may reflect evolutionary trends or mutational pressures.
## Genomic analysis of canine parvovirus KZ
The Canine parvovirus KZ strain also demonstrates high sequence similarity to previously reported CPV genomes. Its complete genome comprises 5,058 nucleotides with a GC content of 35.7%, and contains multiple ORFs encoding structural and nonstructural proteins, including NS1 and VP2. The NS1 gene encodes a 668-amino acid non-structural protein, while the VP2 gene encodes the major capsid protein with 584 amino acids.
The VP2 gene is 1,755 nucleotides long and is characterized by a high AT content (64.3%), with adenine (35.2%) and thymine (29.1%) dominating, particularly in the third codon position (A -40.5%, T -44.3%). This pattern aligns with the codon usage observed in Asian CPV-2c strains. The NS1 gene spans 2,007 nucleotides and also exhibits a strong AT bias (64.2%), with adenine being the dominant nucleotide (38.0%). Compared to reference sequences, several unique amino acid substitutions were identified in the VP2 gene of the KZ strain, potentially contributing to its genetic differentiation from previously reported isolates.
## Comparative analysis of mutations
Genetic analysis of the VP2 gene of the Kazakhstan strain of canine parvovirus (KZ) revealed the following substitutions: 370 (Q → R), 426E (N → E)the key substitutionand 440 (A → T), which are all characteristic of the CPV-2c variant (Fu et al., 2022) (Supplementary Table S1). In the NS gene at position 60, a substitution I → V is observed, a characteristic of the CPV-2c strain (Supplementary Table S2). Similar substitutions have been observed in the reference strains from China and Vietnam. The substitutions that remain in the other positions of the isolated strain (Y544F, E545V, L630P) also correspond to the CPV-2c variant (Wang et al., 2016). The CanineCV genome encodes two major proteins: the conserved Rep, which is responsible for replication, and Cap, a variable structural protein (Kotsias et al., 2019). Analysis of the Cap gene (Supplementary Table S3) revealed a combination of mutations that partially coincided with isolates from China and Thailand at positions 13, 16, 29, 58, 79, 95, and 101. This finding indicates significant differences from European and American strains and demonstrates phylogenetic proximity to Asian strains. Concurrently, no alterations were detected in the pivotal immunogenic positions 24R, 50V, 103R, and 111R, suggesting antigenic conservatism. Position 149G, which is considered rare, has been hypothesized to be a marker of a unique subvariant or point adaptation. A unique mutation, designated Q8H, was identified in the Rep gene (Supplementary Table S4), along with 71C, a characteristic that was exclusively observed in the Kazakh and Thai lines. Despite the phylogenetic proximity to the Asian lines, we have revealed mutations 115R and 149Y, which are characteristic of the European and American lines.
## Discussion
Gastroenteritis is common in domestic carnivores, particularly in animals under one year of age, especially in areas with high animal densities, such as kennels and animal shelters (Capozza et al., 2023). In our study, the affected animals ranged in age from 1 month to 6 years, with the majority being juvenile dogs under 12 months old. Canine diseases with similar symptoms can be caused by CanineCV (Decaro and Buonavoglia, 2011), CPV, and rotavirus (Malik et al., 2020). CPV and canine enteric coronavirus (CECoV) are considered the most common viral agents causing gastroenteritis in dogs (Decaro et al., 2011).
In various studies of the virome of domestic animals with signs of enteritis, in addition to CPV and CanineCV, astroviruses, rotaviruses, bocaviruses, and kobuviruses usually predominate. In the intestinal virome of healthy dogs, the majority of reads are usually bacteriophages, the majority (up to 80% or more) of which are Myoviridae, Siphoviridae, Podoviridae, and Microviridae. In dogs with severe diarrhea, other viruses (herpesviruses, retroviruses, etc.) can be detected at low levels as concomitant or background infections (Moreno et al., 2017;Esposito et al., 2022). In this research, the most abundant viral genera were Circovirus (42.3%) and Parvovirus (37.8%). Other detected viruses included betapapillomavirus and lentivirus, but they were less represented in the virome analysis. CPV was first detected in vaccinated dogs by PCR, and further high-throughput sequencing allowed us to distinguish the positives from vaccine strains by finding mutations not specific to vaccine strains in the virome. PCRpositive samples were pooled for further library preparation. In this regard, we were unable to compare the results obtained in PCR with data obtained by Next-Generation Sequencing.
The BLAST results showed that the CPV sequence we obtained is highly similar to three reference sequences registered in the NCBI database: MF134808.1, isolated from a dog in China in 2017 and belonging to the CPV-2 genotype; OR296281, a partial genome of the CPV-2a isolate CO2_2011, obtained from a fecal swab of a dog in Coimbatore (India) in 2009/2019; and MT165692, a CPV-2b strain, also isolated in China in December 2017. High identity (99.64%) and complete coverage (100%) indicate that our strain is genetically very close to widespread CPV strains, allowing us to assign it to the same phylogenetic lineages represented by both CPV-2a and CPV-2b.
Mutant strains of CPV-2a, CPV-2b, and CPV-2c are gradually pushing the original CPV-2 into the background (Ogbu et al., 2020). For example, the CPV-2a line is the dominant strain of parvovirus circulating in China (61.81% of isolates) (Zhuang et al., 2019). Mutation analysis has shown that our sequence belongs to the CPV-2c strain. Given the geographical proximity, the possibility of parvovirus B19 migration from China to Kazakhstan cannot be excluded. Additionally, service dogs participating in exhibitions and international competitions are at a higher risk of infection.
Comparative analysis with GenBank data showed that the highest similarity to our strain was shared by two canine circovirus strains from China (OQ198057 and OQ198058), obtained from fecal samples of dogs in 2020-2021, and two strains from Thailand (MZ826142 and MZ826143), isolated from nasopharyngeal samples of dogs with respiratory disease in 2020. The high degree of similarity to these isolates may indicate common evolutionary roots or similar routes of virus spread in these regions. Genetic analysis of the CanineCV revealed phylogenetic proximity to Asian lineages and distance from European ones.
Complete genome analysis revealed several amino acid substitutions that distinguish the outbreak strains. In the CPV VP2 gene, multiple unique mutations were identified that are absent from available GenBank records. Since VP2 plays a key role in host cell binding and antigenicity, these substitutions may affect viral properties and require further investigation (Decaro and Buonavoglia, 2012). The VP2 structural protein is responsible for tropism and adaptation to new hosts. Substitutions in its sequence affect the virus's biological properties and also serve as a marker for identifying CPV type 2 variants (Li et al., 2017). The role of NS1 gene mutations remains unclear. Shaohan's study suggests that substitutions at positions 60, 443, 544, 545, and 630 can affect virion replication and packaging (Li et al., 2022).
Similarly, the Canine circovirus KZ strain exhibited four unique amino acid substitutions in the Cap gene and one in the Rep gene, not observed in previously described CanineCV sequences. As Cap and Rep are essential for virus structure and replication, these findings indicate the genetic divergence of the KZ strain from other global isolates (Hsu et al., 2016). The obtained amino acid substitutions suggest potential strain-specific adaptations, but to date, the biological implications of these mutations have not been explored, and further experimental analyses are necessary. The role of mutations in the genes of these proteins has not been thoroughly studied (Liu et al., 2024). However, evolutionary changes in the Cap gene are partly determined by the immunity of the host animal (Dankaona et al., 2024). Being single-stranded DNA viruses, both CPV and CanineCV undergo mutations due to their lack of proofreading mechanisms (da Silva et al., 2025). This results in ongoing genetic diversification.
The detection of both CPV and CanineCV, along with rotavirus in a minority of cases, suggests that viral co-infection may have contributed to the severity and high mortality observed during the outbreak. Compared to previously described co-infection cases of CPV and CanineCV (Thaiwong et al., 2016), the strain in this study led to much more mortality, but clinical symptoms were similar. It was previously hypothesized that clinical symptoms in co-infected dogs are more severe than in those with single infections (Thaiwong et al., 2016). Supposedly, CanineCV was not merely an incidental finding and likely played a pathogenic role, particularly in synergy with CPV. This conclusion is supported by previous compelling molecular findings and established biological mechanisms (Thaiwong et al., 2016). The similarity of KZ_2024 CPV to CPV-2c and CanineCV strains prevalent in China and other parts of Asia indicates possible regional endemicity. However, these strains are now spreading globally, and their distribution patterns may vary depending on geographic location, sampling period, and commercial movements of dogs imported from abroad (Decaro et al., 2007;Decaro and Buonavoglia, 2012).
In dogs, rotaviruses, which are wheel-shaped, double-stranded RNA viruses of the Reoviridae family, chiefly affect juveniles under 12 weeks of age, producing watery diarrhea and dehydration associated with group A (G3P3) rotavirus infections. This condition is generally mild in adults but can be zoonotic, making fluid therapy and strict hygiene the mainstays of care (Ortega et al., 2017). Although metagenomic sequencing confirmation for the rotavirus-positive PCR result was not obtained, the finding was independently verified by the Central Reference Laboratory. An internal, unpublished report further supported this result.
Canine papillomaviruses encompass more than 25 DNA virus types spread across the Lambda-, Tau-, and Chipapillomavirus genera; transmission by direct contact leads to oral or cutaneous warts that usually self-resolve within weeks, although persistent lesions in immunosuppressed dogs may evolve into squamous cell carcinoma, and viral DNA can be found on apparently normal skin (Medeiros-Fonseca et al., 2023). Lentiviruses, by contrast, are not a major natural canine pathogen-only an isolated canine immunodeficiency virus from a leukemic dog has been described (Perk et al., 1992).
The presence of these additional viruses, even if not the primary cause of mass mortality, underscores the importance of comprehensive diagnostics in understanding the full spectrum of pathogens contributing to disease in complex outbreaks. Some of the co-infecting agents can exacerbate clinical signs, particularly in a high-density environment where animals may experience stress or have underlying parasite loads, further compromising their immune systems. This reinforces the critical importance of advanced molecular diagnostics in resolving complex infectious disease outbreaks in canine populations.
Other, non-viral factors also influence susceptibility to parvovirus enteritis: for example, stress from overcrowding and parasite burden may have also exacerbated the outbreak (Grellet et al., 2013); climatic factors also influence susceptibility. The risk of parvovirus infection in dogs increases significantly in spring, late autumn, and early winter. Seasonal changes in temperature and climate may explain the increased likelihood of both CPV (Qi et al., 2020) and other viral infections. Although it was not possible to perform in situ hybridization (ISH) and other histological studies in our case, it is extremely unlikely that the primary agent and cause of mass mortality was precisely the circovirus. Such cases are extremely rare (Liu et al., 2024), since canine circoviruses mainly act as secondary agents associated with CPV (Zaccaria et al., 2016).
The dog center where the outbreak occurred houses several hundred dogs, a factor that is thought to have contributed to the rapid spread of the infection. All dogs in the kennel were vaccinated with the combined polyvalent vaccine Nobivac DP Plus, which is designed to protect against CPV and is expected to provide reliable immunity when administered correctly. Despite claims that vaccines may not provide complete protection against the different CPV variants, numerous studies (Larson and Schultz, 2008;Glover et al., 2012;Reitzenstein et al., 2012) have shown that CPV-2-based vaccines provide reasonable protection against CPV-2a, CPV-2b, and CPV-2c when the appropriate vaccination schedules are followed (Decaro et al., 2020).
CPV is common in many places and can be transmitted from healthy carriers, contributing to the spread of infection in the pet population. Vaccination against parvovirus can induce a good immune response in most dogs; all service dogs were vaccinated in the first two months after birth, but their samples were PCRpositive for parvovirus. This is probably because vaccinations are too early, which may affect maternal antibodies, increasing the susceptibility of juveniles to virus infection (Mazzaferro, 2020).
Maternally derived antibodies (MDA) can block active immunization after the administration of CPV vaccines.
Vaccination of puppies with MDA titers that prevent vaccination may result in failure to seroconvert due to neutralization of the vaccine virus antigen. There is a period known as the "window of susceptibility" or "immunity gap", averaging 2-3 weeks, during which the MDA titer falls below that required for protection but is still capable of neutralizing the vaccine virus. During this period, puppies can be infected and sometimes develop diseases (Decaro et al., 2020). The "immune window" following vaccination is a temporal vulnerability when the body has not yet mounted a consolidated, protective response. This period necessitates meticulously designed, multi-dose primary series and booster campaigns to achieve and maintain durable immunity. Additionally, the continuous evolution of pathogens through mechanisms like antigenic drift and shift necessitates constant vaccine reformulation and strategic booster campaigns to counter immune evasion (Zimmermann and Curtis, 2019). The confluence of these challenges-a population with waning immunity facing a novel, highly transmissible variant-creates a dynamic landscape where vaccine efficacy is a fluid metric, not a fixed value.
Thus, our data indicate that the key role in the outbreak was played by the conditions of confinement, early vaccination, and additional risk factors (high density of animals in the nursery, possible genetic resistance to vaccination, and possible parasite invasion). The sequencing results obtained confirm the importance of comprehensive diagnostics and the need to monitor both the main infection, canine parvovirus, and the concomitant infection, canine circovirus. Based on the trends identified, a possible preventive measure could be a revision of the vaccination schedule for juveniles, taking into account the level of maternal antibodies, as well as an improvement in the conditions of confinement. These measures, together with regular screening and systematic surveillance, can reduce the risk of such outbreaks. Although genetic resistance to CPV vaccination occurs in approximately one dog in a thousand (Day et al., 2007), epidemiological data are lacking, and further prospective studies are needed to determine the prevalence of primary vaccine failure.
This study has some limitations.
The main limitation of this study is the relatively small sample size (n=10 dogs) compared to the total kennel population (more than 200 dogs). At the time of our investigation, only 10 symptomatic animals were available for sample collection. Most of the other affected dogs had either died or were unavailable.
An additional factor was time and resource constraints: the studies were conducted urgently during an outbreak, reducing the opportunities for more detailed or long-term monitoring of the disease situation. Taken together, these factors may limit the extrapolation of the results to the whole kennel population.
The absence of histopathological or ISH evidence to definitively establish the primary pathogenic role of CanineCV is also acknowledged as a limitation.
Although the hemagglutination inhibition (HAI) assay using porcine erythrocytes was performed, we considered the results to be due to the inconsistency in duplicates and did not include them in the article. Also, limitations may include the absence of virus isolation, vaccine efficacy testing, or measurement of maternally derived antibody titers.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
## References
1. Anderson, Hartmann, Leutenegger et al. (2017) "Role of canine circovirus in dogs with acute haemorrhagic diarrhoea" *Veterinary Rec*
2. Capozza, Buonavoglia, Pratelli et al. (2023) "Old and novel enteric parvoviruses of dogs" *Pathog. (Basel Switzerland)*
3. Chung, Kim, Nguyen et al. (2020) "New genotype classification and molecular characterization of canine and feline parvoviruses" *J. veterinary Sci*
4. Dankaona, Nooroong, Poolsawat et al. (2024) "Canine circovirus: emergence, adaptation, and challenges for animal and public health" *BMC veterinary Res*
5. Day, Horzinek, Schultz et al. (2020) "Vaccination Guidelines Group (VGG) of the World Small Animal Veterinary Association (WSAVA)" *J. Small Anim. Pract*
6. Day, Horzinek, Schultz "Guidelines for the vaccination of dogs and cats. Compiled by the Vaccination Guidelines Group (VGG) of the World Small Animal Veterinary Association (WSAVA)" *J. Small Anim. Pract*
7. Decaro, Buonavoglia (2011) "Canine coronavirus: not only an enteric pathogen" *Veterinary Clinics North America Small Anim. Pract*
8. Decaro, Buonavoglia (2012) "Canine parvovirus-a review of epidemiological and diagnostic aspects, with emphasis on type 2c. Veterinary Microbiol"
9. Decaro, Buonavoglia, Barrs (2020) "Canine parvovirus vaccination and immunisation failures: Are we far from disease eradication? Veterinary Microbiol"
10. Decaro, Desario, Addie et al. (2007) "The study molecular epidemiology of canine parvovirus" *Europe. Emerg. Infect. Dis*
11. Decaro, Desario, Billi et al. (2011) "Western European epidemiological survey for parvovirus and coronavirus infections in dogs"
12. Veterinary (1997)
13. Elia, Camero, Losurdo et al. (2015) "Virological and serological findings in dogs with naturally occurring distemper" *J. virological Methods*
14. Esposito, Esposito, Ptashnik (2022) "Phylogenetic diversity of animal oral and gastrointestinal viromes useful in surveillance of zoonoses" *Microorganisms*
15. Faraji, Sadeghi, Mozhgani et al. (2022) "Detection of canine circovirus in dogs infected with canine parvovirus" *Acta Trop*
16. Fu, He, Cheng et al. (2022) "Prevalence and characteristics of canine parvovirus type 2 in Henan province" *China. Microbiol. Spectr*
17. Giraldo-Ramirez, Rendon-Marin, Vargas-Bermudez et al. (2020) "First detection and full genomic analysis of Canine Circovirus in CPV-2 infected dogs in Colombia" *South America. Sci. Rep*
18. Glover, Anderson, Piontkowski et al. (2012) "Canine Parvovirus (CPV) Type 2b Vaccine Protects puppies with Maternal Antibodies to CPV when Challenged with Virulent CPV-2c Virus" *Int. J. Appl. Res. Veterinary Med*
19. Godsall, Clegg, Stavisky et al. (2010) "Epidemiology of canine parvovirus and coronavirus in dogs presented with severe diarrhoea to PDSA PetAid hospitals" *Veterinary Rec*
20. Grellet, Brunopolack, Boucraut-Baralon et al. (2013) "Prevalence, risk factors of infection and molecular characterization of trichomonads in puppies from French breeding kennels. Veterinary Parasitol"
21. Hao, Li, Chen et al. (2022) "Canine circovirus suppresses the type I interferon response and protein expression but promotes CPV-2 replication" *Int. J. Mol. Sci*
22. Hsu, Lin, Wu et al. (2016) "High detection rate of dog circovirus in diarrheal dogs" *BMC veterinary Res*
23. Jensen, Van De Bildt, Dietz et al. (2002) "Another phocine distemper outbreak in Europe" *Science*
24. Kapoor, Dubovi, Henriquez-Rivera et al. (2012) "Complete genome sequence of the first canine circovirus" *J. Virol*
25. Karamendin, Goodman, Kasymbekov et al. (2024) "Viral metagenomic survey of Caspian seals" *Front. veterinary Sci*
26. Kotsias, Bucafusco, Nuñez et al. (2019) "Genomic characterization of canine circovirus associated with fatal disease in dogs in South America" *PloS One*
27. Kumar, Stecher, Li et al. (2018) "MEGA X: Molecular Evolutionary Genetics Analysis across computing platforms" *Mol. Biol. Evol*
28. Larson, Schultz (2008) "Do two current canine parvovirus type 2 and 2b vaccines provide protection against the new type 2c variant? Veterinary therapeuticыs" *Res. Appl. veterinary Med*
29. Li, Chen, Hao et al. (2022) "Characterization of the VP2 and NS1 genes from canine parvovirus type 2 (CPV-2) and feline panleukopenia virus (FPV) in Northern China" *Front. veterinary Sci*
30. Li, Ji, Zhai et al. (2017) "Evolutionary and genetic analysis of the VP2 gene of canine parvovirus" *BMC Genomics*
31. Li, Mcgraw, Zhu et al. (2013) "Circovirus in tissues of dogs with vasculitis and hemorrhage" *Emerging Infect. Dis*
32. Liu, Qin, Hu et al. (2024) "Epidemiological and evolutionary analysis of canine circovirus from 1996 to 2023" *BMC veterinary Res*
33. Malik, Bhat, Dar et al. (2020)
34. "Evolving rotaviruses, interspecies transmission and zoonoses" *Open Virol. J*
35. Mazzaferro (2020) "Update on canine parvoviral enteritis" *Veterinary Clinics North America Small Anim. Pract*
36. Medeiros-Fonseca, Faustino-Rocha, Medeiros et al.
37. Costa (2023) "Canine and feline papillomaviruses: an update" *Front. Veterinary Sci*
38. Miranda, Thompson (2016) "Canine parvovirus: The worldwide occurrence of antigenic variants" *J. Gen. Virol*
39. Moës, Vijgen, Keyaerts et al. (2005) "A novel pancoronavirus RT-PCR assay: frequent detection of human coronavirus NL63 in children hospitalized with respiratory tract infections in Belgium" *BMC Infect. Dis*
40. Moreno, Wagner, Mansfield et al. (2017) "Characterisation of the canine faecal virome in healthy dogs and dogs with acute diarrhoea using shotgun metagenomics" *PloS One*
41. Mylonakis, Kalli, Rallis (2016) "Canine parvoviral enteritis: an update on the clinical diagnosis, treatment, and prevention" *Veterinary Med. (Auckland N.Z.)*
42. Niu, Wang, Zhao et al. (2020) "Detection and molecular characterization of canine circovirus circulating in northeastern China during 2014-2016" *Arch. Virol*
43. Ogbu, Mira, Purpari et al. (2020) "Nearly full-length genome characterization of canine parvovirus strains circulating in Nigeria"
44. Ortega, Martıńez-Castañeda, Bautista-Goḿez et al. (2017) "Identification of co-infection by rotavirus and parvovirus in dogs with gastroenteritis in Mexico" *Braz. J. Microbiol*
45. Payungporn, Phakdeewirot, Chutinimitkul et al. (2004) "Single-step multiplex reverse transcription-polymerase chain reaction (RT-PCR) for influenza a virus subtype H5N1 detection" *Viral Immunol*
46. Perk, Safran, Dahlberg (1992) "Propagation and characterization of novel canine lentivirus isolated from a dog" *Leukemia*
47. Plyusnin, Kant, Jaaskelainen et al. (2020) "Novel NGS pipeline for virus discovery from a wide spectrum of hosts and sample types" *Virus Evol*
48. Qi, Zhao, Guo et al. (2015) "A mini-review on the epidemiology of canine parvovirus in China"
49. Reitzenstein, Ludlow, Marcos et al. (2012) "Cross protection of vanguard 5L4-CV vaccine against virulent canine parvovirus-2c circulating in the USA" *Int. J. Appl. Res. Veterinary Med*
50. Senda, Parrish, Harasawa et al. (1995) "Detection by PCR of wild-type canine parvovirus which contaminates dog vaccines" *J. Clin. Microbiol*
51. Singh, Kaur, Chandra et al. (2021) "Prevalence and molecular characterization of canine parvovirus" *Veterinary World*
52. Thaiwong, Wise, Maes et al. (2016) "Canine circovirus 1 (CaCV-1) and canine parvovirus 2 (CPV-2): recurrent dual infections in a papillon breeding colony. Veterinary Pathol"
53. Wang, Jin, Li et al. (2016) "Isolation and sequence analysis of the complete NS1 and VP2 genes of canine parvovirus from domestic dogs in 2013 and 2014 in China"
54. Zaccaria, Malatesta, Scipioni et al. (2016) "Circovirus in domestic and wild carnivores: An important opportunistic agent?" *Virology*
55. Zhuang, Qiu, Pan et al. (2019)
56. "Genome sequence characterization of canine parvoviruses prevalent in the Sichuan province of China"
57. Zimmermann, Curtis (2019) "Factors that influence the immune response to vaccination" *Clin. Microbiol. Rev*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12737581&blobtype=pdf
|
# Feather Defects in a Juvenile Common Swift (Apus apus) Associated with a Circovirus Infection
Marko Legler, Kristin Heenemann
## Abstract
The common swift (Apus apus), a widespread wild bird in Germany, can reach high population densities in urban areas. For example, the Hanover region hosts approximately 800 to 1200 breeding pairs. The common swift is known for its unique lifestyle, spending most of the year in continuous flight, interrupted only during the breeding season. During extreme weather conditions, such as heat waves, many juvenile swifts in need of help are brought to wildlife rescue centers and require veterinary care and rehabilitation. However, many aspects of their veterinary treatment remain insufficiently understood. Little is known about viral infections and their transmission in swifts. This case report describes a feather disease of a hand-reared juvenile common swift (Apus apus) housed in a wildlife rescue center in the Hanover region. The swift was referred to the clinic due to progressive feather abnormalities. A circovirus was identified from the abnormally growing feathers using circovirus-consensus nested PCR and was considered as the probable cause for the swift's condition. To our knowledge, this is the first report of a circovirus infection in a swift species.
## 1. Introduction
The common swift (Apus apus) represents the most frequently observed species of the order Apodiformes in Germany. As a long-distance migratory bird, it undertakes annual migrations across Europe to spend the non-breeding season in sub-Saharan Africa, typically from September to May. During this period, common swifts remain airborne for over ten months per year, exhibiting highly specialized aerial behavior [1][2][3][4][5][6][7]. Urban environments can support substantial breeding populations, with national estimates ranging between 215,000 and 395,000 breeding pairs in Germany [8,9]. In the summer months, common swifts are regularly admitted to wildlife rescue centers, particularly juveniles. Admission rates tend to increase significantly during periods of extreme weather, such as heatwaves or extended cold periods, during which several hundred individuals may require care and rehabilitation [10,11]. The housing and appropriate care of large numbers of individuals of this avian species, which has specific biological and husbandry requirements, can pose considerable challenges. This is particularly the case in non-specialized wildlife rescue centers, where contact with other bird species is often unavoidable. In this context, current knowledge regarding infectious diseases in swifts and the potential transmission of pathogens from other avian species to common swifts remains limited [10,12,13]. Circovirids are small icosahedral virions of 20 to 25 nm with a small circular single-stranded DNA genome. The genome encodes for two proteins, the replication-associated protein and the capsid protein. Within the family Circoviridae, the Circovirids are classified into two genera, Circovirus and Cyclovirus [14][15][16][17][18]. Viruses of the genus Circovirus are common pathogens in birds worldwide [18]. Particularly, the beak and feather disease virus (BFDV) is well known and examined in parrots in captivity and the wild [14,19,20]. However, circoviruses have also been detected in other wild bird species, including, for example the BFDV in the rainbow bee-eater (Merops ornatus) [21], the pigeon circovirus (PiCV) in the Eurasian collared-dove (Streptopelia decaocto) and other wild pigeons [22], the goose circovirus (GoCV) in different species of wild geese [23], the duck circovirus (DuCV) in wild ducks [24][25][26], the gull circovirus (GuCV) in the herring gull (Larus argentatus) [27], the swan circovirus (SwCV) in the mute swan (Cygnus olor) or the starling circovirus (StCV) in European starlings (Sturnus vulgaris) and spotless starling (Sturnus unicolor) [28,29]. Circovirus species have recently been identified in Adélie penguin (Pygoscelis adeliae; PenCV) [30] American wigeons (Mareca americana; WigFec Circovirus 1 and WigFec Circovirus 2) [31], little bittern (Ixobrychus minutus; BitternCV), a European bee-eater (Merops apiaster; Bee-eaterCV) [32] and a tawny owl (Strix aluco; ToCV) [33]. Circoviruses have also been detected in other members of Passeriformes, particularly in different species of corvids and the blackbird (Turdus merula) [34]. In captivity, Circoviruses are also widespread among finches, e.g., the canary circovirus in Atlantic canary (Serinus canaria; CaCV), the finch circovirus in Gouldian finch (Erythrura gouldiae) and the zebra finch circovirus in Taeniopygia guttata [34][35][36][37]. However, there are little information's about the circulation of circovirus infections in the wild bird populations and the expected pathology associated with such a viral infection [14,18,34]. Especially for swifts worldwide, there is no information available in the literature. Among avian hosts, circoviruses are characterized by a tropism for epithelial cells and lymphatic tissues [38]. Therefore, avian diseases caused by circovirus are mainly characterized by feather abnormalities and causally associated with immunosuppression. Immunosuppression can lead to growth retardation and wasting and increase susceptibility to secondary infections [38]. Morbidity and mortality rates can vary significantly from case to case depending on the bird species affected and the virus involved [38]. Vertical transmission of circoviruses via the egg has also been described [34].
The aim of this case presentation is to describe feather defects associated with a circovirus in the common swift (Apus apus).
## 2. Case Report
## 2.1. Clinical Case
A juvenile common swift with a history of progressively developing feather defects was presented for clinical examination by a wildlife rehabilitator. At the time of presentation, the bird had been kept in a wildlife rescue center for 17 days. During a period of warm weather, the bird had fallen out of the breeding site, but showed no changes in its feather growth at this time and was classified as healthy. The animal was kept in its own box in a room with other juvenile swifts and other different wild bird species with a frequently changing stock. The bird was on a diet with steppe crickets supplemented two times a week with Korvimin ZVT (WDT, Garbsen, Germany). Vitamin B was supplemented every 10 days sub cutaneous by the attending veterinarian. The loss of feathers has already been observed for a week at the time of the investigation.
At the time of presentation, the juvenile swift was about 35 days old and in good physical condition with 36 g body weight. The development corresponded to the age of the common swift, and the bird showed normal begging behavior. The main findings were disturbances in feather development. Body feathers as well as wing, primaries and secondaries, and tail feathers were affected. The affected growing primaries and tail feathers showed retention of feather sheaths. A malformation of the feather vane was recognizable in the areas of the pathologic remaining feather sheaths, described as fault bars or "hunger streaks". Feathers stopped the crowing, showed circumferential constrictions and fractures of the feather shafts and fell out of the feather follicles (Figure 1). Some feathers showed hemorrhage within the pulp cavity.
In order to rule out a viral infection, fresh growing feathers of the shoulder feathers and lost feathers were sampled and examined in a polyomavirus and a circovirus family-specific consensus-nested PCR [28,39].
The bird was euthanized due to the already heavily destroyed plumage at the time of presentation, and the poor prognosis for reintroduction into the wild, and the suspected viral infection with unfavorable prognosis for recovery. Unfortunately, we were not allowed to examine the animal further. In addition, swifts without feather defect and with no known contact with other wild or diseased birds in the region of Hanover were screened for circo-and polyomaviruses. A total of 18 adult and 17 juvenile swifts, aged between 25 and 40 days, were euthanized due to severe injuries that precluded the possibility of successful rehabilitation. Feather samples collected postmortem were used for the virological examination.
## 2.2. Virological Examination
The feathers were stored at 4 • C for further molecular diagnostic investigations. The DNA was extracted from a blood keel using the DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany). In order to investigate for viruses of the polyomavirus and circovirus families, family-specific consensus-nested PCRs were performed as described previously [28,39]. In the first-round of the Circovirus family-specific PCR, a product of approximately 600 bp was generated using the primers Cv-s (5 ′ -AGAGGTGGGTCTTCACNHTBAAYAA) and Cvas (5 ′ -AAGGCAGCCAC-CCRTARAARTCRTC). The reaction mixture, with a total volume of 25.0 µL, consisted of the following components: 5.0 µL of 10× Pfx Amplification Buffer, 0.75 µL of dNTPs (10 mM each), 0.5 µL of MgSO 4 (50 mM), 0.75 µL (10 µM) of each primer, 14.85 µL of DEPC-treated water, 0.4 µL of Platinum ® Pfx DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, USA), and 2.0 µL of plasmid DNA or DNA (≤100 ng).
The thermal profile for the PCR product's production was as follows: an initial denaturation at 94 • C for 5 min, followed by 35 cycles of 94 • C for 15 s, 60 • C for 30 s, and 68 • C for 40 s, and a final elongation at 68 • C for 2 min.
Subsequently, a second PCR was performed using the internal primers Cn-s (5 ′ -AGCAAGGAACCCCTCAYYTBCARGG) and Cn-as (5 ′ -ACGATGACTTCNGTCTT-SMARTCACG). The reaction mixture for this nested PCR, with a total volume of 25.0 µL, consisted of the following components: 5.0 µL of 10× Pfx Amplification Buffer, 0.75 µL of dNTPs (10 mM each), 0.5 µL of MgSO4 (50 mM), 0.75 µL (20 µM) of each primer, 14.85 µL of DEPC-treated water, 0.4 µL of Platinum ® Pfx DNA Polymerase, and 2.0 µL of a 1:10 dilution of the first PCR product.
The thermal profile used to generate the 350 bp PCR product was an initial denaturation at 94 • C for 5 min, followed by 35 cycles of 94 • C for 15 s, 60 • C for 30 s, and 68 • C for 20 s, with a final elongation at 68 • C for 2 min.
The resulting PCR products were analyzed on a 1.5% agarose gel alongside a 100 bp DNA ladder (Thermo Fisher Scientific, Waltham, MA, USA). The gel was stained with ethidium bromide and viewed under UV light. The polyomavirus family-specific PCR was performed using the same protocol as for circoviruses, with primer combination PVs (5 ′ -CCAG-ACCCAACTARRAATGARAA) and Pvas (5 ′ -AACAAGAGACACA-AATNTTTCCNCC) used in the first round and primer combination Pns (5 ′ -ATGAAAATGGGGTTGGCCCNCTNTGYAARG) and Pnas (5 ′ -CCCTCATAAACCCGAACYTCYTCHACYTG) used in the second round. After purification with the GenJet PCR Purification Kit (Thermo Fisher Scientific, Waltham, MA, USA), the positive PCR products were sent to Microsynth Seqlab GmbH (Göttingen, Germany) for sequencing. The Sanger method was applied for sequencing using the internal primers of the circovirus consensus nested PCR. To obtain the full genome, we also employed inverse PCR and rolling circle amplification (RCA) with both degenerate and specific primers [29,40]. The specific primers used for this process were designed based on the initial partial sequence obtained from the consensus PCR. Sequence quality control and editing were performed using the GENtle program (Version 1.9.0; Magnus Manske, University of Cologne, Germany). The identity of the processed sequences was confirmed by using the Basic Local Alignment Search Tool of the National Center for Biotechnology Information (NCBI; https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 7 April 2025). Phylogenetic analysis and the construction of phylogenetic trees were conducted using the software MEGA 12 [41,42]. The sequences were aligned within MEGA using the ClustalW algorithm. Phylogenetic relationships were subsequently determined using the Maximum Likelihood method, based on Kimura 2-parameter model [43]. The statistical support for the tree topology was evaluated through a bootstrap analysis with 1000 replicates.
## 2.2.1. Results of the Virological Examination
The virological examination of the dystrophic feathers revealed a positive result for the circovirus-consensus-nested PCR and a negative result for the polyomavirus-consensusnested PCR. Molecular characterization of the isolate yielded a partial sequence (349 bp) that was deposited in the NCBI-GenBank database under accession no.: PV416848. The NCBI Blast analysis showed the highest identity in 300 base pairs (bp) of 83.11% to a Canary circovirus isolate (NCBI accession no.: NC 003410.1).
No circo-or polyomaviruses were detected in feather samples from adult and juvenile common swifts without feather abnormalities and no history of staying at a wildlife rescue center.
## 2.2.2. Phylogenetic Analysis
For the phylogenetic analysis, we used a comprehensive set of complete avian circovirus genome reference sequences. This set included 17 available reference sequences of avian circoviruses from various hosts from the NCBI-GenBank database. The partial replicase gene sequences of the circovirus from this study (349 nucleotides) and all reference sequences were aligned, and a phylogenetic tree was constructed, representing the tree with the highest log likelihood (-4949.16). The percentage of trees in which the associated taxa clustered together is shown next to the branches. Phylogenetic analysis showed that the circovirus sequence from this study clustered with the Canary circovirus isolate (NCBI accession no.: NC 003410.1). The sequence showed a considerable genetic distance from all other avian circovirus references used in the analysis (Figure 2). To provide a comprehensive analysis, we also included 35 recently published partial sequences [34] from various hosts with a focus on finches and other passerines in a separate phylogenetic reconstruction to ensure our analysis was as state-of-the-art as possible. This analysis, shown in Figure 3, demonstrates that the swift circovirus sequence (PV416848) groups with the sequences of a captive red-fronted serin (Serinus pusillus; PQ299758.1; Redfronted serin circovirus isolate 045) and a common crossbill (Loxia curvirostra; PQ299776.1; Common crossbill circovirus isolate 712). -No.: PV416848, high-lighted with a black circle) and various circovirus isolates from GenBank. Phylogenetic analysis was conducted using the Maximum Likelihood method with the Kimura 2-parameter model. The final dataset of this analysis comprised 37 nucleotide sequences with a total of 220 positions. To ensure high data quality, all positions with less than 95% site coverage were excluded. This means that alignment gaps, missing data, and ambiguous bases were allowed at any position only if they constituted less than 5% of the total data. The tree with the highest log likelihood (-1524.31) is displayed. The common swift's sequence is marked with a black circle. The canary circovirus reference genome sequence is indicated with an empty square. All other sequences are from the study by Ledwo ń et al. (2025) [34]. The scale bar in the phylogenetic analysis shows how many substitutions per nucleotide position have taken place on average to explain the observed sequence differences.
## 3. Discussion
The presented case of a juvenile common swift with feather loss associated with a circovirus is to our knowledge the first detection of a circovirus infection in a swift species [14][15][16][17][18]. Feather defects with circumferential constrictions were the main clinical indication of a possible circovirus infection and were the clinical focus in our case. The feather growing abnormalities that were observed in the juvenile common swift correspond to the disease symptoms in other bird species in connection with a circovirus, particularly in psittacines and especially budgerigars with the BFDV virus [19][20][21]. Similar to budgerigars with BFDV, it can be assumed that the age of a swift also has an influence on the severity of the viral disease [14,19]. The growth of the feathers in the juvenile swift probably also favored the typical feather loss. However, circovirus infections in finches, e.g., in canaries, cause mainly an increase in morbidity and mortality, but not feather defects [34,37,44]. In the swift case, there were no indications of a disturbed general condition in our examinations. The swift was in good nutritional condition with normal begging behavior. Unfortunately, immunocompetence could not be investigated further in this case though immunosuppression by circoviruses is described for other species, such as canaries, pigeons, ducks and geese [19,25,28,37,44].
The analysis of the partial sequence of the detected circovirus in the common swift showed the highest identity to a CaCV isolate and other circoviruses isolated from finches. Viruses are classified into separate species if their genomes share less than 80% genomewide pairwise sequence identity with the classified family members [43]. This requirement is not fulfilled in our case, but we were only able to examine a partial sequence. Further research on circoviruses in swifts is essential to clarify this matter and to illustrate the spread of circoviruses in the common swift population.
Our investigations do not clarify whether the detected virus is a species-specific pathogen of the common swift. If this is the case, natural transmission could potentially occur within the nest during the breeding season. The transmission of circovirus is referred to as oral, intranasal and intra-cloacal routes [19] and the direct contact between the juveniles and the adults in the rearing period allows an easy transmission of the virus in the swift. A vertical transmission of the circovirus via the eggs should also be considered in swifts as a possible transmission route to the juvenile birds [34]. Outside the breeding season, transmission between swifts should be very difficult due to their specific aerial lifestyle. However, the clinical picture with a dystrophic feather loss as described in the presented case is very rare in swifts. Furthermore, it can be assumed that a viral infection causing feather defects would be subject to strong selective pressure and, therefore, is unlikely to occur in swifts. It is therefore quite possible that the swift in our case may have been infected with a circovirus from another bird species and developed a species-specific disease. The analysis of the partial sequence and its close phylogenetic relationship to circoviruses from finches suggest a potential host-jump or a shared reservoir between different bird species. This would be supported by the fact that the bird was kept in the same room with other bird species during the hand rearing. Natural infections between different wild birds could result from the fight for nesting sites and the use of the same nesting site [7]. Direct contact of this kind has been documented between swifts and house sparrows or starlings [7]. In the literature, the incubation period for a circovirus infection is reported to range from a few weeks to several years, with feather defects becoming apparent as early as three weeks after infection [19]. For swifts, however, no data are available. However, according to information in the literature for other bird species [19], it appears that the feather defects can develop after infection in the wildlife rescue center during the period of hand-rearing. Whether natural infections occur in the wild should be investigated in future studies on the prevalence of circoviruses in swifts. The absence of circovirus infection in the other juvenile and adult swifts we examined does not necessarily reflect the true prevalence within the population of the common swift. However, it does indicate that circoviruses are not ubiquitous subclinical within this species. The Hanover area has a population of 800 to 1200 breeding pairs [45].
The potential link between a newly identified feather disorder, termed Paper-Shaft-Syndrome, and a viral etiology remains entirely unexplored [46]. Consequently, further research into the occurrence of circoviruses in common swifts and other swift species, in relation to various diseases, is essential to identify potential correlations.
## References
1. Åkesson, Klaassen, Holmgren et al. (2012) "Migration routes and strategies in a highly aerial migrant, the common swift Apus apus, revealed by light-level Geolocators" *PLoS ONE*
2. Hedenström, Norevik, Warfvinge et al. (2016) "Annual 10-Month Aerial Life Phase in the common swift Apus apus" *Curr. Biol*
3. Holmgren (2004) "Roosting in tree foliage by common swift Apus apus" *Ibis*
4. Muijres, Henningsson, Stuiver et al. (2012) "Aerodynamic flight performance in flap-gliding birds and bats" *J. Theor. Biol*
5. Rattenborg, Voirin, Cruz et al. (2016) "Evidence that birds sleep in mid-flight" *Nat. Commun*
6. Wellbrock, Bauch, Rozman et al. (2017) "Same procedure as last year?"-Repeatedly tracked swifts show individual consistency in migration pattern in successive years" *J. Avian Biol*
7. Glutz Von Blotzheim, Bauer (1997) "Handbuch der Vögel Mitteleuropas"
8. Gedeon, Grüneberg, Mitschke et al. (2014) "Atlas Deutscher Brutvogelarten, Atlas of German Breeding Birds"
9. Tigges (2006) "The breeding cycle in calendar form of the common swift Apus apus across its Eurasian breeding range-A testable hypothesis? Podoces"
10. Haupt, Diagnostik, Mauersegler (1758) "Anatomie und Pathologie des Skeletts und ein Beitrag zur Tierärztlichen Therapie und Prognose"
11. Matthes (2006) "Recovery of a hand-reared common swift (Apus apus)" *APUSlife*
12. Tiyawattanaroj, Jung, Mohr et al. (2021) "Examination of common swifts (Apus apus) for salmonella shedding in the area of Hannover" *Germany. Tierarztl. Prax. Ausg. K Kleintiere Heimtiere*
13. Tiyawattanaroj, Lindenwald, Mohr et al. (2021) "Monitoring of the Infectious Agent Chlamydia psittaci in Common Swifts (Apus Apus) in the Area of Hannover" *Berl. Münchener Tierärztliche Wochenschr*
14. Gavier-Widén, Duff, Meredith (2012) "Infectious Diseases of Wild Mammals and Birds in Europe"
15. Crowther, Berriman, Curran et al. (2003) "Comparison of the structures of three circoviruses: Chicken anemia virus, porcine circovirus type 2, and beak and feather disease virus" *J. Virol*
16. Breitbart, Delwart, Rosario et al. (1997) "Consortium, I.R. ICTV Virus Taxonomy Profile: Circoviridae" *J. Gen. Virol*
17. Delwart, Li (2012) "Rapidly expanding genetic diversity and host range of the Circoviridae viral family and other rep encoding small circular ss DNA genomes" *Virus Res*
18. Varsani, Harrach, Roumagnac et al. (2024) "2024 taxonomy update for the family Circoviridae" *Arch. Virol*
19. Ritchie, Niagro, Latimer et al. (1991) "Routes and prevalence of shedding of psittacine beak and feather disease virus" *Am. J. Vet. Res*
20. Fogell, Martin, Groombridge (2016) "Beak and feather disease virus in wild and captive parrots: An analysis of geographic and taxonomic distribution and methodological trends" *Arch. Virol*
21. Sarker, Das, Helbig et al. (2016) "Identification of beak and feather disease virus in an unusual novel host (Merops ornatus) using nested PCR" *Genome Announc*
22. Kubíček, Taras (2005) "Incidence of Pigeon Circovirus in Eurasian Collared-Dove (Streptopelia decaocto) Detected by Nested PCR" *Acta Vet. Brno*
23. Stenzel, Dziewulska, Muhire et al. (2018) "Recombinant Goose Circoviruses Circulating in Domesticated and Wild Geese in Poland" *Viruses*
24. Wan, Fu, Shi et al. (2011) "Epidemiological investigation and genome analysis of duck circovirus in Southern China" *Virol. Sin*
25. Niu, Liu, Han et al. (2018) "First findings of duck circovirus in migrating wild ducks in China" *Vet. Microbiol*
26. Liu, Li, Sun et al. (2018) "Molecular survey of duck circovirus infection in poultry in southern and southwestern China during" *BMC Vet. Res*
27. Smyth, Todd, Scott et al. (2006) "Identification of circovirus infection in three species of gull" *Vet. Rec*
28. Halami, Nieper, Müller et al. (2008) "Detection of a novel circovirus in mute swans (Cygnus olor) by using nestedbroadspectrum PCR" *Virus Res*
29. Johne, Fernández-De-Luco, Höfle et al. (2006) "Genome of a novel circovirus of starlings, amplified by multiplyprimed rolling-circle amplification" *J. Gen. Virol*
30. Morandini, Dugger, Ballard et al. (1088) "Identification of a novel adelie penguin circovirus at cape crozier" *Viruses*
31. Khalifeh, Custer, Kraberger et al. (2021) "Novel viruses belonging to the family Circoviridae identified in wild American wigeon samples" *Arch. Virol*
32. Fehér, Kaszab, Bali et al. "Novel Circoviruses from Birds Share Common Evolutionary Roots with Fish Origin Circoviruses" *Life*
33. Legnardi, Grassi, Franzo et al. "Detection and Molecular Characterization of a Novel Species of Circovirus in a Tawny Owl (Strix aluco)"
34. Ledwo Ń, Szotowska, Dolka et al. (2025) "Occurrence and vertical transmission of avian polyomavirus and circovirus in captive and wild Passeriformes in Poland" *BMC Vet. Res*
35. Shivaprasad, Hill, Todd et al. (2004) "Circovirus infection in a Gouldian finch (Chloebia gouldiae)" *Avian Pathol*
36. Todd, Scott, Fringuelli et al. (2007) "Molecular characterization of novel circoviruses from finch and gull" *Avian Pathol*
37. Rinder, Schmitz, Peschel et al. (2015) "Complete genome sequence of a novel circovirus from zebra finch" *Genome Announc*
38. Ritchie, Carter (1995) "Avian Viruses Function and Control"
39. Johne, Enderlein, Nieper et al. (2005) "Novel polyomavirus detected in the feces of a chimpanzee by nested broadspectrum PCR" *J. Virol*
40. Johne, Müller, Rector et al. (2009) "Rolling-circle amplification of viral DNA genomes using phi29 polymerase" *Trends Microbiol*
41. Kumar, Stecher, Suleski et al. (2024) "Molecular Evolutionary Genetics Analysis Version 12 for adaptive and green computing" *Mol. Biol. Evol*
42. Kimura (1980) "A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences" *J. Mol. Evol*
43. Rosario, Breitbart, Harrach et al. (2017) "Revisiting the taxonomy of the family Circoviridae: Establishment of the genus Cyclovirus and removal of the genus Gyrovirus" *Arch. Virol*
44. Sheykhi, Sheikhi, Charkhkar et al. (2018) "Detection and Characterization of Circovirus in Canary Flocks" *Avian Dis*
45. Wendt (2006) "Die Vügel der Stadt Hannover; Hannoversche Vogelschutzverein von 1881 e.V"
46. (2025) "German Association for the Protection of Common Swifts: Information for Vets: Plumage Defects"
47. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754636&blobtype=pdf
|
# Cytokine and Endothelial Activation Patterns Related to Severe and Non-Severe Respiratory Viral Infections
Roberto Ferrarese, Sara Boutahar, | Angelo, Paolo Genoni, Gabriele Arcari, Gaia Zambon, Maria Dolci, Federica Perego, | Sara, Serena Delbue, | Mancini, Lucia Signorini, Federica Novazzi
## Abstract
Respiratory viral infections (RVIs) are a major cause of global morbidity and mortality. Severe cases are driven by dysregulated inflammation, impaired interferon (IFN) responses, and thromboinflammation, yet the mechanisms underlying endothelial dysfunction remain poorly defined. We collected 234 leftover material samples from hospitalized patients with PCR-confirmed RVIs. Patients were stratified by viral etiology, differential involvement of the respiratory tract, age and possible co-infections. Cytokines (IL-6, IL-8, IL-1β, TNF-α), IFNs (α/β/γ), and endothelial markers (ICAM-1, VCAM-1) were quantified using microfluidic immunoassays. Routine coagulation parameters were measured in a subset of patients. Compared with controls, RVI patients exhibited significantly elevated systemic cytokines (p < 0.001). IL-6 and IL-8 were higher in patients with lower respiratory tract involvement, particularly in influenza cases. Elderly patients displayed reduced IFN-α/β responses but increased proinflammatory CRP levels. Infants and children had higher ICAM-1 but lower CRP levels. Patients with viralbacterial co-infections showed amplified IFN-γ/IL-1β/ICAM-1 response. Older adults demonstrated prolonged prothrombin times and reduced fibrinogen, indicating coagulopathy. Severe RVIs are characterized by a triad of impaired antiviral IFN responses, hyperinflammation, and endothelial activation, culminating in thromboinflammation. Age, viral type and coinfections critically shape host responses, underscoring the need for biomarker-guided, personalized therapies.
## 1 | Background
Respiratory viral infections (RVIs) are among the most common illnesses worldwide [1][2][3]. Clinically significant viruses include influenza viruses (flu A and B), respiratory syncytial virus (RSV), rhinoviruses (HRV), and human coronaviruses (hCoVs) [4][5][6]. RVIs are generally classified into two categories: severe, which affect the lower respiratory tract (LRT), and non-severe, which remain confined to the upper respiratory tract (URT) [7,8]. Evidence indicates that inflammatory mediators play a dual role in severe respiratory infections conferring protection but also potentially eliciting uncontrolled pathogenic inflammation [9][10][11][12]. Protective inflammation occurs when patternrecognition receptors (PRRs), including Toll-like receptors (TLRs) and RIG-I-like receptors (RIG-I and MDA5), detect viral nucleic acids or structural components. This recognition triggers the timely production of type I and type III interferons (IFN-α/β and IFN-λ), as well as early pro-inflammatory cytokines such as IL-1, TNF, and IL-6. These mediators act cooperatively to limit viral replication, induce antiviral gene expression, and recruit effector immune cells to the site of infection, thereby promoting viral clearance and host protection [12][13][14][15]. On the other hand, pathogenic inflammation arises from dysregulated or excessive immune responses, leading to uncontrolled production of cytokines and chemokines. This hyperinflammatory state results in tissue injury, airway obstruction, and neutrophil-and macrophage-mediated immunopathology. Furthermore, platelet endothelial activation contributes to microthrombi formation and vascular dysfunction, which exacerbate respiratory failure in severe infections [9,11,[16][17][18].
Elevated levels of pro-inflammatory cytokines, such as IL-6, IL-1β and TNF, together with chemokines (CXCL/8/IL-8, CXCL10, CCL2) are strongly correlated with the severity of RVIs [10,[19][20][21][22]. These mediators amplify leukocyte recruitment, vascular permeability, and tissue inflammation, contributing to the development of acute lung injury and respiratory failure.
Aberrant immune activation and cytokine overproduction are central to the pathogenesis of severe RVIs. Such dysregulated immune responses can progress beyond the respiratory tract, leading to systemic inflammatory disease, cytokine release syndrome (CRS), thromboembolic complications, and multiorgan dysfunction. These systemic effects reflect the widespread impact of uncontrolled cytokine and coagulation cascades during severe viral infection [11,22,23]. Innate cytokines and chemokines, particularly type I and type III interferons (IFNs), are critical determinants of disease outcome in RVIs. Delayed or suppressed IFN responses allow uncontrolled viral replication, which in turn triggers hyperinflammatory cascades. Subsequently, elevated levels of IL-6, TNF, IL-1, and chemokines such as CXCL8 (IL-8) and CCL2 (MCP-1) act as amplifiers of systemic inflammation, promoting immune cell recruitment, alveolar epithelial injury, and impaired gas exchange characteristic of severe respiratory disease [21,24,25]. Despite these insights, effective therapies are limited, and the mechanisms of severe RVIs particularly endothelial activation and dysfunction remain incompletely understood. Several viruses, including highly pathogenic hCoVs, can invade endothelial cells, triggering a cytokine storm that disrupts barrier function, drives systemic inflammation, and contributes to thrombosis and multi-organ failure [21,[26][27][28][29].
A comprehensive understanding of the serum and cellular inflammatory/thrombotic profile during RVIs is therefore needed.
The main objective of this project is to investigate on the factors concurring to severe pathogenesis during RVIs, with a focus on mechanisms that trigger systemic inflammation and thrombotic events. We collected leftover material from respiratory and plasma samples belonging to hospitalized patients with respiratory virus (RV)positive results and tested it for a broad panel of RV and microvascular endothelial related inflammatory and thrombotic mediators. This study will comprehensively characterize the inflammatory and thrombotic responses of vascular endothelial tissue during RV infection by quantifying key mediators and markers, including cytokines (IL-6, IL-8, IL-1β, TNF-α), interferons (IFNα, IFN-β, IFN-γ), and endothelial adhesion molecules (ICAM-1, VCAM-1). These insights could improve understanding of severe RV pathogenesis and identify novel therapeutic targets to manage systemic inflammation and thrombosis to be used alongside traditional diagnostic methods.
## 2 | Materials and Methods
## 2.1 | Study Population
Between September 2024 and March 2025, a total of 234 patients samples were analyzed from Respiratory Physiopathology Unit, Intensive Care Unit (ICU), Neonatal Intensive Care Unit (NICU), and Pediatric Ward of the University Hospital of Varese (Italy). Leftover materials were used to perform the study. All patients presented with respiratory symptoms and tested positive to a RV on nasopharyngeal swabs (NPS) and/or on Bronchoalveolar lavage (BAL). Due to the lack of detailed clinical and radiological data on all patients, all cases requesting diagnostic testing on BAL were considered as lower respiratory tract infections (LRTI), which were used as a criterion to distinguish between severe and non-severe infections. The characteristics of subjects are summarized in Table 1. Peripheral venous blood samples were collected at the same time as the respiratory specimen was taken upon admission to the ER for baseline biochemical analyses. All biochemical parameters, including C-Reactive Protein (CRP), Prothrombin Time (PT), Activated Partial Thromboplastin Time (APTT) Test and Fibrinogen were measured in a subpopulation of our cohort as part of routine laboratory analysis using standard laboratory methods.
## 2.2 | Molecular Testing of Respiratory Samples and Blood Laboratory Determinations
Whole blood was collected into commercially available tubes containing ethylene diamine tetraacetic acid (EDTA) on the day of confirmed PCR diagnosis of RV infection. Blood samples were centrifuged at 1,000-2,000 × g for 10 min. The resulting plasma was collected, divided into 0.5 mL aliquots, and stored at -20°C or lower until analysis. Samples that were hemolyzed, icteric, or lipemic were excluded from the study. Stored plasma samples include n = 234 from RV-positive patients and n = 22 from a healthy control group.
## 2.3 | Automated Immunoassay for Chemokine and Cytokine Plasma Assessment
Plasma levels of IL-6, IL-8, IFN-α, TNF-α, IFN-γ, IL-1β, IFN-β, ICAM-1, VCAM-1 were measured in triplicate using ELLA™ microfluidic immunoassays (Protein Simple, Bio-Techne, USA) according with the manufacturer's specifications. Detection limits were: IL-6 (0.626 pg/mL), IL-8 (0.4 pg/mL), IFN-α (3.79 pg/mL), TNF-α (6.23 pg/mL), IFN-γ (2.56 pg/mL), IL-1β (1 pg/mL), IFN-β (2.88 pg/mL), ICAM-1 and VCAM-1 (1.26 pg/mL). Intra-and inter-assay coefficients of variation were all below 11%. Data were analyzed using the Bio-Plex Manager software version 6.0 (Bio-Rad Laboratories, USA).
## 2.4 | Statistical Analysis
Results are reported as median and interquartile range (IQR). Cytokine levels were compared using Mann-Whitney test for two-group comparison while Kruskal-Wallis test, followed by Dunn′s multiple comparison test and Quade's non-parametric ANCOVA were employed for comparisons involving more than two groups. In order to decrease the risk of type I error, we applied False Discovery Rate (FDR) correction and reported adjusted p-values (Benjamini-Hochberg adjusted p-value). Results were considered statistically significant for p-values < 0,05. Statistical analyses were performed with Prism V8.0 (GraphPad Software Inc, La Jolla, CA)and IBM SPSS Statistics Version 31 (IBM, Armonk, NY).
## 3 | Results
## 3.1 | Cytokine Quantification in Plasma Samples
Cytokine quantification was performed on all plasma samples. Levels of IFN-α, IFN-β, IFN-γ, TNF-α, IL-1β, IL-6, IL-8, ICAM-1, and VCAM-1 were evaluated in samples from patients with LRT and URT infections, as well as from control (CTRL) subjects. An initial analysis was conducted without considering the specific type of RV. Statistically significant differences in cytokine levels were observed between both LRT and URT groups compared to the CTRL group for all cytokines analyzed (see Supplementary Summary a ). Notably, IL-6 and IL-8 levels were significantly higher in LRT samples compared to URT samples (IL-6: LRT 63 (24-229)pg/mL vs. URT 21 (6.4-47) pg/mL, p < 0.001; IL-8: LRT 102 (60-177) pg/mL versus URT 45 (25-99) pg/mL, p = 0.002) (Figure 1 A,B). A second round of analysis was performed by clustering samples according to the infecting RV type (FLU, RSV, HRV). Among subjects infected with FLU, a statistically significant difference in IL-6 levels between LRT and URT samples was observed (LRT: 85 (29-234)pg/mL vs. URT: 22 (7.8-51)pg/mL, p = 0.007) (Figure 1C). As in the previous analysis, significant differences in cytokine levels between LRT and URT groups compared to the CTRL group were also observed (see Supplementary Summary b ).
## 3.2 | Cytokines Quantification in Plasma Samples Sorted by Type of RV Infection Type
A second round of analysis was performed clustering samples according to RV type. Interestingly, IFN-α levels were significantly higher in plasma samples of patients infected with FLU compared to those infected withRSV, HRV and ADV/PIV (FLU: 5.1 (1.4-36) pg/mL; RSV: 2.1 (0.84-10)pg/mL; HRV: 1.2 (0.79-2.4)pg/mL;
ADV/PIV: 1.4 (0.84-3.2)pg/mL; p = 0.0003, p = 0.003 and p = 0.0003 respectively). ICAM-1 levels were significantly higher in FLU infected samples compared to RSV infected patients (FLU: 326158 (224364-460878)pg/mL; RSV: 282793 (161726-404148)pg/ mL; p = 0.003). VCAM-1 level was higher in samples positive for FLU compared to RSV, HRV and ADV/PIV infected ones (FLU: 1135318 (707017-1770890)pg/mL; RSV: 705887 (463879-1069542) pg/mL; HRV: 627688 (412200-1013086)pg/mL; ADV/PIV: 742585 (420177-1245975)pg/mL; p = 0.0003, p = 0.0003 and p = 0.04 respectively). Furthermore, IFN-βlevel was higher in samples positive for FLU compared to RSV andHRV (FLU: 0.17 (0.006-0.95)pg/mL; RSV: 0.035 (0.006-0.48)pg/mL; HRV: 0.006 (0.006-0.56) pg/mL; p < 0.001 for both analyses)(Figure 2 A-D). All cytokine levels, except IFN-β, were statistically different between each infecting virus and control group (Supplementary summary c ).
## 3.3 | Cytokines Quantification in Plasma Samples Sorted Byagegroups
Then population was divided into six age ranges (0-1, 2-4, 5-17, 18-59, 60-80, 81-94) to perform a third round of analyses about cytokines levels. IFN-α levels were reduced in the 60-80 age group (1 (0.57-4.2)pg/mL) compared to 5-17 (5.7 (1.7-42)pg/ mL) age range (p = 0.04) (Figure 3A). IFN-β levels were lower in 60-80 age group (0 (0-0.006)pg/mL) compared to 0-1 (0.47 (0.029-0.92)pg/mL), 2-4 (0.75 (0.08-2.8)pg/mL) and a trend was observed with 5-17 (0.33 (0-0.67)pg/mL)age groups (p = 0.01, p = 0.01 and p = 0.07respectively 3C).A trend was observed also for TNF-α levels that were lower in 60-80 age group (14 (8.9-22)pg/mL) compared to 0-1 age group (22 (17-29)pg/mL) (p = 0.07) (Figure 3D). ICAM-1 levels were higher in the 0-1 group (397783 (307951-631969)pg/mL) compared to the 60-80 (233843 (132446-368954)pg/mL) and 81-94
(237499 (155858-331366)pg/mL) groups (p = 0.01 for both analysis) and in the 2-4 group (513274 (361943-584437)pg/mL) compared to 60-80 and 81-94 (p = 0.01 for both analysis). A trend was observed between the 2-4 and the 18-59 (291824 (213503-366415)pg/mL) age groups (p = 0.09) (Figure 3E). No statistically significant differences were observed for IL-1β, IL-6, IL-8 and VCAM-1.
## 3.4 | Analysis of Biochemical Parameters in Samples Sorted by RV Infection Type
Furthermore, the levels of CRP (mg/L), PT (s'), APTT (s′), P-INR (s') and fibrinogen (mg/L) were assessed stratifying the patient population by both infecting viruses and age groups.
When analyzing the cohort based on the infecting virus, the only statistically significant differences were in CRP levels which were elevated across all viruses groups compared to the control group (CTRL: 0.5 (0.5-0.5) mg/L; FLU: 74.4 (17.9-120.3) mg/L; RSV: 22.4 (7.2-95.7) mg/L; HRV: 33.9 (7.9-113)mg/L; Adeno/Para: 52.9 (34.7-91.7)mg/L; all p values are < 0.001) (data not shown).
## 3.5 | Analysis of Biochemical Coagulation Parameters in Samples Sorted by Age Groups
When stratifying by age, lower CRP levels in younger age groups compared to older ones were observed. Specifically, CRP 4A). For PT, APTT, P-INR, and fibrinogen, all samples from patients under 18 years of age were grouped in a single category to ensure comparable sample size across groups and enhance statistical power. Using this approach, significantly higher values of PT were found in the 60-80 (1.17
## 3.6 | Analysis of the Co-Infected Population
The presence of bacterial, viral or combined co-infections and their influence on the cytokine levels and other related parameters was then investigated in the overall patient population. In patients with bacterial coinfections, compared to those without coinfection, we observed significantly higher levels of ICAM-1 (416880 (309023-578837)pg/mL vs 290249 (194780-379760); p = 0.014) and a trend in IFN-β (0.18 (0-0.84) pg/mL vs 0 (0-0.52)pg/mL; p = 0.06), IFN-γ (6.3 (1.5-29)pg/mL vs 3.3 (0.71-13)pg/mL; p = 0.06) and IL-1β (0.44 (0.05-1.2)pg/ mL vs 0.19 (0-0.76) pg/mL; p = 0.06) (Figure 5A-D) (Supplementary summary d ).
Next, differences between plasma samples with and without viral coinfection were analyzed. In those with viral coinfection we observed a trend of higher levels of IFN-γ ( 5 Finally, differences in the parameters levels of samples with both bacterial and viral coinfection and the ones with no coinfection were evaluated. In the double coinfected group a trend of higher levels of IFN-γ (7.94 (2.24-32.6)pg/mL vs 3.16 (0.55-13.95)pg/mL; p = 0.06) and significantly higher levels of ICAM-1 (397783 (265152-633438)pg/mL vs 276071 (187284-367072)pg/mL; p = 0.02) were observed (Figure 6A,B). No statistically significant differences were observed for the remaining parameters.
## 4 | Discussion
Aberrant immune responses and dysregulated cytokine production are central to the pathogenesis of RVIs [30,31]. RVIs are strongly associated with multi-systemic inflammatory disease, overproduction of proinflammatory mediators (cytokine release syndrome), thromboembolic phenomena, and microcirculatory dysfunction, often resulting in severe to critical illness [32][33][34][35]. Severe Flu, for example, is characterized by rapid viral replication and a robust induction of IL-6, TNF, and IL-1β, a phenomenon commonly referred to as a "cytokine storm." This response drives vascular leakage and acute respiratory distress syndrome (ARDS) [24,36,37]. Early type I interferon (IFN) responses to confer protection, whereas late excessive inflammation mediates lung injury. Elevated levels of IL-6, IL-1β, and CRP correlate with disease severity, including respiratory failure and mortality [38,39].
We analyzed expression levels of type-I IFNs (IFN-β1, IFN-α), type-II IFN (IFN-γ), proinflammatory cytokines (IL-6, IL-8, IL-1β), and endothelial markers (ICAM-1, VCAM-1) in leftover blood samples from 234 RV infected subjects with URT and/or LRT involvement. Samples were analyzed both as a single cohort and across six age-dependent groups, stratified by infecting virus.
Regarding compartment-specific inflammation, we observed that blood levels of IL-6 and IL-8 were significantly higher in LRT infections than URT infections. This suggests that systemic inflammation is influenced by the site of immune activation. The elevated levels of IL-6 and IL-8 driven by LRT likely reflect more extensive tissue injury, immune cell infiltration or increased vascular permeability [9,[40][41][42][43][44]. In particular, the influenza-specific response showed that, IFN-α, IFN-βVCAM-1and ICAM-1 were significantly higher in patients infected with influenza than in those infected with other respiratory viruses (RV), reflecting enhanced chemokine-mediated neutrophil recruitment, an antiviral IFN response and endothelial activation [45,46]. We observed reduced antiviral responses in older adults: IFN-α, IFN-β and TNF-α levels were lower in elderly patients than in younger individuals, which is consistent with immunosenescence and delayed interferon responses [47]. Older adults exhibited elevated CRP, reflecting heightened systemic inflammation and neutrophil driven tissue damage [48]. ICAM-1 levels were highest in infants and young children, facilitating leukocyte trafficking and representing a potential vulnerability to RVs, particularly HRV [49]. Infants and children had lower CRP levels than adults and elderly patients, consistent with immature acute-phase responses [50]. Patients aged 60-80 exhibited prolonged prothrombin times (PT), elevated international normalized ratios (P-INR), and lower fibrinogen levels compared to younger cohorts. These changes highlight an increased susceptibility to coagulopathy in older adults, which may contribute to worse clinical outcomes in viral infections, including SARS-CoV-2 [51][52][53].
Patients with viral-bacterial coinfections displayed higher levels of IFN-β, IFN-γ, IL-1β and ICAM-1, than those with viral infection alone. These results imply the synergistic activation of antiviral and antibacterial pathways, resulting in hyperinflammation and endothelial activation [54,55]. Elevated levels of IFN-γ, IL-1β and ICAM-1 could be used as biomarkers to identify patients at risk of severe disease early on. These findings emphasize the importance of immunoprofiling in RVIs as it sheds light on the underlying pathophysiology, helps with risk stratification and could inform personalized therapeutic strategies. The study's limitations include the fact that neither the days from symptom onset nor the respiratory virus Ct values were available.
Cytokine targeted therapies (e.g. IL-6 blockade with tocilizumab) can be beneficial for patients experiencing hyperinflammation. However, the timing of administration must be carefully considered to avoid suppressing the body′s natural antiviral defenses. There is evidence that the virus-specific response to RSV often induces type 2 inflammation and mucus production. In contrast, Flu and SARS-CoV-2 generate neutrophil-rich cytokine storms and endothelial injury [56,57]. This emphasizes the need for tailored therapeutic strategies.
The retrospective design of the study means that there is a lack of clinical patient data. Additionally, the absence of sequential samples means that cytokine kinetics cannot be evaluated. Finally, respiratory samples need to be measured to detect any differences in cytokine expression at the mucosal level.
In conclusion, RV infections trigger distinct, compartmentalized and age-dependent immune responses that influence systemic inflammation, endothelial activation and coagulation. Influenza elicits a robust chemokine and cytokine response, whereas older adults exhibit reduced antiviral IFNs and elevated inflammatory mediators, which explains their increased susceptibility to severe outcomes. Viral-bacterial coinfections amplify immune activation and endothelial involvement, highlighting the need for early, biomarkerguided intervention. The novel therapeutic strategies, to be used in combination with traditional ones, must consider factors such as timing, virus type, infection site, age, and inflammatory profile to maximise efficacy and minimise adverse effects.
## References
1. Cilloniz, Luna, Hurtado et al. (2022) "Respiratory Viruses: Their Importance and Lessons Learned From COVID-19" *European Respiratory Review*
2. Lei, Yang, Lou (2021) "Viral Etiology and Epidemiology of Pediatric Patients Hospitalized for Acute Respiratory Tract Infections in Macao: A Retrospective Study From 2014 to 2017" *BMC Infectious Diseases*
3. Uyeki, Hui, Zambon et al. *Monto*
4. Kim, Song, Ahn (2022) "Shift in Clinical Epidemiology of Human Parainfluenza Virus Type 3 and Respiratory Syncytial Virus B Infections in Korean Children Before and During the COVID-19 Pandemic: A Multicenter Retrospective Study" *Journal of Korean Medical Science*
5. Chen, Zhu, Wang (2015) "A Multi-Center Study on Molecular Epidemiology of Human Respiratory Syncytial Virus From Children With Acute Lower Respiratory Tract Infections in the Mainland of China Between" *Virologica Sinica*
6. Zhang, Reeves, Ma (2025) "Estimating the Respiratory Syncytial Virus-Associated Hospitalisation Burden in Older Adults in European Countries: A Systematic Analysis" *BMC Medicine*
7. Mokrani (2025) "Timsit Role of Respiratory Viruses in Severe Acute Respiratory Failure" *Journal of Clinical Medicine*
8. Cillóniz, Pericàs, Rojas et al. (2022) "Severe Infections Due to Respiratory Viruses" *Seminars in Respiratory and Critical Care Medicine*
9. Newton, Cardani (2016) "Braciale The Host Immune Response in Respiratory Virus Infection: Balancing Virus Clearance and Immunopathology"
10. Short, Kroeze, Fouchier et al. (2014) "Pathogenesis of Influenza-Induced Acute Respiratory Distress Syndrome" *Lancet Infectious Diseases*
11. Merad (2020) "Martin Pathological Inflammation in Patients With Covid-19: A Key Role for Monocytes and Macrophages" *Nature Reviews Immunology*
12. Iwasaki (2014) "Pillai Innate Immunity to Influenza Virus Infection" *Nature Reviews Immunology*
13. Kawai, Akira (2010) "The Role of Pattern-Recognition Receptors in Innate Immunity: Update on Toll-Like Receptors" *Nature Immunology*
14. Lazear, Schoggins (2019) "Diamond Shared and Distinct Functions of Type I and Type III Interferons"
15. Schlee, Hartmann (2016) "Discriminating Self From Non-self in Nucleic Acid Sensing" *Nature Reviews Immunology*
16. Nicholls, Poon, Lee (2003) "Lung Pathology of Fatal Severe Acute Respiratory Syndrome" *Lancet*
17. Zuo, Yalavarthi, Shi (2020) "Neutrophil Extracellular Traps in COVID-19" *JCI Insight*
18. Middleton, He, Denorme (2020) "Neutrophil Extracellular Traps Contribute to Immunothrombosis in COVID-19 Acute Respiratory Distress Syndrome" *Blood*
19. Kombe Kombe, Fotoohabadi, Gerasimova "The Role of Inflammation in the Pathogenesis of Viral Respiratory Infections"
20. Huang, Wang, Li (2020) "Clinical Features of Patients Infected With 2019 Novel Coronavirus in Wuhan"
21. Channappanavar, Perlman (2017) "Pathogenic Human Coronavirus Infections: Causes and Consequences of Cytokine Storm and Immunopathology" *Seminars in Immunopathology*
22. Tisoncik, Korth, Simmons et al. (2012) "Into the Eye of the Cytokine Storm" *Microbiology and Molecular Biology Reviews*
23. Levi, Thachil, Iba (2020) "Levy Coagulation Abnormalities and Thrombosis in Patients With COVID-19" *Lancet Haematology*
24. Li, Huang, Rai et al. (2025) "Innate Immune Role of IL-6 in Influenza A Virus Pathogenesis" *Frontiers in Cellular and Infection Microbiology. Frontiers Media SA*
25. Blanco-Melo, Nilsson-Payant, Liu (2020) "Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19" *Cell*
26. Van Gorp, Suharti, Cate (1999) "Review: Infectious Diseases and Coagulation Disorders" *Journal of Infectious Diseases*
27. Friedman (1989) "Infection of Endothelial Cells by Common Human Viruses" *Clinical Infectious Diseases*
28. Fosse, Haraldsen, Falk et al. (2021) "Endothelial Cells in Emerging Viral Infections"
29. Teijaro, Walsh, Cahalan (2011) "Endothelial Cells Are Central Orchestrators of Cytokine Amplification During Influenza Virus Infection" *Cell*
30. Gogoi, Baruah, Narain (2024) "Immunopathology of Emerging and Re-emerging Viral Infections: An Updated Overview" *Acta Virologica. Frontiers Media SA*
31. Andreychyn, Melnyk, Zavidniiuk et al. (2024) "Interleukins in the Pathogenesis of Influenza and Other Acute Respiratory Viral Infections" *Epidemiology and Infection*
32. Que, Hu, Wan (2022) "Cytokine Release Syndrome in COVID-19: a Major Mechanism of" *International Reviews of Immunology. Taylor and Francis Ltd*
33. Peart Akindele, Kouo, Karaba (2021) "Distinct Cytokine and Chemokine Dysregulation in Hospitalized Children With Acute Coronavirus Disease 2019 and Multisystem Inflammatory Syndrome With Similar Levels of Nasopharyngeal Severe Acute Respiratory Syndrome Coronavirus 2 Shedding" *The Journal of Infectious Diseases*
34. Mir, Almas, Kaur (2021) "Coronavirus Disease 2019 (COVID-19): Multisystem Review of Pathophysiology"
35. Kumar, Rivkin, Raffini (2023) "Thrombotic Complications in Children With Coronavirus Disease 2019 and Multisystem Inflammatory Syndrome of Childhood" *Journal of Thrombosis and Haemostasis*
36. Gu, Zuo, Zhang (2021) "The Mechanism Behind Influenza Virus Cytokine Storm" *Viruses*
37. Nie, Zhou, Tian (2025) "Deep Insight Into Cytokine Storm: From Pathogenesis to Treatment" *Signal Transduction and Targeted Therapy*
38. Schultze (2021) "Aschenbrenner COVID-19 and the Human Innate Immune System" *Cell*
39. Hadjadj, Yatim, Barnabei (2020) "Impaired Type I Interferon Activity and Inflammatory Responses in Severe COVID-19 Patients" *Science*
40. Lucas, Wong, Klein (2020) "Longitudinal Analyses Reveal Immunological Misfiring in Severe COVID-19" *Nature*
41. Koyama, Ishii, Coban et al. (2008) "Innate Immune Response to Viral Infection" *Cytokine*
42. Wang, Kurt-Jones, Finberg (2007) "Innate Immunity to Respiratory Viruses" *Cellular Microbiology*
43. Miles, Jayawardena, Liong (2025) "TLR7 Deficiency Enhances Inflammation in the URT but Reduces LRT Immunity Following Influenza A Infection"
44. Mcgonagle, Sharif, O'regan et al. (2020) "The Role of Cytokines Including Interleukin-6 in COVID-19 Induced Pneumonia and Macrophage Activation Syndrome-Like Disease" *Autoimmunity Reviews*
45. Ramos, Fernandez (2015) "Sesma Modulating the Innate Immune Response to Influenza A Virus: Potential Therapeutic Use of Antiinflammatory Drugs"
46. Xie, Wei, Wang (2025) "The Intersection of Influenza Infection and Autoimmunity"
47. Shaw, Goldstein, Montgomery (2013) "Age-Dependent Dysregulation of Innate Immunity" *Nature Reviews Immunology*
48. Metcalf, Cubas, Ghneim (2015) "Global Analyses Revealed Age-Related Alterations in Innate Immune Responses After Stimulation of Pathogen Recognition Receptors" *Aging Cell*
49. Papi, Johnston (1999) "Rhinovirus Infection Induces Expression of Its Own Receptor Intercellular Adhesion Molecule 1 (ICAM-1) via Increased NF-κB-mediated Transcription" *Journal of Biological Chemistry*
50. Hofer, Zacharias, Müller et al. (2012) "An Update on the Use of C-Reactive Protein in Early-Onset Neonatal Sepsis: Current Insights and New Tasks" *Neonatology*
51. Yuan, Tong, Wang et al. (2020) "Coagulopathy in Elderly Patients With Coronavirus Disease 2019" *Aging Medicine*
52. Von Meijenfeldt, Havervall, Adelmeijer (2021) "Prothrombotic Changes in Patients With COVID-19 Are Associated With Disease Severity and Mortality" *Research and Practice in Thrombosis and Haemostasis*
53. Matsuoka, Koami, Shinada et al. (2022) "Investigation of Differences in Coagulation Characteristics Between Hospitalized Patients With SARS-CoV-2 Alpha, Delta, and Omicron Variant Infection Using Rotational Thromboelastometry (ROTEM): A Single-Center, Retrospective, Observational Study" *Journal of Clinical Laboratory Analysis*
54. Mcnab, Mayer-Barber, Sher et al. (2015) "Garra Type I Interferons in Infectious Disease" *Nature Reviews Immunology*
55. Ivashkiv, Donlin (2014) "Regulation of Type I Interferon Responses" *Nature Reviews Immunology*
56. Potere, Del Buono, Caricchio (2022) "Interleukin-1 and the NLRP3 Inflammasome in COVID-19: Pathogenetic and Therapeutic Implications" *EBioMedicine*
57. Ma, Sahu, Cano (2021) "Increased Complement Activation is a Distinctive Feature of Severe SARS-CoV-2 Infection" *Science Immunology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772256&blobtype=pdf
|
# Development of an RT-qPCR and a Next-Generation Sequencing approach to assess viral shedding of NDV-based SARS-CoV-2 variant vaccines
Marta Boza, Adam Abdeljawad, Stefan Slamanig, Nicholas Lemus, Ying Tsoi, Lai, William Shea, Weina Sun, Peter Palese, Irene González-Domínguez
## Abstract
Live-attenuated intranasal vaccines offer great promise to generate mucosal immunity and reduce breakthrough infections after vaccination. Previously, we have developed a Newcastle disease virus (NDV)-based vaccine expressing the spike (S) protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), named NDV-HXP-S. This vaccine can be administered intranasally as a live-attenuated vector to provide both systemic and mucosal immunity. With the appearance of new SARS-CoV-2 variants of concern, new NDV-HXP-S variant-based vaccines have been developed. These new NDV-HXP-S variant vaccines can be combined in a multivalent formulation to extend their protection against other SARS-CoV-2 variants of concern. One of the main concerns when administering a live-attenuated vector intranasally is the possible vaccine shedding after vaccination. In the present work, we have developed a reverse transcrip tase-quantitative polymerase chain reaction (RT-qPCR) method to evaluate NDV-HXP-S in vitro and in vivo along with an amplicon-based next-generation sequencing technique to quantify the relative percentage of each variant vaccine in the multivalent formulation. Both techniques were validated following the recommendations of the Food and Drug Administration and the European Medicines Agency. Finally, we applied both methods in in vitro cell culture experiments. The quantification of the NDV-based vaccines by RT-qPCR and the relative percentage of each variant vaccine were confirmed using a prototype trivalent formulation expressing the Ancestral (Wuhan), Beta, and Delta SARS-CoV-2 spikes. The present results will help the study of virus shedding in future clinical trials. IMPORTANCE Live-attenuated intranasal vaccines offer great promise to generate mucosal immunity and reduce breakthrough infections after vaccination. In this work, we have developed a combined reverse transcriptase-quantitative polymerase chain reaction and next-generation sequencing-directed approach to assess the viral shedding of multivalent Newcastle disease virus-based severe acute respiratory syndrome coronavirus 2 vaccines. This new formulation will be evaluated as a mucosal booster in phase I clinical trials next year. KEYWORDS viral vector vaccine, live-attenuated vaccine, mucosal immunity S evere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of the coronavirus disease 2019 (COVID-19) (1, 2). In previous work, our groups have developed a live-attenuated viral vector vaccine expressing the spike (S) protein of SARS-CoV-2 based on the Newcastle disease virus (NDV) named NDV-HXP-S (3). This vaccine contains the mammalian codon-optimized sequence of the S protein inserted between the phosphoprotein (P) and the matrix (M) genes of the NDV (Fig. 1). Compared
to other vaccine platforms, NDV-HXP-S presents several advantages: (i) it is based on the lentogenic LaSota strain, an avirulent and non-human pathogen that has a long safety record as veterinary vaccine and human oncolytic vector (4); (ii) NDV can be produced at high yields at a very low cost in embryonated chicken eggs; (iii) it can be administered as live viral vector vaccine via the intranasal route, conferring both humoral and mucosal immunity; and (iv) its high immunogenicity has been demonstrated in several preclinical and clinical studies (3,(5)(6)(7). With the appearance of SARS-CoV-2 variants of concern (VOC) which threaten the protection conferred by the prototype vaccines, new NDV-HXP-S variant-based vaccines have been developed. To date, NDV-HXP-S vaccines expressing the spike protein of Ancestral (Wuhan), Beta, Gamma, Delta, Omicron BA.1, Omicron BA.5, BQ.1.1, and XBB.1.5 have been produced (8)(9)(10). Furthermore, our groups have optimized a multivalent vaccine, combining these vectors in a single formulation, which demonstrated extended protection against mismatched SARS-CoV-2 VOCs, not present in the formulation (8).
Given the great potential of this new formulation as a mucosal multivalent vaccine, we have developed two methods to (i) quantitatively evaluate NDV-HXP-S in vitro and in vivo and (ii) to quantify the different relative percentages of each NDV-HXP-S vaccine in a multivalent formulation. Regulatory guidelines developed by the Food and Drug Administration (FDA) and the European Medicines Agency recommend nucleic acid-based assays to analyze vector biodistribution, vector shedding, and vector-derived gene expression due to their superior sensitivity and specificity at a wide dynamic range (11). Following these recommendations, we developed a one-step reverse transcriptasequantitative polymerase chain reaction (RT-qPCR) method to quantify a small region of the nucleoprotein (NP) gene of the NDV (Fig. 1A). International Council for Harmoni zation (ICH) guideline M10 (12) and the FDA recommendations (13) were followed to validate the RT-qPCR method. The identity, linearity, sensitivity, specificity, precision, and reproducibility were optimized to meet the acceptance criteria. To differentiate each NDV-HXP-S variant vaccine, we developed an amplicon-based next-generation sequencing (NGS) method (Fig. 1B) (14). We selected the receptor-binding domain (RBD) region of the spike protein, which accumulates the highest number of muta tions among VOCs, and performed NGS to quantify the frequency of each variant in a multivalent formulation. The identity, sensitivity, precision, and reproducibility were studied following the acceptance criteria. Finally, both techniques were applied in in vitro cell culture models.
## MATERIALS AND METHODS
## NDV-HXP-S production in embryonated chicken eggs
Ten-day-old specific pathogen-free (SPF) embryonated chicken eggs (AVSBio, Norwich, CT, USA) were infected with vaccine seed viruses to produce NDV-HXP-S variant vaccines. The incubation time and plaque-forming units per egg were optimized for each NDV-HXP-S variant. One hundred microliters of each master virus seed containing 10-300 plaque-forming units per egg was injected into the allantoic cavity of each egg. Eggs were incubated at 37°C, according to their incubation time and cooled at 4°C overnight. The specific growth conditions are reported in Table S1. Allantoic fluid (AF) was harvested from cooled eggs and subsequently clarified by centrifugation at 2,000 × g at 4°C for 30 minutes. NDV-HXP-S production was confirmed by hemagglutination (HA) assays. The clarified AF was aliquoted and stored at -80°C. Live viruses harvested in AF were pelleted through a 20% sucrose cushion in PBS (pH = 7.4) by ultracentrifugation in a Beckman L7-65 ultracentrifuge at 25,000 rpm, which corresponds to 107,000 × g or 113,000 × g at 4°C for 2 hours using a Beckman SW32 or SW28 rotor, respectively (Beckman Coulter, Brea, CA, USA). Supernatants were discarded, and the pellets were resuspended in PBS (pH = 7.4) and stored at -80°C until further use. Vaccine titers were measured by 50% of egg infectious dose (EID 50 ), plaque assay, and RT-qPCR.
## Hemagglutination assay
Turkey red blood cells (Lampire Biological Laboratories, Pipersville, PA, USA) were diluted with PBS to a final concentration of 0.5% (vol/vol) in agreement with previous work (3). In a 96-well V-bottom plate, 100 µL of AF was added to the first well of each row, and 50 µL of PBS was added to the rest of the wells, then serial twofold dilutions were performed by transferring 50 µL in each dilution step. After discarding the excess volume of the final dilution, 50 µL of the red blood cell suspension (0.5%) was added to all wells, and plates were read after 45 minutes of incubation at 4°C (15,16). In-house produced NDV-HXP-S vaccines were used as positive controls.
## NP gene one-step reverse transcriptase-quantitative polymerase chain reaction
NDV primers targeting the NP gene were designed for the RT-qPCR. Primers were designed using the primer-BLAST online tool from NCBI. Forward primer: 5′-AGAGAG CACAGAGATTTGCG-3′ and reverse primer: 5′-GATCCTCTCCAGGGTATCGGT -3′ were used to amplify an amplicon of 128 base pairs (bp). RT-qPCR was performed using the iTaq Universal SYBR Green One-Step Kit (Bio-Rad Laboratories, Inc, Hercules, CA, USA) using the CFX Opus 96 Real-Time PCR Instrument (Bio-Rad Laboratories, Inc, Hercules, CA, USA). The master mix for a single PCR reaction consisted of 10 µL of the iTaq Universal SYBR Green reaction mix, 3.75 µL of nuclease-free water, 0.5 µL of each primer at a concentration of 10 µM, 0.25 µL of the iScript Reverse Transcriptase, and 5 µL of the template RNA. The RT-qPCR was run as follows: 50°C for 10 minutes, 95°C for 1 minute, and 40 consecutive cycles of 95°C for 10 seconds and 55°C for 20 seconds. No template control (NTC) with 5 µL of nuclease-free water as RNA sample was used as negative control. For each RT-qPCR run, a positive control consisting of a defined concentration of the template, derived from the plasmid used to generate the standard curve, was used. From the standard curve, the copies in 5 µL were interpolated, then the copies/μL were calculated and corrected by the amount of RNA sample extracted using the QIAmp Viral RNA Mini Kit. A plasmid containing the sequence of the NP was used as DNA standard (Twist Bioscience, South San Francisco, CA, USA). An 8-point, 10-fold dilution series of standard DNA was used to prepare the standard curve (11). The copies/μL were calculated as follows:
## NP gene reverse transcriptase-polymerase chain reaction
Viral RNA was extracted using the QIAmp Viral RNA Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. A reverse transcriptase-polymerase chain reaction (RT-PCR) was performed using the SuperScript IV One-Step RT-PCR System (Thermo Fisher Scientific, Waltham, MA, USA) to amplify the transgene of the NP gene. Forward primer: 5′ -AGAGAGCACAGAGATTTGCG -3′ and reverse primer: 5′ -GAGAG GCTACTAAGTGCAAGG -3′ were used to amplify an amplicon of 495 bp. Primers were used at a concentration of 10 µM. The RT-PCR was run as follows: 60°C for 10 minutes; 98°C for 2 minutes; 40 consecutive cycles of 98°C for 10 seconds, 63°C for 10 seconds, and 72°C for 20 seconds; and 72°C for 5 minutes. An agarose gel at 1% (wt/vol) with SYBR Safe (Invitrogen, Waltham, MA, USA) was performed to visualize the amplicon. The RT-PCR product was mixed with gel loading dye Orange DNA Loading Dye (1×, final concentration) and loaded to the gel. Ten microliters of Trackit 100 bp DNA Ladder (Thermo Fisher Scientific, Waltham, MA, USA) was added to one well to confirm the size of the resultant amplicon. Gel electrophoresis results were visualized using the Bio-Rad ChemiDoc Imaging System (Bio-Rad Laboratories, Inc, Hercules, CA, USA).
## NP gene RT-qPCR assay validation
RT-qPCR was validated following the ICH guideline M10 on bioanalytical method validation and study sample analysis (12) and the recommendations by the FDA guidance for gene therapy clinical trials and long-term follow-up after administration of human gene therapy products (13). The RT-qPCR assay was validated following established acceptance criteria described in Table 1, including specificity (no ampli fication in no template controls), linearity (R² >0.98), efficiency between 90% and 110%, sensitivity with a lower limit of detection ≤10 copies, and quantification ≤50 copies. Reproducibility was confirmed with intra-assay variation and inter-assay variation. Validation included five RT-qPCR runs performed over a minimum of 3 days by two different operators. Each plate contained one set of standards and NTC as negative control, all of them run in triplicates. The standard curve was used to evaluate the identity, linearity, sensitivity, specificity, and reproducibility of the qPCR assay.
## Specificity
Specificity was determined by the lack of amplification in NTC in all the validation runs as well as evaluated in silico using SnapGene software (GSL Biotech LLC, Boston, MA, USA) by aligning the primer sequences against different NDV variants and virus of the Paramyxoviridae family. NP gene sequences were retrieved from the NCBI nucleo tide database. NP sequences of NDV vaccine strains are described in Table 2, and NP sequence for the different paramyxoviruses is shown in Table S2. NP gene identity scores (%) between each pair of viruses were quantified using the two-sequence alignment tool of SnapGene software (GSL Biotech LLC, Boston, MA, USA).
## Linearity
Linearity was determined using a suitable linear regression analysis of the Cq value vs NP gene log copies. Slope, efficiency, and the coefficient of determination (R 2 ) were calculated using Maestro Software (Bio-Rad).
## Sensitivity
The sensitivity was determined by the lowest limit of quantification (LLOQ) and lowest limit of detection (LLOD). The LLOQ and LLOD were established according to the FDA recommendations as 50 copies/target and 10 copies/target, respectively (13). The upper limit of detection was established by the highest point in the standard curve. To calculate
## Acceptance criteria for RT-qPCR validation
## Specificity, linearity, and sensitivity
No template control wells should be blank Efficiency should be between 90% and 110% R 2 >0.98
Lower limit of detection ≤10 copies Lower limit of quantification ≤50 copies Reproducibility (intra-assay precision) Coefficient of variation of the Cq value between each standard replicate should be <2%
Reproducibility (inter-assay precision) Coefficient of variation of the log (copies) of each standard from different validation runs should be <25% the empirical LLOD, serial dilutions of NDV-HXP-S were performed in technical triplicates and quantified by RT-PCR, RT-qPCR, and HA assay.
## Reproducibility
The intra-assay precision was determined by analyzing the coefficient of variation (CV) of the Cq values of the intra-assay replicates, with an acceptance threshold of <2%, according to the acceptance criteria. The inter-assay precision was determined by analyzing the CV of the log copies between the inter-assay replicates in five independent runs, with an acceptance threshold of <25%, also in-line with these guidelines.
## Fifty percent of egg infectious dose (EID 50 )
EID 50 quantification was performed in 8-10 days old embryonated chicken eggs. NDV-HXP-S samples were 10-fold serially diluted in PBS, resulting in 10 -7 -10 -12 dilutions of the virus. One hundred microliters of each dilution was injected into each egg for a total of six eggs per dilution. The eggs were incubated at 37°C according to each virus growth condition and then cooled at 4°C overnight. AF was collected and analyzed by HA assay. The EID 50 titer of the NDV, determined by the number of HA-positive and HA-negative eggs in each dilution, was calculated using the Reed and Muench method (3).
## Next-Generation Sequencing
Next-Generation Sequencing (NGS) was performed using the Amplicon-EZ service from a commercial laboratory (Genewiz from Azenta Life Sciences, South Plainfield, NJ, USA). Primers targeting a specific region of the RBD spike protein with variability between all the variants were designed including NGS-adapter sequences (forward primer: 5′-ACACTCTTTCCCTACACGACGCTCTTCCGATCTCTACCAAGCTGAACGACCTGTGCTTCAC-3′, reverse primer: 5′-GACTGGAGTTCAGACGTGTGCTCTTCCGATCTAGTTCGAAGCTCAGCACCA CCACTCTGTA -3′. A variable region within the receptor-binding domain of the spike protein, which allows differentiation among all variants of concern, was first identified. Conserved flanking sequences surrounding this region were then selected for primer design. Primer specificity was confirmed in silico by aligning the primer sequences using SnapGene software (GSL Biotech LLC, Boston, MA, USA). An RT-PCR was performed using the SuperScript IV One-Step RT-PCR System (Thermo Fisher Scientific, Waltham, MA, USA). Forward and reverse primers were used for the RT-PCR at a concentration of 10 µM.
RT-PCR was performed as follows: 50°C for 10 minutes; 98°C for 2 minutes; 40 consecutive cycles of 98°C for 10 seconds, 72°C for 10 seconds, and 72°C for 20 seconds; and 72°C for 5 minutes. RT-PCR products were purified using the PCR Clean-up protocol of the NucleoSpin Gel and PCR Clean-up (Macherey-Nagel, Düren, Germany). Samples were then normalized to 20 ng/µL and sent to a commercial laboratory for NGS (14). Full sequence coverage was obtained by paired-end data with two sequencing fastq files per sample. Files were extracted from the compressed folder and analyzed using the SeqMan Ultra software (DNA STAR, Madison, WI, USA), using the variant analysis workflow for NGS amplicon data. Settings were modified to obtain the highest specificity. Read technology was Illumina paired-end data, with multisample experiment setup. Mer size was changed to 31 nucleotides (nt), and the minimum match percentage was changed to 100%. Minimum aligned length was also modified to 200 nt. In the analysis options, detection of single nucleotide polymophisms (SNPs) and other small variants was unchecked, with a 100% match threshold; no mismatches or variants are expected to be present in the aligned reads. The number of sequences of each variant assembly over the total number of sequences was measured to estimate the frequency of each variant in the sample.
## NGS assay validation
NDV-HXP-S variant vaccines expressing the spike of Ancestral, Beta, Gamma, Delta, BA.1, BA.5, BQ.1.1, and XBB.1.5 were used to validate the technique. RNA was extracted using the QIAmp Viral RNA Mini Kit (QIAGEN, Hilden, Germany) following the manufacturer's instructions. RT-PCR was performed as explained above, in the NGS section of Materials and Methods. The specific criteria used in this study are described in Table 3. Each variant vaccine sample was run by technical duplicate.
## Specificity and sensitivity
The specificity of the technique was evaluated by performing in silico and in vitro analyses. Primers were aligned against the S protein of SARS-CoV-2 (NCBI Gene ID: 43740568) and the codon-optimized S protein in the different NDV-HXP-S variant vaccines in silico using SnapGene software (GSL Biotech LLC, Boston, MA, USA). In vitro amplification of individual amplicons for Ancestral, Beta, Gamma, Delta, BA.1, BA.5, a Mean, SD, and coefficient of variation of the log copies at each dilution among the five independent runs. CV of the log copies is less than 25%, meeting the acceptance criteria.
BQ.1.1, and XBB.1.5 vaccines was performed, and the relative percentage of each variant was analyzed by NGS. For each variant, at least one distinguishing nucleotide difference and a relative frequency above 90% in the sequencing data were expected. The relative error was measured using equation 2, calculated by comparing the measured frequency to the expected frequency for each variant in the sample. NGS analysis was performed on samples with known amounts of NDV-HXP-S variant amplicons. Specifically, individu ally sequenced variant vaccines and mixed samples containing defined proportions of Ancestral and XBB.1.5 amplicons were used for the analysis.
(2) Relative error = Expected value -Real value Real value x 100
## Linearity
Linearity was studied by analyzing samples containing different proportions of two amplicons (Ancestral and XBB.1.5). Linear correlation between the predicted and the observed values was analyzed and compared to the acceptance criteria.
## Reproducibility
The intra-assay precision was determined by analyzing the CV of the relative frequency of each variant vaccine measured in duplicate. The inter-assay precision was determined by analyzing the CV of Ancestral amplicon measured at three different times.
## In vitro infections
Vero-E6 (CRL-1586) cells were used to evaluate in vitro replication of NDV-HXP-S monovalent and multivalent formulations. Briefly, Vero-E6 cells were seeded onto 12-well plates in growth media (1% HEPES 1M, 1% penicillin-streptomycin, and 10% heat-inac tivated fetal bovine serum in Dulbecco's modified Eagle medium [DMEM]) and were cultured for 1 day. Once the cells achieved 80%-100% confluency, they were infec ted with monovalent Ancestral, Beta, Delta NDV-HXP-S vaccines or with the trivalent formulation (Ancestral + Beta + Delta) described in previous work (17), at an MOI of 0.1 in biological triplicates. Infection was carried out in infection media containing 1% HEPES 1M, 1% penicillin-streptomycin, and 0.3% Bovine Serum Albumin (BSA) in DMEM, and N-tosyl-L-phenylalanine chloromethyl ketone (TPCK)-trypsin at 1 µg/mL to allow for multicycle infection. Seventy-two hours post infection (hpi), lysates and supernatants were collected. Viral RNA was extracted using the QIAmp Viral RNA Mini Kit (QIAGEN, Hilden, Germany) for the supernatant following the manufacturer's instructions and the PureLink RNA Mini Kit (Thermo Fisher Scientific, Waltham, MA, USA) for the lysates. Kruskal-Wallis analysis between conditions is shown (P > 0.05 ns).
## RESULTS
## NDV nucleoprotein sequence identity among NDV strains
The NP gene was chosen as the target to monitor NDV presence in biological samples. NP is the most abundant component of the nucleocapsid complex (composed of NP, L, and P) and has two conserved functions among paramyxoviruses: protect the viral genomic RNA and assist in the viral RNA synthesis (18). To ensure the specificity of our RT-qPCR method, we studied the NP sequence identity in silico among representative viruses of the Paramyxoviridae family (Fig. S1). Lower identity scores ranging from 46% to 58% were found for the NP gene within the Paramyxoviridae family, supporting only the detection of NDV-like viruses when the NP gene is targeted. We then performed the same in silico analysis among different NDV strains, including several lentogenic (VG/GA, F, and Hitchner B1), mesogenic (Mukteswar or Komarov), and apathogenic enteric strains (VH, Ulster 2C, and V4) (19,20). An NP sequence identity higher than 90% was observed for all strains evaluated, except for the Mukteswar strain (Fig. 2). Next, we designed a set of primers targeting a conserved region in the NP gene of 128 bp. Further in silico analyses with the NDV NP reference strains (Table S3) and other NP paramyxoviruses (Table S4) confirmed only NP primer annealing with NDV sequences and not with the other viruses of the family.
## Validation of the NDV NP gene RT-qPCR method
Using a recombinant plasmid DNA encoding for LaSota NDV NP sequence, we optimized an RT-qPCR assay to meet the acceptance criteria defined in Table 1. We chose to use SYBR Green since it is a quick, cheap, and reliable method that can be easily implemen ted in several clinical settings (21). Results from five independent runs were used to evaluate the linear range, reproducibility, and intra-and inter-assay precision of the method. A dynamic range of 10 8 -10 1 copies was used for these assays. The results from the five runs are depicted in Table 4, where all runs met the linearity acceptance criteria, and no amplification was shown in the NTC. Intra-assay precision was evaluated by quantifying the coefficient of variation of technical triplicates. A CV below 2% was obtained in all samples analyzed in agreement with intra-precision acceptance criteria (Table 5). Inter-assay precision and reproducibility were calculated by analyzing the CV of the NDV genome copies calculated after five independent runs performed in different days. The CV of log copies was below 25% in all standard samples analyzed, supporting the reproducibility of the method (Table 6).
## Determination of RT-qPCR empirical lower limit of detection
Internal LLOD and LLOQ were established as 10 copies and 50 copies per reaction, respectively, following FDA recommendations (13). Empirical LLOD was determined by comparing the minimum detectable NP copies compared to gold-standard NDV analytical methods: PCR, HA assay after infection in SPF eggs, and EID 50 quantification (Fig. 3A). An NDV-HXP-S vaccine preparation was used for these analyses. Eleven 10-fold dilutions were performed in PBS, and the different virus dilutions were quantified with all four methods. In all cases, the LLOD was encountered at the same dilution as shown in Fig. 3B: dilution -7 presented the minimum amount of 10 copies per reaction that translated into 8.57 × 10 10 copies/mL. DNA electrophoresis after RT-PCR presented a positive band until the same dilution, and two out of three eggs were HA positive after SPF infection. EID 50 titration calculated from the HA assay gave a titer of 1.78 × 10 8 EID 50 /mL. Next, we studied the matrix effect of different human biological fluids. Viral RNA was extracted from different biological fluids, including saliva, nasal swabs, and urine samples, and a known concentration of NDV NP gene was spiked into those samples. Neither non-specific amplifications nor inhibition effects were observed in any of the conditions tested (Fig. 3C), confirming the robustness of the method in detecting the presence of the NDV in different biological fluids.
## Development and validation of an RBD-based NGS method to quantify NDV-HXP-S variant vaccines
An NGS method was generated to differentiate between NDV-HXP-S variant vaccines. A highly variable region in the RBD of the S protein was chosen for this end as shown in Fig. 4A. Amplicon-based NGS was performed using Illumina sequencing in a bidirectional way (14). The amplicon-based NGS generated ~50,000 reads in depth sequencing with an average length of 250 bp. Afterward, the reads were aligned against the different reference genomes to calculate the relative percentage of sequences that match with each template. To do so, DNAStar software SeqMan Ultra (DNA STAR, Madison, WI, USA) using the variants analysis method was used. To validate the RBD-amplicon-based NGS, purified NDV-HXP-S variant vaccines were quantified by this method in duplicate. In all cases, NDV-HXP-S variant vaccines presented a frequency greater than 90% against its homologous sequence, confirming the precision and sensitivity of the method. The CV of the technical duplicates was 1%, indicating minimal variability between replicates and confirming the high intra-assay reproducibility of the technique. To determine the empirical relative error, sequencing was performed on samples with known variant proportions: individual NDV-HXP-S variants (100% of a single variant) and mixtures of NDV-HXP-S Ancestral and XBB.1.5 amplicon sequences at defined ratios (Table 7). The relative error for each sample was calculated by comparing the theoretical and empirical variant frequencies, and the average error across all samples was 5% ± 4%. To further evaluate the sensitivity of the technique, purified NDV-HXP-S amplicons from Ancestral, Beta, Gamma, Delta, BA.1, BA.5, BQ.1.1, and XBB.1.5 were mixed in equal amounts, with a final theoretical concentration of 12.5% per variant (Fig. 5B, last column). A percentage of 11.5%, 13.6%, 11.4%, 11.1%, 14.4%, 13.5%, 9.7%, and 14.8% was obtained for each variant, respectively, confirming the differentiation of each variant from the complex pool.
NDV-HXP-S Ancestral and XBB.1.5 amplicon sequences were used to evaluate the linearity of the technique. Samples mixing different proportions of each amplicon were analyzed (Table 7), and a linear correlation analysis between the theoretical and the empirical values was performed. A significant linear correlation with an R 2 of 0.99 was achieved, supporting the linearity of the method. Finally, the robustness (inter-assay of the method) was evaluated by performing three independent runs of the Ancestral amplicon sequences. A CV of 2% was achieved (Table 8), meeting the acceptance robustness criteria as shown in Table 3.
## Application of NDV NP RT-qPCR method and RBD-based NGS in in vitro cell cultures
To evaluate the application of both techniques, Vero E6 cells were infected with monovalent Ancestral, Beta, Delta NDV-HXP-S vaccines or with the trivalent formulation described in previous work (17), at an MOI of 0.1 in biological triplicates (Fig. 5A). Seventy-two hours post infection, cell lysates and supernatants were collected and analyzed by RT-qPCR to measure the NP viral genome copies and by NGS to study the relative percentage of each variant vaccine when administered together.
Comparable NP viral copies/mL were obtained after the monovalent and the trivalent infections 72 hpi (Fig. 5B). Regarding the variant analysis, monovalent vaccines contained its homologous sequence in relative frequencies higher than 90% (Fig. 5C). This relative percentage was similar to the one obtained in the NGS validation experiments, supporting the use of this technique in in vitro models. As for the trivalent infection, a relative frequency of 33% is expected to conclude that the three variants replicate equally when administered together. Results show the relative percentage of reads of each variant was 34.2%, 22.9%, and 42.9% at an MOI of 0.1 in cell supernatants for Ancestral (Wuhan), Beta, and Delta variants (Fig. 5C), supporting the replication of the three variants when co-administered together.
## DISCUSSION
Since the declaration of the COVID-19 pandemic, huge efforts have been undertaken to develop effective vaccines as well as reliable diagnostic methods. In previous work, our groups have developed NDV-based viral vector vaccines (NDV-HXP-S) to prevent SARS-CoV-2 infections (3). NDV is attenuated in mammalian hosts (4,22,23), making it a promising live-attenuated vector that can be administered as mucosal vaccine. With views to study the viral shedding of these new candidates in monovalent and multi valent formulations, we developed two methods to quantitatively assess NDV-based vaccines in vitro and in vivo.
Nucleic acid amplification tests are globally accepted as one of the gold-standard detection methods for respiratory viruses, including SARS-CoV-2 and influenza viruses (24). Like influenza, NDV is a negative-stranded RNA virus; hence, it requires the synthesis of positive-stranded mRNA sequences to produce the different viral proteins (25). With the aim to detect any possible presence of the virus, we designed an RT-qPCR assay, in which the reverse transcriptase reaction will amplify both positive and negative strands. We chose to target a small region of the NP protein, which is a conserved and abundant structural protein among NDV vaccine strains, like LaSota or B1 (26,27). An RT-qPCR efficiency of 100% ± 10% was obtained in the validation with an R 2 >0.98, which confirmed the linearity of the method. The inter-and intra-assay CV values were below 25% and 2%, respectively, supporting the good precision and reproduci bility of the technique. NDV isolation has been classically performed by infection of SPF-embryonated chicken eggs (28). Although the method is known to be highly sensitive, it is labor intensive. The suspected material, which typically comes from avian respiratory secretions or cloacal samples, needs to be mixed with an antibiotic cocktail and incubated for several days in SPF eggs (28). Comparing the technology involving the infections of SPF-embryonated chicken eggs and the current RT-qPCR technique revealed the same value for LLOD. We believe the current RT-qPCR methodol ogy presents great promise to quantify NDV viral copies in a short period of time without the need for antibiotic cocktails or egg supplies.
With the aim of testing a mucosal multivalent NDV-HXP-S vaccine, we developed an NGS method to quantify the relative percentage of different NDV-HXP-S variant vaccines in a single formulation. We designed primers flanking a specific region of the spike protein with high variability between all the current VOCs (Ancestral, Beta, Gamma, Delta, BA.1, BA.5, BQ.1.1, and XBB.1.5). Of note, we have also tested the method with the newly developed EG.5, JN.1, and KP.2 NDV-HXP-S vaccines (data not shown). Importantly, the targeted region includes variant-defining nucleotide differences, and the method can distinguish variants differing by as little as a single nucleotide, which supports its high accuracy even in co-infection scenarios. The assay showed high specificity allowing to differentiate each variant from the rest with an average error of 5% ± 4%. In in vitro cell culture experiments, monovalent NDV-HXP-S infections were detected with a specificity of 93%-98%, whereas trivalent NDV-HXP-S formulations presented similar amounts (22.9%-42.9%) of each spike in the cell supernatants analyzed 72 hours post infection. In conclusion, both techniques were validated to accurately quantify NDV-HXP-S vaccines in monovalent and multivalent formulations. Of note, other NDV-based vaccines against respiratory infections, including influenza A (29), human parainfluenza virus-3 (30) or respiratory syncytial virus (31), among others, are in preclinical development (19). The present results will help the study of virus shedding in future clinical trials to evaluate mucosal NDV-based vaccines.
## References
1. Who (2020) "Naming the coronavirus disease (COVID-19) and the virus that causes it"
2. Ochani, Asad, Yasmin et al. (2021) "COVID-19 pandemic: from origins to outcomes. A comprehensive review of viral pathogenesis, clinical manifestations, diagnostic evaluation, and management" *Infez Med*
3. Sun, Liu, Amanat et al. (2021) "A Newcastle disease virus expressing a stabilized spike protein of SARS-CoV-2 induces protective immune responses" *Nat Commun*
4. Park, García-Sastre, Cros et al. (2003) "Newcastle disease virus V protein is a determinant of host range restriction" *J Virol*
5. Tcheou, Raskin, Singh et al. (2021) "Safety and immunogenicity analysis of a newcastle disease virus (NDV-HXP-S) expressing the spike protein of SARS-CoV-2 in sprague dawley rats" *Front Immunol*
6. Lara-Puente, Carreño, Sun et al. "-Dubernard B. 2021. Safety and immunogenicity of a newcastle disease virus vector-based SARS-CoV-2 Vaccine candidate, AVX/COVID-12-HEXAPRO (Patria)"
7. Fulber, Farnós, Kiesslich et al. (2021) "Process development for newcastle disease virusvectored vaccines in serum-free vero cell suspension cultures" *Vaccines (Basel)*
8. Gonzalez-Dominguez, Martinez, Slamanig et al. (2022) "Trivalent NDV-HXP-S vaccine protects against phylogenetically distant SARS-CoV-2 variants of concern in mice" *Microbiol Spectr*
9. Slamanig, González-Domínguez, Chang et al. (2024) "Intranasal SARS-CoV-2 Omicron variant vaccines elicit humoral and cellular mucosal immunity in female mice" *EBioMedicine*
10. Slamanig, Lemus, Lai et al. (2025) "A single immunization with intranasal Newcastle disease virus (NDV)-based XBB.1.5 variant vaccine reduces disease and transmission in animals against matched-variant challenge" *Vaccine (Auckl)*
11. Ma, Bell, Loker "2021. qPCR and qRT-PCR analysis: regulatory points to consider when conducting biodistribution and vector shedding studies" *Mol Ther Methods Clin Dev*
12. Ema (2022) "European medicines agency ICH guideline M10 on bioanalytical method validation and study sample analysis"
13. Fda (2020) "Long term follow-up after administration of human gene therapy products"
14. Genewiz (2017) "Next Generation Sequencing: Amplicon-EZ"
15. Gilchuk, Bangaru, Kose et al. (2021) "Human antibody recognition of H7N9 influenza virus HA following natural infection" *JCI Insight*
16. Izsve (2020) "Haemagglutination test (SOP IMM-065), on European union reference laboratory for avian influenza and newcastle disease"
17. González-Domínguez, Martínez, Slamanig et al. (2022) "Trivalent NDV-HXP-S vaccine protects against phylogenetically distant SARS-CoV-2 variants of concern in mice" *Microbiol Spectr*
18. Morgan (1991) "Evolutionary relationships of paramyxovirus nucleocapsid-associated proteins"
19. Fulber, Kamen (2022) "Development and scalable production of newcastle disease virus-vectored vaccines for human and veterinary use" *Viruses*
20. Howley, Knipe, Whelan (2020) "Fields virology: emerging viruses"
21. Pereira-Gómez, Fajardo, Echeverría et al. (2021) "Evaluation of SYBR green real time PCR for detecting SARS-CoV-2 from clinical samples" *J Virol Methods*
22. López, García-Sastre, Williams et al. (2003) "Type I interferon induction pathway, but not released interferon, participates in the maturation of dendritic cells induced by negative-strand RNA viruses" *J Infect Dis*
23. Fernandez-Sesma, Marukian, Ebersole et al. (2006) "Influenza virus evades innate and adaptive immunity via the NS1 protein" *J Virol*
24. Dutta, Naiyer, Mansuri et al. (2022) "COVID-19 Diagnosis: a comprehensive review of the RT-qPCR method for detection of SARS-CoV-2" *Diagnostics (Basel)*
25. Ganar, Das, Sinha et al. (2014) "Newcastle disease virus: current status and our understanding" *Virus Res*
26. Mao, Ma, Schrickel et al. (2022) "Review detection of Newcastle disease virus" *Front Vet Sci*
27. Kim, Samal (2016) "Newcastle disease virus as a vaccine vector for development of human and veterinary vaccines" *Viruses*
28. Mao, Ma, Schrickel et al. (2022) "Review detection of Newcastle disease virus" *Front Vet Sci*
29. Nakaya, Cros, Park et al. (2001) "Recombinant Newcastle disease virus as a vaccine vector" *J Virol*
30. Bukreyev, Huang, Yang et al. (2005) "Recombinant newcastle disease virus expressing a foreign viral antigen is attenuated and highly immuno genic in primates" *J Virol*
31. Martinez-Sobrido, Gitiban, Fernandez-Sesma et al. (2006) "Protection against respiratory syncytial virus by a recombinant Newcastle disease virus vector" *J Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12522720&blobtype=pdf
|
# Frontiers Editorial O ce, Frontiers Media SA, Switzerland *CORRESPONDENCE
Ahmed Elolimy, Abd El, Mahmoud Mohamadin, Rashid Manzoor, Mohamed Abdelmegeid, Samah Mosad, Sahar Abd, El Rahman
## A Correction on
Advancements in antiviral approaches against foot-and-mouth disease virus: a comprehensive review by Mohamadin, M., Manzoor, R., Elolimy, A., Abdelmegeid, M., Mosad, S., and Abd El Rahman, S. (
). Front. Vet. Sci. :
. doi: . /fvets. .
In the published article, there was a mistake in the affiliation of one of the co-authors, Dr. Rashid Manzoor. Dr. Rashid Manzoor's affiliation was displayed as " 2 College of Veterinary Medicine, University of Al Dhaid, Sharjah, United Arab Emirates". The correct affiliation is " 3 Veterinary Science Program, Faculty of Health Sciences, Higher Colleges of Technology, Sharjah, United Arab Emirates."
The original version of this article has been updated.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12758128&blobtype=pdf
|
# Fast-acting single-dose vesicular stomatitis virus-Sudan virus vaccine: a challenge study in macaques
Paige Fletcher, Kyle O'donnell, Friederike Feldmann, Joseph Rhoderick, Chad Clancy, Jil Haase, Cecilia Prator, Brian Smith, Andrea Marzi, Bronwyn Gunn, Heinz Feldmann
## Abstract
Background:The Sudan virus (SUDV) outbreaks in recent years, including the most recent outbreak in Uganda, created a public health emergency in the eastern Africa region. There are no licensed vaccines or therapeutics approved against SUDV; however, we have previously developed a vesicular stomatitis virus (VSV)-based vaccine expressing the SUDV glycoprotein and showed its single-dose efficacy in non-human primates (NHPs) with pre-existing Ebola virus (EBOV) immunity. Here, we determined the fast-acting capacity of this vaccine in naive NHPs. In addition, we examined if the licensed VSV-EBOV vaccine would provide any prophylactic benefit against SUDV infection.
Methods:We used four groups of male cynomolgus macaques (n=6 per group) aged 2•5-4•5 years and weighing 2•9-5•2 kg. NHPs were vaccinated by intramuscular injection 28 days before challenge with either 1 × 10 7 plaque-forming units (PFUs) VSV-SUDV, VSV-EBOV, or control vaccine (VSV-LASV). Another group was vaccinated with 1 × 10 7 PFU VSV-SUDV 7 days before challenge. On day 0, all 24 NHPs were challenged with a lethal dose of SUDV (1 × 10 4 50% tissue culture infectious doses [TCID 50 ] SUDV-Gulu, backtitred as 1•1 × 10 4 TCID 50 ). We assessed anaesthetised NHPs on days -28, -21, -14, and -7 before challenge; days 0, 3, 6, 9, 14, 21, 28, and 35 after challenge; and at euthanasia (day 42 for survivors).Findings: All VSV-SUDV-vaccinated NHPs were protected from disease after the lethal SUDV challenge. In contrast, the VSV-EBOV-vaccinated and control NHPs succumbed to disease between days 5 and 7 after challenge presenting with classical signs of Sudan virus disease associated with high-titre viraemia (>1 × 10 8 TCID 50 per mL), high viral organ load (>1 × 10 8 TCID 50 per g), dysregulated cytokine profiles, and typical pathological changes. The humoral immune response in the NHPs vaccinated with VSV-SUDV 1 month before challenge resulted in a profound and sustained serum antibody response (20 000-30 000 U/mL) with a diverse functionality profile (antibody-dependent cellular phagocytosis and antibody-dependent complement), which was not observed to the same extent in NHPs vaccinated 1 week before challenge.
Interpretation:We showed that a single dose of VSV-SUDV protected NHPs from lethal SUDV infection within 1 week. The fast-acting nature highlights VSV-SUDV as an ideal countermeasure for ring vaccination during outbreaks of Sudan virus disease pending further preclinical and clinical assessment. In contrast, VSV-EBOV provided no relevant protection against SUDV infection in NHPs, highlighting the need for species-specific filovirus vaccines.
## Introduction
Filoviruses continue to pose a substantial threat to regional public health and beyond, particularly in their endemic areas in Africa. 1 In the past decade alone, Ebola virus (EBOV) caused the West African epidemic and several large outbreaks in the Democratic Republic of the Congo, resulting in over 30 000 cases and 12 000 deaths. 2 In addition, Marburg virus, Sudan virus (SUDV), and Bundibugyo virus have caused outbreaks, albeit smaller, in west, central and east Africa, highlighting the continued circulation of filoviruses in their respective reservoirs with expanding endemic areas. 2,3 SUDV is the second most important filovirus for public health in Africa; nine outbreaks have occurred mainly in Sudan (now in south Sudan territory) and Uganda, with the largest on record in Gulu, Uganda, causing 425 infections with 224 fatalities. 2 In 2022, an outbreak of Sudan virus disease (SVD) in Uganda resulted in 164 cases and 77 fatalities. 4 In 2025, Uganda had an SVD outbreak in its capital, Kampala. 5 There are no approved vaccine or treatment options available for SUDV; however, a vesicular stomatitis virus (VSV)-based vaccine expressing the SUDV glycoprotein, termed VSV-SUDV, is currently being used in a clinical trial in Uganda. 6 In addition, a monoclonal antibody binding the SUDV glycoprotein, as well as the antiviral remdesivir, are being administered to infected people. 5 Previous studies showed that surviving non-human primates (NHPs) from EBOV infections generally succumb to a subsequent SUDV challenge and vice versa, indicating limited cross-protection between those orthoebolavirus species. 7,8 Thus, licensed EBOV vaccines and direct-acting specific therapeutics such as monoclonal antibodies are unlikely to provide protection against SUDV infection. Previously, we showed that VSV-SUDV is uniformly protective within 28 days when a single high dose was administered to NHPs with preexisting immunity to VSV-EBOV. 8 These results add to the growing body of evidence that pre-existing vector immunity has only limited impact on VSV-filovirus vaccine efficacy. [8][9][10] However, only limited data are available on the cross-protective potential of species-specific VSV-filovirus vaccines in NHP models. [11][12][13] Therefore, we determined the protective efficacy of a single high-dose VSV-SUDV or VSV-EBOV vaccination against lethal SUDV challenge. In addition, we investigated whether the VSV-SUDV vaccine possesses a fastacting potential to be used to control SVD outbreaks. 14,15
## Methods
## Study design
Previously described VSV-based vaccine vectors expressing the EBOV-Kikwit glycoprotein (VSV-EBOV), 14 SUDV-Gulu glycoprotein (VSV-SUDV), 8 or Lassa virus glycoprotein complex (VSV-LASV) 9 were used in this study. For the NHP challenge, SUDV-Gulu (GenBank NC_006432.1) at 10 000 median 50% tissue culture infectious doses (TCID 50 ) was used intramuscularly into the caudal thigh as previously described. 8 24 male cynomolgus macaques of Chinese or Cambodian origin, aged 2•5-4•5 years and weighing 2•9-5•2 kg at the time of vaccination, were used in this study. The NHPs were randomly divided into four study groups with six NHPs each. On day -28, NHPs received a 1 mL intramuscular injection for vaccination containing 1 × 10 7 plaque-forming units (PFUs) of VSV-SUDV, VSV-EBOV, or VSV-LASV (control). On day -7, the last vaccine group received 1 × 10 7 PFU VSV-SUDV intramuscularly. All 24 NHPs were challenged intramuscularly on day 0 with 1 × 10 4 TCID 50 SUDV-Gulu (backtitred as 1•1 × 10 4 TCID 50 ) into two sites in the caudal thigh, as previously described. 8 We assessed anaesthetised NHPs on days -28, -21, -14, and -7 before challenge; days 0, 3, 6, 9, 14, 21, 28, and 35 after challenge; and at euthanasia (day 42 for survivors). On these days, physical examinations, including rectal temperature determination, swab collection, body weight measurements, and blood draws were performed. Blood draws included collection of up to 15% of the circulating blood volume in a 14-day period, as approved by the IACUC, from the femoral vein into EDTA and serum vacutainers for immunological, virological, and serological analyses. Detailed methods are provided in the appendix (pp 2-4).
All work involving SUDV was performed in the maximum containment laboratory at the Rocky Mountain Laboratories, Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, MT, USA. Rocky Mountain Laboratories is an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited institution. All procedures followed Rocky Mountain Laboratories institutional biosafety committee-approved standard operating procedures. NHP work was performed in strict accordance with the recommendations described in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health, the Office of Animal Welfare, and the Animal Welfare Act, US Department of Agriculture. This study was approved by the Rocky Mountain Laboratories animal care and use committee (ASP#2022-051-E), and all procedures were conducted on anaesthetised NHPs by trained personnel under the supervision of board-certified clinical veterinarians. The NHPs were observed at least twice daily for clinical signs of disease according to a Rocky Mountain Laboratories animal care and use committee-approved scoring sheet (appendix p 2), humanely euthanised when they reached endpoint criteria, and samples for analysis were collected at necropsy.
## Statistical analysis
Statistical analysis was performed in GraphPad Prism, version 10.2.0. For statistical analysis, data from the VSV-EBOV group on day 5 were combined with the single day 6 datapoint and compared with all other groups on day 6. Unless otherwise noted, data were analysed by two-way ANOVA with Tukey's multiple comparisons to evaluate statistical significance at all timepoints between all groups. Significant differences in the survival curves were determined by performing log-rank analysis. To analyse cytokine levels, Kruskal-Wallis test with Dunn's multiple comparisons was used to compare all groups. To analyse SUDV soluble glycoprotein (sGP) levels, Mann-Whitney test was used. Statistical significance is indicated as p value less than 0•0001, 0•001, 0•01, or 0•05.
## Role of the funding source
The funders had no role in study design, data collection, data analysis, data interpretation, decision to publish, or writing of the report.
## Results
Vaccine viraemia was short-lived in all vaccinated NHPs regardless of the vaccine construct, peaking 24 h after vaccination and resolving within a week (figure 1A). On day 0, all 24 NHPs were challenged intramuscularly with a lethal dose of 1 × 10 4 TCID 50 of SUDV. Clinical signs of disease were assessed at least once daily during the acute disease phase by assigning a clinical score (figure 1B). NHPs in the VSV-EBOV and the control groups started to display signs of disease on day 3. Disease in the VSV-EBOV group progressed faster compared with the control group, leading to euthanasia of the NHPs on days 5-6 (figure 1C). NHPs in the control group reached endpoint criteria on days 6-7 and were euthanised. The VSV-EBOV and control NHPs presented with greater than 1 × 10 8 TCID 50 per mL of SUDV in the blood (figure 1D,E), which corresponded with concentrations of greater than 100 000 ng/mL SUDV sGP in serum (figure 1F). All non-protected animals displayed a decrease in thrombocytes (figure 1G), increase in white blood cell (WBC), specifically neutrophil, counts (appendix p 5), and elevated liver (figure 1H; appendix p 5) and kidney (figure 1I; appendix p 5) enzymes as well as a decrease in serum albumin, total protein, and calcium (appendix p 5), all characteristic parameters for SVD in NHPs. The VSV-EBOV group showed a stress leukogram as evidenced by increases in WBCs and neutrophils on day 3, together with a decrease in lymphocytes. The control NHPs did not develop this immune cell signature. In contrast, NHPs vaccinated with VSV-SUDV at either days 28 or 7 survived the lethal challenge with minimal signs of disease and presented with low and temporary SUDV in the blood on days 6 and 9 (figure 1D,E).
High levels up to 1 × 10 9 TCID 50 per g of SUDV were detected in tissue samples (figure 1J,K). Histological lesions consistent with classic SVD were observed in the liver and spleen in all non-protected NHPs (figure 2). Mild to severe lymphoid necrosis with lymphoid depletion was observed in the spleen of every control NHP (figure 2A,D). Immunohistochemical evaluation showed follicular presence of SUDV antigen in sections of spleen of all non-protected NHPs (figure 2G,J). Hepatic lesions included single-cell hepatocellular necrosis, random or centrilobular coagulative necrosis, and sinusoidal fibrin thrombi (figure 2B,E). Minimal to mild lymphohistiocytic periportal hepatitis was observed in all control NHPs (figure 2E). Additionally, Kupffer cells and some hepatocytes showed SUDV antigen in these groups (figure 2H,K). In contrast to our previous study with VSV-SUDV and SUDV challenge, 8 a prominent lesion observed in at least one lung lobe of every non-protected NHP was pyogranulomatous vasculitis extending into interstitial pneumonia (figure 2C,F). Pyogranulomatous inflammation was categorised as moderate in at least one lung lobe of five of six VSV-EBOV-vaccinated NHPs, whereas moderate inflammation was only detected in the lung lobes of one of six control NHPs. Prominent fibrin thrombi and thrombo-emboli ranging from capillary to large-calibre arteries were observed in at least one evaluated lung lobe in each of the non-protected NHPs (appendix p 6). Evaluation of lung sections showed SUDV antigen in pneumocytes, pulmonary macrophages, endothelial cells, and smooth muscle cells of at least one evaluated section of lung in all non-protected NHPs (figure 2I,L). Additional classic histological changes observed in these NHPs included lymphohistiocytic duodenitis associated with haemorrhage and oedema (appendix p 6).
In VSV-SUDV-vaccinated NHPs on day 42, the only histological changes observed in peripheral nodes and sections of the spleen included lymphoid hyperplasia and sinus histiocytosis (figure 2M,P) consistent with a response to occasional remaining low levels of up to 1 × 10 4 TCID 50 per g of SUDV (figure 1J,K). No clinically relevant histopathological lesions were observed in the liver of any VSV-SUDV-vaccinated NHP (figure 2N,Q). Evaluation of lung sections of VSV-SUDV-vaccinated NHPs revealed lymphohistiocytic vasculitis extending into alveolar septa (figure 2O,R). SUDV antigen was not observed in evaluated sections of lymph node, spleen, liver, or lung in any VSV-SUDV-vaccinated NHPs (figure 2S-X).
SVD is associated with a dysregulated cytokine response; 8 therefore, serum concentrations of GM-CSF, MCP-1, IL-10, IL-12p70, IFN-a2a, IFNg, IL-4, IL-6, IL-15, IP-10, IL-1β, and TNFα were measured for 1 week after vaccination and for 2 weeks after challenge. After vaccination, all groups had a similar response to the replicating viral vaccine (appendix p 7). Only the control NHPs developed a significant increase in MCP-1 1 day after vaccination (2600-5500 pg/mL; p<0•0324). After challenge, only the VSV-EBOV and control NHPs developed elevated cytokine and chemokine responses indicative of a cytokine storm (figure 3). MCP-1, IL-15, and IP-10 were substantially increased in the serum of NHPs during end-stage SVD. In contrast, all VSV-SUDV-vaccinated NHPs maintained normal cytokine responses up to day 14 (figure 3).
Cytokine and chemokine expression as well as the amount of SUDV sGP present in lung samples of all NHPs collected at the time of euthanasia were assessed. Only lung samples from NHPs in the VSV-EBOV and control groups had substantially higher cytokine and chemokine expression profiles. Notably, GM-CSF, TNFα, IL-6, IP-10, and MCP-1 were only upregulated in lung samples from the VSV-EBOV group compared with both VSV-SUDV groups, but not compared with the control NHPs (figure 4A). Although there was not a substantial difference in the lung cytokine expression profile or SUDV lung burden (figure 1J,K) between the VSV-EBOV and control groups, SUDV sGP levels were significantly higher in the VSV-EBOV group compared with control NHPs (12 000-700 000 ng/g vs 5600-11 800 ng/g; p=0•0022; figure 4B). This difference was not observed in serum SUDV sGP (figure 1E). The NHP with the highest cytokine and SUDV sGP responses in the lung also had the highest SUDV lung titre (figure 1J,K), but no other examined parameters, including viraemia or serum SUDV sGP, were elevated beyond the variability of the VSV-EBOV group. In VSV-SUDV-vaccinated NHPs, cytokines in the lungs on day 42 remained within normal range, but, for ethical and logistical reasons, the study did not include a group for an earlier timepoint or day 42 naive NHPs. Since there was no indication of SUDV viraemia or SUDV sGP in the serum during the acute disease phase (figure 1D,E), SUDV sGP was not assessed as it correlates with viral RNA expression. 16 The total antigen-specific IgG response has been identified as the main mediator of protection for VSV-filovirus vaccines. 17,18 Therefore, SUDV glycoprotein-specific IgG serum titres were determined longitudinally from days -28 to 42 and, as expected, revealed the highest responses for the VSV-SUDV day -28 group at the time of challenge (1400-3600 U/mL; figure 5A). The boosting effect of the challenge virus was lower in the VSV-SUDV day -7 group (2500-11 600 U/mL, day 6) compared with the VSV-SUDV day -28 group (20 900-34 100 U/mL, day 9), likely due to the short time between vaccination and challenge (figure 5A). The VSV-EBOV-vaccinated NHPs developed low titres of IgG cross-reactive with SUDV glycoprotein (330-2900 U/mL, day 0), which decreased after challenge likely due to IgG consumption from SUDV replication (figure 5A). In contrast, these NHPs developed high titres of EBOV glycoprotein-specific IgG (2800-5000 U/mL, day 0; appendix p 8). The day -28 VSV-SUDV-vaccinated NHPs developed cross-reactive IgG only after challenge (appendix p 8). Analysis of serum samples from the control NHPs did not result in substantial differences in EBOV glycoprotein-specific or SUDV glycoprotein-specific IgG (figure 5A; appendix p 8). A 2023 publication highlighted the importance of EBOV sGP-specific antibodies for durable protection in NHPs. 19 When we compared the IgG titres specific to SUDV sGP in serum at selected timepoints between the groups, we found that it was associated with the serum titres of SUDV glycoprotein-specific IgG (figure 5A,B). This was not the case for IgG cross-reactive to EBOV sGP in the VSV-SUDV day -28 group where titres increased until day 14 and then remained constant after day 14 (appendix p 8), but IgG titres cross-reactive to EBOV glycoprotein spiked between days 6 and 14 and then remained constant (appendix p 8).
Next, serum neutralisation titres were evaluated using a previously established VSV-SUDV-GFP neutralisation assay 8 with results presented as 50% fluorescence reduction (FRNT 50 ) of GFP-positive cells at selected timepoints. Only VSV-SUDV day -28 vaccinated NHPs had developed a significant neutralising response (1:40-1:320; p<0•0099) on day 0, which was boosted by the SUDV challenge (figure 5C). In contrast, the neutralising antibody titres in the day 7 group only showed a significant increase on day 42 (figure 5C). NHPs in the control as well as VSV-EBOV-vaccinated group developed no neutralising response (figure 5C).
We also investigated if the EBOV glycoprotein-specific immunity, which is unspecific to the challenge virus, could have resulted in antibody-dependent enhancement (ADE) of SUDV infection as there is in vitro evidence for ADE in EBOV infection. 20,21 We compared serum samples from day -28 (vaccination) and day 0 (challenge) of the VSV-EBOV and control groups in this assay using SUDV-GFP. Serum samples in both groups did not reach titres associated with ADE described by others for filoviruses (appendix p 8). 21 Antibody Fc-effector functions were investigated next, including the antibody-dependent cellular phagocytosis (ADCP), antibody-dependent complement deposition (ADCD), antibody-dependent neutrophil phagocytosis (ADNP), and antibody-dependent natural killer cell activation (ADNKA). NHPs vaccinated with VSV-SUDV at day -28 showed significantly more ADCP on day 0, which continued through day 6 (34-57 phagocytic score; p<0•0230; figure 5D). We investigated whether neutrophils would facilitate phagocytosis. NHPs vaccinated at either timepoint with VSV-SUDV showed significantly higher ADNP levels (2-29 phagocytic score; p<0•0353) on day 6 when the control and VSV-EBOV groups succumbed to disease (figure 5E). On day 42, only the levels in the VSV-SUDV day 28 group had further increased (figure 5E). Similarly, NHPs vaccinated with VSV-SUDV at either timepoint showed a strong induction of ADCD on days 0 and 6 (figure 5F). On day 0, ADCD was significantly increased in the VSV-SUDV day 28 group compared with the cohort vaccinated on day 7 with VSV-SUDV (54 000-80 000 vs 9600-26 500 C3 deposition; p<0•0001; figure 5F). ADNKA analysis (appendix p 8) yielded few differences, with only the control group showing a significantly lower amount of IFNγ induction (7•75-9•39% positive cells; p<0•0437) on day 0 (appendix p 8).
Despite the limited impact of T-cell responses on the protective efficacy of VSV-filovirus vaccines, [20][21][22] CD4 T-cell responses were analysed to evaluate the changes in T-cell polarity over time. Peripheral blood mononuclear cells were stimulated with a SUDV glycoproteinspecific peptide pool and analysed for CD4 + effector memory re-expressing (EM-RE) T cells (appendix p 9). CD69, IFNg, and IL-4 expression was analysed at selected timepoints. No substantial differences were observed between any groups at all timepoints. However, upon further investigation, a shift was observed in the T-cell polarity of the antigen-specific CD4 + EM-RE cells for the VSV-SUDV groups (appendix p 9). Initially, this memory cell cohort was primarily Th1-driven in nature, expressing IFNg on day 0. As the cellular immunity memory matured over time, the CD4 + EM-RE cells shifted to a balanced Th1/Th2 phenotype expressing either IFNg or IL-4, likely contributing to the maintenance of the humoral immune response.
## Discussion
Uganda declared its sixth outbreak of SVD on Jan 30, 2025; it ended on April 26, 2025, with 14 cases (12 confirmed, two probable) with four fatalities (two confirmed, two probable). 5 Despite the lack of approved vaccines, several vaccine candidates 23 have shown efficacy in preclinical NHP studies including a VSV-SUDV vaccine following the success of the related EBOV vaccine, VSV-EBOV (Merck, Rahway, NJ, USA). This vaccine is in clinical development by IAVI, has shown single-dose efficacy in NHPs within 4 weeks, and proved to be immunogenic and safe in a phase 1 clinical trial. 24 In addition, VSV-SDUV administration 20-30 min post-challenge resulted in 100% survival in NHPs. 25 Uganda started a ring vaccination trial with this vaccine in February, 2025. 6 However, the fast-acting prophylactic potential of this vaccine has not been investigated.
We previously updated VSV-SUDV by using a more current SUDV glycoprotein from the SUDV-Gulu strain and showed uniform protection with a single dose within 4 weeks in NHPs with pre-existing EBOV immunity. 8 Single-dose VSV-SUDV vaccination 28 days or 7 days before lethal SUDV challenge protected naive NHPs from disease and lethality independent of vaccination timepoint, highlighting the fast-acting potential of the VSV-SUDV vaccine. We show the lack of cross-protection from the VSV-EBOV vaccine against SUDV infection despite the many similarities between the viruses including the development of cross-reactive IgG responses after vaccination. This contrasts with a 2023 study exploring cross-protection of VSV-EBOV against SUDV in the guinea pig disease model, which resulted in approximately 60% survival. 26 However, the NHP model is regarded as the gold standard in filovirus countermeasure research, and efficacy results seem more predictive for human outcome than results from rodent challenge studies. 27 This highlights the continued need for NHP studies on filovirus countermeasure development and approval following the animal rule 28 when SVD outbreaks occur infrequently and with low case numbers, presenting a challenge for the conduct of human efficacy trials.
In a previous study, we showed that VSV-EBOV vaccination approximately 1 year before SUDV challenge only resulted in limited cross-reactivity, but no protective efficacy. 8 Our results indicate that the time between VSV-EBOV vaccination and SUDV exposure might impact the disease course as the VSV-EBOV cohort succumbed earlier compared with the control group. We examined serum antibodies for infection-enhancing properties, but could only detect background levels of approximately 4% of GFP-positive cells, which are negligible in comparison with previous studies describing filovirus ADE using monoclonal antibodies at approximately 60%. 21 Similarly, for dengue virus, ADE levels are much higher at 20-40%, 29 leading us to believe that ADE is not the primary mechanism for the differences in disease progression we observed after SUDV challenge. A secondary mechanism for ADE of disease is through the binding of the complement C1q receptor. 30 However, we found no substantial increase in complement deposition for the VSV-EBOV vaccinated NHPs compared with the controls (figure 5F). One might speculate that 4 weeks after vaccination is likely the time when the immune response specific to EBOV glycoprotein is high yet not specific to the challenge virus and, therefore, might contribute to the accelerate disease course we observed. Cross-reactive immunity might wane over time or might never be elicited at a level relevant to confer protection. We have yet to determine why this disease phenotype was observed 28 days after VSV-EBOV vaccination, but not 1 year after VSV-EBOV vaccination and EBOV challenge. 8 Notably, the VSV-EBOV-vaccinated NHPs developed lung pathology after SUDV infection, which was less severe in control NHPs at the time of necropsy. Another significant difference between the VSV-EBOV and control groups is the neutrophilia observed on day 3 for the VSV-EBOV group only. The induction of aberrant neutrophils early after SUDV challenge has previously been described as a contributing factor to fatal disease and might have had the same effect here. 31,32 Collectively, our data indicate that an aberrant innate immune response in expanded tissue compartments early in the disease course might contribute to the accelerated time to death of the VSV-EBOV-vaccinated NHPs.
A recent study by van Tol and colleagues showed that single dose as well as prime or boost vaccination with an adenovirus-based vector encoding both the EBOV glycoprotein and the SUDV glycoprotein did not result in protection from SUDV challenge. 32 Like the data presented here for the VSV-EBOV group, all NHPs vaccinated with this adenovirus-based vector had cross-reactive binding antibodies, but failed to protect from SUDV challenge. It seems that there is an importance to the specificity of the protective immune respons, at least against infection with filoviruses, that warrants further investigation.
In addition, both VSV-SUDV vaccine groups showed substantially elevated C3 deposition indicative of ADCD compared with the control and VSV-EBOV groups at day 6. The role of complement activation in filovirus pathogenesis until this point has both positive and negative effects. Overactivation of the complement pathway via the classical pathway and mannose binding lectin route has contributed to increased pathogenesis. 33 However, in several monoclonal antibody treatment studies, it has been shown that the engineering of antibodies to induce ADCD increased their therapeutic potential. 34,35 NHPs in the VSV-SUDV groups had significantly higher C3 deposition on day 6, but did not succumb to disease like the control and VSV-EBOV groups, which did not show high levels of ADCD at that time. Our results indicate that a diverse antibody functionality repertoire is desired to inhibit disease progression after SUDV infection.
Although this study provides insight into the protective immunity elicited by the VSV-SUDV vaccine, it also presents its limitations regarding the investigation of cross-protection. If a day -7 group vaccinated with VSV-EBOV would have been included, a more thorough investigation of the timing for the onset of a potentially cross-protective immune response could have been performed. Likewise, a control group vaccinated on day -7 was also not included; however, the impact of the data from a VSV-EBOV day -7 group would have likely been greater. Furthermore, T-cell responses and ADE could have been investigated in more detail; however, the sampling timepoints and amounts of samples collected limited us in pursuing this further. Future studies investigating cross-protection between filoviruses will be designed with additional analyses in mind.
## Research in context
## Evidence before this study
We searched PubMed for articles describing Sudan virus (SUDV)-specific vaccination approaches with protective efficacy in non-human primate (NHP) models published between Jan 1, 2000, and March 4, 2025, with no language restrictions using the MeSH search terms "Sudan virus", "Sudan ebolavirus", "nonhuman primate" and "vaccine". We identified several preclinical vaccine studies with protective efficacy against SUDV. However, we did not find any study describing fast-acting prophylactic activity that conferred uniform protection against SUDV in NHPs.
## Added value of this study
Vaccination with a single high dose of the VSV-SUDV uniformly protects NHPs when administered 1 week before lethal SUDV challenge. The FDA-approved Ebola virus vaccine (VSV-EBOV) does not confer protective efficacy against SUDV challenge in NHPs.
## Implications of all the available evidence
Our results show the onset of fast-acting protective efficacy of the VSV-SUDV vaccine within 1 week in the NHP model. They further show that VSV-EBOV does not provide any cross-protection against SUDV disease, highlighting the need for species-specific vaccine development against filoviruses. Finally, our data provide strong support for the use of the VSV-SUDV vaccine for ring vaccination during outbreaks of SUDV disease.
## References
1. Feldmann, Sprecher, Geisbert et al. (2020) *N Engl J Med*
2. (2024) "Ebola (Ebola virus disease) -Outbreaks"
3. Who (2025) "Marburg virus disease"
4. Who (2023) "Ebola disease caused by Sudan Ebolavirus -Uganda"
5. Who (2025) "Sudan virus disease -Uganda"
6. (2025) "Adepoju P Uganda launches vaccine trial for Sudan virus disease" *Lancet Microbe*
7. Jones, Feldmann, Ströher (2005) "Live attenuated recombinant vaccine protects nonhuman primates against Ebola and Marburg viruses" *Nat Med*
8. Marzi, Fletcher, Feldmann et al. (2023) "Species-specific immunogenicity and protective efficacy of a vesicular stomatitis virus-based Sudan virus vaccine: a challenge study in macaques" *Lancet Microbe*
9. Marzi, Feldmann, Geisbert et al. (2015) "Vesicular stomatitis virus-based vaccines against Lassa and Ebola viruses" *Emerg Infect Dis*
10. Marzi, Feldmann, Donnell et al. (2023) "Preexisting immunity does not prevent efficacy of vesicular stomatitis virus-based filovirus vaccines in nonhuman primates" *J Infect Dis*
11. Geisbert, Geisbert, Leung (2009) "Single-injection vaccine protects nonhuman primates against infection with Marburg virus and three species of Ebola virus" *J Virol*
12. Falzarano, Feldmann, Grolla (2011) "Single immunization with a monovalent vesicular stomatitis virus-based vaccine protects nonhuman primates against heterologous challenge with Bundibugyo Ebolavirus" *J Infect Dis*
13. Mire, Geisbert, Marzi et al. (2013) "Vesicular stomatitis virus-based vaccines protect nonhuman primates against Bundibugyo Ebolavirus" *PLoS Negl Trop Dis*
14. Marzi, Robertson, Haddock (2015) "EBOLA vaccine. VSV-EBOV rapidly protects macaques against infection with the 2014/15 Ebola virus outbreak strain" *Science*
15. Marzi, Jankeel, Menicucci (2021) "Single dose of a VSV-based vaccine rapidly protects macaques from Marburg virus disease" *Front Immunol*
16. Furuyama, Marzi (2020) "Development of an enzyme-linked immunosorbent assay to determine the expression dynamics of Ebola virus soluble glycoprotein during infection" *Microorganisms*
17. Marzi, Engelmann, Feldmann (2013) "Antibodies are necessary for rVSV/ZEBOV-GPmediated protection against lethal Ebola virus challenge in nonhuman primates" *Proc Natl Acad Sci*
18. Marzi, Menicucci, Engelmann (2018) "Protection against Marburg virus using a recombinant VSV-vaccine depends on T and B cell activation" *Front Immunol*
19. Gunn, Mcnamara, Wood (2023) "Antibodies against the Ebola virus soluble glycoprotein are associated with long-term vaccine-mediated protection of non-human primates" *Cell Rep*
20. Furuyama, Marzi, Carmody (2016) "Fcgamma-receptor IIa-mediated Src signaling pathway is essential for the antibody-dependent enhancement of Ebola virus infection" *PLoS Pathog*
21. Kuzmina, Younan, Gilchuk (2018) "Antibody-dependent enhancement of Ebola virus infection by human antibodies isolated from survivors" *Cell Rep*
22. Menicucci, Sureshchandra, Marzi et al. (2017) "Transcriptomic analysis reveals a previously unknown role for CD8 + T-cells in rVSV-EBOV mediated protection" *Sci Rep*
23. Marzi, Feldmann (2024) "Filovirus vaccines as a response paradigm for emerging infectious diseases" *NPJ Vaccines*
24. Iavi (2025) "Preclinical immunogenicity and efficacy of a vesicular stomatitis virus-based Sudan virus vaccine and an update on its performance in a phase 1 clinical trial"
25. Geisbert, Daddario-Dicaprio, Williams (2008) "Recombinant vesicular stomatitis virus vector mediates postexposure protection against Sudan Ebola hemorrhagic fever in nonhuman primates" *J Virol*
26. Cao, He, Liu (2023) "The rVSV-EBOV vaccine provides limited cross-protection against Sudan virus in guinea pigs" *Vaccines*
27. Geisbert, Strong, Feldmann (2015) "Considerations in the use of nonhuman primate models of Ebola virus and Marburg virus infection" *J Infect Dis*
28. Finch, Dowling, King (2022) "Bridging animal and human data in pursuit of vaccine licensure" *Vaccines*
29. Pantoja, Pérez-Guzmán, Rodríguez (2017) "Zika virus pathogenesis in rhesus macaques is unaffected by pre-existing immunity to dengue virus" *Nat Commun*
30. Furuyama, Nanbo, Maruyama et al. (2020) "A complement component C1qmediated mechanism of antibody-dependent enhancement of Ebola virus infection" *PLoS Negl Trop Dis*
31. Eisfeld, Halfmann, Wendler (2017) "Multi-platform 'omics analysis of human Ebola virus disease pathogenesis" *Cell Host Microbe*
32. Van Tol, Fletcher, Feldmann (2024) "A bivalent adenovirus-vectored vaccine induces a robust humoral response, but does not protect cynomolgus macaques against a lethal challenge with Sudan virus" *J Infect Dis*
33. Liu, Speranza, Muñoz-Fontela (2017) "Transcriptomic signatures differentiate survival from fatal outcomes in humans infected with Ebola virus" *Genome Biol*
34. Gunn, Yu, Karim (2018) "A role for Fc function in therapeutic monoclonal antibody-mediated protection against Ebola virus" *Cell Host Microbe*
35. Gunn, Lu, Slein (2021) "A Fc engineering approach to define functional humoral correlates of immunity against Ebola virus" *Immunity*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548401&blobtype=pdf
|
# Respiratory syncytial virus phosphoprotein has NTPase and helicase-like activities
Ting Shu, Xiaotong Wang, Yiyang Li, Jiaoling Su, Haiwu Zhou, Mengyu Hu, Puyu Yang, Chao Shan, Yang Qiu, Xi Zhou
## Abstract
Respiratory syncytial virus (RSV), a non-segmented, negative-sense RNA virus (NNSV) in the family Pneumoviridae, represents a significant global health burden causing severe lower respiratory tract infections in infants and immunocompromised adults. While RNA helicases are essential for viral replication through their RNA remod eling functions, the presence of such enzymatic activities in RSV remains unclear.Here, we reveal that the RSV phosphoprotein (P), despite lacking canonical helicase motifs, demonstrates nucleoside triphosphatase (NTPase) activity and directional 5′-to-3′ RNA helix-unwinding capacity in an NTP-dependent manner. Through mutagenesis assays, we establish a functional coupling between NTP hydrolysis and helicase-like activity. Importantly, reverse genetics experiments, RSV minigenome, and antiviral-effect assays demonstrate the essentiality of RSV P's helicase-like activity for viral viability and replication. These findings identify P as an enzymatic component critical for RSV replication, providing new insights into the mechanisms of pneumovirus propagation.IMPORTANCE RNA helicases and helicase-like viral proteins are crucial for viral RNA replication and are prime targets for antiviral development. RSV infects nearly all children by age two, causing over 30 million acute lower respiratory infections, 3.6 million hospitalizations, and 100,000 deaths annually in children under five, while also posing a significant threat to immunocompromised adults and the elderly. In this study, we demonstrate for the first time that the RSV P has NTPase activity and unwinds RNA helices in an NTP-dependent manner. Mutagenesis and reverse genetics experiments confirm that these enzymatic activities are essential for RSV viability. These findings not only redefine RSV P as a multifunctional protein but also expand our understanding of the RSV replication machinery, highlighting the potential of targeting P for antiviral therapy.
The RSV genome spans approximately 15.2 kilobases (kb) and encodes 11 proteins through 10 genes (1). Among these, the nucleoprotein (N), phosphoprotein (P), and large polymerase protein (L) form the minimal replication complex essential for viral propagation. The N protein encapsidates viral RNAs to form a ribonucleoprotein (RNP) complex, shielding it from host innate immune sensors and nucleases. In contrast, P protein serves as a critical cofactor for the L polymerase, anchoring it to the N-RNA template while preventing premature binding of nascent N to host RNAs (3)(4)(5).
RNA helicases, a class of enzymes conserved across diverse RNA viruses, are indispensable for viral RNA replication and transcription. These enzymes utilize nucleoside triphosphate (NTP) hydrolysis to unwind structured RNA duplexes, resolve secondary structures, and separate genomic strands during replication (6)(7)(8). Currently, a wide range of RNA viruses have been reported to encode viral proteins with RNA helicase or helicase-like activities, including picornavirus 2C (9), enterovirus 71 2C (10), norovirus NS3 (11), flavivirus NS3 (12,13), alphavirus nsP2 (14), coronavirus Nsp13 (15,16), cypovirus VP5 (17), and filovirus VP35 (18). These RNA helicase or helicase-like viral proteins play pivotal roles in the viral RNA replication and therefore are recognized to be ideal targets for developing antivirals (19)(20)(21)(22). However, since bioinformatic analyses of RSV-encoded proteins reveal no conserved canonical NTPase/helicase motifs, it is still unknown whether RSV or any member in the family Pneumoviridae encodes proteins with RNA helicase or helicase-like activity, representing a critical gap in understanding RSV replication strategies and underscoring the need to re-evaluate the functional repertoire of RSV proteins.
Our previous study reported that VP35 encoded by Ebola virus (EBOV), a highrisk pathogenic virus belonging to NNSV, contains RNA helicase-like activity that can hydrolyze all types of NTPs and unwind RNA helices in an NTP-dependent manner (18). Interestingly, EBOV VP35 also has no conserved canonical NTPase/helicase motifs. Although EBOV VP35 is different from other NNSV-encoded P proteins, they are considered functional analogs in NNSVs due to conserved functional modules and roles in viral replication (20,21). Therefore, it would be intriguing to exploit whether RSV P also has a helicase-like activity similar to that of EBOV VP35.
In the current study, we demonstrate, for the first time, that RSV P exhibits intrinsic NTPase activity and unwinds RNA helices in a 5′-to-3′ direction, dependent on NTP hydrolysis. Crucially, mutagenesis experiments confirm that the NTPase function is indispensable for the helicase-like activity of RSV P, while reverse genetics, RSV mini genome, and antiviral-effect assays reveal that disrupting these enzymatic properties severely compromises RSV viability. These findings not only redefine RSV P as a multifunctional protein but also extend existing paradigms of RSV replication machinery.
## RESULTS
## RSV P has NTPase activity
To investigate whether RSV P has RNA helicase-like activity, we first examined its NTPase activity, as the ability to hydrolyze NTPs is a common feature of conserved viral RNA helicases. To this end, we expressed and purified RSV P fused with maltose-binding protein (MBP) at its N-terminal (MBP-P) using a baculovirus expression system (Fig. S1) and tested the NTPase activity of MBP-P via using a canonical colorimetric assay. In this assay, the activity of hydrolyzing NTPs (ATP, GTP, UTP, and CTP) was measured by the total amounts of free orthophosphate released after NTP was hydrolyzed. As shown in Fig. 1A, we found that RSV P can efficiently hydrolyze ATP, GTP, and UTP, with a preference for GTP. Since ATP is the common hydrolytic substrates in colorimetric assay, we used it in the subsequent assays. We showed that MBP-P can hydrolyze ATP in a dose-dependent manner (Fig. 1B). Moreover, we then tested the optimal biochemical reaction conditions for RSV P, including different salt content and Mg 2+ concentrations. We found that 1 mM Mg 2+ or Zn 2+ can support the ATPase activity of MBP-P (Fig. 1C). And the ATPase activity of MBP-P was dependent on divalent metal ions, as its ATP-hydrolyzing activity was undetectable in the absence of Mg 2+ (0 mM). MBP-P reached the highest ATPase activity at a concentration of 0.7 mM Mg 2+ , while higher concentrations of Mg 2+ showed certain inhibitory effects on the ATPase activity (Fig. 1D).
Taken together, our data indicated that RSV P has NTPase activity that is dependent on the presence of certain divalent metallic ions.
## RSV P contains dsRNA-binding activity and RNA helix-unwinding activity
After identifying that RSV P has NTPase activity, we then examined whether it possesses dsRNA-binding activity via incubating MBP-P with digoxin (DIG)-labeled dsRNA derived from 1-to 200-nt egfp ORF. Our results showed that MBP-P can bind to dsRNA in a dose-dependent manner, while the negative control (MBP alone) does not exhibit dsRNA-binding activity (Fig. 2A andB).
Because RSV P possesses NTPase and dsRNA-binding activity, we then examined whether it has RNA helix-unwinding activity using a classical helix-unwinding assay as described previously (18). In this assay, a hexachlorofluorescein (HEX)-labeled RNA helix substrate with both 5′ and 3′ single-strand protrusions was constructed (Fig. 3A; Table S2) and subjected to incubation with MBP-P in the standard unwinding reaction containing ATP and Mg 2+ and then separated the substrate via gel electrophoresis. As shown in Fig. 3B, the HEX-labeled RNA strand was efficiently released from the RNA helix substrate in the presence of MBP-P (lane 5), while the same substrate remained stable when MBP alone was added in the reaction as negative control (lane 3). The boiled reaction mixture (lane 2) and the addition of MBP-fusion EBOV VP35 (MBP-VP35, lane 4), a well-characterized viral RNA helicase-like protein, were used as positive controls. We further examined the helix-unwinding activity of MBP-P under four types of NTPs. The results showed that MBP-P can efficiently unwind the RNA helix substrate in the presence of ATP and GTP, whereas its RNA helix-unwinding efficiency was relatively weak under UTP and CTP (Fig. 3C). Moreover, we found that the helix-unwinding activity of MBP-P can be stimulated by ATP in a dose-dependent manner (Fig. 3D).
We further characterized the roles of divalent metallic ions in RNA helix-unwinding of RSV P. From the results, we found that Mg 2+ and Mn 2+ can support MBP-P in unwinding RNA helix substrate, especially Mg 2+ (Fig. 3E). In addition, the RNA helix unwinding of RSV P required the presence of divalent metallic ions, as no RNA strand can be released by MBP-P in the absence of Mg 2+ (Fig. 3F, lane 3). Moreover, MBP-P exhibited optimal helixunwinding activity in the presence of 0.7 mM Mg 2+ , consistent with the previous results (Fig. 1D).
Taken together, our results demonstrated that RSV P has dsRNA-binding activity and RNA helix-unwinding activity. The RNA helix-unwinding activity of RSV P is dependent on ATP and divalent metal ions.
## Characterization of the RNA helix-unwinding activity of RSV P
After identifying that RSV P has RNA helix-unwinding activity, we then sought to determine its helix-unwinding directionality, as the directionality of helix-unwinding is a fundamental characteristic for helicases (16,18). We generated three different RNA helix substrates: one containing a 3′ single-strand protrusion (Fig. 4A), one containing a 5′ single-strand protrusion (Fig. 4B), and the last one with blunt ends (Fig. 4C). Then, MBP-P was reacted with these three different RNA helices using the standard helix-unwinding assay. Interestingly, our data showed that MBP-P could unwind both 3′-tailed and 5′tailed RNA helix (Fig. 4D andE), but not the blunt-ended one (Fig. 4F). Furthermore, MBP-P unwound 5′-tailed RNA helix more efficiently than the 3′-tailed one (Fig. 4E lane 3 and 4D lane 3). Moreover, the increasing concentrations of ATP failed to promote the unwinding rate of 3′-tailed RNA helix (Fig. 4G), but efficiently promoted the unwinding activity of RSV P with 5′-tailed RNA helix in a dose-dependent manner (Fig. 4H). Together, our findings showed that the RNA helix-unwinding activity of RSV P tends to be in the 5′to-3′ direction. We sought to examine whether RSV P can also unwind DNA helices or RNA-DNA hybrids. To this end, we constructed four different nucleic acid helix substrates: RNA helix (R*/R) with RNA strand protrusions, DNA helix (D*/D) with DNA strand protrusions, and RNA-DNA hybrids (D/R* or R*/D) with RNA or DNA strand protrusions (Fig. 5A through D, upper panel). Our results showed that MBP-P could unwind both R*/R (Fig. 5A) and D/R* (Fig. 5B), which have longer RNA strand protrusions. On the other hand, MBP-P could not unwind substrates with longer DNA strands, including D*/D (Fig. 5C) and R*/D (Fig. 5D).
Together, our findings indicate that RSV P requires the presence of protruded singlestranded RNA to unwind nucleic acids from 5′-to-3′ direction.
## The NTPase activity is required for the helix unwinding of RSV P
The interaction between P and N is essential for RSV replication (5). The N-terminal 1-20 amino acids (a.a.) of the P contain multiple residues critical for N-P interaction, and deletion of residues 220-241 at the C-terminus of P abolishes its binding to N (23,24). We sought to determine the key regions responsible for RNA helix-unwinding activity of RSV P. Given that P-N binding is essential for viral replication, we deliberately avoided these regions and generated a series of deletion mutations (25) (as illustrated in Fig. 6A; Fig. S2A) and found that RSV P mutant with the deletion of amino acids 21-30 (△21-30) failed to hydrolyze ATP, indicating that this region is important for the ATPase activity of RSV P. The a.a. GK is the conserved NTPase active-site signature of the classical superfam ily 3 (SF3) helicases (10,11). Based on this result, we further generated the point mutant (KGK/AAA) containing three crucial sites within a.a. 21-30 (K25, G26, and K27) being mutated to alanine. We found that the KGK/AAA mutant also completely lost its ATPase activity like △21-30 (Fig. 6B). Furthermore, we examined the dsRNA-binding and helixunwinding activities of the KGK/AAA mutant and found that, although the KGK/AAA mutant retained its dsRNA-binding activity (Fig. 6C), it completely lost the RNA helixunwinding activity (Fig. 6D; Fig. S2B). Our results indicate that NTPase activity is required for the helix-unwinding activity of RSV P.
## The RNA helicase activity of P is critical for the viability and replication of RSV
To determine whether the loss of the helicase activity of P has any consequence on the viral life cycle of RSV, we introduced the KGK/AAA mutation, which disrupts the NTPase and helicase activities, into the P coding region of the infectious clone of the recombi nant full-length cDNA encoding RSV A2. Then, WT and mutant RSV cDNA were transfec ted into BSRT7/9 cells. Virus production was detected via immunofluorescence for the gglycoprotein (GP) in cells. Strikingly, compared to the WT virus, the KGK/AAA mutation resulted in the loss of RSV viability (Fig. 7A), suggesting that the NTPase and the RNA helicase activities of P are critical for RSV viability. In addition, we generated mutant plasmids and confirmed that the KGK mutation does not affect P protein expression (Fig. 7B) and N-P interaction (Fig. 7C). Building on these findings, we used the mutant plasmids to assess the impact of P and the KGK mutation on RSV minigenome activity and viral replication. We found that the KGK mutation markedly inhibited both minige nome replication (Fig. 7D) and infectious virus production (Fig. 7E), suggesting that the NTPase and the RNA helicase activities of P are critical for RSV replication.
## DISCUSSION
The RNA remodeling proteins, including RNA helicases and RNA chaperones, which can promote RNA molecule folding and refolding, are generally thought to play important roles in RNA metabolism (7). Currently, numerous RNA viruses have been found to encode their own RNA helicase and/or RNA chaperone (10,11,16,18). In this study, we report for the first time that RSV-encoded protein P has NTPase and dsRNA-binding activities, along with RNA-helicase-like activity that efficiently unwinds RNA helix from 5′-to-3′ direction in the presence of NTP and divalent metallic ions, especially ATP and Mg 2+ . In addition, we also identified the key points P for its NTPase activities, and the NTPase activity is required for the helix-unwinding activity of RSV P. Notably, the RNA helicase activity of P is critical For the viability of RSV. Our current study has added RSV P to this growing list of virus-encoded RNA remodeling proteins. RSV P plays critical roles in regulating RNA replication and transcription. It is intriguing to ask how the helicase activity of P functions in the life cycle of RSV. The RSV replication complex is the functional unit for viral transcription and replication, comprising the N, L, P, M2-1, and the viral RNA. During transcription and replication, the viral RNA associates with N to form N-RNA, which serves as a template for the P-bound L polymerase complex (P-L). The attachment of P-L to N-RNA occurs through the interaction between P and N proteins. Additionally, P is thought to travel along the RNP together with L, facilitating the polymerase's translocation. In viral transcription, the interaction between P and M2-1 (P-M2-1) is essential for M2-1 to enhance L's elongation and termination activities on the viral RNA. Although the RSV genome is a single-stranded negative-sense RNA, it has been demonstrated to generate dsRNA replication intermediates that trigger innate immunity (26,27). P is an essential polymerase cofactor that tethers L, the RNA-depend ent RNA polymerase, to the RNA replication complex (3). P also serves as a chaperone that associates with newly synthesized nascent N (28). Our dsRNA-binding assays suggest that the P may directly recognize RSV replication intermediates. Therefore, it is postulated that P can work together with L and N to mediate the unwinding of dsRNA replicative intermediates and facilitate the correct folding or refolding of the RSV mRNA and genomic RNA, thereby promoting the transcription, translation, and encapsidation of RSV.
Previous studies have shown that RSV infection activates TLR3 and RIG-I, which sense viral dsRNA or dsRNA-containing replication intermediates (26,27). RSV P possesses canonical helicase activity and functions as a dsRNA-binding protein that can both bind and unwind dsRNA. This dual activity raises a provocative question: can a virus-encoded helicase subvert host immunity by destabilizing viral dsRNAs, thereby preventing them from being sensed by RNA sensors? Notably, RIG-I itself harbors an RNA-helicase domain, whose unwinding activity is critical for sensing viral RNA and igniting innate signaling. We therefore speculate that P antagonizes RIG-I by competing for or directly interfering with its helicase domain. Consequently, RSV P appears to be a bifunctional helicase: it simultaneously accelerates viral replication by resolving replication intermediates and dampens innate immunity by destabilizing viral dsRNA that would otherwise trigger robust antiviral responses.
RNA helicases are classified into six superfamilies, termed SF1 to SF6, according to the conserved motifs. Interestingly, RSV P does not have any traditional helicase conserved motif, but it contains the fundamental biochemical characteristics of canonical RNA helicases, including the dependence on NTP and divalent metal ions, as well as the directionality of RNA helix unwinding. Consistently, our previous study also uncovered that EBOV VP35 is an untraditional RNA helicase-like protein without any conventional helicase motif (18). Our findings suggest that some proteins associated with RNA helicase activities may not be strictly dependent on the conserved motifs in the linear amino acid sequence, but are probably functionally determined by more complex regions/sites formed by protein higher-order structures.
In summary, our work provides the first demonstration of the NTPase and helicaselike activities associated with RSV P. These findings uncover novel functions of RSV P, extend the view of RNA remodeling proteins, and shed light on the understanding of RSV replication. Furthermore, elucidating the helicase-like activity of RSV P carries profound implications for antiviral development. As a central coordinator of viral transcription and replication, P represents an attractive therapeutic target. The discovery of its enzymatic functions opens new avenues for structure-guided drug design, particularly for small molecules inhibiting NTPase/helicase activity. Furthermore, this work provides a framework for re-evaluating accessory proteins in related viruses, potentially uncover ing conserved mechanisms of RNA metabolism. By bridging molecular virology and enzymology, our findings advance the fundamental understanding of RSV biology while offering tangible strategies to combat this pervasive pathogen.
## MATERIALS AND METHODS
## Plasmid and recombinant baculovirus construction
The construction of pFastBac HTB-MBP and pFastBac HTB-MBP-RSV P and pFastBac HTB-MBP-EBOV VP35 has been described previously (21). The cDNA fragments of RSV P (GenBank accession no. MK816924.1) and EBOV VP35 (GenBank accession no. AF086833.2) were cloned into the vector pFastBac HTB-MBP, where the maltose binding protein (MBP) was fused to the N-terminus. The mutations were conducted as previously described (29). The resulting plasmid was subjected to Bac-to-Bac baculovirus system to express the recombinant protein. The primers used in this study are shown in Table S1.
## Expression and purification of recombinant fusion protein
The expression and purification of recombinant MBP-P and negative control MBP from Bac-to-Bac system were conducted as previously described (9,30). Briefly, SF9 cells were infected with the recombinant baculoviruses and harvested 72 h post-infection. Cell pellets were resuspended, lysed by sonication, and subjected to centrifugation for 15 min at 12,000 × g to remove debris. The protein in the supernatant was purified using amylose affinity chromatography (New England BioLabs, Ipswich, MA), according to the manufacturer's protocol. Then the protein was concentrated using Ultra-15 filters (Millipore, Schwalbach, Germany), and the store buffer was exchanged to 50 mM 2-[4-(2-hydroxyethyl)-1-piperaziny] ethanesulfonic acid (HEPES)-KOH (pH 7.5). All proteins were quantified using the Bradford method and stored at -80℃ in aliquots. Proteins were separated by 10% SDS-PAGE and visualized with Coomassie blue.
## Preparation of oligonucleotide helix substrates
In brief, the RNA helix, DNA helix, and RNA-DNA hybrid helix were prepared by annealing two complementary nucleic acid strands. One strand was labeled at 5′ end with hexachloro-fluorescein (HEX), and the other strand was unlabeled. HEX-labeled oligonucleotide strands were purchased from TaKaRa (Dalian, China). Unlabeled DNA strands were synthesized by Invitrogen, and unlabeled RNA strands were in vitro transcribed using T7 RNA polymerase (Promega, Madison, WI). The transcribed RNA strands were purified by Poly-Gel RNA Extraction Kit (Omega Bio-Tek, Guangzhou, China) according to the manufacturer's instructions. The two strands were mixed in a proper ratio and annealed through heating and gradually cooling as previously described (9,17). The resulting duplexes were examined by 15% native-PAGE gel to ensure that all single-stranded RNA and DNA were annealed in a 1:1 ratio.
The standard RNA helix substrate (R/R*) was annealed with a 42-nt HEX-labeled single-stranded RNA1 and a 24-nt unlabeled single-stranded RNA2. The 3′-tailed and 5′-tailed RNA substrates were constructed by annealing RNA1 with a 48-nt non-labeled single-stranded RNA3 (3′-tailed) and RNA4 (5′-tailed), respectively. The blunt-ended RNA substrate was prepared by annealing RNA1 with a 42-nt non-labeled single-stranded RNA5. (D*/D) was prepared by annealing a 28-nt HEX-labeled single-stranded DNA1 with a 49-nt non-labeled single-stranded DNA2. (D/R*) was prepared by annealing RNA1 with a 30-nt non-labeled single-stranded DNA3. (R*/D) was prepared by annealing RNA1 with a 54-nt non-labeled single-stranded DNA4. All oligonucleotides used in this study are listed in Table S2.
## NTPase assay
NTPase activities were determined by measuring the inorganic phosphate released during NTP hydrolysis using a direct colorimetric assay, following the standard proce dure as previously described (9). All of the results obtained from this quantitative assay represent the average of three repeated experiments.
## Gel mobility shift assay
Gel mobility shift assay was performed in 50 mM HEPES-KOH (pH 7.5), 100 mM NaCl, 2 mM MgCl 2 , 1 mM tRNA, 2 mM DTT, and 20 U RNase inhibitor (Promega) in a total volume of 10 µL reaction, with the indicated amount of protein and 0.1 pmol of dsRNA. The dsRNA was labeled with DIG-UTP (Roche) by in vitro transcription and derived from 200 nt EGFP. Reactions were incubated for 30 min at 25°C and terminated by the addition of 2.5 µL of 5 × sample buffer (20 mM Tris-HCl [pH 7.5], 36% glycerol, and 0.1% bromophenol blue). The nucleic acid-protein complexes were separated by electropho resis on a 1.5% protein agarose gel and transferred to Hybond-A nylon membrane (GE Healthcare). After that, the membrane was subjected to cross-linking at 120°C and was incubated with anti-DIG-alkaline phosphatase antibody (Roche), followed by incubation with CDP-STAR (Roche) for 15 min at 37°C. The signals were then detected by X-ray film (Fujifilm, Tokyo, Japan).
## Nucleic acid helix-unwinding assay
The standard helix destabilizing assay was performed as previously described (17). Briefly, 20 pmol of recombinant protein and 0.1 pmol of HEX-labeled helix substrate were added to a mixture containing 50 mM HEPES-KOH (pH 7.5), 1 mM MgCl 2 , 100 mM NaCl, and 20 U RNase inhibitor (Promega). Reactions were incubated for 60 min at 37°C and terminated by adding 5 × loading buffer (100 mM Tris-HCl, 1% SDS, 50% glycerol, and bromophenol blue [pH 7.5]). The mixtures were then electrophoresed on 15% native-PAGE gels (15% acrylamide:bis [
## Construction and recovery of RSV mutant virus
To construct the plasmid for the RSV A2 strain P protein (KGK/AAA mutation), we used wild-type RSV A2 as the template. With the pCC1 vector as the backbone, we sequentially inserted a T7 promoter, three additional guanine (ggg) bases to enhance viral expres sion, the A2 strain genome sequence (with the KGK/AAA mutation introduced in the P gene), HDV ribozyme, and a T7 terminator to obtain the mutated viral cDNA. Mean while, we constructed four auxiliary plasmids encoding the N, P, L, and M2-1 proteins using the eukaryotic expression vector pCDNA3.1, resulting in pCDNA3.1-N, pCDNA3.1-P, pCDNA3.1-M2-1, and pVSV-RSV-L.
For recovery of the RSV WT and mutant (KGK/AAA) virus, BSRT7/9 cells were cultured overnight in six-well plates the day before transfection. When the cell density reached approximately 70%, the cells were transfected with the PCC1-RSV WT or KGK/AAA plasmid, along with the four helper plasmids (pCDNA3.1-N, pCDNA3.1-P, pCDNA3.1-M2-1, and pVSV-RSV-L), according to the Lipofectamine 3000 reagent manual. The transfection mixture included PCC1-RSV (WT or KGK/AAA) (1.25 µg), pCDNA3.1-N (1 µg), pCDNA3.1-P (1 µg), pCDNA3.1-M2-1 (0.25 µg), and pVSV-RSV-L (0.5 µg). After incubating the DNA-Lipofectamine-OptiMEM (Gibco) mixture at room temperature for 30 min, it was added to each well of cells. Six to 8 h post-transfection, the medium was replaced with 2% FBS DMEM. The cells were incubated at 37°C in a 5% CO₂ incubator, and cytopathic effects were observed daily. For blind passage, 4 days post-transfection, the supernatant was collected and centrifuged at 4500 × g for 15 min. The supernatant was then used to infect BSRT7/9 cells to increase the viral titer. The virus was subsequently transferred to Hep-2 cells for further passage and amplification.
## Immunofluorescence assay
Hep2 cells were plated in Glass Bottom Culture Dishes (NEST, 801002). After 24 h, cells were infected with WT or mutant RSV A2 for 96 h. The cells were then fixed with 4% paraformaldehyde for 45 min and permeabilized with 0.2% Triton X-100 for 20 min. After that, the cells were blocked with phosphate-buffered saline (PBS) containing 5% bovine serum albumin (BSA) for 1 h and then incubated with anti-RSV-GP antibody (1:1,000 in 5% BSA) overnight, followed by staining with FITC-labeled goat anti-mouse IgG (ABclonal) (1:1,000 in 5% BSA). Nuclei were stained with 49,6-diamidino-2-phenylin dole (DAPI; Beyotime) for 5 min at 37°C in the dark. Cells were photographed under a confocal microscope (A1R; Nikon, Japan). The fluorescence intensity was analyzed using ImageJ software.
## Western blotting and antibodies
Cells were washed twice with cold PBS and lysed in lysis buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% NP40, 0.25% deoxycholate and a protease inhibitor cocktail [MCE]). The lysates were subjected to 10% SDS-PAGE gels and transferred to polyvinylidene fluoride membranes (Bio-Rad). The blots were incubated with primary antibodies in 5% nonfat milk in TBS with 0.05% Tween 20 (TBST) overnight at 4°C. Membranes were washed in TBST and incubated in HRP-coupled secondary antibodies at room temperature for 1 h. Proteins were detected by chemiluminescence using ECL (Bio-Rad) in a Bio-Rad ChemiDoc Imager and were quantified using Image J software. The antibodies used for Western blots are as follows: the anti-HA and anti-FLAG antibodies were purchased from CST, and the anti-Tubulin and anti-GAPDH were purchased from Proteintech.
## Co-Immunoprecipitation
For the co-immunoprecipitation (co-IP) assay, cells were harvested and then lysed with IP buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 10% glycerol, 2 mM EDTA, 0.5% Nonidet P-40, 0.5% Triton X-100) with 1% protease and phosphatase inhibitor (MCE). The insoluble component was removed by centrifugation at 12,000 × g for 10 min at 4°C, and the supernatant was collected. For each sample, 600 µL protein lysate was incubated with 1 µg antibody and 30 µL protein A/G magnetic beads (MCE) overnight at 4°C. The beads were washed five times with 1 mL of IP buffer, and then the precipitates were detected by Western blotting.
## RSV minigenome assay
BHK-21 cells were seeded in 24-well plates. After the cells have completely adhered, infect with vaccinia virus vTF7-3 and perform transfection 1 h post-infection. Cells were co-transfected with 31.25 ng pGEM4-RSV N, 62.5 ng pGEM4-RSV P, 15.625 ng pGEM4-RSV M2-1, 7.8125 ng pGEM4-RSV L, 31.25 ng pGEM4-RSV Minigenome (Firefly luciferase), and 5 ng pRL-TK (Renilla luciferase). The thymidine kinase promoter-driven Renilla lucifer ase plasmid (TK) was included as a transfection control. The minigenome system was driven by vTF7-3, which expresses T7 RNA polymerase. At 24 h post-transfection, cells were lysed with Passive Lysis Buffer (Promega) at room temperature for 15 min. RSV minigenome activity was quantified using the Dual-Luciferase Reporter Assay System (Promega).
## References
1. Duan, Liu, Zang et al. (2024) "Landscape of respiratory syncytial virus" *Chin Med J (Engl)*
2. Munro, Martinón-Torres, Drysdale et al. (2023) "The disease burden of respiratory syncytial virus in Infants" *Curr Opin Infect Dis*
3. Asenjo, González-Armas, Villanueva (2008) "Phosphorylation of human respiratory syncytial virus P protein at serine 54 regulates viral uncoating" *Virology (Auckl)*
4. Grosfeld, Hill, Collins (1995) "RNA replication by respiratory syncytial virus (RSV) is directed by the N, P, and L proteins; transcription also occurs under these conditions but requires RSV superinfection for efficient synthesis of full-length mRNA" *J Virol*
5. Yu, Hardy, Wertz (1995) "Functional cDNA clones of the human respiratory syncytial (RS) virus N, P, and L proteins support replication of RS virus genomic RNA analogs and define minimal trans-acting requirements for RNA replication" *J Virol*
7. Bleichert, Baserga (2007) "The long unwinding road of RNA helicases" *Mol Cell*
8. Musier-Forsyth (2010) "RNA remodeling by chaperones and helicases" *RNA Biol*
9. Rajkowitsch, Chen, Stampfl et al. (2007) "RNA chaperones, RNA annealers and RNA helicases" *RNA Biol*
10. Cheng, Yang, Xia et al. (2013) "The nonstructural protein 2C of a picorna-like virus displays nucleic acid helix destabilizing activity that can be functionally separated from its ATPase activity" *J Virol*
12. Xia, Wang, Wang et al. (2015) "Human enterovirus nonstructural protein 2CATPase functions as both an RNA helicase and ATP-independent RNA chaperone" *PLoS Pathog*
13. Li, Hosmillo, Schwanke et al. (2018) "Human norovirus NS3 has RNA helicase and chaperoning activities" *J Virol*
14. Jain, Coloma, García-Sastre et al. (2016) "Structure of the NS3 helicase from Zika virus" *Nat Struct Mol Biol*
15. Lam, Keeney, Eckert et al. (2003) "Hepatitis C virus NS3 ATPases/helicases from different genotypes exhibit variations in enzymatic properties" *J Virol*
16. Karpe, Aher, Lole (2011) "NTPase and 5'-RNA triphosphatase activities of Chikungunya virus nsP2 protein" *PLoS One*
17. Lee, Kwon, Park et al. (2010) "Cooperative translocation enhances the unwinding of duplex DNA by SARS coronavirus helicase nsP13" *Nucleic Acids Res*
18. Shu, Huang, Wu et al. (2020) "SARS-coronavirus-2 nsp13 possesses NTPase and RNA helicase activities that can be inhibited by bismuth salts" *Virol Sin*
19. Yang, Cheng, Zhang et al. (2014) "A cypovirus VP5 displays the RNA chaperonelike activity that destabilizes RNA helices and accelerates strand Full-Length Text Journal of Virology October"
20. *Nucleic Acids Res*
21. Shu, Gan, Bai et al. (2019) "Ebola virus VP35 has novel NTPase and helicaselike activities" *Nucleic Acids Res*
22. Pfister, Wimmer (1999) "Characterization of the nucleoside triphos phatase activity of poliovirus protein 2C reveals a mechanism by which guanidine inhibits poliovirus replication" *J Biol Chem*
23. Habchi, Mamelli, Darbon et al. (2010) "Structural disorder within Henipavirus nucleoprotein and phosphoprotein: from predictions to experimental assessment" *PLoS One*
24. Karlin, Ferron, Canard et al. (2003) "Structural disorder and modular organization in Paramyxovirinae N and P" *J Gen Virol*
25. Kadaré, Haenni (1997) "Virus-encoded RNA helicases" *J Virol*
26. Galloux, Gabiane, Sourimant et al. (2015) "Identification and characterization of the binding site of the respiratory syncytial virus phosphoprotein to RNA-free nucleoprotein" *J Virol*
27. Mason, Aberg, Lawetz et al. (2003) "Interaction between human respiratory syncytial virus (RSV) M2-1 and P proteins is required for reconstitution of M2-1-dependent RSV minigenome activity" *J Virol*
28. Cao, Gao, Chen et al. (2024) "Structures of the promoter-bound respiratory syncytial virus polymerase" *Nature*
29. Groskreutz, Monick, Powers et al. (2006) "Respiratory syncytial virus induces TLR3 protein and protein kinase R, leading to increased double-stranded RNA responsiveness in airway epithelial cells" *J Immunol*
30. Liu, Jamaluddin, Li et al. (2007) "Retinoic acid-inducible gene I mediates early antiviral response and Tolllike receptor 3 expression in respiratory syncytial virus-infected airway epithelial cells" *J Virol*
31. Esneau, Raynal, Roblin et al. (2019) "Biochemical characterization of the respiratory syncytial virus N 0 -P complex in solution" *J Biol Chem*
32. Qi, Cai, Qiu et al. (2011) "RNA binding by a novel helical fold of B2 protein from wuhan nodavirus mediates the suppression of RNA interference and promotes B2 dimerization" *J Virol*
33. Wang, Han, Qiu et al. (2012) "Identification and characterization of RNA duplex unwinding and ATPase activities of an alphatetravirus superfamily 1 helicase" *Virology (Auckl)*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12811238&blobtype=pdf
|
# Author Correction: Transcriptional synergy between Tat and PCAF is dependent on the binding of acetylated Tat to the PCAF bromodomain
Alexander Dorr, Veronique Kiermer, Angelika Pedal, Hans-Richard Rackwitz, Peter Henklein, Ulrich Schubert, Ming-Ming Zhou, Eric Verdin, Melanie Ott
The authors contacted the journal after being made aware of blot re-use within Fig. 5A. The authors were able to locate the original scans and the correct blots. After reviewing the data provided by the authors, the journal retracts and replaces the following figure. Associated original scans are published with this correction.
Author statement.
During assembly of Fig. 5, the researcher provided a guide file, '01.0324E.Grant.pdf,' which contained gel section IDs corresponding to the file names of the linked images. The compositor inadvertently copied the incorrect gel for Fig. 5, 2nd row, 3rd column, using '01-0324D (01-0323P GelSec)' instead of the correct '01-0324D (01-0323Q GelSec).'
The corrigendum does not affect the findings and conclusions of the manuscript.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12826961&blobtype=pdf
|
# Bibliometric analysis of global research output on viral vaccines and antivirals in the 21st century
Lili Ma, Wei Pan, Miaomiao Yan, Jiali Si, Virologica Sinica
## Abstract
Human knowledge of viruses has experienced explosive growth in the 21st century. This leap forward is reflected not only in the deepening of basic research on viruses but also in addressing public health challenges posed by viral diseases. Key advancements include the rapid identification of emerging viruses and variants, the revelation of diverse viromes and evolutionary patterns, the elucidation of viral pathogenesis * Corresponding author.
and antiviral targets, as well as development of novel vaccines and antiviral drugs through far more advanced techniques and pipelines (Holmes et al., 2024). Altogether, these breakthroughs are reshaping our understanding of viruses and our strategies to combat viral infections at an unprecedented pace. To assess the scale and current development status of antiviral vaccine and drug research within the broader landscape of virus studies, we analyzed papers on the "vaccine and antiviral" topic published in SCIEindexed journals between 2000 and 2024. To precisely capture studies directly related to vaccine and antiviral development, we designed a comprehensive but targeted search strategy. The search consists of three separate query combinations, each pairing the topic search (TS) with one of the three fields (Title [TI], Author Keywords [AK], and Keywords Plus [KP]). Specifically: (1): "TS = ("*virus*" or "*viral*") and TI = ("vaccin*" or "antiviral*" or "drug*" or "prodrug" or "derivative*" or "compound*" or "agent*" or "fusion inhibitor*" or "entry inhibitor*" or "nanobod*" or "neutrali* antibod*" or "therap* antibod*"). ( 2): Same as the first, but with "TI" replaced by "AK" (Author Keywords). (3): Same as the first, but with "TI" replaced by "KP" (Keywords Plus). Ensure that the asterisk wildcard appears as a superscript in the query. By combining results from the TI, AK, and KP fields and restricting them to the "Article" type of publication, a total of 191,965 research articles were retrieved from the Web of Science Core Collection SCI-EXPANDED Editions (Web of Science Core Collection, 2025) for subsequent analysis. Derwent data analyzer was used for data processing.
We first analyzed global research output trends from 2000 to 2024. As shown in Fig. 1, the annual number of articles (represented by the purple histogram) generally exhibited steady growth over this 25-year period, starting at around 3500 and gradually rising to 9000 by 2019. A significant surge began in 2020, peaking at 16,000 articles in 2022, followed by a gradual decline to 12,000 in 2024. This sharp increase in 2020 is closely linked to the outbreak of COVID-19. Although the number decreased in 2024 as the pandemic's impact waned, it remained substantially higher than pre-pandemic levels.
To further characterize the major contributing countries, the annual publication counts of the top 10 countries are illustrated in the line charts of Fig. 1. Over these 25 years, the United States (USA) began with approximately 1000 annual publications and maintained the highest output among individual countries, exceeding 4000 during the COVID-19 years. Most other top countries also experienced more than a threefold increase in publications, mirroring the growth trend of the USA. Notably, China and India, two countries that started with the weakest foundations among the top 10 with fewer than 100 annual publications each, achieved particularly robust growth, reaching over 3000 and approximately 1000 articles respectively during the pandemic period.
To contextualize these trends, this study further assesses the proportion of vaccine and antivirals publications within the entire field of virology research. Using the method reported in the precious paper (Ma et al., 2024) To identify virus types that have been the focus of research over the years, we extracted virus types from the entire dataset (191,965 articles) using Author Keywords. Since the initial extraction yielded keywords referring to the same virus but under different terms, such as SARS-CoV-2, 2019-nCoV, and COVID, we manually clustered these synonymous terms. This process enabled us to identify 43 virus types that have received the most research attention, collectively accounting for 97,561 occurrences of author keywords. Fig. 4A illustrates the annual publication trends of 16 viruses that each exceeded 100 articles in at least one historical year, while Fig. 4B presents the remaining 27 viruses. Occurrence frequencies are annotated in parentheses after each virus type. As shown, HIV accounts for the largest cumulative number of publications, while SARS-CoV-2, despite its recent emergence, has been the subject of the most intensive research. Over time, we observe clear shifts in global research focus toward emerging viral pathogens, highlighting how the scientific community responds to combat these viral threats.
We further explored how scientists strategically address challenges posed by these 43 virus types. Using the same method as described in Fig. 3, these viruses were categorized into the vaccine research and antiviral research subfields, with results presented in Fig. 5. We can see distinct patterns on different viruses. Several viruses, such as SARS-CoV-2, influenza virus, and HBV have received substantial research efforts in both areas, with slight biases toward one or the other subfield. In contrast, HIV and HCV research leans more toward therapeutic strategies, while interventions for HPV and rotavirus are predominantly vaccine-focused. Notably, due to a 5% overlap in total publications, a small number of publications were dually classified in the vaccineversus-antiviral breakdown. However, this overlap does not affect the overall findings.
We further dissected the antiviral strategies of these 43 viruses among the top 10 countries. Fig. 6 visualizes the research attention from these countries toward each virus, highlighting national specialization and geographic variations in antiviral research priorities.
This study utilized bibliometric methods to assess global trends in antiviral vaccine and drug research and development, identifying key contributing countries and their temporal variations, uncovering divergent dynamics between vaccine and drug subfields, and exploring strategic preferences for different virus types. These analyses provide a comprehensive overview of global antiviral landscape in the first quarter of the 21st century, revealing research dynamics, regional contributions, and strategic shifts in response to emerging viral threats. We hope this work may offer valuable insights for policymakers, funders, and general readers, facilitating their understanding of researchers' efforts in addressing global antiviral needs.
## References
1. Holmes, Krammer, Goodrum (2024) "Virology-the next fifty years"
2. Ma, Pan, Si (2024) "Bibliometric analysis of virology advancements in the 21st century" *Virol. Sin*
3. (2025) "Web of Science Core Collection SCI-EXPANDED (SCIE)"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724340&blobtype=pdf
|
# Phosphorylation of Shiftless by casein kinase 1 δ/ε is required for its antiviral activity
Yongle Wang, Shaozu Fu, Xinlu Wang, Guangxia Gao
## Abstract
Shiftless (SHFL) is a host antiviral factor that inhibits the viral -1 program med ribosomal frameshifting (-1PRF). How the antiviral activity of SHFL is regulated is not known. Here, we report that the phosphorylation of SHFL by the casein kinase 1 delta (CK1δ) and epsilon (CK1ε) regulates its antiviral activity. The casein kinases were identified as SHFL-interacting proteins. Downregulation of the expression of the endogenous casein kinases or pharmacological inhibition of kinase activity significantly impaired the activity of SHFL to inhibit HIV-1 -1PRF and viral production. The residues T250 and T253 were identified as critical phosphorylation sites; substitution of either of the residues with alanine, resulting in the mutants SHFL-T250A or SHFL-T253A, abolished the antiviral activity of SHFL against HIV-1. The T250A mutation did not affect SHFL interaction with target RNA or the ribosomal proteins uL5 and eS31 but signifi cantly reduced its interaction with IGF2BP proteins. Furthermore, SHFL-T250A displayed reduced association with the actively translating polysomes. These results are consistent with the notion that SHFL interaction with ribosomes is critical for its activity to inhibit -1PRF. Taken together, our results indicate that phosphorylation of SHFL by CK1δ/ε is required for its antiviral activity. IMPORTANCE Innate immune responses and antiviral defense mechanisms are regulated through a variety of mechanisms, among which post-translational modifi cations, especially phosphorylation, play important roles. Deciphering the regulatory mechanisms helps understand the innate immune responses more comprehensively. SHFL is an interferon-stimulated host antiviral factor that inhibits the replication of multiple viruses. The present study revealed that phosphorylation of SHFL by casein kinase 1 δ/ε is required for its antiviral activity against HIV-1. These results provide an additional example for how immune responses are regulated. Furthermore, given that SHFL inhibits -1PRF, the present studies provide tools for further exploring the mechanisms of -1PRF. KEYWORDS antiviral factor, -1 programmed ribosomal frameshifting, casein kinase 1, phosphorylation, Shiftless S hiftless (SHFL) is an interferon-stimulated host antiviral factor that inhibits the replication of multiple viruses. It inhibits the -1 ribosomal frameshifting (-1PRF) of HIV-1 and thus inhibits the replication of the virus (1). In addition to HIV-1, SHFL has been reported to inhibit the -1PRF of other viruses including SARS-CoV-2 and Japanese encephalitis virus (2-4). SHFL has also been reported to inhibit the replication of some viruses that do not have obvious -1PRF ( 5), but the underlying mechanism is not yet elucidated. Recent reports suggest that SHFL also participates in ribosome surveillance (6).-1PRF is a translational recoding mechanism widely used by viruses to expand their genome coding capacity and regulate viral protein expression. The frameshifting occurs
at specific -1PRF site, which typically consist of slippery sequences and downstream RNA high-order structures such as pseudoknots or stem-loops. The translating ribosomes pause at the -1PRF site because of the downstream high-order structure block. Although most ribosomes manage to pass the block to continue translation in the original frame, some ribosomes shift one nucleotide backward at the slippery sequence and continue translation in the -1 frame (7). Precise control of -1PRF efficiency is essential for viral replication (8,9). For instance, HIV-1 uses -1PRF to synthesize Gag (0 frame) and Gag-Pol (-1 frame) with the frameshifting efficiency 5%-10%, resulting in Gag to Gag-Pol ratio to be 10-20:1. Such a controlled ratio is critical for HIV-1 virion assembly and replication (10)(11)(12). SHFL interacts directly with the -1PRF signals of target mRNAs and with ribosomal proteins uL5 and eS31 (1,13). It was proposed that SHFL interaction with the translating ribosomes and -1PRF signals causes ribosome stalling, thus inhibiting frameshifting (1).
Phosphorylation is a post-translational modification mechanism that regulates a variety of biological processes by modulating protein activity, localization, and proteinprotein interactions (14,15). This process is controlled by approximately 568 kinases and 156 phosphatases in human cells (16). Among these kinases, casein kinase 1 (CK1) is a conserved serine/threonine kinase family comprising seven isoforms (α, β, γ1-3, δ, and ε) (17,18). CK1δ and CK1ε, two closely related isoforms with overlapping biological functions, regulate multiple biological processes such as circadian rhythms (19,20), Wnt signaling (21), protein translation (22), and 40S ribosome biogenesis (23). CK1 preferentially phosphorylates substrates that contain a priming phosphate and typically recognizes a consensus sequence of pS/pT-X-X-S*/T*, where X represents any amino acid and S*/T* denotes the target phosphorylation site. Certain substrates can be directly phosphorylated by CK1 without priming (24)(25)(26)(27)(28). It has been reported that CK1-inter acting proteins are prone to be phosphorylated, and such interactions can serve as supporting evidence for the identification of CK1 substrates (27).
Here, we identified CK1δ and CK1ε as SHFL-interacting proteins. SHFL phosphoryla tion by CK1δ/ε is required for its ability to inhibit -1PRF and HIV-1 replication. We identified the T250 and T253 of SHFL as critical phosphorylation sites by CK1δ/ε. The phosphorylation at T250 and T253 is required for its antiviral activity against HIV-1.
## RESULTS
## SHFL interacts with CK1δ/ε
In an attempt to better understand the mechanisms underlying the antiviral activity of SHFL, we set out to use HA-Flag tandem affinity purification coupled with mass spectrometry (TAP-MS) to identify SHFL-interacting proteins (Fig. 1A). HA-Flag-SHFL was expressed in HEK293 cells in a doxycycline-inducible manner. The cell lysates were first treated with the nuclease Benzonase to disrupt RNA-or DNA-mediated protein-pro tein interactions. The proteins were sequentially purified with anti-HA and anti-Flag conjugated beads. A fraction of the purified proteins was subjected to mass spectrom etry analysis and the rest was analyzed by SDS-PAGE. Silver staining analysis revealed multiple distinct protein bands specific to the SHFL sample (Fig. 1B). Mass spectrome try analysis combined with label-free quantification identified multiple SHFL-associated proteins, including RNA-binding proteins, a ribosomal protein, and CK1ε (Fig. 1C). We focused on CK1 in the following studies.
Although CK1δ was not detected in the TAP-MS analysis, we speculated that SHFL may interact with both CK1δ and CK1ε, considering that they share substantial sequence similarity and functional redundancy and that the rigorous TAP procedure might disrupt weak or transient interactions (27,29). Two CK1δ isoforms have been identified, CK1δ and CK1δ2, with CK1δ2 lacking six amino acids at the C-terminus (30). The interactions of CK1ε, CK1δ, and CK1δ2 with SHFL were analyzed by the co-immunoprecipitation (Co-IP) assay. Flag-tagged CK1 isoforms and Myc-tagged SHFL were co-expressed in 293T cells, followed by immunoprecipitation using anti-Flag affinity gel. The nuclease Benzonase was added into the cell lysates to disrupt RNA-or DNA-mediated protein interactions. SHFL co-precipitated with all the CK1 isoforms (Fig. 1D).
To assess the interaction between SHFL and endogenous CK1δ/ε, Flag-tagged SHFL was transiently expressed in 293T cells. Cell lysates were immunoprecipitated with anti-Flag affinity gel and analyzed by western blotting. Both CK1δ and CK1ε co-precipitated with SHFL (Fig. 1E), further demonstrating the interaction between SHFL and CK1δ/ε.
## CK1δ/ε phosphorylates SHFL
The interaction between a kinase and its binding partner often suggests a potential kinase-substrate relationship (31). We reasoned that SHFL may be a phosphorylation substrate of CK1δ and CK1ε. To examine whether CK1δ/ε is involved in SHFL phos phorylation, HEK293 cells transiently expressing Flag-tagged SHFL were treated with increasing concentrations of CK1δ/εspecific inhibitor PF-670462 (32). The phosphoryla tion of SHFL was monitored by its electrophoretic mobility shift on SDS-PAGE. The protein with less phosphorylation is expected to move faster. Indeed, SHFL expressed in the cells treated with the inhibitor moved faster than that expressed in the mocktreated cells (Fig. 2A). To show that the faster-moving SHFL was less phosphorylated, we generated antisera for western analysis that specifically recognize SHFL phosphoryla ted at T250 and T253, which were later identified as critical phosphorylation sites for CK1δ/ε (see below for more information). Consistent with the above mobility shift results, treatment with the inhibitor reduced the phosphorylation of SHFL in a dose-dependent manner (Fig. 2A).
We next examined whether the endogenous SHFL is phosphorylated. The SHFL knockout (KO) or control (NC) THP-1 cells were treated with interferon-alpha to upregulate the expression of the endogenous SHFL (1,33). The cells were treated with DMSO or PF-670462. Since the phosphoSHFLspecific antisera were not sensitive enough to detect the endogenous SHFL in the cell lysate, SHFL was first immunopreci pitated with SHFLspecific polyclonal antibody and then analyzed by western blotting. Treatment with PF-670462 led to increased electrophoretic mobility of SHFL (Fig. 2B). Consistently, phosphorylation of the endogenous SHFL was barely detected in the PF-670462-treated cells (Fig. 2B). These results further show that SHFL is phosphorylated, and the phosphorylation can be suppressed by the CK1δ/ε inhibitor PF-670462.
To test which kinase is involved in the phosphorylation of SHFL, we generated CK1δ and CK1ε knockout HEK293 cells. Myc-tagged SHFL was transiently expressed in these cells and analyzed by western blotting. Knockout of either kinase alone did not alter the electrophoretic mobility of SHFL or its phosphorylation at T250 and T253 (Fig. 2C), suggesting that CK1δ and CK1ε function redundantly. We thus downregulated the expression of both CK1δ and CK1ε in HEK293 cells. Since simultaneous knockout of these two kinases was lethal to the cells, CK1δ was knocked out by the CRISPR/ Cas9 method, and CK1ε was knocked down by the siRNA method. Myc-tagged SHFL was transiently expressed in these cells and analyzed by western blotting. Consistent with the inhibitor treatment results, downregulation of CK1δ/ε rendered SHFL mov ing faster, indicative of reduced phosphorylation. Furthermore, the phosphospecific antisera detected less SHFL phosphorylated at T250 and T253 in the cells in which CK1δ and CK1ε were downregulated. To confirm the specificity of the downregulation, a CK1δ or CK1ε-expressing rescue construct was used, which was not targeted by the sgRNA or siRNA. Expression of the rescue construct restored SHFL migration pattern and phosphorylation level (Fig. 2D). Collectively, these results indicate that CK1δ and CK1ε are required for the phosphorylation of SHFL.
## CK1δ/ε phosphorylation of SHFL is required for its antiviral activity
We next analyzed the function of CK1δ/ε in SHFL inhibition of -1PRF. A well-established HIV-1 dual-luciferase reporter system was employed to assess -1PRF efficiency (1). In this system, the -1PRF signal from HIV-1 was inserted between the coding sequences of Renilla luciferase (Rluc) and Firefly luciferase (Fluc) in the pDual-HIV(-1) reporter, with Fluc in the -1 reading frame. The reporter pDual-HIV(0) was used as a control, in which a nucleotide was added in the slippery sequence such that Rluc and Fluc are in the same reading frame. The -1PRF efficiency was calculated as the Fluc/Rluc ratio from pDual-HIV(-1) divided by that from pDual-HIV(0). To assess the role of CK1δ/ε in SHFL inhibition of -1PRF, we first examined whether the knockout of CK1δ or CK1ε affected SHFL function. HEK293 cells were co-transfected with the HIV-1 dual-luciferase -1PRF reporters and a construct expressing Myc-tagged SHFL, and frameshifting efficiency was measured in the presence or absence of SHFL. Knockout of either kinase alone did not obviously affect SHFL suppression of -1PRF (Fig. 3A), consistent with the above observation that phosphorylation at T250 and T253 was not altered by the knockout of either CK1δ or CK1ε alone. We next analyzed SHFL activity in the cells in which CK1δ was knocked out and CK1ε was knocked down. In the control cells, SHFL markedly suppressed -1PRF efficiency (Fig. 3B). CK1δ/ε downregulation significantly impaired the inhibitory effect, and expression of the CK1δ or CK1ε rescue construct restored the ability of SHFL to suppress -1PRF (Fig. 3B). In line with these results, the CK1δ/ε inhibitor PF-670462 attenuated SHFL inhibition of frameshifting in a dose-dependent manner, and the SHFL activity well correlated with its phosphorylation level (Fig. 3C).
We further analyzed the function of CK1δ/ε in regulating the antiviral activity of SHFL against HIV-1. The pNL4-3 Env-Luc plasmid is a modified HIV-1 NL4-3 construct in which the coding sequence of Firefly luciferase is inserted in-frame within the coding sequence of Nef (33). Myc-tagged SHFL was co-expressed with pNL4-3 Env-Luc in 293T cells and treated with either DMSO or the CK1δ/ε inhibitor PF-670462. The expression levels of the viral proteins in these cells were measured by western analysis. The culture supernatants were used to infect recipient cells to evaluate the production of the virus. In the cells were co-transfected with the dual-luciferase reporter with an empty vector (EV) or a construct expressing Flag-tagged SHFL. The cells were cultured in medium containing increasing concentrations of PF-670462 for 24 h. The frameshift efficiency, SHFL expression, and phosphorylation status were analyzed as described above. Data presented are means ± SD of two independent experiments. ** denotes P < 0.01; n.s. denotes not significant. (D and E) 293T cells were transfected with the HIV-1 producing vector pNL4-3 Env-Luc, together with an EV or a construct expressing Myc-tagged SHFL. The cells were cultured in medium containing either DMSO or PF-670462. At 48 h post-transfection, the culture supernatants were collected and used to infect HOS-CD4/CCR5 recipient cells.
Producer cells were lysed and luciferase activity was measured. HIV-1 Gag and Gag-Pol expression in the producer cells was analyzed by western blotting (D). At (Continued on next page) DMSO-treated cells, SHFL markedly reduced Gag-Pol protein level (Fig. 3D). The inhibi tory effect was nearly abolished upon PF-670462 treatment (Fig. 3D). Without PF-670462 treatment, SHFL significantly inhibited the production of the virus (Fig. 3E). In compari son, PF-670462 treatment dramatically compromised the antiviral activity of SHFL (Fig. 3E). These results further demonstrate that CK1δ/ε-mediated phosphorylation is required for the antiviral activity of SHFL against HIV-1.
## CK1δ/ε-mediated phosphorylation at T250 and T253 is essential for the antiviral activity of SHFL
We next investigated which phosphorylation sites in SHFL are functionally important for the regulatory activity. The candidate phosphorylation residues from the Phospho SitePlus database were each substituted with alanine. The mutants were tested for their ability to inhibit -1PRF using the dual-luciferase reporters in 293T cells. Although most mutations had little effect, the T250A and T253A mutations nearly abolished the function of SHFL to inhibit frameshifting (Fig. 4A). These results and the foregoing results suggested that the phosphorylation at T250 and T253 by CK1δ/ε is essential for SHFL to inhibit -1PRF. To substantiate this notion, we generated two phosphomimetic mutants of SHFL, T250E and T253E, and analyzed their ability to inhibit -1PRF. As expected, these two mutants displayed comparable activity as the wild-type SHFL (Fig. 4B).
To demonstrate that the phosphorylation of SHFL at T250 and T253 is important for its antiviral activity, we analyzed the effect of these mutants on the expression of HIV-1 Gag-pol and virus production. 293T cells were transiently co-transfected with pNL4-3 Env-Luc and Myc-tagged SHFL variants. Consistent with the previous results, the wildtype SHFL inhibited the expression of Gag-pol (Fig. 4C) and the virus production (Fig. 4D). Although the antiviral activity of the SHFL-T250A and -T253A mutants was markedly reduced, the SHFL-T250E and -T253E mutants exhibited comparable antiviral activity as the wild-type protein (Fig. 4C andD). These results indicate that the phosphorylation at T250 and T253 is critical for the antiviral activity of SHFL. Considering that the SHFL-T250A and -T253A mutants displayed very similar phenotypes, we focused on the SHFL-T250A mutant in the following studies.
## Phosphorylation at T250 modulates SHFL association with polysomes
To investigate how phosphorylation at T250 influences SHFL inhibition of -1PRF, we analyzed the distribution patterns of HIV-1 gag-pol mRNA in the polysome profiling assay. HOS-CD4/CCR5 cells expressing Myc-tagged GFP (control) or SHFL variants in a doxycycline-inducible manner were infected with VSV-G pseudotyped HIV-1 vector NL4-3 GFP and treated with doxycycline to induce protein expression. The cell lysates were subjected to polysome profiling analysis (see Materials and Methods for detailed procedure). In the GFP-expressing control cells, a peak of the gag-pol RNA level was detected in the light polysome fractions (Fig. 5A), which is consistent with the results published previously (34). The expression of SHFL or SHFL-T250E led to a reduction of the gag-pol RNA level in the light polysome fractions (Fig. 5A). In the SHFL-T250A-expressing cells, the RNA distribution pattern was very similar to that in the control cells (Fig. 5A). We also analyzed the distribution patterns of the viral nef RNA and host β-actin mRNA, which do not have -1PRF elements, and no obvious difference was observed (Fig. 5A).
The above results suggested that the phosphorylation at T250 may affect SHFL interaction with the translation machinery. To further investigate this possibility, we examined the association with polysomes of the SHFL variants. The 293T cells transiently expressing the SHFL variants, including SHFL, SHFL-T250A, and SHFL-T250E, were subjected to polysome profiling analysis, and the SHFL protein levels in each fraction were measured by western blotting. The wild-type SHFL was easily detected in most polysome fractions (Fig. 5B). In comparison, SHFL-T250A exhibited markedly reduced association with the polysomes, whereas the association of SHFL-T250E with the polysomes was similar to that of the wild-type protein (Fig. 5B). Collectively, these results indicate that the phosphorylation of SHFL at T250 is important for its association with polysomes.
## The T250A mutation impairs SHFL interaction with IGF2BPs
To explore the mechanism by which the phosphorylation at T250 affects SHFL associa tion with polysomes, we first assessed whether the T250A mutation influences the RNAbinding capacity of SHFL. Two Cy3-labeled RNA probes were used, HIV-1 FSE and Poly (CAA). The HIV-1 FSE probe consisted of 100-nt encompassing the frameshift element (including the slippery site and downstream stem-loop) of HIV-1. The Poly (CAA) probe was used as a negative control (Fig. 6A). Flag-tagged SHFL proteins transiently expressed in 293T cells were immunoprecipitated and then incubated with the RNA probes, followed by Urea-PAGE. The interaction of the RNA probes with the SHFL proteins was detected by the fluorescence signals precipitated with the SHFL proteins. Although none of the proteins interacted with the Poly (CAA) control RNA, they displayed comparable binding to the HIV-1 FSE probe (Fig. 6B). These results indicated that the phosphorylation of SHFL at T250 does not compromise its RNA-binding ability. We next examined whether T250 phosphorylation affects SHFL interaction with ribosomal proteins. The ribosomal proteins uL5 and eS31 have been reported to interact with SHFL (1). Myc-tagged SHFL variants and Flag-tagged uL5 or eS31 were co-expressed in 293T cells. The interactions of the proteins were assessed by the Co-IP assay. No significant difference was observed in the interaction with uL5 or eS31 of SHFL, SHFL-T250A, and SHFL-T250E (Fig. 6C).
IGF2BPs were identified as SHFL-interacting proteins in the TAP-MS analysis (Fig. 1C). Previous studies have shown that IGF2BPs associate with polysomes (35,36). To assess whether SHFL phosphorylation affects its interaction with IGF2BP proteins, Flag-tagged SHFL, SHFL-T250A, and SHFL-T250E were expressed in 293T cells. Co-IP analysis revealed that the T250A mutant exhibited markedly reduced interaction with the endogenous IGF2BP proteins compared with wild-type SHFL (Fig. 6D). In comparison, SHFL-T250E interacted with the endogenous IGF2BPs (Fig. 6D). These results indicate that the phosphorylation of SHFL at T250 is essential for its interaction with IGF2BP proteins and FIG 5 The T250A mutation reduces SHFL association with polysomes. (A) HOS-CD4/CCR5 cells expressing the Myc-tagged protein indicated in a doxycyclineinducible manner were infected with VSV-G pseudotyped NL4-3 GFP for 3 h and treated with doxycycline to induce protein expression. At 24 h postinfection, the cell lysates were fractionated through a 10%-50% sucrose gradient with continuous monitoring of absorbance at 260 nm (upper panel). The RNA levels in each fraction were measured by RT-qPCR. Relative RNA level was calculated as the RNA level in each fraction divided by the total RNA level in all the fractions. Data presented are representative of two independent experiments. (B) 293T cells were transfected with an empty vector (EV) or a plasmid expressing the Myc-tagged SHFL indicated. At 24 h post-transfection, the cells were treated with cycloheximide and lysed in polysome lysis buffer. The cell lysates were fractionated through a 10%-50% sucrose gradient with continuous monitoring of absorbance at 260 nm (upper panel). The protein levels in each fraction were measured by western analysis (middle panel). Relative protein levels of SHFL variants in the middle panel were quantified using ImageJ by dividing the signal intensity of each fraction by the total signal intensity across all fractions (bottom panel). Data presented are representative of two independent experiments. suggest the association of SHFL with the translating ribosomes is at least partially mediated by the IGF2BP proteins.
## DISCUSSION
Here, we identified CK1δ and CK1ε as kinases that interact with and phosphorylate SHFL. CK1δ/CK1ε phosphorylation of SHFL at T250 and T253 is essential for SHFL to inhibit -1PRF and to suppress HIV-1 production.
How the phosphorylation of SHFL regulates its antiviral activity is not clear yet. Phosphorylation can introduce steric bulk and negative charges, alter the local physicochemical environment, and influence protein stability, conformational dynamics, and molecular interactions (37)(38)(39)(40). Based on AlphaFold3 predictions, SHFL comprises three structured regions: an N-terminal domain, a Zinc Finger domain containing three zinc finger motifs, and a C-terminal polyglutamate (poly-E) tail. T250 and T253 are located within a predicted intrinsically disordered region embedded between these domains (Fig. 7A). Phosphorylation within such core regions could influence protein folding and structural stability (39). Consistently, ConSurf analysis, a computational method that estimates and visualizes evolutionary conservation in macromolecules (41), revealed that T250, T253, and their neighboring residues are highly conserved (Fig. 7B), suggesting that this region plays a functionally important role maintained under evolutionary constraint. Based on our experimental data, CK1δ and CK1ε are supported as kinases that interact with and phosphorylate SHFL, consistent with their identified roles in regulating its antiviral activity. Structural modeling was performed to gain further insight into these processes. Considering the high structural similarity between CK1δ and CK1ε ( 27), we performed AlphaFold3 multimer prediction to evaluate the potenCOMtial interaction between SHFL and CK1δ. The predicted complex suggests a direct interaction between SHFL and CK1δ (Fig. 7C, left panel). In this model, the kinase domain of CK1δ (orange) is positioned close to SHFL (blue), and the T250 and T253 residues of SHFL (red) are located near the catalytic cleft (magenta) containing the catalytic loop (pink) of CK1δ (Fig. 7C, right panel). Given that the kinase domains of CK1δ and CK1ε are highly conserved (42), this model may also explain the observed interaction of SHFL with both kinases. Together with our biochemical data, this structural model provides computational support for a direct kinase-substrate relationship between CK1δ/ε and SHFL.
SHFL-T250A showed reduced association with polysome fractions (Fig. 5). The T250A mutation did not affect SHFL interaction with target RNA (Fig. 6B) nor its interaction with the ribosomal proteins uL5 and eS31 (Fig. 6C). However, the T250A mutation significantly reduced SHFL interaction with IGF2BP proteins, whereas the T250E mutation had little effect. These results suggest that the phosphorylation at T250 is essential for SHFL interaction with IGF2BPs. The reduced interaction between the T250A mutant and IGF2BPs, together with the known roles of IGF2BPs in mRNA localization and translation regulation (35), suggests that phosphorylation-dependent association with IGF2BPs is important for SHFL to achieve spatial and functional coupling with the translation machinery, as similar potential recruitment mechanisms have been suggested (43,44). Further investigation is needed to decipher the precise mechanism underlying this process.
In HIV-1-infected control cells, polysome profiling revealed a peak signal of gag-pol mRNA in the light polysome fractions, consistent with previous observations (34). The lower ribosome density in these fractions may permit RNA structures to refold or remain intact, thereby promoting ribosome pausing and enabling -1PRF. Upon SHFL expression, the gag-pol mRNA peak in the light polysome fractions was selectively diminished, suggesting that SHFL effectively suppressed -1PRF within this region. SHFL has been reported to promote ribosome dissociation by recruiting eukaryotic release factors at the frameshifting site (1). The reduction of gag-pol RNA levels in the polysome fractions is in line with the notion that SHFL caused premature ribosome dissociation. SHFL-T250E exhibited a polysome distribution pattern of gag-pol mRNA similar to the wildtype SHFL, whereas SHFL-T250A mirrored the GFP control. These mRNA distribution patterns are consistent with the polysome association pattern of SHFL; the wild-type SHFL and the T250E mutant exhibited comparable association with the polysome fractions, whereas the T250A mutant showed a markedly reduced distribution in the polysome fractions. These results emphasize the critical role of T250 phosphorylation in enabling SHFL association with polysomes, which is required for its ability to suppress -1PRF.
In the polysome profiling analysis, SHFL was enriched in ribosomal fractions, includ ing the actively translating polysomal fractions (1-5) and the non-translating ribosomal fractions (80S/60S/40S; 5-8), with a smaller portion detected in the free fractions (9, 10) (Fig. 5B). This distribution pattern is consistent with SHFL interaction with ribosomes. SHFL detected in the free fractions may represent the protein partially phosphorylated or unphosphorylated at T250 and T253, as CK1δ/ε-mediated phosphorylation is often incomplete due to limited kinase accessibility and the reversible nature of this modifica tion (45)(46)(47). The T250A mutation impaired SHFL interaction with IGF2BPs, resulting in reduced association with polysomes and increased accumulation in the non-translating ribosomal fractions. The phosphomimetic T250E mutant behaved more like the WT protein in the polysome fractions, which is in line with its antiviral activity. More T250E was detected in fractions 5-7 than the WT protein. One possible explanation is that more T250E was detected in fractions 5-7 because less T250E was detected in fractions 9 and 10, which are supposed to contain only the free SHFL protein. T250E could be considered fully phosphorylated, while the WT protein phosphorylation is expected to be heteroge neous, and thus there is always a fraction of the WT protein being unphosphorylated and thereby not associated with the ribosome.
In summary, our results indicate that CK1δ/ε phosphorylation at T250 is essential for SHFL to engage the translational machinery and to suppress -1PRF, thereby restrict ing viral replication. These findings highlight phosphorylation at T250 as a molecular requirement for the antiviral activity of SHFL.
## MATERIALS AND METHODS
## Plasmids and cell culture
The coding sequences of CK1δ (NM_001893.6), CK1δ2 (NM_139062.4), and CK1ε (NM_152221.3) were PCRamplified from 293T cell-derived cDNA and cloned into the pCMV-HA-Flag (pCMV-HF) expression vector as described previously (48). Site-direc ted mutagenesis of pCMV-HF-SHFL or pLPCX-SHFL-Myc was performed as described previously (1). The dual-luciferase reporters pDual-HIV(-1) and pDual-HIV(0) were constructed by inserting the HIV-1 -1PRF signal sequence or the modified control sequence between the Renilla luciferase (Rluc) and Firefly luciferase (Fluc) coding regions, as described by Léa Brakier-Gingras (49).
To generate doxycycline-inducible SHFL-expressing cell lines for tandem affinity purification (TAP) assays, the coding sequence of HA-Flag-SHFL was cloned into the lentiviral vector pTRIPZ-puro (Horizon). The resulting vector was transduced into HEK293 cells. To express GFP-Myc, SHFL-Myc, and its mutants in HOS-CD4/CCR5 cells in a doxycycline-inducible manner, the coding sequences were cloned into a modified pTRIPZ-mCherry vector, in which the puromycin resistance gene was replaced with the mCherry fluorescence marker (50). CK1δ/ε knockout cell lines were generated using sgRNA sequences cloned into the LentiCRISPRv2-puro vector, which was kindly provided by Dr. Mingzhao Zhu of the Institute of Biophysics, Chinese Academy of Sciences.
HEK293T, HEK293, and HOS-CD4/CCR5 were cultured in Dulbecco's modified Eagle medium (DMEM, Invitrogen) supplemented with 10% fetal bovine serum (Gibco). Puromycin-resistant cell lines were selected in medium containing 5 µg/mL puromycin (Sigma-Aldrich) and maintained in medium containing 1 µg/mL puromycin.
Plasmid DNA transfection was carried out using Xpregen following the manufac turer′s instructions (Beijing Yu-Feng Biotechnology, Cat no. ND01). Small interfering RNA (siRNA) transfection was performed with Lipofectamine 2000 (Thermo Fisher Scientific) following the manufacturer's recommended protocol.
## Antibodies
To generate antisera specifically recognizing SHFL phosphorylated at T250 and T253, a synthetic phosphopeptide (SGS-pT-VA-pT-SLS-C) conjugated to Keyhole Limpet Hemocyanin (SciLight Biotechnology) was used to immunize rabbits. The antisera were harvested following standard immunization protocols and validated for specificity before use. SHFLspecific polyclonal antibodies were kindly provided by Dr. Xuan Yifang. The p24specific antibody was kindly provided by Prof. Yongtang Zheng (51). Commercial antibodies used in this study included: mouse monoclonal anti-β-actin (GSGB-BIO, China; TA-09), mouse monoclonal anti-Myc (9E10, Santa Cruz Biotechnology; sc-40), mouse monoclonal anti-Flag M2 (Sigma-Aldrich; F3165), rabbit polyclonal anti-IGF2BP1 (Proteintech; 22803-1-AP), rabbit monoclonal anti-IGF2BP2 (Abclonal; A5189), rabbit monoclonal anti-IGF2BP3 (Abclonal; A23295), anti-CK1ε (Abcam; ab302638), and rabbit polyclonal anti-CK1δ (Proteintech; 14388-1-AP).
## TAP-MS and protein immunoprecipitation assays
For TAP-MS analysis, HEK293 cells expressing HA-Flag-SHFL in a doxycycline-inducible manner were treated with 1 µg/mL doxycycline for 24 h to induce protein expression, washed twice with ice-cold PBS, and lysed in Co-IP buffer (150 mM NaCl, 25 mM Tris-HCl pH 7.4, 2 mM MgCl₂, 0.5% IGEPAL CA-630 [Sigma-Aldrich]) supplemented with Complete EDTA-free protease inhibitor cocktail (Sigma-Aldrich), PhosSTOP phosphatase inhibitor (Sigma-Aldrich) and 250 U/mL Benzonase (Sigma-Aldrich). The cell lysates were clarified through centrifugation at 4°C for 15 min. For Flag-tag affinity purification, anti-Flag M2 affinity gel (Sigma-Aldrich) was washed three times with PBS, incubated with the clarified cell lysates at 4°C for 2 h with gentle rotation, followed by washing three times with PBS. Proteins were eluted by incubation with Flag elution buffer (200 µg/mL 3 × FLAG peptide [SciLight] in Co-IP buffer) at 4°C for 30 min. After centrifugation at 5,000 × g for 1 min, the eluates were transferred to new tubes. For HA-tag affinity purification, anti-HA magnetic beads (Thermo Fisher Scientific) were washed three times with PBS, incubated with the above eluates at 4°C for 2 h with gentle rotation, and collected using a magnetic stand. Proteins were eluted with acid elution buffer (0.1 M Glycine-HCl, pH 2.5) at room temperature for 10 min. Eluates were neutralized immediately using neutralization buffer (15 µL per 100 µL eluate; 1 M Tris-HCl, pH 8.5, 1.5 M NaCl), aliquoted, snap-frozen in liquid nitrogen, and stored at -80°C for subsequent western blotting and mass spectrometry analyses. For mass spectrometry analysis, protein bands visualized by silver staining were excised and analyzed by the National Center for Protein Sciences as described previously (1). The TAP-MS data have been deposited in the iProx database (ProteomeXchange accession number PXD068760) (52).
Co-immunoprecipitation (Co-IP) assays using Flag-tagged SHFL variants were performed as described in the TAP-MS analysis. Briefly, 293T cells were transfected with plasmids expressing Flag-tagged SHFL variants. At 24 h post-transfection, the cells were lysed in Co-IP buffer, and lysates were subjected to immunoprecipitation using anti-Flag M2 affinity gel. The immunoprecipitants were analyzed by Western blotting using the indicated antibodies to detect interacting proteins.
For endogenous SHFL immunoprecipitation and phosphorylation analysis, THP-1 cells were treated with interferon-alpha for 24 h, in the presence or absence of the CK1δ/ε inhibitor PF-670462 at 40 µM. Cells were lysed in RIPA buffer (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% IGEPAL CA-630 [Sigma-Aldrich], 1% sodium deoxycholate), supplemented with Complete EDTA-free protease inhibitor cocktail (Sigma-Aldrich) and PhosSTOP phosphatase inhibitor (Sigma-Aldrich). Lysates were clarified by centrifugation and incubated with SHFLspecific polyclonal antibodies and protein G beads (Amersham Pharmacia) at 4°C for 2 h. The immunoprecipitants were analyzed by western blotting using antisera specific for SHFL phosphorylated at T250 and T253.
## Downregulation of CK1
To generate CK1δ or CK1ε knockout cells, HEK293 cells were transduced with VSV-G-pseudotyped lentiviral vectors based on LentiCRISPRv2-puro expressing sgRNAs targeting CK1δ or CK1ε at CDS region, two sgRNAs were designed for each gene. Given that CK1δ has two functional isoforms (27), the sgRNAs targeting CK1δ were selected to disrupt both CK1δ and CK1δ2 by recognizing sequences at the N terminus. A non-target ing sgRNA (sgNC) was included as a control. Transduced cells were selected in medium containing 5 µg/mL puromycin, and the pool of resistant cells was used. Knockout efficiency was confirmed by western blotting using CK1δ-or CK1εspecific antibodies. The sgRNA sequences used are listed below.
sgNC: 5′-AAATGTGAGATCAGAGTAAT-3′ sgCSNK1D-1: 5′-GGACTACAACGTCATGGTGA-3′ sgCSNK1D-2: 5′-TGAGAGTCGGGAACAGGTAC-3′ sgCSNK1E-1: 5′-CCGCAAATTCAGCCTCAAGA-3′ sgCSNK1E-2: 5′-GTCCTTCGGAGATATCTACC-3′ To knock down CK1ε in CK1δ knockout cells, the cells were transfected twice with scrambled siRNA (SCR) or CK1εspecific siRNA (GenePharma), with or without a CK1ε rescue construct, using Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer's instructions. In the second-round transfection, plasmids required for subsequent assays were co-transfected. The siRNA sequences have been previously described (4) and are listed below.
Scrambled siRNA (SCR): 5′-UUCUCCGAACGUGUCACGUTT-3′ siCSNK1E: 5′-CCUCCGAAUUCUCAACAUATT-3′ To confirm the specificity of CK1δ and CK1ε downregulation, rescue expression constructs of CK1δ and CK1ε that were not targeted by the CK1δ sgRNAs or CK1ε siRNA were generated. The nucleotide sequences recognized by the CK1δ sgRNAs and CK1ε siRNA were synonymously mutated and cloned into expression vectors. The mutated sequences of CK1δ were 5′-GGCGATTATAATGTGATGGTCATGGAA-3′ and 5′-CTGCGGGTGG The culture supernatants were used to infect recipient HOS-CD4/CCR5 cells. At 48 h postinfection, luciferase activity in the recipient cells was measured and normalized with the luciferase activity in the producer cells. Fold inhibition of SHFL against HIV-1 production was calculated as the normalized luciferase activity from the control cells divided by that from the SHFL-expressing cells.
## Statistical analysis
Statistical analyses were performed using GraphPad Prism and Proteome Discoverer. Unless otherwise specified, data are presented as arithmetic mean ± standard deviation (SD) from three independent experiments. For pairwise comparisons, P-values were determined using a two-tailed paired Student's t-test. For multiple group comparisons, one-way analysis of variance (ANOVA), followed by Tukey's post hoc test, was applied. The number of replicates and additional details relevant to data reliability are provided in the corresponding figure legends.
## References
1. "GCAATAGATATAGA-3′, and the mutated sequence of CK1ε was 5′-CCTAGCGAGTTTAGCAC CTAT-3′. REFERENCES"
2. Wang, Han, Ding et al. (2019) "Regulation of HIV-1 Gag-Pol expression by shiftless, an inhibitor of programmed -1 ribosomal frameshifting" *Cell*
3. (2025) *Full-Length Text Journal of Virology*
4. Zimmer, Kibe, Rand et al. (2021) "The short isoform of the host antiviral protein ZAP acts as an inhibitor of SARS-CoV-2 programmed ribosomal frameshift ing" *Nat Commun*
5. Schmidt, Lareau, Keshishian et al. (2021) "The SARS-CoV-2 RNA-protein interactome in infected human cells" *Nat Microbiol*
6. Yu, Zhao, Pan et al. "2021. C19orf66 inhibits Japanese encephalitis virus replication by targeting -1 PRF and the NS3 protein" *Virol Sin*
7. Suzuki, Chin, Han et al. (2016) "Characterization of RyDEN (C19orf66) as an interferon-stimulated cellular inhibitor against dengue virus replication" *PLoS Pathog*
8. Kelly, Dinman (2023) "Shiftless is a novel member of the ribosome stress surveillance machinery that has evolved to play a role in innate immunity and cancer surveillance" *Viruses*
9. Dinman (2012) "Control of gene expression by translational recoding" *Adv Protein Chem Struct Biol*
10. Hill, Brierley (2023) "Structural and functional insights into viral programmed ribosomal frameshifting" *Annu Rev Virol*
11. Riegger, Caliskan (2022) "Thinking outside the frame: impacting genomes capacity by programmed ribosomal frameshifting" *Front Mol Biosci*
12. Jacks, Power, Masiarz et al. (1988) "Characterization of ribosomal frameshifting in HIV-1 gag-pol expression" *Nature*
13. Biswas, Jiang, Pacchia et al. (2004) "The human immunodeficiency virus type 1 ribosomal frameshifting site is an invariant sequence determinant and an important target for antiviral therapy" *J Virol*
14. Marcheschi, Tonelli, Kumar et al. (2011) "Structure of the HIV-1 frameshift site RNA bound to a small molecule inhibitor of viral replication" *ACS Chem Biol*
15. Dinman (2019) "Scaring ribosomes shiftless" *Biochemistry*
16. Li, Wilmanns, Thornton et al. (2013) "Elucidating human phosphatase-substrate networks" *Sci Signal*
17. Sacco, Perfetto, Castagnoli et al. (2012) "The human phosphatase interactome: an intricate family portrait" *FEBS Lett*
18. Ardito, Giuliani, Perrone et al. (2017) "The crucial role of protein phosphorylation in cell signaling and its use as targeted therapy (Review)" *Int J Mol Med*
19. Knippschild, Gocht, Wolff et al. (2005) "The casein kinase 1 family: participation in multiple cellular processes in eukaryotes" *Cell Signal*
20. Gross, Anderson (1998) "Casein kinase I: spatial organization and positioning of a multifunctional protein kinase family" *Cell Signal*
21. Vielhaber, Eide, Rivers et al. (2000) "Nuclear entry of the circadian regulator mPER1 is controlled by mammalian casein kinase I epsilon" *Mol Cell Biol*
22. Narasimamurthy, Virshup (2021) "The phosphorylation switch that regulates ticking of the circadian clock" *Mol Cell*
23. Price (2006) "CKI, there's more than one: casein kinase I family members in Wnt and Hedgehog signaling" *Genes Dev*
24. Francisco, Virshup (2022) "Casein kinase 1 and human disease: insights from the circadian phosphoswitch" *Front Mol Biosci*
25. Zemp, Wandrey, Rao et al. (2014) "CK1δ and CK1ε are components of human 40S subunit precursors required for cytoplasmic 40S maturation" *J Cell Sci*
26. Flotow, Graves, Wang et al. (1990) "Phosphate groups as substrate determinants for casein kinase I action" *J Biol Chem*
27. Meggio, Perich, Marin et al. (1992) "The comparative efficiencies of the Ser(P)-, Thr(P)-and Tyr(P)-residues as specificity determinants for casein kinase-1" *Biochem Biophys Res Commun*
28. Meggio, Perich, Reynolds et al. (1991) "A synthetic β-casein phosphopeptide and analogues as model substrates for casein kinase-1, a ubiquitous, phosphate directed protein kinase" *FEBS Lett*
29. Fulcher, Sapkota (2020) "Functions and regulation of the serine/ threonine protein kinase CK1 family: moving beyond promiscuity" *Biochem J*
30. Narasimamurthy, Hunt, Lu et al. (2018) "CK1δ/ε protein kinase primes the PER2 circadian phosphoswitch" *Proc Natl Acad Sci*
31. Decaprio, Kohl (2019) "Tandem immunoaffinity purification using anti-FLAG and anti-HA antibodies" *Cold Spring Harb Protoc*
32. Fustin, Kojima, Itoh et al. (2018) "Two Ck1δ transcripts regulated by m6A methylation code for two antagonistic kinases in the control of the circadian clock" *Proc Natl Acad Sci*
34. Sekar, Li, Schlessinger et al. (2024) "A web portal for exploring kinase-substrate interactions" *NPJ Syst Biol Appl*
35. Sunkari, Meijer, Flajolet (2022) "The protein kinase CK1: Inhibition, activation, and possible allosteric modulation" *Front Mol Biosci*
36. Zhang, Yang, Wang et al. (2018) "Identification of new type I interferon-stimulated genes and investigation of their involvement in IFN-β activation" *Protein Cell*
37. Kibe, Buck, Gribling-Burrer et al. (2025) "The translational landscape of HIV-1 infected cells reveals key gene regulatory principles" *Nat Struct Mol Biol*
38. Huang, Weng, Sun et al. (2018) "Recognition of RNA N6-methyladenosine by IGF2BP proteins enhances mRNA stability and translation" *Nat Cell Biol*
39. Jiang, Liu, Nie et al. (2021) "The role of m6A modification in the biological functions and diseases" *Signal Transduct Target Ther*
40. Newcombe, Delaforge, Hartmann-Petersen et al. (2022) "How phosphorylation impacts intrinsically disordered proteins and their function" *Essays Biochem*
41. Nishi, Shaytan, Panchenko (2014) "Physicochemical mechanisms of protein regulation by phosphorylation" *Front Genet*
42. Kamacioglu, Tuncbag, Ozlu (2021) "Structural analysis of mammalian protein phosphorylation at a proteome level" *Structure*
43. Sternburg, Da Silva, Dormann (2022) "Post-translational modifications on RNA-binding proteins: accelerators, brakes, or passengers in neurodegeneration?" *Trends Biochem Sci*
44. Ashkenazy, Abadi, Martz et al. (2016) "ConSurf 2016: an improved methodology to estimate and visualize evolutionary conservation in macromolecules" *Nucleic Acids Res*
45. (2025) *Full-Length Text Journal of Virology*
46. Xu, Ianes, Gärtner et al. (2019) "Structure, regulation, and (patho-)physio logical functions of the stress-induced protein kinase CK1 delta (CSNK1D)" *Gene*
47. Kraushar, Thompson, Wijeratne et al. (2014) "Temporally defined neocortical translation and polysome assembly are determined by the RNA-binding protein Hu antigen R" *Proc Natl Acad Sci*
48. Corley, Burns, Yeo (2020) "How RNA-binding proteins interact with RNA: molecules and mechanisms" *Mol Cell*
49. Wu, Haas, Dephoure et al. (2011) "A large-scale method to measure absolute protein phosphorylation stoichiometries" *Nat Methods*
50. Von Stechow, Francavilla, Olsen (2015) "Recent findings and technological advances in phosphoproteomics for cells and tissues" *Expert Rev Proteomics*
51. Needham, Parker, Burykin et al. (2019) "Illuminating the dark phosphoproteome" *Sci Signal*
52. Mu, Fu, Zhu et al. (2015) "HIV-1 exploits the host factor RuvB-like 2 to balance viral protein expression" *Cell Host Microbe*
53. Dulude, Berchiche, Gendron et al. (2006) "Decreasing the frameshift efficiency translates into an equivalent reduction of the replication of the human immunodeficiency virus type 1" *Virology (Auckl)*
54. Huang, Zhang, Nie et al. (2024) "Assembly and activation of EBV latent membrane protein 1" *Cell*
55. Liu, Wang, Xiao et al. (2007) "Preparation and characterization of three monoclonal antibodies against HIV-1 p24 capsid protein" *Cell Mol Immunol*
56. Ma, Chen, Wu et al. (2019) "iProX: an integrated proteome resource" *Nucleic Acids Res*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12568068&blobtype=pdf
|
# Correction: Holbrook et al. Updated and Validated Pan-Coronavirus PCR Assay to Detect All Coronavirus Genera. Viruses 2021, 13, 599
Myndi Holbrook, Simon Anthony, Isamara Navarrete-Macias, Theo Bestebroer, Vincent Munster, Neeltje Van Doremalen
## References
1. Holbrook, Anthony, Navarrete-Macias et al. (2021) "Updated and Validated Pan-Coronavirus PCR Assay to Detect All Coronavirus Genera" *Viruses*
2. (1364) "MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content" *Viruses*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12749442&blobtype=pdf
|
# npj | vaccines Article
Juhong Rao, Jin Li, Tingting Jiang, Weiwei Guo, Xuekai Zhang, Zihan Zhang, Jiaoling Su, Mingqing Lu, Xue Hu, Xingpo Liu, Rong Qu, Tao Zhu, Chenlong Hu, Kunpeng Liu, Haomeng Wang, & Chao, Haomeng Wang Or Chao
## Abstract
Crimean-Congo hemorrhagic fever virus (CCHFV) is a tick-borne virus of the Orthonairovirus genus, Nairoviridae family, that causes severe febrile hemorrhagic disease in humans with a case fatality rate ranging from approximately 3-30%. This zoonotic pathogen is distributed across a broad geographic area spanning Asia, Europe, and Africa. Despite its significant public health threat and outbreak potential, no licensed vaccines are available. In this study, we developed and systematically assessed the immunogenicity and protective efficacy of mRNA vaccines encoding either CCHFV nucleoprotein (NP) or glycoprotein precursor (GPC) in mouse models. Vaccination with the NP-encoding mRNA alone provided complete protection against lethal cross-genotype CCHFV challenge. Moreover, combined vaccination with both the NP and GPC mRNAs elicited robust immune responses and conferred protection against CCHFV infection. Notably, a single-dose immunization with 2 μg mRNA-NP was sufficient to confer protection against lethal challenge. Furthermore, the passive transfer of NP-immune serum provided partial protection, supporting the role of NP-specific antibodies in mediating protection. Overall, these mRNA vaccines demonstrate protective efficacy against CCHFV, with combined antigenic protection and dose-sparing potential, highlighting their potential for outbreak preparedness and further clinical development.
Crimean-Congo hemorrhagic fever (CCHF) is a tick-borne disease with high fatality rates in humans ranging from approximately 3-30%, and it is endemic across Asia, Europe, Africa, and the Middle East 1,2 . The causative agent, Crimean-Congo hemorrhagic fever virus (CCHFV), is a negativesense, enveloped RNA virus that is classified within the genus Orthonairovirus, family Nairoviridae in the family genus Orthonairovirus, family Nairoviridae [3][4][5][6] and has a tri-segmented genome. CCHFV is transmitted primarily by ticks of the genus Hyalomma within the Ixodidae family 3 . In humans, CCHF infection occurs most commonly in agricultural workers through tick bites, as well as in slaughterhouse employees and healthcare workers via exposure to blood, tissues, or bodily fluids from infected animals or patients, respectively 3 . The disease manifests with highly variable clinical outcomes, spanning from silent infections to critical and potentially lethal illness, progressing through incubation, pre-hemorrhagic, hemorrhagic, and convalescent phases 7 . Initial symptoms include a nonspecific febrile syndrome with acute onset of fever, headache, dizziness, and myalgia 8,9 . The hemorrhagic stage is characterized by extensive ecchymoses and uncontrolled multiorgan hemorrhage 1 . Owing to its endemicity, annual case reports, and a lack of licensed vaccines or effective therapies, CCHFV has been recognized by the World Health Organization (WHO) as a highpriority pathogen requiring the development of effective antiviral strategies 10 .
The genome of CCHFV consists of small (S), medium (M), and large (L) RNA segments encoding the viral nucleoprotein (NP), glycoprotein precursor (GPC), and RNA-dependent RNA polymerase (RdRp), respectively 11 . Regarding vaccine development, NP and GPC have been prioritized as key targets because of their essential roles in viral replication and entry. Several CCHFV vaccine candidates targeting these antigens have been evaluated in rodent and nonhuman primate models, with reported protective efficacies ranging from 0% to 100% [12][13][14][15] . While GPC serves as a primary immunogen targeted by both neutralizing and non-neutralizing antibodies, NP is also capable of eliciting humoral responses. In addition, both antigens can stimulate T-cell immunity, and the relatively high conservation of NP among various CCHFV strains makes it particularly suitable for development of vaccines that can mediate protection in against multiple genotypes. These immunogenic and structural characteristics make NP and GPC promising targets for vaccine development. Compared with traditional vaccines, mRNA vaccines provide several advantages, including a robust immune response, rapid development, design flexibility, and improved safety, as they do not involve live pathogens or genomic integration risks 16 . Currently, various mRNA vaccines targeting infectious diseases, including SARS-CoV-2 17 , influenza virus 18 , respiratory syncytial virus (RSV) 19,20 , rabies virus 21,22 , mpox virus 23 , and Zika virus 24,25 , have either been approved for use or are being studied in clinical trials. This highlights the promise of mRNA vaccines in combating emerging epidemics and acute outbreaks.
In this study, we developed two mRNA vaccines encoding the fulllength NP (mRNA-NP) and GPC (mRNA-GPC) of CCHFV, each of which are encapsulated separately in lipid nanoparticles (LNPs) to increase stability and delivery. We evaluated the immunogenicity of mRNA-NP, mRNA-GPC, and mRNA-NP + mRNA-GPC by assessing CCHFVspecific antibody and T-cell responses and subsequently evaluated their protective efficacy in a lethal Ifnar1 -/-(C57BL/6JGpt background) mouse challenge model. Our results revealed that a two-dose regimen of mRNA-NP or mRNA-NP + mRNA-GPC conferred complete protection against lethal cross-genotype CCHFV challenge, and even a single dose of as low as 2 μg mRNA-NP resulted in 100% survival. This protection was attributed primarily to NP-specific immunity, as mRNA-NP alone provides complete survival. However, combined immunization with mRNA-NP and mRNA-GPC reduced the proportion of viral RNA positivity in tissues compared with mRNA-NP alone, especially at low dose, highlighting a complementary role for GPC in enhancing protective efficacy. Additionally, passive transfer of serum from mRNA-NP-immunized mice conferred partial protection (33.33%) to naïve mice, indicating a role for NP-specific antibodies. These findings support the potential of mRNA vaccines targeting NP and GPC to elicit protective immune responses and significantly improve survival in a lethal model of CCHFV infection.
## Results
## The mRNA vaccination elicits robust B-and T-cell responses in mice
To systematically evaluate the CCHFV-specific immunogenicity of the mRNA vaccine candidates, C57BL/6 J or Ifnar1 -/-mice (n = 6 per group, aged 6-8 weeks) were divided into seven groups and immunized (Fig. 1A-B). The mice were subcutaneously (S.C.) injected with either mRNA-NP (1 or 5 μg), mRNA-GPC (1 or 5 μg), or mRNA-NP + mRNA-GPC (1 μg of each, totaling 2 μg; or 5 μg of each, totaling 10 μg), with empty LNPs serving as controls. All the groups were subjected to a prime-boost immunization regimen, receiving the second vaccination with the same dose 21 days later.
Humoral immune responses were assessed using enzyme-linked immunosorbent assays (ELISAs) and virus neutralization assays. The ELISA results revealed that all vaccine regimens effectively elicited antigen-specific IgG responses that targeted their respective immunogens (NP or GPC). Booster immunization significantly increased antibody titers, which peaked between 5-7 weeks post-priming and remained detectable for up to 38 weeks (Fig. 1C-D). High-dose (5 μg) regimens induced significantly more robust and sustained antibody responses than low-dose (1 μg) regimens, as evidenced by higher NP-specific IgG titers at weeks 3, 6, 7, 23, and 38, and higher Gc-specific IgG titers at weeks 3, 4, 11, and 23 (Fig. 1C,D; Supplementary Data 1). Neutralizing antibody responses were assessed using both pseudovirus and live virus neutralization assays in C57BL/6 and Ifnar1 -/-mice. Pseudovirus neutralization assays demonstrated that mRNA-GPC and mRNA-GPC + mRNA-NP vaccination elicited neutralizing antibodies in a dose-dependent manner, with higher titers in the 5 μg-vaccinated groups than in the 1 μg-receiving groups (Fig. 1E,F).
Live virus neutralization assays against cross-genotype CCHFV strains (YL16070 and IbAr10200) revealed enhanced neutralization at the higher vaccine dose (5 μg), despite the vaccine being designed on the basis of the Turkey strain (Fig. 1G, H; Supplementary Fig. 2A,B). These findings indicate that mRNA-GPC vaccination elicits potent neutralizing antibody responses capable of cross-reacting with genetically distinct CCHFV strains that exhibit significant glycoprotein sequence divergence from the vaccine strain. Based on S-segment phylogeny, the vaccine sequence from strain Turkey (Europe 1 lineage) and neutralization test using strains YL16070 (Asia 2 lineage) and IbAr10200 (Africa 3 lineage) represent distinct CCHFV lineages, with GPC amino acid identities of approximately 74.9% (Turkey vs. YL16070) and 84.5% (Turkey vs. IbAr10200), respectively 26,27 .
The T-cell responses to vaccination were assessed via interferon-γ (IFN-γ) ELISpot assays using splenocytes collected 7 days after booster immunization in each group (Fig. 1A). Compared with the LNP control mice, a statistically significant increase in spot-forming cells (SFCs) was observed in the mRNA-NP (1 μg or 5 μg) and mRNA-GPC (1 μg or 5 μg) groups in response to the NP-or GPC-specific peptide pools, respectively (Fig. 1I-K). In the mRNA-NP + mRNA-GPC groups, both the 1 μg and 5 μg dose groups presented significantly increased IFN-γ-positive SFCs compared with the NP, GPC, and NP + GPC peptide pools, except for the 1 μg dose group, which did not show a statistically significant increase in response to the NP peptide pools (Fig. 1K). Taken together, these data demonstrate that mRNA vaccination induces a robust antigen-specific cellular immune response.
Two-dose mRNA vaccination protects Ifnar1 -/-mice from crossgenotype CCHFV infection We next evaluated the protective efficacy of the vaccine candidates against lethal cross-genotype challenge with the CCHFV strain YL16070 in vivo. Comparative sequence analysis revealed antigenic divergence between the vaccine reference strain (Turkey) and the challenge isolate YL16070, with 96.06% amino acid identity in the NP and 74.93% in the GPC. To assess cross-protection, Ifnar1 -/-mice (n = 12 per group) were subjected to a subcutaneous prime-boost immunization regimen with a 21-day interval between the two doses (Fig. 1B). Two weeks after the booster dose, the mice were intraperitoneally inoculated with a lethal dose of CCHFV YL16070 (3,000 TCID 50 ). Of the twelve mice in each group, six were monitored daily for survival and body weight changes over a 14-day period, while the remaining six were sacrificed at 5 days post infection (d.p.i.) for viral load quantification and histopathology analysis (Fig. 2A). Following infection, the LNP control mice began losing body weight at 2 dpi and reached humane endpoints between 4-5 dpi (Fig. 2B-G). In contrast, mice subjected to either the high-dose or low-dose regimen of mRNA-NP or mRNA-NP + mRNA-GPC resulted in complete (100%) survival throughout the 14-day observation period (Fig. 2B,D), with vaccinated mice exhibiting no obvious clinical signs and maintaining relatively stable body weights after CCHFV infection (Fig. 2E,G). In the high-dose (5 μg) mRNA-GPC group, 33.33% (2/6) of the mice died from infection between 5-7 dpi, whereas the remaining 66.67% (4/6) survived. The surviving mice exhibited mild, transient weight loss (2.35-4.53%) between 4-7 dpi and fully recovered by 8-14 dpi (Fig. 2F). In contrast, the low-dose (1 μg) mRNA-GPC group presented reduced protective efficacy, with 83.33% (5/6) mortality occurring at 5-7 dpi (Fig. 2C). While the survival difference between dose groups was not statistically significant, both doses conferred significant protection compared to the LNP control (p = 0.0178 and p = 0.0034, respectively), and the higher dose achieved greater survival. Collectively, these findings demonstrate that mRNA-NP immunization elicits protective immunity against clinical disease progression in response to lethal cross-genotype CCHFV challenge.
Viral RNA loads in the blood, liver, and spleen of the 5 dpi euthanized mice were subsequently quantified. The data revealed that CCHFV RNA was undetectable in the blood of all the mice vaccinated with mRNA-NP or mRNA-NP + mRNA-GPC, with significantly lower viral loads than those of the LNP controls (Fig. 2H,J). In contrast, viral RNA was detected in the blood of 5/6 mice in the low-dose mRNA-GPC group and 3/6 in the highdose group, although both groups presented significantly lower viral loads than the control group (Fig. 2I). The results of tissue viral RNA quantification revealed a similar trend. In the high-dose mRNA-GPC group, all recipients (6/6) presented high viral loads in the liver, and viral RNA was detectable in the spleens of 50% (3/6) of these mice (Fig. 2L). The low-dose mRNA-GPC group presented relatively high viral loads in both the liver and spleen, indicating insufficient suppression of viral replication in these tissues. In contrast, mRNA-NP vaccination led to markedly reduced viral RNA levels in both dose groups (Fig. 2K). High-dose recipients presented no detectable viral RNA in either the liver or spleen, whereas low-dose recipients presented residual viral RNA in the liver (3/6) but undetectable levels in the spleen (0/6) (Fig. 2K). Notably, both the high-and low-dose mRNA-NP + mRNA-GPC regimens resulted in viral RNA levels reduced to below the detection limit in the liver and spleen (Fig. 2M). These findings suggest that while mRNA-GPC vaccination alone provides only partial control of CCHFV replication (Fig. 2L), mRNA-NP vaccination provides robust suppression of viral loads (Fig. 2K). Furthermore, the combination of mRNA-NP and mRNA-GPC further reduced the frequency of detectable viral RNA in liver tissue (3/6 in mRNA-NP vs. 0/6 in mRNA-NP + mRNA-GPC group; Fig. 2K,M), although this difference did not reach statistically significant.
To determine whether the observed reduction in viral loads correlated with decreased tissue damage, histopathological examinations and immunohistochemistry (IHC) staining for CCHFV NP antigen were performed on liver and spleen samples. Consistent with the viral load results, animals vaccinated with mRNA-NP or mRNA-NP + mRNA-GPC presented little to no histopathology signs of infection, with no detectable viral antigen. The only exception was mild inflammatory infiltration in the livers of the lowdose mRNA-NP group (Fig. 3A-D). In contrast, LNP control and mRNA-GPC-vaccinated mice presented obvious liver and spleen pathology at 5 dpi. Liver histopathology revealed hepatocellular necrosis, inflammatory cell infiltration, steatosis, and edema (Fig. 3A), accompanied by strong viral antigen positivity via IHC staining (Fig. 3B). Similarly, spleen sections exhibited disrupted splenic nodule architecture, reduction of small lymphocytes within white pulp, infiltration of a small number of neutrophils, and hemosiderin deposition (Fig. 3C). Viral antigens were also detected in the spleens, with a lower antigen-positive rate identified in the high-dose mRNA-GPC group than in the low-dose and LNP control groups, which was consistent with the viral load data (Fig. 3D). These findings indicate that mRNA-NP + mRNA-GPC vaccination effectively suppresses viral replication, significantly alleviates liver and spleen pathology, and provides robust protection against lethal cross-genotype CCHFV challenge.
## Prime-only mRNA vaccination protects against cross-genotype CCHFV infection
To further explore the correlation between the immunization regimen, RNA dosage, and protective effectiveness against cross-genotype CCHFV infection, the mice received a single immunization with 10, 5, or 2 μg mRNA and were challenged with 3000 TCID 50 CCHFV YL16070 five weeks later (Fig. 4A,B). Serum samples were collected up to 35 days postvaccination, and ELISA results revealed that mRNA-NP and mRNA-NP + mRNA-GPC vaccination induced NP-specific IgG antibody titers ranging from 10 4 -10 5 (Fig. 4C). The challenge results revealed that all vaccinated mice achieved 100% survival (Fig. 4D). Although the 2 μg mRNA-NP group experienced slight weight loss from 4-8 dpi, body weight began to recover by 9 dpi (Fig. 4E). Viral RNA quantification at 5 dpi revealed significantly reduced viral loads in the blood, livers, and spleens of all immunized groups (Fig. 4F-H), indicating that a single low-dose immunization effectively suppresses CCHFV replication. Among the regimens tested, 10 μg mRNA-NP + 10 μg mRNA-GPC resulted in the most effective viral suppression, with undetectable levels of viral RNA observed in both blood and tissues. The 10 μg mRNA-NP and 5 μg mRNA-NP + 5 μg mRNA-GPC groups also demonstrated low proportions of samples with detectable viral RNA (1/6 or 2/6, Fig. 4F-H). In contrast, the lower-dose groups (2 μg mRNA-NP, 1 μg mRNA-NP + 1 μg mRNA-GPC, and 5 μg mRNA-NP) presented moderate reductions in viral load. Notably, viral RNA levels in the liver at 5 dpi were significantly lower in the 10 μg mRNA-NP + 10 μg mRNA-GPC group than in the 2 μg mRNA-NP group (p = 0.0218, Fig. 4F). By 14 dpi, viral RNA was nearly undetectable in all immunized groups, except for a residual presence in the 2 μg mRNA-NP group (Fig. 4F-H). These findings highlight the superior efficacy of high-dose mRNA-NP vaccination for suppressing viral replication, with further enhancement observed when it is combined with mRNA-GPC.
Passive transfer of anti-NP antibodies confers partial protection in Ifnar1 -/-mice To evaluate the protective efficacy of NP-specific and GPC-specific antibodies induced by vaccination, passive immunization experiments were conducted. Serum was collected from Ifnar1 -/-mice vaccinated with mRNA-NP, mRNA-GPC, or mRNA-NP + mRNA-GPC, yielding antibody titers of approximately 10 6 , as determined by ELISA (Fig. 5B,C). Serumrecipient mice received 300 μL heat-inactivated immune serum via intraperitoneal injection, and 18 h later, blood was collected before challenge with 3000 TCID 50 CCHFV YL16070 (Fig. 5A). ELISA analysis revealed a substantial decrease in NP-specific IgG (Fig. 5D) and Gc-specific IgG (Fig. 5E) titers in the recipient mice. After challenge, the disease progression of the mice that received serum from mice in the mRNA-NP or mRNA-NP + mRNA-GPC groups was delayed, with the mean time to death (MTD) increased from 4 dpi in the LNP control animals to 5 dpi (Fig. 5F). Additionally, survival rates significantly increased from 0% in the control group to 33.33% (2/6) in the mRNA-NP and mRNA-NP + mRNA-GPC serumtreated groups (Fig. 5F). Two of the six mice in these groups experienced transient weight loss starting at 2 dpi but showed signs of recovery by 7-8 dpi (Fig. 5G). In contrast, all other mice either succumbed to infection or reached a humane endpoint (Fig. 5G). Although there were no statistically significant differences in viral loads among the blood, spleens, or livers between the groups, compared with the LNP control mice, the mice receiving mRNA-NP-immune serum presented 8.8-fold and 11.2-fold reductions in the viral loads in their livers and spleens, respectively (Fig. 5I). These data indicate that the passive transfer of mRNA-NP vaccinationinduced serum anti-NP antibodies provides partial but detectable protection against CCHFV infection.
## Discussion
Despite recent progress in CCHFV vaccine development, no mRNA-based candidate has been approved to date 28,29 . Two previous studies explored CCHFV mRNA vaccines: one used naked mRNA without delivery systems 28 , and the other applied LNP-encapsulated mRNA encoding separate NP or Gn/Gc antigens. However, both studies were limited by highdose regimens, and did not assess single-dose efficacy or cross-genotype protection. In our study, we developed LNP-formulated, nucleosidemodified mRNA vaccines encoding either NP, full-length GPC or both. Compared with previous approaches, our design incorporates several advancements. First, our GPC construct encodes the full glycoprotein precursor, which undergoes natural processing into Gn and Gc, potentially A Experimental scheme. C57BL/6 J and Ifnar1 -/-mice received a prime dose followed by a booster three weeks later. Blood and spleens were collected at different time points to assess humoral and cellular immune responses. B Study design. A two-dose regimen specifying antigen composition, dosage, and administration route. C, D CCHFV-specific IgG responses. NP-specific (C) and Gc-specific (D) IgG levels in C57BL/6 J mice were quantified by ELISA. E, F Pseudovirus neutralization. Serum collected two weeks post-boost from C57BL/6 J (E) and Ifnar1 -/-(F) mice was tested for neutralization against pseudotyped CCHFV (*GPC-VSVΔG/GFP). Neutralization efficacy was determined by quantifying GFP-expressing cells, and IC 50 values were calculated from serum dilution curves and presented on a logarithmic scale. G, H Live virus neutralization. Serum from C57BL/6 J (G) and Ifnar1 -/ -(H) mice was tested for neutralization against live CCHFV YL16070 in Vero E6 cells. IC 50 values were calculated from serum dilution curves based on viral inhibition, expressed on a logarithmic scale. I-K T-cell responses. IFN-γ-secreting splenocytes in C57BL/6 J mice immunized with mRNA-NP (I), mRNA-GPC (J), or mRNA-NP + mRNA-GPC (K) were quantified by ELISpot. Data represent cumulative spot-forming cells (SFCs) against the NP (I), GPC (J), and NP + GPC (K) peptide pools after background subtraction (DMSO-only controls), normalized to 10 6 splenocytes. Data are shown as mean ± SEM. Statistical significance for (I) was determined by one-way ANOVA with Tukey's multiple comparisons test. Statistical significance for (E-H, J, and K) was determined by Kruskal-Wallis test followed by Dunn's multiple comparisons test. Exact p values are shown.
improving antigen authenticity and epitope presentation. Second, we demonstrate robust protection using as little as 1-2 μg total mRNA, revealing clear dose-sparing potential not evaluated in earlier work. Third, we report strong efficacy of single-dose immunization, even a prime-only with low-dose mRNA-NP + mRNA-GPC or mRNA-NP provided complete survival, underscoring the potency of this platform for dose-sparing and rapid-response strategies. Fourth, we evaluated cross-genotype protection by immunizing with a Europe 1 strain-derived vaccine and challenging with a genetically distant Asia 2 strain, thus demonstrating crossgenotype efficacy not addressed in prior studies. Additionally, the passive transfer of vaccination-induced anti-NP antibodies conferred partial protection, further highlighting the role of the NP in immune defense.
These findings indicate that mRNA-NP, encoding the conserved NP antigen, plays a pivotal role in protective immunity against CCHFV. Moreover, coadministration of this formulation alongside mRNA-GPC further enhances vaccine efficacy, suggesting a complementary role for GPC in immune protection. Collectively, these findings provide valuable insight into the feasibility of mRNA-based strategies for CCHFV prevention and underscore the potential of NP-targeting mRNA vaccines as a foundation for future vaccine development.
Our data highlight the distinct protective roles of NP and GPC in mRNA vaccine-induced immunity against CCHFV. The GPC undergoes proteolytic processing to generate structural proteins (Gn and Gc), as well as non-structural proteins such as NSm, the mucin-like domain (MLD), and GP38 30,31 . The protein Gc, which is essential for viral attachment and entry 32,33 , is the primary target of neutralizing antibodies (nAbs) 34 , while GP38 has also been explored as a protective target in antibody-based therapies and vaccine development against CCHFV [35][36][37][38] . However, CCHFV GPC exhibits significant genetic diversity, with less than 75% amino acid conservation among strains 3 . This high level of genetic variability raises concerns about potential immune evasion. Consistent with the results of prior studies on DNA and replicating RNA (repRNA) vaccines 14,38 , our mRNA-GPC vaccine provides only partial protection against crossgenotype CCHFV challenge. In contrast, a vesicular stomatitis virus (VSV)-vector vaccine targeting GPC has demonstrated complete crossgenotype protection 39 . Moreover, modified vaccinia virus Ankara (MVA)and adenovirus-vector vaccines have conferred full protection against homologous CCHFV challenge 13,40 , although their cross-protective potential remains unclear. Notably, even when DNA vaccines match the challenge strain, complete protection is not always achieved 41 , suggesting that GPCbased immunity may be influenced by platform-specific immune responses. Interestingly, a virus-like particle (VLP)-based vaccine induced higher neutralizing antibody titers than a DNA vaccine but exhibited a 60% lower survival rate 37 , whereas subunit vaccines targeting Gn or Gc ectodomains elicited strong neutralizing antibody responses but failed to protect against lethal challenge 42 . These findings indicate that protection does not directly correlate with neutralizing antibody titers and that humoral immunity alone is insufficient for effective CCHFV protection. Consistent with previous studies in mice, our data confirm that the mRNA-GPC vaccine alone induces a strong humoral response but fails to provide adequate protection against CCHFV challenge. To further assess the protective potential of vaccination-induced antibodies, heat-inactivated serum from mRNA-GPC-vaccinated mice was passively transferred to naive mice. However, no protective effect was observed post-challenge, which aligns with findings from MVA-GP vaccine studies 43 . Notably, despite donor serum exhibiting ELISA titers as high as 10⁶, recipient mice displayed a sharp decline in antibody levels post-transfer, with titers decreasing to 50-450. While this reduction likely results from antibody dilution and clearance, it complicates assessment regarding whether the lack of protection is solely due to the insufficiency of humoral immunity. Furthermore, studies regarding the MVA-GP vaccine have demonstrated that effective protection requires the transfer of both antibodies and T-cells from vaccinated mice 43 , indicating essential roles of both humoral and cellular immunity in vaccine-mediated protection.
Compared with GPC, CCHFV NP exhibits a greater degree of conservation 3 . Consistent with our findings and those of prior studies on DNA-based and repRNA vaccines, NP-based immunization alone is sufficient to confer complete protection in mice 14,44 . Notably, repNP + repGPC vaccination has been shown to provide full protection independent of CD8⁺ T-cells, whereas the survival rate decreases to 40% in B-cell-deficient (μMT) mice, suggesting a crucial role for humoral immunity 14 . In support of this, recent studies have demonstrated that a monoclonal antibody targeting the NP confers protection against lethal CCHFV challenge 45 , and the passive transfer of NP-immune sera achieves 40%-50% protection, with higher doses correlating with improved survival 46 . Unlike traditional viral neutralizing antibodies, NP-specific antibodies do not block viral entry but instead restrict viral replication intracellularly through Fc-dependent, TRIM21-mediated proteasomal degradation of NP-containing virus complexes 46 . The results of our passive transfer experiments align with these findings. Despite a significant decrease in NP antibody titers post-transfer (from 10⁶ in donor serum to 10³ in recipient serum), survival improved significantly (~33.33%). This finding suggests that a portion of the NP antibodies may have entered cells and contributed to intracellular antiviral effects. However, the precise mechanism of cellular entry remains unclear and requires further investigation.
Although our study provides important insights into mRNA vaccineinduced immunity against CCHFV, several limitations should be acknowledged. First, our experiments were conducted in Ifnar1 -/-mice, which lack type I interferon signalling. While this model is widely used for CCHFV research because of its susceptibility to infection, it may not fully recapitulate human immune responses. Future studies should assess vaccine efficacy in more physiologically relevant models, such as nonhuman primates, to better predict human immunogenicity and protection. Second, although we evaluated the persistence of vaccination-induced IgG antibodies, we did not perform longitudinal challenge studies to assess the durability of protection. Protection against viral challenge is a critical factor in vaccine development, and further studies are necessary to determine whether long-term immunity requires booster doses. Finally, while our findings suggest that mRNA-NP confers robust protection, the exact mechanisms underlying NP-mediated immunity remain incompletely understood. Elucidating these mechanisms will be critical for optimizing CCHFV mRNA vaccine design and evaluating its importance for broad and durable protection.
Overall, our findings demonstrate that mRNA vaccines encoding NP and GPC elicit strong immune responses and protect against lethal CCHFV challenge. Both the two-dose and prime-only regimens were effective, and the passive transfer of anti-NP antibodies provided partial protection. Collectively, these results highlight the potential of mRNA vaccines for CCHFV prevention and establish a foundation for further preclinical and clinical development.
## Methods
## Ethics statement
All animal studies were reviewed and approved by the Animal Ethics Committee of the Wuhan Institute of Virology, Chinese Academy of Sciences (Ethics number: WIVA42202306). Experimental procedures involving CCHFV were conducted within the Animal Biosafety Level 3 (ABSL-3) facility at the National Biosafety Laboratory (Wuhan), Chinese Academy of Sciences. All procedures, including subcutaneous and intraperitoneal injections, retro-orbital blood collection, intracerebral injection in neonatal mice, and euthanasia, were performed under appropriate anesthesia. Adult mice were anesthetized with isoflurane before all procedures. Neonatal mice (1-2 days old) undergoing intracerebral injection were anesthetized via brief hypothermia, induced by placement on ice for 2-3 min, in accordance with institutional animal care protocols. Euthanasia of adult mice was conducted under deep isoflurane anesthesia followed by cervical dislocation, while neonatal mice were euthanized by decapitation following hypothermia-induced anesthesia. All procedures were performed in accordance with institutional ethical guidelines and the AVMA Guidelines for the Euthanasia of Animals (2020 Edition), ensuring animals were fully unconscious and insensible prior to death. All efforts were made to minimize animal suffering.
Cells, viruses, and antibody HEK293T (ATCC: ACS-4500) and Vero E6 (ATCC: CRL-1586) cells were maintained in Dulbecco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% Fetal bovine serum (FBS, Gibco), 100 units/mL penicillin, and 100 μg/mL streptomycin, at 37 °C in a 5% CO 2 . 293 Freestyle (293 F, a kind gift from Dr. Rui Gong; Wuhan Institute of Virology, CAS) cells were maintained in FreeStyle 293 expression medium supplemented with 100 units/mL penicillin and 100 μg/mL streptomycin, in shaker incubators set at 150 rpm, 37 °C, and 8% CO₂. The CCHFV strain YL16070 (GenBank accession numbers KY354080, KY354081, KY354082) was propagated by intracranial injection into 1-2-day-old KM suckling mice. Once all symptomatic mice were identified, they were immediately euthanized. The brains were harvested and stored in 50% glycerol at -80 °C. Following the three washes with PBS, the brains were homogenized in 1 mL of PBS, and the virus was subsequently extracted and purified by centrifugation 6000 × g for 10 min at 4 °C. Viral titers (expressed as TCID 50 /mL) were quantified by an indirect immunofluorescence assay (IFA) on Vero E6 cells. Briefly, tenfold serial dilution of the virus were inoculated onto Vero E6 monolayer cells in the 96-well plate and incubated at 37 °C with 5% CO₂ for 4 days. The cells were then fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 5% BSA for 30 min, followed by incubation with CCHFV-specific primary antibodies and subsequently with fluorescent dye-labeled secondary antibody. Fluorescent foci were visualized under an EVOS fluorescence microscope (Invitrogen), and TCID₅₀ values were calculated by the Reed-Muench method. The primary antibody against CCHFV NP was produced in-house by immunizing mice with mRNA-NP, followed by collection of blood three weeks after the booster dose; serum containing anti-NP polyclonal antibodies was obtained by centrifugation at 3000 × g for 10 min at 4 °C. The highly cross-adsorbed goat anti-mouse IgG (H + L) secondary antibody was obtained from Thermo Scientific (Cat# A-11001).
## Synthesis and characterization of CCHFV mRNA
CCHFV mRNA vaccine was prepared as previously described 47 . In brief, the mRNAs encoded codon-optimized open reading frames of the CCHFV GPC (GenBank accession no. ARB18232) or NP (GenBank accession no. ARB18231), respectively. The constructs included 5' untranslated region (UTR), 3' UTR (see Supplementary Data 2 for sequences), and a 100-nucleotide poly(A) tail. mRNAs were synthesized via in vitro transcription (all reagents purchased individually from Hongene) using a linearized DNA template, with 100% substitution of uridine by N1-methylpseudouridine and incorporation of a cotranscriptional a cap structure (m7G (5') ppp (5') (2'OMeA) pG). The synthesized mRNA was then purified using lithium chloride precipitation and resuspended in RNase-free water for further analysis and application. The concentration and integrity of CCHFV mRNA were measured using a NanoDrop One (Thermo) and 5200 Bioanalyzer (Agilent), respectively.
Formulation and characterization of mRNA-LNP mRNA-LNPs were prepared using a modified procedure. Lipid components were dissolved in ethanol at a molar ratio of 47:10:41.5:1.5 (ionizable lipid CS21001: DSPC: cholesterol: PEG-lipid). The lipids were rapidly mixed with mRNA dissolved in 25 mM sodium acetate buffer (pH4.0) at an N/P ratio of 6, using a 3:1 (aqueous: ethanol) volume ratio at a flow rate of 12 mL/ min, via a microfluidic system (INanoTML from Micro&Nano Biologics). The resulting sample was diluted with 20 mM Tris, 10.7 mM sodium acetate buffer (pH7.5), then concentrated by 30 kDa Ultra Centrifugal Filters and sterile-filtered through a 0.2 μm filter. Dynamic light scattering (Malvern Panalytical Zetasizer Pro) was employed to measure particle size, polydispersity index (PDI), and zeta potential. Encapsulation efficiency was determined using the Quant-iT RiboGreen RNA assay kit (Invitrogen) as previously described 47 .
Mice, vaccinations, and infection C57BL6/J mice aged 6-8 weeks were purchased from Vital River Laboratories (Beijing, China). In both single-dose and double-dose groups, mice were inoculated subcutaneously with varying immunogens (NP, GPC, and NP + GPC) (n = 6 per group) at different doses of 1 μg, 5 μg, or empty LNP as a control. For double-dose groups, a booster of the same dose was given on day 21 post-initial immunization. To assess cellular immune response, a dedicated cohort of mice (n = 6 per group) were euthanized on day 7 after the boost, and spleens were collected for assessment of cellular immune responses by ELISpot (IFN-γ). To evaluate humoral immune responses, a separate cohort of immunized mice (n = 6 per group) was monitored longitudinally, with blood samples collected at designated timepoints for ELISA and neutralization assay.
Ifnar1 -/-mice (C57BL/6 J background) were purchased from Gem-Pharmatech Co., Ltd. (Nanjing, China). In the CCHFV challenge experiment, both single-dose and double-dose immunized, along with LNP control (n = 12) were bled and intraperitoneally infected with 3,000 TCID 50 of the CCHFV YL16070 strain in a total volume of 100 μL. This challenge occurred four weeks after the first immunization in the single-dose groups and two weeks after the second immunization in the double-dose groups. Following infection, mice were randomly divided into two subgroups (n = 6): one subgroup was euthanized at 5 dpi for the collection of blood, liver, and spleen samples for viral loads quantification and pathological analysis, while the other subgroup was monitored for survival and body weight changes over a 14-day period. Mice were weighed daily and monitored for clinical symptoms. Humane endpoint criteria were defined as a ≥ 20% body weight loss from baseline or the presence of severe clinical signs such as lethargy, hunched posture, ruffled fur, or impaired mobility. Animals meeting these criteria were humanely euthanized in accordance with institutional animal welfare guidelines. In the single-dose groups, samples were also collected on day 14 post-infection from surviving animals.
## Western blot
For protein expression analysis, HEK293T cells were seeded into 6-well plates and transfected with 2 μg of LNP-formulated mRNA-NP or mRNA-GPC by directing adding it to the medium. After 24 h of incubation at 37 °C in 5% CO 2 , cells were washed with PBS and lysed in 200 μL of RIPA buffer (Beyotime, China) supplemented with a protease inhibitor cocktail. The lysates were incubated on ice for 30 min and then clarified by centrifugation at 12,000 × g for 15 min at 4 °C. Equal amounts of total proteins were resolved on a 10% SDS-PAGE gel and electrotransferred onto PVDF membrane (Millipore) using a semi-dry transfer system. Membrane was blocked with 5% skim milk in TBST (20 mM Tris-HCl, 150 mM NaCl, 0.1% Tween-20) at room temperature for 1 h, followed by incubation overnight at 4 °C with mouse anti-CCHFV NP (in-house generated by mRNA-NP immunization) or anti-Gc primary antibodies (a generous gift from Dr. Fei Deng, Wuhan Institute of Virology, CAS). After three washes in TBST, the membrane was incubated with HRP-conjugated goat anti-mouse IgG secondary antibody (Proteintech) for 1 h at room temperature. Signal was developed using an enhanced chemiluminescence (ECL) detection kit (Thermo Scientific) and imaged with a Tanon 5200 imaging system (Tanon, China). GAPDH (Abcam) was used as a loading control.
## Expression and purification of CCHFV Gc and NP proteins
Codon-optimized coding sequences for the Gc ectodomain (residues 1053-1573 of the full-length M segment; GenBank accession no. KR864902.1) and full-length NP (GenBank accession no. KY362517.1) of the CCHFV Turkey strain were cloned into the mammalian expression vector pcDNA3.4, each bearing a C-terminal His-tag for affinity purification. To enable secretion of the normally intracellular NP protein, an N-terminal interleukin-10 (IL-10) signal peptide was added to the NP construct. Plasmid constructs were transiently transfected into 293 Freestyle (293 F) suspension cells using polyethyleneimine (PEI; Sigma-Aldrich) in accordance with the manufacturer's instructions. The cells were cultured in FreeStyle™ 293 Expression Medium (Gibco) under conditions of 37 °C, 8% CO 2 , and 150 rpm shaking. Five days post-transfection, the supernatants were collected by centrifugation at 1000 rpm for 30 min to remove cellular debris. The clarified supernatants were passed through a 0.45 μm filter. Gc and NP proteins were purified via Ni Sepharose High Performance resin (Cytiva), following the manufacturer's protocol for His-tagged protein purification.
## ELISA
The binding levels of serum IgG antibodies specific to CCHFV Gc and NP proteins was assessed by ELISA. Briefly, 96-well EIA/RIA microplates (Corning) were coated with 200 ng per well of Gc or NP protein in 1× ELISA coating buffer (Solarbio, Cat# C1055), prepared by diluting the 10× stock with deionized water, and incubated overnight at 4 °C. Plates were blocked the next day with 100 μL of PBS containing 5% skim milk at 37 °C for 1 h. After washing with PBS supplemented with 0.1% Tween-20 (PBST), serial three-fold dilutions of mouse serum (starting at 1:50) were prepared in blocking buffer and added to the wells, followed by incubation at 37 C for 2 h. Subsequently, plates were washed four times with PBST and incubated with HRP-conjugated goat anti-mouse IgG secondary antibody (Abcam) at 37 °C for 1.5 h. After a final wash, 100 μL of TMB substrate solution (Beyotime) was added and incubated for 8 min in the dark for color development. The substrate reaction was stopped by adding 50 μL of stop solution (Beyotime), and absorbance was measured at 450 nm (OD 450 ) using a microplate reader. The endpoint titer is defined as the reciprocal of the highest serum dilution at which the OD₄₅₀ value equals or exceeds the positivity threshold. The threshold was calculated as 2.1 times the OD₄₅₀ of blank wells (wells with no serum), consistent with several published studies 48 .
## Enzyme-linked immunospot (ELISpot) assay
The Mouse IFN-γ ELISpot kit (Mabtech) was conducted to evaluate CCHFV-specific T-cell responses according to the manufacturer's instructions. The CCHFV NP peptide pool (comprising eight peptides, each with a length of 18-20 amino acids, corresponding to NP pools #1 and #4) and the GPC peptide pool derived from the Gc region only (comprising four peptides, each with a length of 18-20 amino acids, corresponding to Gc pool #2), as described in the reference 29 , were obtained from Genscript and dissolved in DMSO (Sigma). Splenocytes were harvested from C57BL/6 J mice on day 7 after the booster immunization with mRNA vaccines. Freshly isolated splenocytes (5 × 10 5 cells per well) were plated into pre-coated ELISpot plate and stimulated with 250 ng of peptide pool. The plates were incubated at 37 °C with 5% CO 2 for 20 h. For positive controls, cells were treated with phorbol 12-myristate 13-acetate (PMA) and ionomycin (Dakewe), while unstimulated cells served as negative controls. Spotforming cells (SFCs) were analyzed using a CTL ImmunoSpot Analyzer and ImmunoSpot software (Cellular Technology Ltd). SFCs counts were normalized to 10 6 splenocytes.
Neutralization assay based on pseudotyped CCHFV CCHFV Hoti GPC (accession number EU037902.1), with a C-terminal deletion of 53 amino acids, was subcloned into the pcDNA3.1 expression vector to generate pcDNA3.1-Hoti-GPCdel53aa plasmid. HEK 293 T cells were plated in 6-well plate (NEST) in DMEM supplemented with 10% FBS and incubated at 37 °C with 5% CO 2 for 18 h prior to transfection. Cells were then transfected with 3 μg of pcDNA3.1--Hoti-GPCdel53aa per well using Lipofectamine 3000 Transfection Reagent (Thermo Scientific). Following a 24-hour transfection period, cells were infected with *G-VSVΔG/GFP (a generous gift from Dr Manli Wang, Wuhan Institute of Virology, CAS) and incubated at 37 °C for 2 h. The monolayers were subsequently washed three times with serum-free DMEM and replenished with fresh complete DMEM (containing 10% FBS). After 48 h, Supernatants were collected, centrifuged, and filtered through a 0.22 μm filter to remove debris. The pseudotyped CCHFV (*GPC-VSVΔG/GFP) was stored at -80 °C and titrated using an end-point dilution assay. For neutralization assays, serum samples were heat-inactivated at 56 °C for 30 min, then serially diluted threefold in DMEM containing 2% FBS and incubated with an equal volume of pseudovirus at 37 °C for 1 h. After incubation, 100 μL of each serum-virus mixtures were added to confluent Vero E6 cell monolayers in black 96-well plates (Corning). Plates were incubated for 24 h at 37 °C with 5% CO 2 . Infection was assessed by quantifying GFP-positive cells using a fluorescence imaging system (EVOS M7000), and neutralizing antibody titers were determined accordingly.
Neutralization assay based on live CCHFV Vero E6 cells were seeded into 24-well plates (NEST) with DMEM supplemented with 10% FBS and incubated at 37 °C with 5% CO 2 for 18 h to allow monolayer formation. Serum samples were heat-inactivated at 56 °C for 30 min and prepared as a series of threefold dilutions (starting from a 1:50 dilution) in DMEM containing 2% FBS. Equal volumes of diluted sera were mixed with 100 PFU of CCHFV strain YL16070 and incubated at 37 °C for 1 h to enable neutralization. After incubation, 100 μL of each serumvirus mixture was added to Vero E6 monolayers and adsorbed for 1 h at 37 °C. Following this, the inoculum was removed, and wells were overlaid with DMEM containing 2% FBS and 0.8% carboxymethylcellulose sodium (Sigma) at 37 °C with 5% CO 2 for 6 days. Cells were fixed with 8% formaldehyde (Sigma) for 30 min, permeabilized with PBS containing 0.1% Triton X-100 for 25 min, and then blocked in PBS containing 5% bovine serum albumin (BSA) for 30 min. Wells were then incubated with mouse anti-CCHFV NP antibody diluted in PBS with 1% BSA for 2 h at room temperature. After three washes with TBST, plates were incubated with HRP-conjugated goat anti-mouse IgG (Beyotime) in PBS with 1% BSA for 1 h. Following additional TBST washes, plaques were developed using DAB substrate and hydrogen peroxide from the Enhanced HRP-DAB Chromogenic Kit (TIANGEN), with incubation at room temperature in the dark for 5-30 min. Plaques were visualized and counted, and neutralization titers were determined accordingly.
## CCHFV RNA detection
Whole blood, spleen, and liver samples were collected from infected mice for viral RNA quantification. Each tissue was weighed and mechanically homogenized in 1 mL of DMEM using a tissue grinder. The homogenates were clarified by centrifugation at 5000 × g for 20 min at 4 °C. A 140 μL aliquot of the supernatant was used for RNA isolation using the QIAamp Viral RNA Mini Kit (Qiagen), following the manufacturer's instructions. For quantitative detection, 2 μL of RNA was used as a template in a one-step real-time reverse transcription PCR (qRT-PCR), performed using the HiScript II One Step qRT-PCR SYBR Green Kit (Vazyme). Primers specific to the CCHFV S segment were used as follows: Forward primer: 5'-TCAAGTGGAGGAAGGACATAGG-3'; Reverse primer: 5'-TCCA-CATGTTCACGGCTCACTGGG'. Amplification reactions were run on a CFX96 Real-Time PCR System (Bio-Rad) under the following cycling conditions: reverse transcription at 50 °C for 15 min, initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. To quantify viral RNA copy numbers, a standard curve was generated using an RNA standard prepared by in vitro transcription of a known-length CCHFV NP fragment. The RNA standard was serially diluted 10-fold (10¹-10⁹ copies/μL) and used as the template in qRT-PCR to construct the standard curve. Sample Ct values were converted to RNA copy numbers based on this curve. Results were expressed as copies per μL of blood or copies per gram of tissue.
## Histopathology and immunohistochemistry
Liver and spleen tissues were fixed in 8% formaldehyde (Sigma) for 7 days at room temperature, followed by routine dehydration and paraffin embedding. Tissue blocks were sectioned at a thickness of approximately 4 μm and mounted onto glass slides. Hematoxylin and eosin (H&E) staining was performed for histopathological evaluation under a light microscope.
Multiple regions from each organ were examined to ensure representative sampling. For immunohistochemical (IHC) detection of viral antigen, tissue sections were deparaffinized and subjected to antigen retrieval prior to incubation with an in-house mouse monoclonal antibody against CCHFV NP. A biotinylated goat anti-mouse IgG (SeraCare) was applied as the secondary antibody. Visualization was achieved using DAB chromogen followed by hematoxylin counterstaining. Images were acquired using the Pannoramic MIDI digital slide scanner (3DHISTECH, Budapest, Hungary).
## Statistical analysis
All statistics were done using GraphPad Prism 9.0 software. Data normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was assessed using the Brown-Forsythe test. For datasets that satisfied both assumptions (normal distribution and equal variances), one-way ANOVA with Tukey's multiple comparisons test was used. For non-normally distributed or heteroscedastic datasets, the Kruskal-Wallis test with Dunn's multiple comparisons correction was applied. For longitudinal IgG antibody response comparisons between dose groups, two-way mixed-effects models (REML) with Geisser-Greenhouse correction and Tukey's multiple comparisons test were employed. Survival data were analyzed using the Logrank (Mantel-Cox) test. A p-value < 0.05 was considered statistically significant.
## References
1. Hawman, Feldmann (2018) "Recent advances in understanding Crimean-Congo hemorrhagic fever virus" *F1000Res*
2. Aslam, Abbas, Alsayeqh (2023) "Distribution pattern of Crimean-Congo Hemorrhagic Fever in Asia and the Middle East" *Front Public Health*
3. Bente (2013) "Crimean-Congo hemorrhagic fever: history, epidemiology, pathogenesis, clinical syndrome and genetic diversity" *Antivir. Res*
4. Whitehouse (2004) "Crimean-Congo hemorrhagic fever" *Antivir. Res*
5. Ergönül (2006) "Crimean-Congo haemorrhagic fever" *Lancet Infect. Dis*
6. Garrison (2020) "ICTV virus taxonomy profile: Nairoviridae" *J. Gen. Virol*
7. Hoogstraal (1979) "The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe, and Africa" *J. Med. Entomol*
8. Ergönül (2004) "Characteristics of patients with Crimean-Congo hemorrhagic fever in a recent outbreak in Turkey and impact of oral ribavirin therapy" *Clin. Infect. Dis*
9. Swanepoel (1989) "The clinical pathology of Crimean-Congo hemorrhagic fever" *Rev. Infect. Dis*
10. Who (2018) "Prioritizing diseases for research and development in emergency contexts"
11. Overby, Pettersson, Grünewald et al. (2008) "Insights into bunyavirus architecture from electron cryotomography of Uukuniemi virus" *Proc. Natl. Acad. Sci. USA*
12. Hawman (2023) "Accelerated DNA vaccine regimen provides protection against Crimean-Congo hemorrhagic fever virus challenge in a macaque model" *Mol. Ther*
13. Saunders (2023) "Adenoviral vectored vaccination protects against Crimean-Congo Haemorrhagic Fever disease in a lethal challenge model" *EBioMedicine*
14. Leventhal (2022) "Replicating RNA vaccination elicits an unexpected immune response that efficiently protects mice against lethal Crimean-Congo hemorrhagic fever virus challenge" *EBioMedicine*
15. Tipih, Burt (2020) "Crimean-Congo hemorrhagic fever virus: advances in vaccine development" *Biores Open Access*
16. Pardi, Hogan, Porter et al. (2018) "mRNA vaccinesa new era in vaccinology" *Nat. Rev. Drug Discov*
17. Chen (2022) "Safety and immunogenicity of the SARS-CoV-2 ARCoV mRNA vaccine in Chinese adults: a randomised, double-blind, placebo-controlled, phase 1 trial" *Lancet Microbe*
18. Feldman (2019) "mRNA vaccines against H10N8 and H7N9 influenza viruses of pandemic potential are immunogenic and well tolerated in healthy adults in phase 1 randomized clinical trials" *Vaccine*
19. Wilson (2023) "Efficacy and Safety of an mRNA-Based RSV PreF Vaccine in Older Adults" *N. Engl. J. Med*
20. Qiu (2022) "Development of mRNA vaccines against respiratory syncytial virus (RSV)" *Cytokine Growth Factor Rev*
21. Alberer (2017) "Safety and immunogenicity of a mRNA rabies vaccine in healthy adults: an open-label, non-randomised, prospective, firstin-human phase 1 clinical trial" *Lancet*
22. Aldrich (2021) "Proof-of-concept of a low-dose unmodified mRNAbased rabies vaccine formulated with lipid nanoparticles in human volunteers: a phase 1 trial" *Vaccine*
23. Mucker (2024) "Comparison of protection against mpox following mRNA or modified vaccinia Ankara vaccination in nonhuman primates" *Cell*
24. Chaudhary, Weissman, Whitehead (2021) "mRNA vaccines for infectious diseases: principles, delivery and clinical translation" *Nat. Rev. Drug Discov*
25. Essink (2023) "The safety and immunogenicity of two Zika virus mRNA vaccine candidates in healthy flavivirus baseline seropositive and seronegative adults: the results of two randomised, placebocontrolled, dose-ranging, phase 1 clinical trials" *Lancet Infect. Dis*
26. D'addiego (2024) "Development of targeted whole genome sequencing approaches for Crimean-Congo haemorrhagic fever virus (CCHFV)" *Virus Res*
27. Guo (2017) "A new strain of Crimean-Congo hemorrhagic fever virus isolated from Xinjiang" *Virol. Sin*
28. Aligholipour Farzani (2019) "Immunological analysis of a CCHFV mRNA vaccine candidate in mouse models" *Vaccines*
29. Appelberg (2022) "Nucleoside-modified mRNA vaccines protect IFNAR(-/-) mice against crimean-congo hemorrhagic fever virus infection" *J. Virol*
30. Sanchez, Vincent, Erickson et al. (2006) "Crimeancongo hemorrhagic fever virus glycoprotein precursor is cleaved by Furin-like and SKI-1 proteases to generate a novel 38-kilodalton glycoprotein" *J. Virol*
31. Freitas (2020) "The interplays between Crimean-Congo hemorrhagic fever virus (CCHFV) M segment-encoded accessory proteins and structural proteins promote virus assembly and infectivity" *PLoS Pathog*
32. Hulswit, Paesen, Bowden et al. (2021) "Recent Advances in Bunyavirus Glycoprotein Research: Precursor Processing, Receptor Binding and Structure"
33. Sanchez, Vincent, Nichol (2002) "Characterization of the glycoproteins of Crimean-Congo hemorrhagic fever virus" *J. Virol*
34. Bertolotti-Ciarlet (2005) "Cellular localization and antigenic characterization of crimean-congo hemorrhagic fever virus glycoproteins" *J. Virol*
35. Li (2024) "Neutralizing monoclonal antibodies against the Gc fusion loop region of Crimean-Congo hemorrhagic fever virus" *PLoS Pathog*
36. Golden (2019) "GP38-targeting monoclonal antibodies protect adult mice against lethal Crimean-Congo hemorrhagic fever virus infection" *Sci. Adv*
37. Hinkula (2017) "Immunization with DNA plasmids coding for crimeancongo hemorrhagic fever virus capsid and envelope proteins and/or virus-like particles induces protection and survival in challenged mice" *J. Virol*
38. Suschak (2021) "A CCHFV DNA vaccine protects against heterologous challenge and establishes GP38 as immunorelevant in mice" *Vaccines*
39. Rodriguez (2019) "Vesicular stomatitis virus-based vaccine protects mice against Crimean-Congo hemorrhagic fever" *Sci. Rep*
40. Buttigieg (2014) "A novel vaccine against Crimean-Congo Haemorrhagic Fever protects 100% of animals against lethal challenge in a mouse model" *PLoS One*
41. Garrison (2017) "A DNA vaccine for Crimean-Congo hemorrhagic fever protects against disease and death in two lethal mouse models" *PLoS Negl. Trop. Dis*
42. Kortekaas (2015) "Crimean-Congo hemorrhagic fever virus subunit vaccines induce high levels of neutralizing antibodies but no protection in STAT1 knockout mice. Vector Borne Zoonotic Dis"
43. Dowall (2016) "Protective effects of a modified vaccinia ankarabased vaccine candidate against Crimean-Congo haemorrhagic fever virus require both cellular and humoral responses" *PLoS One*
44. Aligholipour Farzani (2019) "Co-delivery effect of CD24 on the immunogenicity and lethal challenge protection of a DNA vector expressing nucleocapsid protein of Crimean Congo hemorrhagic fever virus" *Viruses*
45. Garrison (2024) "Nucleocapsid protein-specific monoclonal antibodies protect mice against Crimean-Congo hemorrhagic fever virus" *Nat. Commun*
46. Leventhal (2024) "Antibodies targeting the Crimean-Congo Hemorrhagic Fever Virus nucleoprotein protect via TRIM21" *Nat. Commun*
47. Wang (2022) "mRNA based vaccines provide broad protection against different SARS-CoV-2 variants of concern" *Emerg. Microbes Infect*
48. Lu (2023) "Both chimpanzee adenovirus-vectored and DNA vaccines induced long-term immunity against Nipah virus infection" *NPJ Vaccines*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724335&blobtype=pdf
|
# The glycosylation variant at residue 381 of the spike protein contributes to virulence shifts in porcine epidemic diarrhea virus during both natural field transmission and laboratory cell passaging with poor cross-protection
Zhiwei Li, Zhiqian Ma, Yongqi Li, Xiaojing Zhao, Yonghui Zheng, Yang Li, Yingtong Feng, Xuyang Guo, Zifang Zheng, Lele Xu, Jianwu Zhang, Haixue Zheng, Shuqi Xiao, Yongqi Li
## Abstract
The virulence and immunogenicity of porcine epidemic diarrhea virus (PEDV) vary during field circulation and cell culture passage (such as when the GI CV777 strain is attenuated through serial passaging). This study revealed that the glycosylation site mutation at position 381 (N381K) of the S protein is associated with these phenomena. Compared with piglets inoculated with P13 virus, piglets inoculated with P100 (N381K) of virulent GX223 exhibited delayed diarrhea, viral shedding, and mortality. Using the virulent rCH/SX/2016-S HNXP strain (rPEDV-S wt ) as the backbone, we generated rPEDV-S N381K . While both the wild-type and mutant strains showed similar growth in vitro and in 2-day-old piglets, rPEDV-S N381K caused milder diarrhea and lower mortality. In 5-day-old piglets, the mutant strain also induced delayed viral shedding and milder diarrhea. At 21 days after post-infection, all the piglets were challenged with the parental strain. The pigs in both the rPEDV-S N381K and rPEDV-S wt groups produced high IgA/IgG levels, but the rPEDV-S N381K -inoculated piglets presented higher fecal viral loads and lower neutralizing antibody titers against the parental strain. Molecular modeling suggests that N381K alters antigenic epitope interactions, which may affect virulence and immunogenicity. While this study has the limitation of a relatively small sample size in the animal studies, the results collectively demonstrate that S protein glycosy lation mutations influence PEDV virulence and contribute to reduced cross-protective vaccine efficacy, offering important insights for PEDV pathogenesis research and vaccine development.IMPORTANCE Porcine epidemic diarrhea virus (PEDV) continues to cause substantial economic losses in the global swine industry, with emerging strains challenging existing vaccine strategies. This study identifies the N381K glycosylation site mutation in the S protein of PEDV as a factor involved in variations in virulence during natural transmis sion and laboratory adaptation. Crucially, the mutant induces suboptimal neutralizing immunity against the prevalent strain, revealing a mechanism by which classical-strain vaccines may provide limited protection against currently circulating strains. Our findings reveal how a single glycan modification modulates both pathogenicity and immunogenicity, providing critical insights for the development of effective vaccines against circulating PEDV variants.
P orcine epidemic diarrhea virus (PEDV) is a single-stranded positive-sense RNA virus of the genus Alphacoronavirus (1). The approximately 28-kb viral genome consists of the 5′ untranslated region (5′ UTR), open reading frame 1a/1b (ORF1a/1b) encoding replicase polyproteins pp1a and pp1ab, the spike (S) gene, the accessory protein-encod ing gene (ORF3), the envelope (E) gene, the membrane (M) gene, the nucleocapsid (N) gene, the 3′ untranslated region (3′ UTR), and the poly(A) tail (2). PEDV has evolved into two major genogroups: GI (classical) and GII (variant). These genogroups further diversified into subgroups: GI split into GI-a and GI-b, while GII subdivided into GIIa, GII-b, and GII-c (3). The GI-a subgroup predominantly comprised early European strains, including the virulent CV777 and DR13 prototypes. In contrast, GI-b contains cell culture-adapted vaccine strains (attenuated CV777 and DR13) along with pandemic classical strains circulating in Asia, and most GI-b PEDV strains are vaccine strains or vaccine-like strains with attenuated virulence (3,4).
Coronaviruses, as RNA viruses, exhibit high genomic variability, which significantly influences their transmissibility, virulence, and immunogenicity (5)(6)(7)(8). PEDV was first identified in Belgium and the United Kingdom in 1976, with subsequent reports emerging from China during the 1980s, where it exhibited a sporadic and geographically limited distribution (9,10). However, severe epidemics occurred in 2010, when highly virulent variant PEDV strains emerged in China (11). These highly virulent variants rapidly disseminated globally, resulting in near-catastrophic impacts on the swine industry, but the mechanism underlying the enhanced virulence of the variant strains remains unclear. Live attenuated vaccines (LAVs) are widely adopted for disease prevention because of their superior immunogenicity and protective efficacy (12). Commercial PEDV LAVs (such as those from the CV777 and DR13 strains) are typically generated through serial passaging in nonporcine cell lines (such as VERO cells), although the molecular mechanisms underlying their attenuation remain poorly characterized. However, the efficacy of classical-strain vaccines against currently circulating epidemic strains has been proven inadequate (13). The mechanisms responsible for this limited cross-protection remain to be elucidated.
The S protein of coronaviruses is a highly glycosylated multifunctional protein that plays critical roles in viral entry, replication, virulence, and tropism (14)(15)(16)(17). Deletion of both N331 and N343 glycosylation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) S protein reduced viral infectivity. The N234Q and N165Q mutations altered viral sensitivity to neutralizing antibodies (18). Mutations in the N-terminal domain (NTD) of the SARS-CoV-2 XEC variant affect immune evasion, cell-cell fusion, and S protein stability (19). However, whether glycosylation mutations in the PEDV S protein contribute to variations in viral virulence remains to be elucidated.
In this study, the N381K glycosylation mutation in the PEDV S protein, which was identified in both cell culture-attenuated GI vaccine strains, low-virulence GI-b field variants, and cell culture-attenuated GII strains, was shown to reduce pathogenicity but compromise immunogenicity. These results, which link S protein glycosylation to viral attenuation during both natural field transmission and laboratory cell passaging with poor cross-genogroup protection, provide a molecular basis for improving vaccine design.
## RESULTS
## Analysis of the variation pattern of the glycosylation site at position 381 in the S protein of PEDV during field epidemics and cell passaging and verifica tion of its glycosylation
Glycosylation modification of viral proteins plays crucial roles in viral infection and pathogenesis (20). To study the mechanisms of PEDV virulence variation during both natural epidemics and cell culture serial passaging, we focused on the glycosylation patterns of the PEDV S protein. Sequence alignment of diverse PEDV strains revealed that the predicted glycosylation site mutation (N381K) in the S protein NTD predominantly occurs in low-virulence GI-b strains but not in highly virulent GII strains (Fig. 1A through C). Viruses typically undergo attenuation during high-passage serial propagation, a characteristic also shared by cell culture-attenuated GI vaccine strains. Surprisingly, an identical glycosylation mutation emerged in the 100th passage (P100) of the GII strain GX223 (Fig. 1B), which was a particularly unexpected finding given the stochastic nature of viral mutations. These observations collectively suggest that the N381K mutation in the S protein NTD may play a regulatory role in viral pathogenicity. Mass spectrometry and Western blotting analyses confirmed that the asparagine residue at position 381 (N381) is indeed glycosylated (Fig. 1D andE). These results demonstrate that the N381 glycosylation mutation (N381K) in the PEDV S protein likely contributes to variations in virulence during both natural transmission and laboratory cell passaging.
## Pathogenicity evaluation of low-passage and high-passage generations of strain GX223
To investigate potential changes in virulence following serial cell passaging, we conducted a pathogenicity assessment of strain GX223 at passage 13 (P13) and passage 100 (P100). Compared with P13-infected piglets, piglets challenged with P100 (N381K) presented markedly attenuated clinical symptoms, including milder diarrhea (a significant difference in fecal scores was observed on the first day), delayed time to death, and delayed fecal virus shedding (Fig. 2A through C). Immunohistochemical (IHC) staining showed extensive and comparable viral antigen distribution in the small intestines of both P13 and P100 groups (Fig. 2D). Histopathological analysis confirmed that compared with P13 infection, P100 infection induced milder intestinal lesions (Fig. 2E andF). These results collectively demonstrate that cell culture passage leads to attenuation of strain GX223. Consistent with our initial hypothesis, the S protein N381K mutation appears to contribute to partial attenuation of high-passage generation in strain GX223.
## Rescue of a recombinant strain with the N381K mutation in the PEDV S protein and characterization of the rPEDV-S N381K
A DNA-launched reverse genetics system has been successfully developed by our group (21). To explore the effect of the N381K mutation in the PEDV S on the viral behaviors, a recombinant strain rPEDV-S N381K was constructed using the rPEDV-S wt strain as the backbone. Immunofluorescence assay (IFA) (Fig. 3A) and Sanger sequencing (Fig. 3B) indicated that rPEDV-S N381K was successfully rescued. Compared with the parental rPEDV-S wt strain, the rPEDV-S N381K strain formed plaques of comparable size and exhibited similar growth kinetics (Fig. 3C through E). These results indicated that a recombinant strain with the N381K mutation in the PEDV S protein was obtained and that the recombinant strain exhibited similar growth kinetics in comparison with the parent strain in vitro.
## The rPEDV-S N381K strain was partially attenuated in 2-day-old piglets
To assess the pathogenesis of rPEDV-S N381K , eighteen 2-day-old piglets were randomly divided into three groups, with six pigs in each group. The pigs were orally inoculated with the rPEDV-S wt and rPEDV-S N381K viruses at a dose of 10 5 TCID 50 or mock infected with Dulbecco's modified Eagle medium (DMEM). rPEDV-S N381K -inoculated pigs showed a lower mean diarrhea score than rPEDV-S wt -inoculated pigs from Day 3 post-infection, with a significant difference observed on Day 3 (Fig. 4A). However, diarrhea scores showed no significant differences on the other days, accompanied by large error bars, which may reflect the limitation of small animal numbers. Besides, two pigs (2/6) in the rPEDV-S N381K group survived until the endpoint, whereas all animals (0/6) in the rPEDV-S wt group died (Fig. 4B). Compared to the rPEDV-S wt -inoculated group, piglets in the group rPEDV-S N381K showed comparable viral fecal RNA shedding (Fig. 4C). IHC staining showed extensive and comparable viral antigen distribution in the small intestines of both rPEDV-S wt and rPEDV-S N381K groups (Fig. 4D). H&E) staining revealed that the rPEDV-S N381K strain caused milder histopathological lesions to intestinal villi compared to the rPEDV-S wt -inoculated group, but more serious intestinal villi damage than the mock group (Fig. 4D andE). These data suggested that the rPEDV-S N381K strain was partially attenuated in 2-day-old piglets.
## The rPEDV-S N381K strain was partially attenuated in 5-day-old piglets
To further assess the pathogenesis of rPEDV-S N381K , fifteen 5-day-old piglets were randomly divided into three groups, with five pigs in each group. The pigs were orally inoculated with the rPEDV-S wt and rPEDV-S N381K viruses at a dose of 10 4.5 TCID 50 or mock infected with DMEM. Compared to the rPEDV-S wt groups, although there was no significant difference in fecal scores, piglets in the rPEDV-S N381K group exhibited a lower mean diarrhea score (Fig. 5A), while no pigs died in either group (Fig. 5B). Additionally, piglets in the rPEDV-S N381K group exhibited lower mean fecal viral loads than those in the rPEDV-S wt group on all days except Day 4, which suggests a delayed onset of shedding in the former group (Fig. 5C). These results suggested that the rPEDV-S N381K strain was partially attenuated in 5-day-old piglets. However, a limitation of these results The rPEDV-S N381K strain elicited suboptimal protective immunity against the parental strain At 21 dpi, all the pigs were challenged with the virulent rPEDV-S wt at a high dose of 10 6 TCID 50 /pig. Pigs in the mock group developed severe diarrhea, whereas no diarrheal symptoms or statistically significant differences in fecal scores were observed between groups rPEDV-S N381K and rPEDV-S wt (Fig. 6A). No pigs in the three groups died (Fig. 6B). Unexpectedly, while pigs in both the rPEDV-S N381K and rPEDV-S wt groups exhibited reduced fecal viral shedding compared to the mock group, pigs in the rPEDV-S N381K group showed higher mean fecal viral shedding versus the rPEDV-S wt group, with a statistically significant difference emerging on Days 4 and 5 (Fig. 6C). This observation correlated with the findings: although comparable levels of serum IgG, serum IgA, and fecal IgA were detected in both groups (Fig. 6D through F), the rPEDV-S N381K group showed significantly lower neutralizing antibody titers against the parental strain at 14 and 21 days post-infection (Fig. 6G). However, at the 21 days post-infection, the immune memory cells in the piglets of both groups may be comparable, and there is no difference in the neutralizing antibody levels between the two groups after the parental virus challenge (Fig. 6G). In summary, these data indicate that rPEDV-S N381K reduced protective immunity against the challenge of virulent rPEDV-S wt .
## The N381K mutant in the S protein may alter intermolecular interactions between amino acids within antigenic epitopes
To analyze the molecular mechanism by which the N381K mutant attenuated virulence and reduced immunogenicity, trimeric S proteins were built. Based on the modeling and the experimental results above, we hypothesized the mechanism by which the N381K mutation affects viral virulence and immunogenicity. As shown in Fig. 7A, the polar but uncharged asparagine 381 of the S wt protein fails to form polar interactions with surrounding residues. Upon mutation to lysine, the positively charged side chain orients its hydrophobic aliphatic carbons to engage in interactions with the aromatic ring of phenylalanine 609 (F609). The F609 is located in the CO-26K equivalent region (COE), a neutralizing epitope widely used in the development of PEDV subunit vac cines (22,23). Furthermore, the C-terminal segment of COE (amino acids 575-639) was identified as a conformational neutralizing epitope (CNE) containing F609 (24) (Fig. 7B). Moreover, synthetic peptides corresponding to the predicted epitopes ("CFLKVDTYNST VYK, " "KIVYGVVDTYNSTVYK, " and "GYPEFGGG") were shown to inhibit the neutralizing activity of PEDV-positive sera, suggesting that these predicted antigenic sites were likely potential neutralizing antibody binding regions (pNABR) (25). N381 is located within epitopes "CFLKVDTYNSTVYK" and "KIVYGVVDTYNSTVYK, " while F609 resides in epitope "GYPEFGGG" (Fig. 7B). Collectively, both N381 and F609 are situated within putative neutralizing epitopes or CNEs. The N381 mutation itself, or its altered interactions with neighboring residues in these antigenic epitopes, may influence viral immunogenicity and other characteristics.
## DISCUSSION
Coronaviruses, as single-stranded positive-sense RNA viruses, are prone to genomic mutations during natural epidemics, which may alter their virulence and immunogenicity. For example, the T492I mutation in nsp4 of SARS-CoV-2 was shown to increase viral replication and transmissibility while improving immune evasion (7), and the R203M and D377Y mutations in the N protein of SARS-CoV-2 promoted viral replication and infectivity by inhibiting interferon production (26). The sudden emergence of virulent PEDV variants (GII strains) in 2010 caused devastating losses to the global swine industry, yet the mechanisms of PEDV virulence variation remain poorly understood (27). Through comprehensive sequence analysis, we revealed that during the evolution from low-virulence, less epidemic GI-b strains to high-virulence, highly epidemic GII strains, a novel glycosylation site emerged at position 381 of the spike protein (from KSTV to NSTV) (Fig. 1) and that the strain carrying the N381K mutation exhibits reduced virulence (Fig. 4). This study demonstrates that the glycosylation site at position 381 (N381) in the PEDV S protein partially contributed to the increased viral pathogenicity observed in epidemic strains since 2010. LAVs represent a crucial strategy for controlling disease. The development of PEDV LAVs, such as the classical CV777 and DR13 strains, is typically achieved through serial passaging in VERO cells, during which the viral genome accumulates numerous mutations accompanied by reduced virulence and sometimes altered immunogenicity (28), a phenomenon also observed in our study (Fig. 2). The findings of this study are entirely consistent with the observed limitations of classical vaccines against circulat ing field strains, wherein highly virulent parental strains (e.g., pathogenic CV777 and DR13) are attenuated through serial passaging to develop vaccine strains that often demonstrate poor protection against prevalent epidemic variants (29). The unexpected discovery suggests that the N381K mutation in the PEDV spike protein may contribute to a reduction in the cross-protective efficacy of the vaccine.
Posttranslational modifications, particularly glycosylation, play pivotal roles in regulating viral virulence, replication efficiency, and tissue tropism (30,31). For example, Li et al. revealed that evolutionary glycosylation mutations in the S protein of SARS-CoV-2 significantly impact antigenicity, neutralization, and S protein stability (19). However, these studies focused primarily on how mutations in S protein glycosyla tion sites during field epidemics affect viral infectivity, virulence, and neutralization susceptibility. Our study provides comprehensive evidence that N381K glycosylation site mutation contributes to variations in virulence and immunogenicity among circulating field strains and passages in nonporcine cell cultures (Fig. 4 and6). These multifaceted findings establish the N381 site as an evolutionary hotspot of PEDV. The NTD of the coronavirus S protein likely has conserved functions in modulating viral evolution (30). Mutations in the NTD of SARS-CoV-2 are correlated with immune evasion and viral fitness (32). For example, ΔH69/V70 in the SARS-CoV-2 S NTD region regulates viral infectivity (33), and ΔY144 or K147E+W152R in the SARS-CoV-2 S NTD region is related to neutralizing antibody evasion (34,35). The N381 glycosylation site is also located in the NTD region of the PEDV S protein (Fig. 1). We found that N381 had limited impact on viral replication, which is consistent with the findings of previous reports (36). However, N381 is associated with immune evasion and virulence evolution (Fig. 4 and6). These results indicated that the modulation of coronavirus evolution by the NTD region of the S protein is likely conserved.
In summary, the glycosylation site mutation at position 381 (N381) in the PEDV S protein contributes to driving virulence evolution during both natural field transmis sion and cell culture adaptation while simultaneously affecting cross-protective efficacy among heterologous strains. Our findings significantly advance the understanding of the molecular mechanisms underlying PEDV virulence evolution and immune evasion and provide new insights for the rational design of PEDV LAVs.
Despite our findings, this study has several limitations. First, the sample sizes were relatively small (n = 5-6 piglets per group). The considerable variability in the dispersion observed in some of our data sets, although common in biological studies, further reduces the ability to detect statistically significant differences with confidence. Second, we propose a mechanism by which the N381 mutation alters interactions with neighbor ing residues in these antigenic epitopes, which may influence viral immunogenicity and other characteristics. Our hypothesis is primarily derived from bioinformatics predictions that suggest that the N381 mutation occurs in a key functional domain. While this is a plausible mechanism that aligns with our phenotypic observations, it is critical to emphasize that this mechanistic explanation remains highly speculative.
## MATERIALS AND METHODS
## Cells and viruses
VERO cells were preserved in our laboratory (37). VERO cells were cultured in DMEM (Gibco, CA) supplemented with 10% fetal bovine serum ( TransGen Biotech, China), 100 U/mL penicillin, and 100 µg/mL streptomycin. Previously, our group constructed and successfully rescued the recombinant strain rCH/SX/2016-S HNXP (rPEDV-S wt ) (21). The GX223 strain (GenBank accession number: PV982377) was isolated and preserved by our group. The propagation of PEDV was performed in VERO cells.
## Western blotting
HEK-293T cells seeded in six-well plates were transfected with plasmids expressing full-length S proteins (pCMV6-S wt and pCMV6-S N381K ). At 36 h after transfection, the cells were lysed with 200 µL of ice-cold RIPA buffer for 30 min on ice, and then, the proteins in the supernatant were collected after centrifugation 12,000 × g at 4°C. The detection of N381 glycosylation using Western blotting was performed as previously reported (31). The samples were separated on a 7% Tris-acetate gel (P0534S, Beyotime) using the BeyoGel Tris-acetate SDS running buffer (P0749, Beyotime) and transferred onto PVDF membranes using western transfer buffer (P0021B, Beyotime). The membranes were blocked with 5% nonfat milk in PBST for 2 h at room temperature and then incubated with the PEDV S monoclonal antibody at 4°C overnight. The membranes were washed with PBST and then incubated with goat anti-mouse IgG (H+L) HRP-conjugated secondary antibody (31430, Thermo Fisher Scientific) for 1 h at room temperature. After washing, the target proteins were detected with the enhanced WesterBright ECL Kit (K-12045-D50, Advansta).
## Strategies for constructing and rescuing chimeric full-length cDNA clones of PEDV
The rPEDV-S N381K strain was generated using a similar strategy as before with slight modification (21). In brief, VERO cells were grown to 80% confluency in a six-well plate, and 2.5 µg of the recombinant BAC plasmids was transfected into VERO cells using the Lipofectamine 3000 transfection reagent. The CPE was monitored daily after transfection. When CPE was obvious, the cells and supernatants were collected and freeze-thawed for propagation.
## Sequencing analysis identified the N381K mutation of the PEDV S protein
The N381K mutation of the PEDV S protein was analyzed by sequencing. VERO cells were infected with rPEDV-S wt and rPEDV-S N381K and collected at 36 hpi. The RNA of the samples was extracted using TRIzol reagent (TaKaRa, Japan) and reverse transcribed using HiScript II Q RT SuperMix for qPCR (Vazyme, China) according to the manufac turer's instructions. For sequencing, the fragments were amplified using the primers PEDV-S-F and PEDV-S-R. The sequences of the primers used are listed in Table 1.
## Indirect IFA
IFA was performed as before with slight modification (21). VERO cells were infected with rPEDV-S wt , rPEDV-S N381K , or mock. Cells were washed with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde for 10 min at 37°C, followed by membrane permeabilization with 0.25% Triton X-100 in PBS for 10 min at 37°C. Cells were blocked with 1% BSA at 37°C for 30 min and then incubated with mouse anti-N polyclonal antibody at a dilution of 1:1,000 at 37°C for 1 h. Cells were washed with PBS three times and incubated with Fluorescein (FITC)-AffiniPure Goat Anti-Mouse IgG (H+L) at 1:200 at 37℃ for 1 h. Cells were washed with PBS three times and stained with DAPI for 10 min at room temperature. The images were captured with a fluorescence microscope.
## Growth kinetics
Multistep growth kinetics were calculated as before with slight modifications (21). VERO cells in 12-well plates were infected with PEDV at an MOI of 0.1. After 1 h of absorption, the cells were washed with PBS three times and maintained in maintenance medium. The virus titers of the supernatants at the indicated time points were determined by the TCID 50 assay.
## References
1. Li, Ma, Li et al. (2020) "Porcine epidemic diarrhea virus: molecular mechanisms of attenuation and vaccines" *Microb Pathog*
2. Wang, Tong, Qin et al. (2025) "Midnolin inhibits coronavirus proliferation by degrading viral proteins" *J Virol*
3. Guo, Fang, Ye et al. (2019) "Evolutionary and genotypic analyses of global porcine epidemic diarrhea virus strains" *Transbound Emerg Dis*
4. Huan, Pan, Fu et al. (2020) "Characterization and evolution of the coronavirus porcine epidemic diarrhoea virus HLJBY isolated in China" *Transbound Emerg Dis*
5. Plante, Liu, Liu et al. (2021) "Spike mutation D614G alters SARS-CoV-2 fitness" *Nature*
6. Korber, Fischer, Gnanakaran et al. (2020) "Tracking changes in SARS-CoV-2 spike: evidence that D614G increases infectivity of the COVID-19 virus" *Cell*
7. Lin, Sha, Trimpert et al. (2023) "The NSP4 T492I mutation increases SARS-CoV-2 infectivity by altering non-structural protein cleavage" *Cell Host Microbe*
8. Chen, Zhang, Li et al. (2025) "Bat-infecting merbecovirus HKU5-CoV lineage 2 can use human ACE2 as a cell entry receptor" *Cell*
9. Pensaert, De Bouck (1978) "A new coronavirus-like particle associated with diarrhea in swine" *Arch Virol*
10. Li, Li, Liu et al. (2011) "New variants of porcine epidemic diarrhea virus" *Emerg Infect Dis*
11. Zhang, Zou, Peng et al. (2023) "Global dynamics of porcine enteric coronavirus PEDV epidemiology, evolution, and transmission" *Mol Biol Evol*
12. Li, Yang, Qian et al. (2025) "Molecular characteristics of the immune escape of coronavirus PEDV under the pressure of vaccine immunity" *J Virol*
13. Lin, Chen, Gao et al. (2016) "Epidemic strain YC2014 of porcine epidemic diarrhea virus could provide piglets against homologous challenge" *Virol J*
14. Fleckenstein, Stephan, Hasse (2024) "Elucidating the behavior of the SARS-CoV-2 virus surface at vapor-liquid interfaces using molecular dynamics simulation" *Proc Natl Acad Sci U S A*
15. Lusvarghi, Stauft, Vassell et al. (2023) "Effects of N-glycan modifications on spike expression, virus infectivity, and neutralization sensitivity in ancestral compared to Omicron SARS-CoV-2 variants" *PLoS Pathog*
16. Kapoor, Chen, Tajkhorshid (2022) "Posttranslational modifications optimize the ability of SARS-CoV-2 spike for effective interaction with host cell receptors" *Proc Natl Acad Sci U S A*
17. Paiardi, Ferraz, Rusnati et al. (2024) "The accomplices: heparan sulfates and N-glycans foster SARS-CoV-2 spike:ACE2 receptor binding and virus priming" *Proc Natl Acad Sci U S A*
18. Li, Wu, Nie et al. (2020) "The impact of mutations in SARS-CoV-2 spike on viral infectivity and antigenicity" *Cell*
19. Li, Faraone, Hsu et al. (2025) "Role of glycosylation mutations at the Nterminal domain of SARS-CoV-2 XEC variant in immune evasion, cell-cell fusion, and spike stability" *J Virol*
20. Shimojima, Sugimoto, Taniguchi et al. (2024) "N-glycosylation of viral glycoprotein is a novel determinant for the tropism and virulence of highly pathogenic tick-borne bunyaviruses" *PLoS Pathog*
21. Li, Ma, Dong et al. (2022) "Molecular mechanism of porcine epidemic diarrhea virus cell tropism" *mBio*
22. Chang, Bae, Kang et al. (2002) "Identification of the epitope region capable of inducing neutralizing antibodies against the porcine epidemic diarrhea virus" *Mol Cells*
23. (2025) *Full-Length Text Journal of Virology*
24. Jia, Liu, Sun et al. (2025) "Effective preparation and immunogenicity analysis of antigenic proteins for prevention of porcine enteropathogenic coronaviruses PEDV/TGEV/PDCoV" *Int J Biol Macromol*
25. Chang, Cheng, Chang et al. (2019) "Identification of neutralizing monoclonal antibodies targeting novel conformational epitopes of the porcine epidemic diarrhoea virus spike protein" *Sci Rep*
27. Polyiam, Ruengjitchatchawalya, Mekvichitsaeng et al. (2021) "Immunodominant and neutralizing linear b-cell epitopes spanning the spike and membrane proteins of porcine epidemic diarrhea virus" *Front Immunol*
28. Li, Li, Xiao et al. (2025) "The R203M and D377Y mutations of the nucleocapsid protein promote SARS-CoV-2 infectivity by impairing RIG-I-mediated antiviral signaling" *PLoS Pathog*
29. Liu, Aryal, Niu et al. (2025) "Engineering a recombinationresistant live attenuated vaccine candidate with suppressed interferon antagonists for PEDV" *J Virol*
30. Zhuang, Zhao, Shen et al. (2025) "Advances in porcine epidemic diarrhea virus research: genome, epidemiology, vaccines, and detection methods" *Discov Nano*
31. Lei, Miao, Guan et al. (2023) "A porcine epidemic diarrhea virus isolated from a sow farm vaccinated with CV777 strain in yinchuan"
32. Focosi, Alfonsi, Bernasconi (2025) "Is SARS-CoV-2 spike evolution being retargeted at the n-terminal domain" *Discov Med*
33. Ma, Li, Li et al. (2025) "Changes in the motifs in the D0 and SD2 domains of the S protein drive the evolution of virulence in enteric coronavirus porcine epidemic diarrhea virus" *J Virol*
34. Carabelli, Peacock, Thorne et al.
35. Genomics, Consortium, Peacock et al. (2023) "SARS-CoV-2 variant biology: immune escape, transmission and fitness" *Nat Rev Microbiol*
36. Meng, Kemp, Papa et al. (2021) "Recurrent emergence of SARS-CoV-2 spike deletion H69/V70 and its role in the Alpha variant B.1.1.7" *Cell Rep*
37. Cao, Song, Wang et al. (2022) "Characterization of the enhanced infectivity and antibody evasion of Omicron BA.2.75" *Cell Host Microbe*
38. Mccarthy, Rennick, Nambulli et al. (2021) "Recurrent deletions in the SARS-CoV-2 spike glycoprotein drive antibody escape" *Science*
39. Zhu, Feng, Sun et al. (2025) "N-glycosylation of the PEDV spike protein modulates viral replication and pathogenicity" *Vet Res*
40. Li, Ma, Zhao et al. (2025) "The effect of asparagine-13 in porcine epidemic diarrhea virus envelope protein on pathogenicity" *Vet Res*
42. Deng, Buckley, Pillatzki et al. (2020) "Inactivating three interferon antagonists attenuates pathogenesis of an enteric coronavirus" *J Virol*
43. Hou, Ke, Kim et al. (2019) "Engineering a live attenuated porcine epidemic diarrhea virus vaccine candidate via inactivation of the viral 2'-O -methyltransferase and the endocytosis signal of the spike protein" *J Virol*
44. Song, Li, Wang et al. (2024) "Efficacy evaluation of a bivalent subunit vaccine against epidemic PEDV heterologous strains with low cross-protection" *J Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12582365&blobtype=pdf
|
# ConceptDrift: leveraging spatial, temporal and semantic evolution of biomedical concepts for hypothesis generation
Amir Shariatmadari, Alireza Jafari, Sikun Guo, Sneha Srinivasan, Nathan Sheffield, Aidong Zhang, Kishlay Jha
## Abstract
Motivation: Hypothesis generation is a fundamental problem in biomedical text mining that aims to generate ideas that are new, interesting, and plausible by discovering unexplored links between biomedical concepts. Despite significant advances made by existing approaches, they do not fully leverage the evolutionary properties of biomedical concepts. This is limiting because scientific knowledge continually evolves over time, with new facts being added and old ones becoming obsolete. Thus, it is crucial to capture the evolutionary properties of biomedical concepts from multiple perspectives (e.g. spatial, temporal, and semantic) to generate hypotheses that reflect the up-to-date information landscape of the biomedical domain.
Results:We introduce a novel framework, ConceptDrift, that models the hypothesis generation task as a sequence of temporal graphlets and simultaneously encodes spatial, temporal, and semantic change. Unlike existing approaches that treat these dimensions independently, ConceptDrift is the first to provide a holistic understanding of concept evolution by integrating them into a unified framework. Grounded in the theories of the Distributional Hypothesis and Conceptual Change, our method adapts these principles to the unique challenges of large-scale biomedical literature. We conduct extensive experiments across multiple datasets and demonstrate that ConceptDrift consistently outperforms state-of-the-art baselines in generating accurate and meaningful hypotheses. Our framework shows immediate practical benefits for webbased literature mining tools in life sciences and biomedicine, offering more robust and predictive feature representations.
## 1 Introduction
Scientific repositories such as PubMed (Jin et al. 2023) host more than 35 million articles and add around 3000 articles daily (Lee et al. 2020). While this swift availability of scientific information has acted as an impetus for rapid innovation, it has also overwhelmed researchers trying to survey published studies and construct novel ideas. For example, consider a researcher whose objective is to generate/validate a new hypothesis for the treatment of Parkinson disease. A PubMed query such as "treatment of Parkinson disease" retrieves more than 80 000 articles. Manually reviewing the literature of this size is impractical for individual researches, and thus it posits a bottleneck to their scientific productivity. To address the above challenge, it is imperative to develop novel computational approaches that can read, reason over the large-scale scientific corpus and help biomedical researchers in forging new, interesting and likely to be true hypotheses for possible in-vitro clinical trials. Central to this is the problem of biomedical hypothesis generation (HG) (Akujuobi et al. 2020a, Zhao et al. 2021, Zhou et al. 2024b) that aims to discover cross-silo connections between two disjoint biomedical concepts by chaining together the already known and established scientific facts that remain dispersed across the disparate research fields.
Specifically, given an input concept of interest (e.g. disease or gene), HG aims to find implicit connections (e.g. potential drug target or novel indicator of disease mechanism) that link them in a previously unknown but semantically meaningful way. As an illustration, consider the example of Raynaud's disease and Fish Oils discussed in the HG literature (Akujuobi et al. 2020a, Zhou et al. 2024a, 2022). Prior to 1985, there was no direct connection known between Raynaud's disease and Fish Oils. However, in 1986, after manually inspecting the titles of articles on both topics separately, researchers inferred and later clinically validated (Swanson, 1986) an association between them. Finding such meaningful implicit links is the crux of the problem that this paper attempts to address.
Over the past decade, approaches ranging from distributional statistics (Srinivasan 2004, Spangler et al. 2014), graph theoretic measures (Pu et al. 2023, Cameron et al. 2015), embedding-based methods (Jha et al. 2018, Wang et al. 2021a) to most recently large language models (Qi et al. 2024, Pelletier et al. 2024) have been proposed in the HG literature. Despite significant advance, many of the existing HG approaches are designed for static settings and fail to leverage the temporal evolution of biomedical entities. This is problematic because the biomedical domain is highly dynamic with new facts emerging and old ones being obsolete every now and then. For example, consider the biomedical concept Parkinson disease. In the early 2000s, within the virology literature, its meaning used to be associated with concepts such as "Amyotrophic lateral sclerosis", "Lysosomal storage diseases" and "Neoplasms"; however, in the 2020s it started to become associated with concepts like "Cytarabine" and "Cycloheximide". Modeling such temporal evolution of biomedical concepts is crucial for generating hypotheses that are accurate and reflect up-to-date information landscape of the biomedical domain.
Recently, some approaches (Akujuobi et al. 2020a, 2020b, Zhou et al. 2022, 2024b, Jha et al. 2018) have proposed modeling the temporal change of biomedical concepts from the diachronic biomedical corpus. For example, Akujuobi et al. (2020a) models temporal dynamics using positiveunlabeled data, while Jha et al. (2018) encodes semantic drift through temporal matrix factorization. Nevertheless, no existing approach unifies spatial, temporal, and semantic evolution, even though integrating all three is critical for capturing the full scope of how biomedical concepts change over time.
Integrating these dimensions into a single coherent framework is fundamentally non-trivial. Each dimension requires distinctly different computational approaches: semantic modeling demands nuanced language understanding and representation, spatial modeling necessitates accurately learning structural graph representations, and spatio-temporal modeling requires capturing evolving patterns over time. Unifying these divergent computational strategies involves addressing significant methodological challenges, such as how best to align evolving semantic representations with temporal graph dynamics, efficiently encoding complex relational information, and ensuring seamless interaction among these diverse representations. To address this, we propose a novel method that captures the evolving nature of biomedical knowledge by modeling changes in a concept's semantic meaning. Our approach is motivated by two key principles: the Distributional Hypothesis, which posits that word meanings emerge from the context of co-occurring concepts (Sahlgren 2008), and Conceptual Change (Posner et al. 1982, Treagust andDuit 2008), which describes how concepts evolve as they encounter new information. These principles guide us in tracking how biomedical concepts change meaning as they interact with new concepts over time.
A key contribution of our work is to adopt these principles to model the semantic, spatial, and temporal dimensions of biomedical concepts in a unified approach. We model the HG task as a sequence of graphlets which spans over discrete time steps. Each graphlet represents a snapshot of biomedical concepts and their connections with each other at discrete time steps. This enables us to track the evolution of spatial relationships between concepts over time, thereby guiding how our approach interprets changes in the meaning of these concepts. By modeling these dimensions in a unified manner, our approach provides a rich understanding of how biomedical concepts evolve. As a result, our approach generates more accurate predictions, yielding hypotheses with potential for scientific discovery.
The main contributions of our work can be summarized as follows:
� We propose a new approach to HG that exploits spatial, temporal, and semantic changes in biomedical concepts for accurate HG. This has immediate practical benefits for web-based literature mining tools designed for applications in life science and biomedicine. � Unlike existing HG approaches that model spatial, temporal, and semantic changes independently, our proposed approach is the first to leverage them collaboratively in a unified framework. This holistic modeling enables our approach to learn feature representations that are not only more robust but also have superior predictive power, leading to more reliable and meaningful hypothesis generation. � Extensive experiments on multiple datasets show that unifying spatial, temporal, and semantic changes in concepts with our proposed approach consistently outperforms baseline methods. Moreover, case studies show that the proposed approach generates hypotheses that are medically sensible and of potential interest to the domain experts.
## 2 Related works
Existing approaches in biomedical HG have predominantly emphasized modeling the semantic, spatial, or spatio-temporal dimensions independently, without integrating all three dimensions comprehensively. Below, we categorize existing methods and discuss their advancements and limitations. Semantic Methods. Semantic approaches focus on representing biomedical concepts by capturing their meanings through embeddings derived from textual contexts. Traditional semantic methods rely on static representations to encode semantic context, facilitating tasks such as link prediction and literature-based discovery (Spangler et al. 2014, Srinivasan, 2004, Gopalakrishnan et al. 2019, Jha et al. 2019, Ding et al. 2024). These approaches typically utilize static statistical methods or co-occurrence measures to infer implicit associations. Recent advances in semantic modeling have leveraged transformer-based pre-trained language models (PLMs) such as BERT (Devlin, 2018) and BioBERT (Lee et al. 2020), significantly enhancing the semantic representation of biomedical text by capturing deep contextual meanings. These PLMs excel at nuanced semantic extraction, dramatically improving downstream tasks like biomedical link prediction and relation extraction (Yao et al. 2019). However, semantic methods are inherently limited due to their static nature. The semantic embeddings produced by PLMs like BioBERT do not evolve with new biomedical knowledge. Consequently, they cannot capture temporal shifts in the meanings of concepts as new contexts emerge over time, rendering them ineffective in dynamically evolving biomedical knowledge domains. This is a critical limitation for HG, as forecasting future associations requires accommodating changing conceptual semantics.
Spatial Methods. Spatial methods primarily involve representing biomedical concepts through structural relationships encoded within static graphs. Prior HG works in spatial modeling (Cameron et al. 2015, Pu et al. 2023) have focused on graph-based representations, utilizing concept cooccurrence and structural connectivity to identify potential novel relationships. Advanced spatial models such as Graph Convolutional Networks (GCNs) (Kipf and Welling, 2016), Graph Attention Networks (GATs) (Veli� ckovi� c et al. 2017), andGraph Transformers (Shi et al. 2020) have substantially improved graph-based representation learning. These models aggregate information from neighboring nodes and have shown promise in capturing complex structural patterns across diverse domains (Jafari et al. 2024, Li et al. 2023). Despite their strengths in modeling structural relationships, spatial methods face significant limitations. Primarily, they rely on static snapshots of knowledge, neglecting the temporal dynamics critical for understanding biomedical concept evolution. Without temporal context, these models cannot effectively predict emerging relationships as structural patterns evolve. Additionally, purely spatial methods fail to incorporate rich semantic contexts, limiting their ability to accurately capture nuanced biomedical relationships necessary for highquality hypothesis generation.
Spatio-Temporal Methods. Spatio-temporal methods aim to overcome limitations of purely semantic or spatial methods by modeling both structural and temporal changes. These approaches integrate graph structures with temporal dynamics, enabling predictions of evolving relationships. Prior HG research employing spatio-temporal modeling (Akujuobi et al. 2020b(Akujuobi et al. , 2020a;;Zhou et al., 2022Zhou et al., , 2024aZhou et al., , 2024b) ) has effectively captured temporal variations in structural connections. Notable general spatio-temporal frameworks include the Diffusion Convolutional Recurrent Neural Network (DCRNN) (Li et al. 2017) and Temporal Graph Network (TGN) (Rossi et al. 2020). DCRNN integrates graph convolutions with recurrent neural networks to model the temporal evolution of relationships, whereas TGN uses temporal memory states to dynamically update node interactions, significantly enhancing predictive capabilities. Spatio-temporal models lack explicit representation of semantic evolution. By focusing solely on structural relationships and temporal cooccurrences, these methods inherently discard detailed semantic context of concepts as they appear in the literature, crucial for nuanced hypothesis generation. Consequently, they struggle to accurately predict future relationships that rely heavily on evolving concept meanings, potentially leading to irrelevant or incorrect hypotheses.
## 3 Hypothesis generation problem
We formulate hypothesis generation as a temporal link prediction task on a dynamic biomedical concept graph. Our goal is to predict whether a link (i.e. co-occurrence) between two biomedical concepts will appear at a future time step by leveraging historical data mined from biomedical research publications. This involves identifying when two concepts, which may not have co-occurred previously, are likely to cooccur in future research, thus signaling a new relationship that can inspire further investigation or experimentation.
Inspired by recent HG methods (Akujuobi et al. 2020b, 2020a, Zhou et al. 2022, 2024a, 2024b), we represent this task using a dynamic graph, denoted as G ¼ fG 0 ; . . . ; G t ; . . . ; G T g, which spans discrete time steps t 2 f0; . . . ; Tg, where T is the total number of time steps. Each graphlet G t ¼ ðV; E t Þ represents the state of biomedical knowledge at time step t. The set V ¼ f1; . . . ; jVjg consists of a total number of jVj biomedical concepts, specifically Medical Subject Headings (MeSH) terms as defined by the National Library of Medicine (Leblanc et al. 2024). Each node i 2 V corresponds to a unique biomedical concept, which remains constant across all time steps. The set of edges E t at time t captures the co-occurrences between concepts. An edge ði; jÞ 2 E t indicates that concepts i and j co-occurred in at least one biomedical article within the time frame represented by t. At the initial time step (t ¼ 0), no co-occurrences are present, i.e., E 0 ¼ ;. Formally, the HG task is to predict future co-occurrences between pairs of concepts ði; jÞ at time T using the historical graph information up to time T -1. By accurately predicting these co-occurrences, our model can uncover new and emerging relationships, leading to potentially novel hypotheses.
We construct dynamic graphs from PubMed articles by extracting concept co-occurrences using, PubTator3 (https:// ftp.ncbi.nlm.nih.gov/pub/lu/PubTator3/), a state-of-the-art tool published by the National Institute of Health that outperforms BioBERT and GPT-4 for biomedical concept extraction (Wei et al. 2024). To minimize noise, we apply three filtering steps: (i) remove entries with missing concept IDs, (ii) restrict concepts to standardized MeSH terms only, and (iii) require concepts to co-occur within the same article to form edges, which filters potential cross-article noise. Additional construction details are provided in the Appendix.
## 4 Methodology
We introduce ConceptDrift, a unified approach that models how the meaning of concepts change over time by collaboratively integrating spatial, temporal, and semantic information. Unlike prior efforts that treat these separately, ConceptDrift unifies them with our novel framework called Temporal Semantic Contextualization (TSC), enabling accurate discovery of future co-occurrences that serve as potential hypotheses. By using TSC, ConceptDrift is able to accurately predict future co-occurrences between biomedical concepts and articulate the connection between the two concepts as a hypothesis with potential for scientific discovery. We illustrate an overview of how ConceptDrift models the change of meaning in biomedical concepts using TSC in Fig. 1.
## 4.1 Conceptual drift
Our approach is driven by two key principles: the Distributional Hypothesis and Conceptual Change as defined and discussed in the Introduction. The Distributional Hypothesis argues that the meaning of a biomedical concept can be inferred by the neighboring concepts it co-occurs with (Sahlgren 2008). Conceptual Change refers to the way people's understanding of concepts can shift when exposed to new knowledge (Posner et al. 1982, Treagust andDuit, 2008). Building on these ideas, we introduce Conceptual Drift, which extends the Distributional Hypothesis to account for how the meanings of biomedical concepts evolve over time. Conceptual Drift states that the meaning of a concept changes when it begins to co-occur with new concepts.
For a given time t, we define the neighborhood of a concept i as the set of concepts that have co-occurred with i at least once from time step 1 up to time step t. Two biomedical concepts, i and j, are said to be similar at time t if their neighborhoods are similar, meaning their respective sets of co-occurring concepts overlap to a some extent. In the Appendix, we empirically validate conceptual drift occurring in real-world data by comparing the overlap of neighborhoods of co-occurring concepts with the neighborhoods of non-co-occurring concepts to see how much overlap similar neighborhoods have in practice.
In the context of the hypothesis generation, this definition implies that we are predicting whether two concepts will cooccur at a future time step based on how their neighborhoods evolve dynamically over time.
## 4.2 Temporal semantic contextualization
We propose Temporal Semantic Contextualization (TSC), a framework for modeling Conceptual Drift by unifying the spatial, temporal, and semantic changes of biomedical concepts. In our TSC framework, each biomedical concept has a semantic state that will evolve over time. They are initialized with concept embeddings that reflect the concept's semantics. Each time a concept co-occurs with another concept, the new context of the concept is captured and is integrated with the concept's meaning to update its semantic state. Finally, before a prediction is made on whether two concepts will co-occur, special embeddings will be generated with temporal multi-head cross attention to enrich the concept's representation with the temporal evolution of the concept's neighborhood.
## 4.2.1 Semantic states
We represent each biomedical concept using a semantic state that evolves over time. For a given concept i at a time t, its semantic state is denoted as s i ðtÞ. Semantic states are central to our framework's integration of spatial, temporal, and semantic changes in concepts, as they capture how each concept's meaning shifts through new co-occurrences. Specifically, as concepts interact and co-occur with others, their semantic states are updated to reflect changes in context.
To initialize these semantic states, we use a pre-trained language model (PLM) (Lee et al. 2020) since it can provide rich semantics for text embeddings. By inputting all the concepts in V into a PLM, we generate the concept embeddings for all concepts, resulting in a matrix X 2 R jVj × d , where d is the dimension of the embedding space. Each row in X corresponds to the concept embedding of a concept i 2 V. At the initial time step (t ¼ 0), these embeddings are used to initialize the semantic state of each concept in the graph. Specifically, the semantic state s i ð0Þ for concept i is set as its corresponding embedding from the matrix X. By grounding the initial state of each concept in its concept embedding, we provide the foundation for modeling how concepts change meaning over time as they interact with other concepts within the graph. In our implementation, we use BioBERT (Lee et al. 2020) since it is trained on a vast corpus of biomedical literature, which enables it to encapsulate the semantic nuances of biomedical concepts based on their contextual usage in scientific texts.
## 4.2.2 Contextual integration
At a given time step t, the graphlet G t indicates when i cooccurs with another concept j, introducing new context for i's evolving meaning. When this occurs, i's previous semantic state must be updated to integrate the context. This contextual integration process involves two steps. First i's new context must be captured. Then i's semantic state is updated by integrating the captured context with i's previous semantic state. To capture i's new context, we propose the Context Aggregation Function (CAF). CAF aggregates the previous semantic state of concept i, s i ðt p Þ, where t p represents the last time step the semantic state of the concept was updated, with the previous semantic state of the new co-occurring concept j, s j ðt p Þ, and a time difference vector Δt (Kazemi et al. 2019). While prior spatio-temporal graph methods (Rossi et al. 2020, Wang et al. 2021b) use temporal encodings to weight structural recency, CAF leverages Δt to model semantic receptivity, i.e. how long a concept has maintained its current meaning determines how strongly new co-occurrences influence its semantic evolution. This is crucial in biomedicine where concepts like 'Cytarabine' undergo gradual semantic shifts as the field advances, requiring us to capture not just when interactions occur, but how temporal distance affects meaning change. By aggregating these three elements, we are able to combine the meaning of both concepts along with the recency of i's interaction, allowing us to comprehensively capture i's new context. CAF captures the new context as follows:
where CAF i ðtÞ is the aggregation of
In practice, CI can be any learnable function that can integrate new context while maintaining important historical information. In our experiments, we define CI to be a Gated Recurrent Unit (GRU) because it provides an effective balance between expressiveness and efficiency. GRUs are designed to capture long-term dependencies while avoiding the parameter overhead of more complex recurrent architectures such as LSTMs (Cho, 2014). While alternative architectures such as Transformer encoders or feed-forward networks could in principle be applied, the former introduces substantially higher computational cost, and the latter lacks the temporal gating needed to retain long-range contextual information.
$$CAF i ðtÞ ¼ s i ðt p Þjjs j ðt p ÞjjΔt (1)$$
$$s i ðt p Þ, s j ðt p Þ,$$
## 4.3 Hypothesis generation
At a future time step t þ 1, we want to predict whether two concepts i and j will co-occur with each other. We first generate predictive embeddings out of the concept's semantic states and their neighborhood of concepts using temporal multihead cross attention. These embeddings are then used to predict whether a co-occurrence will exist between the two concepts. If a co-occurrence is predicted, it can be articulated as a potential hypothesis in natural language.
To predict future co-occurrences between any two concepts i and j, it is important to consider the relevant upto-date semantic states of the concepts' neighborhoods and when each neighboring concept first co-occurred with i or j in addition to i and js current semantic state. Although a concept's semantic state reflects past co-occurrences, neighboring concepts may update their semantic states over time, possibly having new information beneficial for prediction. We use temporal multi-head cross-attention to get a weighted sum of a concept's neighbors based on their most recent semantic states and when they first interacted with the concept. It ensures that the model captures relevant information from the most recent representations of concepts and the temporal relevance of their interactions, leading to more accurate future predictions. The multi-head cross-attention mechanism introduces a query q, keys K, and values V. The query holds what information we want to focus on and is used to see which elements in list of potential matches are most relevant to it. The keys represent the potential matches. The values are the same as the keys, however, the keys relevancy with the query is used to output a weighted sum of the values. In our context, a concept i represents the query, while i's neighborhood represents the keys and values. Multi-head cross-attention is used to weigh a concept i's neighbors based on their semantic and temporal relevance to i. By taking the i and its neighbors' most recent semantic states and concatenating them with temporal encodings (Kazemi et al. 2019), we accomplish this.
We define the query as the semantic state of concept i at time t, concatenated with a temporal encoding of the current time step:
where k denotes the concatenation operator, and ϕðt) represents the temporal encoding of the current time step. We also define the keys and values as the most recent semantic states of i's neighbors and concatenate them with temporal encodings of the time they first co-occurred with i:
Here t 1 ; . . .; t N denotes the time steps when each neighbor ð1; . . .; NÞ first co-occurred with i. We use first co-occurrence times rather than recency based times because the evolving semantic states already capture recent interactions. First cooccurrence times provide a distinct signal marking potential discovery events, enabling the model to distinguish between established relationships and emerging ones.
The multi-head cross-attention inputs the query, keys and values and outputs a vector h i ðtÞ that holds the most relevant temporal and semantic information in concept i's neighborhood:
To integrate this information with the concept's current semantic state, we concatenate h i ðtÞ with s i ðtÞ and pass the result through a Multi-Layer Perceptron (MLP) to produce the predictive embedding z i ðtÞ:
The predictive embedding has a concept's current semantic state with relevant information from the up-to-date semantic states of the concept's neighborhood and the time of its past co-occurrences, leading it to produce more accurate predictions.
After generating the concepts i and j's predictive embeddings z i ðtÞ and z j ðtÞ, we compute the probability of whether i and j will co-occur at a future time step t þ 1. The probability is denoted as pði; jjt þ 1Þ. To do this we pass the predictive embeddings through an MLP and use sigmoid function to compute the probability of co-occurrence:
where σð�Þ is the sigmoid function.
$$q ¼ s i ðtÞjjϕðtÞ (3)$$
$$K ¼ V ¼ ½s 1 ðtÞjjϕðt 1 Þ; . . . ; s N ðtÞjjϕðt N Þ� (4)$$
$$h i ðtÞ ¼ MultiHeadCrossAttentionðq; K; VÞ(5)$$
$$z i ðtÞ ¼ MLPðs i ðtÞjjh i ðtÞÞ (6)$$
$$pði; jjt þ 1Þ ¼ σðMLPðz i ðtÞ þ z j ðtÞÞÞ (7)$$
## 5 Results
Datasets. Following standard hypothesis generation practices (Akujuobi et al. 2020a, 2020b, Zhou et al. 2022, 2024a, 2024b), we train models on historical PubMed data and evaluate their ability to rediscover future publications. We construct dynamic graph datasets from PubMed articles in Virology, Neurology, and Immunology spanning 2000-2024, with each year as a separate time step, similar to Tyagin and Safro (2024)'s approach. This annual resolution improves upon prior work using multi-year periods (Akujuobi et al. 2020a, 2020b, Zhou et al. 2022, 2024a, 2024b Experimental Setup. To evaluate the efficacy of ConceptDrift, we compare the performance of ConceptDrift on predicting co-occurrences in 2024 with a variety of baselines. We evaluate each method's performance using metrics that indicate how well the methods can discriminate between true and false future co-occurrences while balancing precision and recall. When training and evaluating each method, we generate a 1:1 ratio of negative (false) concept co-occurrences by uniform sampling. Following established retrospective validation protocols (Akujuobi et al. 2020a, 2020b, Zhou et al. 2022, Tyagin and Safro 2024, Zhou et al. 2020a, 2024b), we train on historical literature and test on held-out future publications. To ensure the reliability of our findings, we ran ConceptDrift and TGN (with PLM features) 10 times each, reporting mean and standard deviation. Statistical significance is assessed using a two-sample t-test.
Baselines and Metrics. Prior HG methods typically fall into three categories: semantic, spatial, and spatio-temporal. For fair comparison, we select strong baselines from each category. We evaluate BERT (Devlin 2018) and BioBERT (Lee et al. 2020), two widely used PLMs known for their effectiveness in biomedical semantic representation and link prediction (Yao et al. 2019). We additionally evaluate ClinicalBERT (Huang et al. 2019), BiomedNLP (Gu et al. 2020) and SciBERT (Beltagy et al. 2019). These models provide a strong benchmark for purely semantic approaches. For structural modeling, we include GCN (Kipf and Welling 2016), GAT (Veli� ckovi� c et al. 2017), and Graph Transformer (Shi et al. 2020), which are powerful GNNs commonly used across different domains and tasks to learn static graph structures. We experiment against DCRNN (Li et al. 2017) and TGN (Rossi et al. 2020), two advanced models for dynamic graph representation, which effectively capture structural changes over time. We also evaluate STHG (Zhou et al. 2024b), a state-of-the-art HG method that outperforms other prevalent HG works (Akujuobi et al. 2020b, Zhou et al. 2022). Further details about each baseline we use are provided in the Appendix.
We use Area Under the ROC Curve (AUC) and Average Precision (AP) to evaluate predictive performance. AUC measures the model's ability to rank true co-occurrences higher than false ones, while AP assesses the balance between precision and recall. Together, these metrics offer a robust evaluation of hypothesis prediction accuracy.
## 5.1 Main results
Our results, reported in Table 2, show that ConceptDrift consistently outperforms all baselines across all three datasets, underscoring how simultaneously modeling semantic, spatial, and temporal evolution yields stronger predictive accuracy than treating these dimensions in isolation. The results from 10 independent runs demonstrate that ConceptDrift achieves a higher mean AUC and AP across all PLMs and datasets, and the low standard deviation indicates the stability and reliability of our model's performance. To ensure ConceptDrift's superiority is not by chance, we compare ConceptDrift with 10 independent runs of TGN (with PLM embeddings as features) and see ConceptDrift outperforms TGN. In the Table 3, we report two-sample t-tests that shows these improvements are statistically significant.
By modeling the spatial, temporal, and semantic dimensions in an integrated manner, ConceptDrift captures the evolving relationships and contextual meaning of biomedical concepts, which is critical for accurately predicting future cooccurrences. In contrast, baselines that only leverage one or two dimensions fall short, highlighting the importance of a unified approach. ConceptDrift performs robustly across different PLMs. These results demonstrate the model's effectiveness and generalizability, independent of any specific PLM. ConceptDrift also demonstrates robustness in performance even with smaller training samples. The Virology data set has under 3 million co-occurrences from 2000-2020, compared to over 10 million for Neurology and Immunology. Despite the smaller training data, ConceptDrift gets the best AUC and AP of 0.880 and 0.859, respectively. This is due to its ability to model Conceptual Drift, which captures how biomedical concepts evolve over time. By dynamically updating concept meanings and generating temporal embeddings, ConceptDrift can accurately predict future co-occurrences, even with limited data.
ConceptDrift's consistent outperformance across datasets suggests that modeling conceptual drift helps filter spurious associations. By requiring concepts to maintain evolving semantic relationships through their neighborhood dynamics over multiple time steps, our approach implicitly favors stable, meaningful connections over transient co-occurrences that may arise from publication artifacts or trending topics.
## 5.2 Ablation study
In this section we analyze the contribution of key components in ConceptDrift and subsequently discuss how each dimension (semantic, spatial, and temporal) collectively drives our performance. Our results are reported in Table 4. First, we assess the impact of semantic initialization by removing the initialization of embeddings for concepts. Table 4 shows that performance deteriorates on all three datasets without semantic initialization, highlighting the synergy of the semantic dimension with spatial and temporal evolution of concepts within the datasets. In this ablation, ConceptDrift lacks a semantic foundation and thus can only capture spatial and temporal changes without grounding concept representations in rich semantic features.
Second, we examine how updating semantic states as concepts interact affects performance by ablating the contextual integration process. This removal yields a substantial drop on all three datasets, underscoring the need for a collaborative approach to unify the spatial, temporal, and semantic dimensions. Simply incorporating each dimension in a naive manner is insufficient for capturing the evolving nature of biomedical concepts.
Third, we examine the effectiveness of our design choice for the contextual integration module (Equation 2) that inputs both the aggregated context (Equation 1) and previous previous semantic state s i ðt p Þ, which is already in aggregated context. We do this by modifying the the contextual integration module to only input the aggregated context. This modification results in a slight drop in performance indicating the importance of feeding the s i ðt p Þ as an additional input. Altogether, this strengthens the signals of past representation of concepts to the contextual integration module, helping the historical meaning of concepts while still incorporating new information about the concepts' evolving contexts.
Finally, we intend to assess how well temporal multi-head cross-attention improves ConceptDrift by using predictive embeddings that enrich a concept with relevant information from the up-to-date semantic states of the concept's neighborhood and the time of its past co-occurrences. To test this, we remove temporal multi-head cross-attention and rely solely on the semantic states for prediction. Similar to the semantic initialization ablation, we observe that temporal multi-head cross-attention is essential for maintaining high performance. These results highlight the critical importance and non-trivial nature of our methodological integration. Removing key integrated components, such as semantic initialization or temporal cross-attention, substantially degrades
## 5.3 Data scarcity robustness
To demonstrate ConceptDrift's robustness to low-data scenarios and its strong generalization ability, we experiment how well ConceptDrift performs when trained on randomly sampled smaller proportions of training data. We report the performance of ConceptDrift (with BioBERT concept embeddings) when trained on 10%, 30%, 60%, 90%, 100% of the co-occurances from the Neurology training data and report the results in Fig. 2. We see that as fraction of training data decreases, the performance of ConceptDrift gracefully decreases. In particular, even with only 30% of the training data, ConceptDrift is able to achieve a reasonable AUC of 0.851, outperforming many of its baselines trained on the entire dataset. This demonstrates ConceptDrift's strong generalization capabilities and robustness to low-data settings.
## 5.4 Conceptual drift analysis
We tracked the semantic trajectories of Cytarabine and Parkinson Disease using the neurology dataset, analyzing how their semantic state embeddings evolved toward their 2024 real-world connection (Li et al. 2024). temporal evolution filters spurious associations. Concepts must maintain evolving semantic relationships through neighborhood dynamics over multiple time steps, potentially favoring persistent co-occurrences that correlate with meaningful biological relationships rather than incidental mentions. In the appendix we provide more analysis on conceptual drift.
## 5.5 Scalability
The computational cost of ConceptDrift is dominated by two components: contextual integration via GRU updates for each observed co-occurrence and temporal multi-head crossattention during prediction. For contextual integration, each edge event incurs Oðd 2 Þ operations per GRU update, where d is the hidden dimension. For prediction, the cross-attention mechanism requires Oðd 2 Þ operations for linear projections and OðH � � N � dÞ for computing attention over � N average neighbors with H heads. With jEj edge events processed during training, the overall complexity is OðjEj � d 2 þ H � jEj � dÞ, which simplifies to OðjEj � d 2 Þ as jEj dominates. Empirically, ConceptDrift demonstrates near-linear scalability with respect to the number of edges. On the Virology dataset, the model trains within one and half hours on a single NVIDIA A6000 GPU, while Neurology and Immunology require less than two and three hours, respectively. These results confirm that ConceptDrift can efficiently scale to tens of millions of temporal co-occurrences while maintaining state-of-the-art predictive performance.
## 6 Conclusion
We introduced ConceptDrift, a framework that unifies semantic, spatial, and temporal dimensions for biomedical hypothesis generation. By modeling Conceptual Drift through evolving neighborhood dynamics, ConceptDrift captures how concept meanings shift over time. Experiments on multiple datasets show consistent outperformance of baseline methods, underscoring the value of integrating all three dimensions.
## References
1. Akujuobi, Chen, Elhoseiny (2020) "Temporal positive-unlabeled learning for biomedical hypothesis generation via risk estimation" *Adv Neural Inf Process Syst*
2. Akujuobi, Spranger, Palaniappan (2020) "T-pair: temporal nodepair embedding for automatic biomedical hypothesis generation" *IEEE Trans Knowl Data Eng*
3. Beltagy, Lo, Cohan (2019) "Scibert: a pretrained language model for scientific text"
4. Cameron, Kavuluru, Rindflesch (2015) "Context-driven automatic subgraph creation for literature-based discovery" *J Biomed Inform*
5. Cho (2014) "Learning phrase representations using rnn encoder-decoder for statistical machine translation"
6. Devlin, Bert (2018) "pre-training of deep bidirectional transformers for language understanding"
7. Ding, Dahal, Adhikari (2024) "Learning to rank complex biomedical hypotheses for accelerating scientific discovery"
8. Gopalakrishnan, Jha (2019) "A survey on literature based discovery approaches in biomedical domain" *J Biomed Inform*
9. Gu, Tinn, Cheng (2020) "Domain-specific language model pretraining for biomedical natural language processing"
10. Huang, Altosaar, Ranganath et al. (2019) "Modeling clinical notes and predicting hospital readmission"
11. Jafari, Fox, Rundle (2024) "Time series foundation models and deep learning architectures for earthquake temporal and spatial nowcasting"
12. Jha, Xun, Wang (2018) "Concepts-bridges: uncovering conceptual bridges based on biomedical concept evolution"
13. Jha, Xun, Wang (2019) "Hypothesis generation from text based on co-evolution of biomedical concepts"
14. Jin, Leaman, Lu (2023) "Pubmed and beyond: biomedical literature search in the age of artificial intelligence"
15. Kazemi, Goel, Eghbali (2019) "Time2vec: learning a vector representation of time"
16. Kipf, Welling (2016) "Semi-supervised classification with graph convolutional networks"
17. Leblanc, Hamroun, Bentegeac et al. (2024) "Added value of medical subject headings terms in search strategies of systematic reviews: comparative study" *J Med Internet Res*
18. Lee, Yoon, Kim (2020) "Biobert: a pre-trained biomedical language representation model for biomedical text mining" *Bioinformatics*
19. Li, Nian, Sun (2023) "Advancing biomedicine with graph representation learning: recent progress, challenges, and future directions" *Yearb Med Inform*
20. Li, Zhang, Chen (2024) "Cytarabine prevents neuronal damage by enhancing ampk to stimulate pink1 / parkin-involved mitophagy in Parkinson's disease model" *Eur J Pharmacol*
21. Li, Yu, Shahabi (2017) "Diffusion convolutional recurrent neural network: Data-driven traffic forecasting"
22. Pelletier, Ramirez, Adam (2024) "Explainable biomedical hypothesis generation via retrieval augmented generation enabled large language models"
23. Posner, Strike, Hewson (1982) "Accommodation of a scientific conception: toward a theory of conceptual change" *Sci Educ*
24. Pu, Beck, Verspoor (2023) "Graph embedding-based link prediction for literature-based discovery in alzheimer's disease" *J Biomed Inform*
25. Qi, Zhang, Tian (2024) "Large language models as biomedical hypothesis generators: a comprehensive evaluation"
26. Rossi, Chamberlain, Frasca (2020) "Temporal graph networks for deep learning on dynamic graphs"
27. Sahlgren (2008) "The distributional hypothesis" *Ital J Linguist*
28. Shi, Huang, Feng (2020) "Masked label prediction: unified message passing model for semi-supervised classification"
29. Spangler, Wilkins, Bachman (2014) "Automated hypothesis generation based on mining scientific literature"
30. Srinivasan (2004) "Text mining: generating hypotheses from medline" *J Am Soc Inf Sci*
31. Swanson (1986) "Fish oil, raynaud's syndrome, and undiscovered public knowledge" *Perspect Biol Med*
32. Treagust, Duit (2008) "Conceptual change: a discussion of theoretical, methodological and practical challenges for science education" *Cult Stud of Sci Educ*
33. Tyagin, Safro (2024) "Dyport: dynamic importance-based biomedical hypothesis generation benchmarking technique" *BMC Bioinf*
34. Van Der Maaten, Hinton (2008) "Visualizing data using t-sne" *J Mach Learn Res*
35. Cucurull, Casanova (2017) "Graph attention networks"
36. Wang, Wang, Wang "Interhg: an interpretable and accurate model for hypothesis generation"
37. Wang, Chang, Liu (2021) "Inductive representation learning in temporal networks via causal anonymous walks"
38. Wei, Allot, Lai (2024) "Pubtator 3.0: an ai-powered literature resource for unlocking biomedical knowledge" *Nucleic Acids Res*
39. Yao, Mao, Luo (2019) "Kg-bert: Bert for knowledge graph completion"
40. Zhao, Su, Lu (2021) "Recent advances in biomedical literature mining" *Brief Bioinform*
41. Zhou, Jiang, Yao (2022) "Learning temporal difference embeddings for biomedical hypothesis generation" *Bioinformatics*
42. Zhou, Jiang, Wang (2024) "Temporal attention networks for biomedical hypothesis generation" *J Biomed Inform*
43. Zhou, Wang, Yao "Generating biomedical hypothesis with spatiotemporal transformers" *IEEE J Biomed Health Inform*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12728268&blobtype=pdf
|
# Cytomegalovirus Vertical Transmission Rate According to IgG Avidity Value and Valacyclovir Treatment of Maternal Primary Infection in the First Trimester of Pregnancy
Jacques Fourgeaud, Mariame Barry, Nicolas Bourgon, Nicolas Veyrenche, Tiffany Guilleminot, Yves Ville, Marianne Leruez-Ville
## Abstract
Background. Cytomegalovirus (CMV) serology in the first trimester (T1) of pregnancy relies on immunoglobulin (Ig) G (IgG) and IgM testing, followed by IgG avidity measurement when both IgG and IgM are positive. Valacyclovir is proposed to prevent vertical transmission in maternal primary infection (MPI) during T1. Our objective was to assess the risk of transmission, with and without valacyclovir, based on IgG avidity values and MPI timing.Methods. Cases (2012Cases ( -2024) ) with positive or equivocal IgM, intermediate or low IgG avidity (VIDAS and/or LIAISON) in T1, and a CMV PCR performed in amniotic fluid were retrieved from the laboratory software. The MPI date was estimated using a logarithmic model based on VIDAS avidity results.Results. Five hundred and sixty one women (58% untreated, 42% treated) were included. In cases with low avidity and no treatment, transmission was 34% and 31% with VIDAS and LIAISON assays, respectively. Intermediate avidities and no treatment were also associated with high transmission rates (14% and 22%). VIDAS avidity values ≥0.60 or ≥0.50 in samples collected before 12 weeks were associated with 0% and 5% transmission rates, respectively. Without treatment, transmission occurred in 26% (9/34) of women with a LIAISON avidity value between 0.250 and 0.350. Valacyclovir significantly reduced transmission to 2%, 5%, and 16% in preconceptional, periconceptional, and T1 periods, respectively (P = .034, .003, and <.001).Conclusions. Our results suggest that the LIAISON threshold for high avidity should be >0.250. Except for VIDAS avidity values ≥0.60 before 12 weeks, treatment is beneficial for women with positive IgM and low/intermediate avidity, regardless of MPI timing.
Cytomegalovirus (CMV) is the leading pathogen responsible for congenital infections, with a neonatal prevalence of 0.40% to 0.60% in Western Europe and in the United States [1]. Congenital CMV (cCMV) infection may follow a maternal primary or nonprimary infection. Maternal primary infection (MPI) occurs in 1%-2% of pregnancies. When fetal infection follows a MPI in the first trimester of pregnancy (T1) or in the periconceptional period, approximately 50% of infected fetuses develop long-term sequelae. These long-term sequelae encompass a wide neurological and/or auditory spectrum [1]. For MPI in the T1 or the periconceptional period, the risk of fetal transmission is estimated to be 36% and 10%, respectively [2,3].
Since 2019, the management of T1 and periconceptional MPIs has moved towards secondary prevention using valacyclovir, which has proven effective in preventing materno-fetal transmission [4][5][6][7]. The validation of laboratory algorithms to correctly identify women with MPI in their T1 or periconceptional period is therefore crucial. The current state-of-the-art practice is to test for immunoglobulin (Ig) G (IgG) and IgM followed by IgG avidity in cases with positive IgG and IgM to best estimate the date of the MPI [8,9]. The interpretation of avidity results according to the manufacturers' recommendations are 2-fold: a high avidity excludes a recent MPI in the last 3 to 4 months, whereas a low or intermediate avidity cannot exclude a recent MPI. In order to better counsel pregnant women on the risk of fetal infection, we aimed to refine the interpretation of low and intermediate IgG avidity values in women with positive IgG and IgM in their first trimester. The objective was to obtain a better estimate of the risk of vertical transmission based on the value of these avidity indexes and on the date of sampling. We compared results obtained with 2 commonly used avidity assays (LIAISON CMV IgG Avidity II [Diasorin] and VIDAS CMV IgG Avidity II [Biomérieux]) [10][11][12][13] in women with or without valacyclovir prophylaxis.
## METHODS
## CMV Primary Infection Samples
This study was carried out in a population of pregnant women managed in the Prenatal Diagnosis Unit of our center and who were tested for CMV serology in our Virology laboratory from 2012 to 2024. Gestational age was determined by ultrasound, based on the measurement of fetal crown-rump length before 14 weeks of gestation. In our center, screening for CMV serology (IgG and IgM) is offered to all women at the booking visit in the first trimester at a median of 11 weeks. In cases with positive or equivocal IgM and positive IgG, IgG avidity is measured using 2 commercial assays (LIAISON CMV IgG Avidity II and VIDAS CMV IgG avidity II). If the avidity result is low or intermediate (as defined in Supplementary Table 1) with at least 1 of the 2 avidity assays, the woman is considered at risk of fetal infection. Since 2019, we offer valacyclovir prophylaxis to women from the diagnosis of MPI up until prenatal diagnosis by amniocentesis [9]. In all cases, amniocentesis was performed from 17 weeks of gestation onwards. Diagnosis of fetal infection relies on a positive CMV polymerase chain reaction (PCR) on amniotic fluid.
The inclusion criteria in the study were women at T1 with positive IgG, positive or equivocal IgM, low or intermediate IgG avidity values, and a result of CMV PCR in amniotic fluid.
Women included in the study gave their consent for the use of their clinical and laboratory test data. The study was reviewed and approved by the Institutional Review Board (IRB) of Necker-Enfants Malades Hospital under no. 20191120134940.
## CMV IgG and IgM Testing
IgG and IgM testing was performed on the LIAISON XL platform with LIAISON CMV IgG II and LIAISON CMV IgM II. Avidity assays with both the VIDAS CMV IgG Avidity II and LIAISON CMV IgG Avidity II assays were performed in sera with positive or equivocal IgM and IgG levels at least twice the recommended threshold value [9].
## CMV IgG Avidity Testing
IgG avidity data with the VIDAS CMV IgG Avidity II and LIAISON CMV IgG Avidity II assays were obtained prospectively. Both avidity techniques were performed in parallel at the time of diagnosis if the volume was sufficient. When the serum volume was limited and did not allow parallel testing, the VIDAS technique was preferred. In cases where the serum volume did not allow the VIDAS test to be performed, the LIAISON test, which required a lower volume, was performed. Results were interpreted according to the manufacturers' recommendations (see Supplementary Table 1).
## Estimate of the Date of Primary CMV Infection Based on VIDAS Avidity
The timing of MPI was determined using a logarithmic model based on VIDAS avidity results (MyCMV clinical trial NCT 06694428). This model was trained on 208 MPI cases for which the exact timing was known from serial serological profile including positive IgM with negative IgG or IgG seroconversion between 2 sera fewer than 4 weeks apart. In cases with positive IgM and negative IgG, we estimated the date of MPI 10 days earlier. In case of seroconversion, the date of MPI was estimated at the middle date between the IgG-negative serum and the IgG-positive serum. The result of the model estimate was provided with a 97.5% CI.
Preconceptional, periconceptional, and T1 periods were defined as ≤5 weeks preconception, 4 weeks preconception to 3 weeks of gestation (1 week postconception), and 4 to 14 weeks of gestation (2 to 12 weeks postconception), respectively.
## Statistical Analysis
The comparison of 2 or more observed proportions was performed using a 1-sided Z-test for proportions, with 95% CIs calculated using a chi-square test. All statistical analyses were performed using GraphPad Prism 10, applying a 1-sided approach with a significance level set at P < .05. To analyze transmission according to avidity values and to the date of sampling, we applied a threshold value at 12 weeks to compare groups with equivalent size.
## RESULTS
## Population
A total of 561 cases with the inclusion criteria were retrieved from the laboratory database (Figure 1). Of these, 237 received valacyclovir prophylaxis in the T1 at a median of 10.9 weeks, while the remaining 324 did not.
## Vertical Transmission Rate According to Avidity Values
## In the Group Without Valacyclovir
The positive-predictive values (PPVs) of a low avidity were 34% and 31% with the VIDAS and LIAISON assay, respectively (P = .28) (Figure 2A). The PPV of an intermediate avidity was 14% for the VIDAS assay and 22% with the LIAISON XL assay. The risk of transmission was significantly lower in cases with intermediate VIDAS avidity than in cases with low VIDAS avidity (P < .001), while the risk of transmission was not significantly lower in cases with intermediate LIAISON avidity compared with those with low LIAISON avidity (P = .065).
One serum sample had a high VIDAS avidity and an intermediate LIAISON avidity; there was no transmission in that case. Forty-one sera had an intermediate or low VIDAS avidity and a high LIAISON avidity. Among these 41 cases, vertical transmission occurred in 9 (Supplementary Table 2); all 9 cases had a LIAISON avidity greater than 0.250 but less than 0.350.
## In the Valacyclovir Group
In the valacyclovir group, the overall vertical transmission was significantly lower than in the untreated group: 9% (22/237) versus 27% (89/324) (P < .001) (Figure 2B). For low and intermediate VIDAS avidities, the transmission rate was significantly lower in treated than in untreated patients: 12% (17/138) versus 34% (53/156) (P < .001) and 3.4% (3/87) versus 14% (18/127) (P = .005), respectively. For low and intermediate LIAISON avidities, the transmission rate was significantly lower in treated than in untreated patients: 11% (13/117) versus 31% (39/127) (P < .001) and 9% (7/77) versus 22% (21/97) (P = .013), respectively. Of note, a high avidity value with 1 of the 2 assays was identified in 29 cases-27 with LIAISON and 2 with VIDAS. Among these, only 1 case of transmission was identified: a case with a high LIAISON avidity value (0.309) and a low VIDAS avidity value (0.32) at 12 weeks (see Supplementary Table 3).
## Transmission Rate in Untreated Women According to VIDAS Avidity Values and Gestational Age at the Time of Sampling
The population was divided into 2 groups of similar size and analyzed according to whether the serum was collected before (n = 138) or after (n = 145) 12 weeks. The median gestational age at sampling was 8 weeks (IQR: 6-10 weeks) and 13 weeks (IQR: 12-14 weeks) in the pre-and post-12-week groups, respectively. When the avidity was low in sera collected before or after 12 weeks, the transmission rate was 36% and 32%, respectively (Figure 3). When avidity was intermediate in sera collected before 12 weeks, the rate of transmission was at its lowest at 8.7% (6/69; see Figure 3 and Supplementary Table 4). Among the 58 cases with intermediate avidity ranging from 0.40 to 0.64 after 12 weeks, transmission occurred in 12 (21%) (see Supplementary Table 5). In cases with an avidity measured between 0.40 and 0.50, the transmission rates were 27% and 13% when the serum was collected after or before 12 weeks. In cases with an avidity measured between 0.50 and 0.65, the transmission rates were 17% and 5% when the serum was collected after or before 12 weeks (see Figure 4 and Supplementary Table 6).
The risk of transmission was calculated for cases with intermediate VIDAS avidity values in the "high-intermediate" range of 0.50 to less than 0.65 according to gestational age at the time of sampling (see Supplementary Table 6). The risk of transmission was 0% (0/6) for avidity values between 0.60 and less than 0.65 and 5% (2/38) for avidity values between 0.50 and less than 0.65.
## Rate of Vertical Transmission According to Date of MPI in Treated and Untreated Women
Without valacyclovir, the transmission rate was 10% (8/86), 22% (17/76), and 40% (54/136) in the preconceptional (< 4 weeks prior conception), periconceptional, and T1 periods, respectively (Figure 5). With valacyclovir, the transmission rate was 2% (1/57), 5% (3/58), and 16% (18/115) in the preconceptional, periconceptional, and T1 periods, respectively. The difference was significant at all periods (P = .034, .003, and <.001, respectively).
## DISCUSSION
The recent demonstration that antiviral treatment with highdosage oral valacyclovir is efficient in reducing CMV vertical transmission in women with MPI in T1 is a major breakthrough [3][4][5][6]. This progress must be accompanied by efforts to better assess the risk of viral transmission in women with positive and equivocal IgM in the T1. This study provides refined estimates of the risk of vertical transmission in women with positive IgM at T1 and a low or intermediate avidity according to their avidity value, the time of sampling, and whether valacyclovir was prescribed or not in a large population of patients with MPIs with well-documented gestational dates.
As expected, patients with low avidity in their T1 represent a high-risk group for vertical transmission. In untreated women, low avidity values had the highest PPV for fetal transmission (34% [53/156] and 31% [39/127] for the VIDAS and LIAISON assays, respectively).
For women with intermediate avidity, the risk of vertical transmission was lower at 14% (18/127) and 22% (21/97) with the VIDAS and the LIAISON assays, respectively. This rate of transmission in cases with intermediate avidity is moderate but not negligible and these women need to be treated by valacyclovir.
In women with a low IgG avidity with the VIDAS assay, transmission rates were 32% and 36% when the serum sample was collected before or after 12 weeks, respectively. An intermediate avidity obtained in a serum sample collected after 12 weeks led to a higher transmission rate than when obtained in a serum sample collected before 12 weeks (21% [12/58] vs 9% [6/69], P = .027). With regard to the "high intermediate" values obtained in sera collected before 12 weeks, the risk of transmission was zero (0/6) for avidity values between 0.60 and less than 0.65 and 5% (2/38) for avidity values between 0.50 and less than 0.65. This 5% risk of transmission is in the range of that reported in women seropositive before their pregnancy and experiencing nonprimary infection [14][15][16]. Therefore, whether pregnant women presenting before 12 weeks with positive IgG and IgM and an intermediate VIDAS avidity value, particularly if this value is between 0.60 and less than 0.65, should receive valacyclovir prophylaxis remains a matter of discussion. We also detected a flaw in the LIAISON CMV IgG Avidity II technique. The threshold value for high avidity claimed by the manufacturer is greater than 0.250. In our series, among 34 patients with LIAISON avidity values greater than 0.250 but less than 0.350, 26% (9/34) of patients transmitted the virus to their fetus. Vauloup-Fellous et al [11], using the first version Rate of vertical CMV transmission in women treated with valacyclovir and untreated women according to date of maternal primary infection. Distant preconceptional, period < 8 weeks prior conception; preconceptional, 5 to 8 weeks prior conception; periconceptional, 4 weeks prior conception to 3 weeks of gestation (1 week after conception); early T1, 4 to 7 weeks of gestation (2 to 5 weeks after conception); late T1, 8 to 13 weeks of gestation (6 to 11 weeks after conception). Abbreviations: CMV, cytomegalovirus; T1, first trimester.
of the LIAISON CMV IgG Avidity kit, reported high avidity values (>0.250) in 4% of sera from women with recent MPI, while no high avidity values were reported with the VIDAS assay in the same population. We would like to draw the attention of clinical virologists to this shortcoming and suggest using a higher cutoff value (eg, >0.350) as the threshold for high LIAISON avidity. In cases with LIAISON avidity values greater than 0.250 and less than 0.350, we also suggest ensuring that IgG levels are more than twice the threshold value and retesting using a different avidity assay.
As anticipated, the risk of vertical transmission was significantly lower in treated women than in untreated women (9.3% [22/237] vs 27% [89/324]; P < .001). This 66% reduction is similar to that reported in the literature [7]. Valacyclovir significantly decreased the transmission rate for patients with low and intermediate avidity obtained with VIDAS and LIAISON (P < .001 and P = .005 and P < .001 and 0.013, respectively). Without treatment, the transmission rate was 10% and 22% for MPI that occurred in the preconceptional (< 4 weeks prior conception) and periconceptional periods, respectively. Interestingly, in women treated with valacyclovir, transmission rates were significantly reduced irrespective of the timing of MPI, including a reduction in MPI that occurred in the preconceptional and periconceptional periods (P = .034 and 0.003, respectively), suggesting the efficacy of valacyclovir also in these cases. The efficacy of valacyclovir, implemented at a median of 10.6 weeks for an MPI that occurred 4 to 8 weeks before conception, could be explained by its efficacy in avoiding delayed fetal infections. Indeed, without valacyclovir, the interval between the onset of MPI and fetal infection is less than 8 weeks in most cases [17]; however, in 10% of cases, a longer interval has been observed and this interval may be as long as 19 weeks [17,18].
A strength of this study is the large population of treated and untreated women with documentation of (1) avidity results in the T1 with 2 avidity assays largely used worldwide (VIDAS CMV IgG Avidity II kit and LIAISON CMV IgG Avidity II) and (2) PCR findings of CMV in amniotic fluid. Moreover 10 women transmitted the virus with high LIAISON avidity of more than 0.250 and less than 0.350, suggesting that the threshold for the avidity assay should be higher (eg, >0.350), more data are needed to accurately define a new threshold.
## CONCLUSION
This study provides important information on the risk of CMV vertical transmission according to the avidity value obtained by 2 different commercial assays (VIDAS CMV IgG Avidity II and LIAISON CMV IgG Avidity II). A low avidity with the VIDAS assay had the highest PPV for fetal transmission. However, the risk of vertical transmission was still high (14% to 22%) in cases with intermediate avidity values with both the VIDAS and LIAISON assays, suggesting that valacyclovir is indicated in those cases. With the exception of the subgroup of women presenting before 12 weeks with a VIDAS intermediate avidity above 60%, our results suggest the benefits of treating women with MPI in the pre-and periconceptional periods.
## Supplementary Data
Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
## Notes
Author contributions. All authors were involved in patient care, data analysis and interpretation. J. F., M. B., and M. L.-V. drafted the first version of the manuscript, and all authors participated in the revision of the manuscript and approved the manuscript as submitted. All authors agree to be responsible for all aspects of the work.
Data availability. The data that support the findings of this study are available from the corresponding author upon reasonable request.
## References
1. Leruez-Ville, Foulon, Pass et al. (2020) "Cytomegalovirus infection during pregnancy: state of the science" *Am J Obstet Gynecol*
2. Picone, Vauloup-Fellous, Cordier (2009) "A 2-year study on cytomegalovirus infection during pregnancy in a French hospital" *BJOG Int J Obstet Gynaecol*
3. Chatzakis, Ville, Makrydimas et al. (2020) "Timing of primary maternal cytomegalovirus infection and rates of vertical transmission and fetal consequences" *Am J Obstet Gynecol*
4. Shahar-Nissan, Pardo, Peled (2020) "Valaciclovir to prevent vertical transmission of cytomegalovirus after maternal primary infection during pregnancy: a randomised, double-blind, placebo-controlled trial" *Lancet*
5. Faure-Bardon, Fourgeaud, Stirnemann et al. (2021) "Secondary prevention of congenital cytomegalovirus infection with valacyclovir following maternal primary infection in early pregnancy" *Ultrasound Obstet Gynecol*
6. Egloff, Sibiude, Vauloup-Fellous (2023) "New data on efficacy of valacyclovir in secondary prevention of maternal-fetal transmission of cytomegalovirus" *Ultrasound Obstet Gynecol*
7. Chatzakis, Shahar-Nissan, Faure-Bardon (2023) "The effect of valacyclovir on secondary prevention of congenital cytomegalovirus infection, following primary maternal infection acquired periconceptionally or in the first trimester of pregnancy. An individual patient data meta-analysis" *Am J Obstet Gynecol*
8. Fourgeaud, Nguyen, Guilleminot et al. (2023) "Comparison of two serological screening strategies for cytomegalovirus primary infection in the first trimester of pregnancy" *J Clin Virol*
9. Leruez-Ville, Chatzakis, Lilleri (2024) "Consensus recommendation for prenatal, neonatal and postnatal management of congenital cytomegalovirus infection from the European Congenital Infection Initiative (ECCI)" *Lancet Reg Health Eur*
10. Lagrou, Bodeus, Van Ranst et al. (2009) "Evaluation of the new architect cytomegalovirus immunoglobulin M (IgM), IgG, and IgG avidity assays" *J Clin Microbiol*
11. Vauloup-Fellous, Berth, Heskia et al. (2013) "Re-evaluation of the VIDAS(®) cytomegalovirus (CMV) IgG avidity assay: determination of new cut-off values based on the study of kinetics of CMV-IgG maturation" *J Clin Virol*
12. Vauloup-Fellous, Lazzarotto, Revello (2014) "Grangeot-Keros L. Clinical evaluation of the Roche Elecsys CMV IgG avidity assay" *Eur J Clin Microbiol Infect Dis*
13. Delforge, Desomberg, Montesinos (2015) "Evaluation of the new LIAISON(®) CMV IgG, IgM and IgG avidity II assays" *J Clin Virol*
14. Barbosa, Yamamoto, Duarte (2018) "Cytomegalovirus shedding in seropositive pregnant women from a high-seroprevalence population: the Brazilian Cytomegalovirus Hearing and Maternal Secondary Infection Study" *Clin Infect Dis*
15. Zelini, Angelo, Cicco (2022) "Human cytomegalovirus non-primary infection during pregnancy: antibody response, risk factors and newborn outcome" *Clin Microbiol Infect*
16. Gatta, Rochat, Weber (2022) "Clinical factors associated with cytomegalovirus shedding among seropositive pregnant women" *Am J Obstet Gynecol MFM*
17. Enders, Daiminger, Exler et al. (2017) "Amniocentesis for prenatal diagnosis of cytomegalovirus infection: challenging the 21 weeks' threshold" *Prenat Diagn*
18. Revello, Furione, Zavattoni (2008) "Human cytomegalovirus (HCMV) DNAemia in the mother at amniocentesis as a risk factor for iatrogenic HCMV infection of the fetus" *J Infect Dis*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12833410&blobtype=pdf
|
# EDITED AND REVIEWED BY
Rustam Aminov, Avinash Karpe, Pooja Hegde, Coralia Bletotu, Zhigang Qiu
## Abstract
KEYWORDSantibiotic resistance genes (ARG), conjugative transfer genes, multi drug resistant (MDR), mechanism of action, microbiome and resistome, systems biology Editorial on the Research Topic Antimicrobial resistance: causes, mechanisms and mitigation strategies for gut dysbiosis
Antimicrobial resistance (AMR) continues to escalate as a major global health challenge, and studies over the last 10 years indicate that gut microbiome-and its expansive resistome-play a key role here. The gastrointestinal tract, a densely populated microbial ecosystem, is both vulnerable to antimicrobial disruption and capable of serving as a dynamic reservoir for resistance genes (Karpe et al., 2023;Ho et al., 2025;Nazir et al., 2025). This Research Topic, Antimicrobial Resistance: Causes, Mechanisms and Mitigation Strategies for Gut Dysbiosis, was conceived to address this dual role of the gut: as an ecological victim of antimicrobial pressures and as a driver of the evolution, persistence and dissemination of resistance. The Research Topic brings together diverse contributions that collectively highlight the mechanisms underpinning dysbiosis, the pathways through which resistance emerges and propagates, and the possibilities for novel mitigation strategies.
The Research Topic contains, within it, two mechanistic studies using murine models, each providing detailed insights into how antimicrobial stress shapes gut ecology. In the work by Ding et al., the authors investigate conjugative transfer of antibiotic resistance genes mediated by plasmids across different intestinal segments. Their findings reveal striking anatomical specificity: the small intestine emerges as a primary site for donor strains and plasmid-borne antibiotic resistance genes (ARG) localization. This is accompanied by shifts in microbial richness-rising in the duodenum, jejunum and large intestine, but declining notably in the ileum. The concurrent expansion of Proteobacteria under plasmid-transfer pressure offers further evidence of how mobile genetic elements can reshape intestinal microecology. Complementing this, Ma et al. examined the combined impact of multidrug-resistant Escherichia coli and multi-antibiotic exposure. In their study, they observed dramatic reductions in operational taxonomic unit richness, with communities shrinking from over 200 OTUs in controls to just a few dozen in antibiotic-challenged groups. Again, the small intestine was shown to be the key colonization site for resistant strains. Together, these studies expose the profound and segment-specific ecological consequences of antimicrobial and ARG pressures within the gut, pointing to a complex interplay between microbial competition, niche restructuring and gene mobility.
From animal models, the Research Topic shifts to early-life human ecology with the study by Qiu et al. In this study, the authors explored how neonatal antibiotic therapy-specifically amoxicillin-clavulanic acid and moxalactam-affected the initial establishment of the gut microbiota and key functional taxa, notably butyrate-producing bacteria. Their findings of reduced microbial diversity and depletion of butyrate producers in antibiotic-exposed infants highlight how early-life antimicrobial exposure may have enduring consequences. Perturbations during this critical developmental window may influence immune maturation, metabolic programming and the long-term architecture of the gut resistome. This work underscores the importance of stewardship in pediatric settings, where microbiome trajectories are still being established and may be particularly vulnerable to pharmacological disruption.
The Research Topic also includes two comprehensive reviews that provide conceptual frameworks essential for interpreting these experimental findings, and future aspects in building resilience to AMR. In Deshpande et al., the authors frame the gut microbiome as an emerging epicenter of AMR, synthesizing evidence on how reservoirs of resistance genes within commensal communities interact with pathogens via horizontal gene transfer. Their review positions the gut not merely as a site affected by antibiotics but as a crucial engine of resistance evolution. Furthermore, Al-Kuwari et al. presented a perspective pointing toward future translational possibilities: microbiome-based interventions that might complement or even reduce reliance on antibiotics. By exploring strategies such as dietary modulation, probiotics, fermented foods and microbiota restoration, the authors highlight the potential for leveraging microbial diversity to counteract resistance pressures. Their contribution reflects a growing recognition that rebuilding or preserving ecological stability may be as critical as combating individual resistant pathogens.
Finally, the One Health dimension of AMR was presented by Lertwatcharasarakul et al. through their retrospective analysis of antimicrobial resistance in Salmonella spp. isolated from livestock and their environment in Thailand. By profiling resistance patterns across animal and environmental samples, they illustrated how antimicrobial use within agricultural and veterinary settings shapes resistance in zoonotic pathogens and their surrounding niches. This work underscores the continuity between gut-associated resistomes in animals and broader environmental reservoirs. Their study also highlighted the risk of bidirectional transmission of resistant organisms between livestock, the food chain and humans. In doing so, it links the gut-centered themes of this Research Topic to wider questions of agricultural practice, environmental stewardship and cross-species transmission.
Altogether, these articles build a coherent narrative: that AMR cannot be solved by pharmaceutical innovation alone, but demands an ecological understanding of the gut microbiome and its role in resistance evolution. The findings reinforce that antimicrobial exposures-whether through plasmid-mediated transfer, resistant organisms or early-life therapies-can dramatically reconfigure gut communities in ways that promote ARG retention and dissemination. At the same time, the reviews and perspectives highlight the promise of resilience-based and microbiome-centric interventions that move beyond conventional antibiotic strategies.
As this Research Topic demonstrates, addressing AMR requires integration across mechanistic biology, microbial ecology, clinical practice and One Health policy. The gut microbiome, long overlooked in AMR frameworks, now stands as a central player. We hope that this Research Topic not only advances understanding of the complex interplay between antimicrobials, gut ecology and resistance, but also inspires future work that blends ecological insight with translational ambition to preserve both microbiome health and antimicrobial effectiveness.
## References
1. Ho, Wong, Aung et al. (2025) "Antimicrobial resistance: a concise update" *Lancet Microbe*
2. Karpe, Beale, Tran (2023) "Intelligent biological networks: improving anti-microbial resistance resilience through nutritional interventions to understand protozoal gut infections" *Microorganisms*
3. Nazir, Nazir, Zuhair et al. (2025) "The global challenge of antimicrobial resistance: mechanisms, case studies, and mitigation approaches" *Health Sci. Rep*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12797184&blobtype=pdf
|
# Cytomegalovirus IL-10 in Plasma as a Marker of Active Infection in Allogeneic Hematopoietic Transplant Recipients: An Exploratory Study
Ángela Sánchez-Simarro, Eliseo Albert, Estela Giménez, Ester Colomer, | Ariadna Pérez, José Piñana, | Solano, David Navarro
## Abstract
We investigated whether plasma cytomegalovirus (CMV) IL-10 (cmvIL-10) levels could serve as a biomarker of active CMV replication in allogeneic hematopoietic transplant recipients (allo-HCT) in the presence or absence of letermovir (LMV) prophylaxis. A total of 189 leftover plasma samples that tested positive for CMV DNA (Alinity m CMV assay), representing 33 episodes of CMV DNAemia were run on a laboratory-developed enzyme-linked immunosorbent assay for cmvIL-10 quantification. Eighteen episodes developed during LMV prophylaxis. Overall, 16 episodes of CMV DNAemia were classified as clinically significant (CsCMVi). There was an overall very weak correlation between the two biomarkers (Rho = 0.10; p = 0.16). Overall, the median cmvIL-10 area under the curve (AUC) until CMV DNA levels reached their peak was significantly higher (p < 0.001) in CsCMVi episodes than in non-CsCMVi episodes. cmvIL-10 AUC between Days 14 and 23 after allo-HCT (AUC₁₄₋₂₃) values were significantly higher in CsCMVi episodes compared with non-CsCMVi episodes among patients receiving LMV therapy (p = 0.008). An AUC₁₄₋₂₃ cutoff value of log 10 3.06 discriminated anticipately between CsCMVi and non-CsCMVi with a sensitivity and specificity of 100%. Plasma cmvIL-10 levels may reflect true CMV replication and thus provide a unique perspective on viral dynamics, serving as an ancillary marker to CMV DNA monitoring.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
## 1 | Introduction
Monitoring CMV DNA load in blood is a cornerstone of cytomegalovirus (CMV) infection management in allogeneic hematopoietic transplant recipients (allo-HCT) recipients, as it enables timely initiation of antiviral therapy to control viral replication and thereby minimize the risk of CMV end-organ disease [1]. However, the presence of CMV DNA in blood does not always indicate an ongoing episode of viral replication, since free viral DNA may transiently leak into the bloodstream and be rapidly cleared (i.e., "blips" or self-resolving episodes). This phenomenon is particularly frequent among patients receiving letermovir (LMV) prophylaxis [2,3], as LMV inhibits the production of infectious viral particles but does not block CMV DNA replication [4]. While the kinetic pattern of CMV DNA in plasma or whole blood often provides valuable information in this context [5,6], novel biomarkers of active CMV infection, such as CMV UL25.1 RNAemia, have been proposed [7][8][9]. The CMV gene UL111a encodes a homolog of human interleukin-10 (hIL-10), referred to as cmvIL-10, which is predominantly expressed during lytic infection [10]. cmvIL-10 exerts a broad range of immunosuppressive functions through its interaction with the human IL-10 receptor (IL-10R1) [11,12]. Deep sequencing of the UL111a gene directly from clinical samples has revealed the presence of viral variants that may differentially modulate host immune responses [13]. Furthermore, it was recently shown that expression of cmvIL-10 RNA in peripheral blood from kidney transplant recipients was positively associated with an increase in viral DNA detection in subsequent specimens, suggesting that monitoring cmvIL-10 may be ancillary to viral DNA to allow early detection of active CMV infection in transplant recipients [14]. In this study, we investigated whether plasma cmvIL-10 levels could serve as a biomarker of active CMV replication in patients with CMV DNAemia, occurring either during or outside LMV prophylaxis.
## 2 | Materials and Methods
## 2.1 | Patients and Specimens
A convenience panel of 189 leftover plasma samples that tested positive for CMV DNA using the Alinity m CMV assay (Abbott Molecular Inc., Des Plaines, IL, USA) [5] was assembled. These samples represented 33 episodes of CMV DNAemia occurring in unique CMV-seropositive allo-HCT recipients that underwent primary LMV prophylaxis (Table 1). Eighteen episodes developed during LMV prophylaxis (n = 110 specimens), and 15 occurred outside LMV prophylaxis (n = 79 specimens). Sixteen episodes of CMV DNAemia were classified as clinically significant (CsCMVi), defined as those in which peak CMV DNA levels exceeded the institutional threshold for preemptive antiviral therapy (1500 IU/mL, regardless of LMV prophylaxis status). Seven of these 16 CsCMVi episodes occurred in patients receiving LMV therapy. No CsCMVi episodes involved CMV end-organ disease. All plasma samples had been cryopreserved at -80°C within 24 h of collection between January 2023 and February 2025 and had not been thawed before cmvIL-10 testing. The current study was approved by the Institutional Review Board (IRB) Research Ethics Committee of Hospital Clínico Universitario INCLIVA (2024/153). The IRB issued an informed consent waiver.
## 2.2 | Quantification of cmvIL-10 in Plasma
Quantification of cmvIL-10 in plasma was performed using an enzyme-linked immunosorbent assay (ELISA) as previously described [15]. Briefly, 96-well Nunc MaxiSorp plates were coated with 100 μL of capture antibody (cmvIL-10 Affinity Purified Polyclonal Antibody, Goat IgG; R&D Systems, Minneapolis, MN) at a final concentration of 2 μg/mL and incubated overnight at 4°C. Following removal of the capture antibody, plates were washed three times with wash buffer (1 × PBS + 0.05% Tween 20) and blocked with blocking buffer (PBS + 3% bovine serum albumin) for 1 h at room temperature. After three additional washes, 100 μL of plasma samples (diluted 1:5) and standards were added in duplicate and incubated for 2 h at room temperature. Protein standards were prepared by serial twofold dilutions of recombinant cmvIL-10 (R&D Systems) in protein standard dilution buffer (PBS + 0.1% BSA), ranging from 1000 to 15.625 pg/mL. Following incubation and washing, detection antibody (cmvIL-10 Biotinylated Affinity Purified Polyclonal Antibody, Goat IgG; R&D Systems) was added at a final concentration of 0.2 μg/mL and incubated for 2 h at room temperature. After three washes, streptavidin-HRP (R&D Systems) was added at working concentration and incubated for 20 min in the dark. Following three final washes, substrate solution (R&D Systems) was added and incubated in the dark for 20 min. The reaction was stopped with 50 μL of stop solution (1 M H₂SO₄), and optical density was measured at 450 nm using a Virclia instrument (Vircell, Granada, Spain). Standard curves were generated using GraphPad Prism software by plotting optical density values against recombinant cmvIL-10 concentrations using four-parameter logistic curve fitting. Plasma cmvIL-10 concentrations (pg/mL) were interpolated from the standard curve.
## 2.3 | Statistical Analyses
Differences between medians were compared using the Mann-Whitney U test. The degree of correlation between continuous variables was analyzed using Spearman's Rank test. The cmvIL-10 area under the curve (AUC) was calculated when appropriate, and required two or more specimens/patient. The Youden index was used to determine the optimal AUC threshold to maximize the difference between the true positive rate (sensitivity) and the false positive rate (1-specificity). Two-sided exact p-values are reported. A p-value < 0.05 was considered statistically significant. The analyses were performed using the GraphPad Prism 9.0.2 statistical package.
## 3 | Results
## 3.1 | Correlation Between cmvIL-10 and CMV DNA Levels in Plasma
We first investigated whether plasma cmvIL-10 levels correlated with CMV DNA loads as measured by a real-time PCR assay. A total of 189 plasma specimens with quantifiable CMV DNA, representing 33 episodes of active CMV infection that developed a median of 26 days after allo-HCT (IQR, 8.5-65), were available for cmvIL-10 testing (median of 5 samples per episode; range, 3-12). The timing of specimen collection relative to allo-HCT is presented in Table S1. Among the 189 specimens, 135 tested positive for cmvIL-10 (median value, 190 pg/mL; IQR, 82.3-373.2), while 54 were undetectable. Median cmvIL-10 peak value was 396.5 pg/mL (range, 83-5657). As shown in Figure 1, there was an overall very weak correlation between the two biomarkers (Rho = 0.10; p = 0.16). The degree of correlation was poor regardless of whether patients were under LMV prophylaxis (Figure 1B) or not (Figure 1C).
## 3.2 | cmvIL-10 as a Marker of Active Infection
To assess whether cmvIL-10 measurement could identify CsCMVi episodes, we calculated the AUCs using cmvIL-10 values from specimens collected until the CMV DNA peak level. This parameter was available for 29 of the 33 episodes (CsCMVi, n = 15; non-CsCMVi, n = 14). As shown in Figure 2A, the median cmvIL-10 AUC was significantly higher (p < 0.001) in CsCMVi episodes than in non-CsCMVi episodes (log 10 3.1 vs. log 10 0.56). Notably, the number of specimens used for AUC calculations was comparable across groups (median, three specimens; p ≥ 0.5), as was the timing of the first specimen collection after allo-HCT (median, 18 days for CsCMVi vs. 19 days for non-CsCMVi). A similar pattern was observed when only episodes occurring during LMV prophylaxis were analyzed separately (Figure 2B).
We next investigated whether cmvIL-10 AUCs could predict the occurrence of CsCMVi. To this end, we calculated AUCs between days 14 and 23 post-allo-HCT (cmvIL-10 AUC₁₄₋₂₃), encompassing measurements from plasma specimens collected prior to the CMV DNAemia peak (median, 7 days; range, 6-10 days earlier). As shown in Figure 3A, there was a trend toward higher cmvIL-10 AUC₁₄₋₂₃ values in CsCMVi episodes (median, log 10 3.03) compared with non-CsCMVi episodes (median, log 10 2.4; p = 0.08). The median number of specimens used for AUC calculations was similar across groups (n = 3). In this setting, an AUC cutoff value of log 10 2.45 provided the most efficient discrimination between CsCMVi and non-CsCMVi, with a sensitivity of 50% and a specificity of 89% (AUC = 0.73; p = 0.06). Interestingly, cmvIL-10 AUC₁₄₋₂₃ values were significantly higher in CsCMVi episodes compared with non-CsCMVi episodes among patients receiving LMV therapy (p = 0.008). In this setting, an AUC₁₄₋₂₃ cutoff value of log 10 3.06 discriminated perfectly across comparison groups, with a sensitivity and specificity of 100% (AUC = 1; p = 0.009).
## 4 | Discussion
The kinetics of CMV DNA in blood has proven to be a useful parameter for anticipating the occurrence of clinically significant CMV infection (CsCMVi) in allo-HCT recipients irrespective of the real-time PCR assay employed [16,17]. It is well documented that the presence of CMV DNA in blood may not necessarily reflect active CMV infection in tissues or in the blood compartment due to the mechanism of action of LMV [2,3]. In this context, identifying episodes of CMV DNAemia that represent true viral replication, as opposed to abortive infection, is particularly relevant in patients receiving LMV, to avoid unnecessary interruption of therapy. Our group has demonstrated the potential value of CMV DNA doubling time (CMV dt) for the early identification of true episodes of active CMV infection, both in the presence and absence of LMV treatment [5,6,16,17]; nevertheless, the clinical performance of this parameter, in terms of its predictive value, is not maximally accurate.
Monitoring of CMV UL25.1 RNAemia using a commercially available assay has been proposed as a reliable marker of active CMV infection, since this late CMV transcript appears to be virion-associated in plasma [7,8]. Nevertheless, because LMV does not inhibit either CMV DNA replication or the synthesis of late CMV mRNAs, we believe that this marker does not substantially enhance the information already provided by quantitative CMV DNA testing in patients receiving LMV [9]. Here, we reasoned that monitoring plasma cmvIL-10 levels might aid in the early identification of CsCMVi, as this viral humanhomolog cytokine is mainly synthesized during viral replication and exerts immunosuppressive effects that may vary across strains [10][11][12][13]. We found a negligible correlation between CMV DNA and cmvIL-10 levels, in contrast to what has been reported for CMV UL25.1 RNAemia. Although speculative, this observation may indicate that the detection of cmvIL-10 in blood at certain levels more accurately reflects the presence of infectious viral particles in the bloodstream as a result of productive infection in tissues or even within the blood compartment, compared with the detection of particle-free viral DNA in plasma. Furthermore, monitoring cmvIL-10 levels could reliably identify and, more importantly, anticipate clinically significant CMV infection (CsCMVi) both in the presence and absence of LMV therapy. Notably, an AUC cutoff value of log₁₀ 3.06 perfectly discriminated between CsCMVi and non-CsCMVi groups, with this distinction occurring a median of 1 week before CMV DNAemia reached levels sufficient for classification as such. We acknowledge two main limitations of this study: first, the very small sample size, and second, the lack of CMV end-organ disease cases among CsCMVi episodes. Taken together, our findings support the notion that plasma cmvIL-10 levels may reflect true CMV replication and thus provide a unique perspective on viral dynamics, serving as an ancillary marker to CMV DNA monitoring. Nevertheless, well-designed prospective studies are warranted to further assess the clinical utility of cnvIL-10 testing, particularly in LMV-treated patients.
## References
1. Ljungman, Alain, Chemaly (2025) "Recommendations From the 10th European Conference on Infections in Leukaemia for the Management of Cytomegalovirus in Patients After Allogeneic Haematopoietic Cell Transplantation and Other T-Cell-Engaging Therapies"
2. Cassaniti, Colombo, Bernasconi (2021) "Positive HCMV DNAemia in Stem Cell Recipients Undergoing Letermovir Prophylaxis Is Expression of Abortive Infection" *American Journal of Transplantation*
3. Giménez, Guerreiro, Torres (2023) "Features of Cytomegalovirus DNAemia and Virus-Specific T-Cell Responses in Allogeneic Hematopoietic Stem-Cell Transplant Recipients During Prophylaxis With Letermovir" *Transplant Infectious Disease*
4. Piret, Boivin (2025) "Antiviral Drugs Against Herpesviruses" *Advances in Experimental Medicine and Biology*
5. Esteve, Albert, Giménez (2024) "Assessment of Cytomegalovirus DNA Doubling Time and Virus-Specific T-Cell Responses in the Management of CMV Infection in Allogeneic Hematopoietic Stem Cell Transplant Recipients Undergoing Letermovir Prophylaxis" *Bone Marrow Transplantation*
6. Gimenez, García Cadenas, Piñana (2025) "Cytomegalovirus DNA Doubling Time for Early Identification of Clinically Significant Infection Episodes in Allogeneic Hematopoietic Stem Cell Transplant Recipients Undergoing Primary Letermovir Prophylaxis: A Multicenter Study" *Transplant Infectious Disease*
7. Piccirilli, Lanna, Gabrielli (2024) "CMV-RNAemia as New Marker of Active Viral Replication in Transplant Recipients" *Journal of Clinical Microbiology*
8. Giardina, Paolucci, Mele (2025) "Human Cytomegalovirus Virion-Associated mRNA as a Marker of Productive Infection in Immunocompromised Patients" *Journal of Medical Virology*
9. De La Asunción, Ocete, Giménez (2025) "Cytomegalovirus UL21-5 mRNAemia as a Marker of Active Virus Replication in Allogeneic Hematopoietic Transplant Recipients" *Journal of Medical Virology*
10. Kotenko, Saccani, Izotova et al. (2000) "Human Cytomegalovirus Harbors Its Own Unique IL-10 Homolog (cmvIL-10)" *Proceedings of the National Academy of Sciences*
11. Poole, Neves, Oliveira et al. (2020) "Human Cytomegalovirus Interleukin 10 Homologs: Facing the Immune System" *Frontiers in Cellular and Infection Microbiology*
12. Jones, Logsdon, Josephson et al. (2002) "Crystal Structure of Human Cytomegalovirus IL-10 Bound to Soluble Human IL-10R1" *Proceedings of the National Academy of Sciences*
13. Waters, Lee, Ariyanto (2022) "Sequencing of the Viral UL111a Gene Directly From Clinical Specimens Reveals Variants of HCMV-Encoded IL-10 That Are Associated With Altered Immune Responses to HCMV" *International Journal of Molecular Sciences*
14. Almeida, Oliveira, Martines (2025) "Expression Profile of Human Cytomegalovirus UL111A cmvIL-10 and LAcmvIL-10 Transcripts in Primary Cells and Cells From Renal Transplant Recipients" *Viruses*
15. Young, Mariano, Faure et al. (2021) "Detection of Cytomegalovirus Interleukin 10 (cmvIL-10) by Enzyme-Linked Immunosorbent Assay (ELISA)"
16. Giménez, Muñoz-Cobo, Solano et al. (2014) "Early Kinetics of Plasma Cytomegalovirus DNA Load in Allogeneic Stem Cell Transplant Recipients in the Era of Highly Sensitive Real-Time PCR Assays: Does It Have Any Clinical Value?" *Journal of Clinical Microbiology*
17. Vinuesa, Giménez, Solano et al. (2016) "Would Kinetic Analyses of Plasma Cytomegalovirus DNA Load Help to Reach Consensus Criteria for Triggering the Initiation of Preemptive Antiviral Therapy in Transplant Recipients?" *Clinical Infectious Diseases*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12689982&blobtype=pdf
|
# Whole genome sequencing as a reliable alternative for Salmonella serotyping: a comparative study with the gold-standard method
Swayam Prakash, Duolong Zhu, Patricia Alba, Hiroshi Kaneko, Israel Nissan, Greta Romano, Irene Mileto, Stefano Gaiarsa, Giulia Grassia, Jessica Bagnarino, Antonio Piralla, Vincenzina Monzillo, Patrizia Cambieri, Fausto Baldanti, Marta Corbella
## Abstract
Introduction: Salmonella is a major foodborne pathogen of significant global health concern, leading to conditions like gastroenteritis, with symptoms such as diarrhea, abdominal cramps, and fever. Although most cases are self-limiting, Salmonella enterica can cause bacteremia or typhoid fever in some cases. There are over 2600 known serotypes of S. enterica. Each serotype is distinguished by differences in its surface antigens (somatic O and flagellar H), which is the basis for their classification. Considering that traditional phenotypic typing is laborious and time-consuming, the goal of this study is to evaluate whole genome sequencing for S. enterica classification.Methods: From January 2021 to December 2024, 282 isolates of S. enterica were collected at the Microbiology and Virology department of Fondazione IRCCS Policlinico San Matteo (Italy). The isolates were serotyped for O and H antigens with the gold-standard serum agglutination method. Whole genome sequencing. WGS was performed using Illumina MiSeq and quality-filtered reads were used to perform serotype prediction using SeqSero2 and SISTR tools.
Results:The genomic typing of both tools shows an optimal concordance level with the gold-standard method, accurately predicting the serotype (both O and H antigens) for 236 out of 282 analyzed isolates. The discordance rate was 12 isolates out of 282. In addition, 34 isolates were correctly typed by the genomic methods, whereas typing with the gold-standard was unsuccessful. Genomic typing was able to classify all isolates while staying as close as possible to the present classification of Salmonella into subspecies and serovars.Discussions: Whole genome sequencing has proven to be a robust and reliable method for Salmonella typing. It allows for the simultaneous analysis of multiple samples and shortens processing times. Moreover, WGS enabled the recovery of identifications that could not be obtained using the gold-standard method.
## Introduction
Salmonella enterica is a major global cause of foodborne illness, with clinical manifestations influenced by bacterial load, host immune status, and serotype. Human infection is primarily associated with consumption of contaminated food (Carrera et al., 2023; European Centre for Disease Prevention and Control [ECDC], 2022; Huedo et al., 2017;Pagani et al., 2023;Petrin et al., 2024).
The genus Salmonella comprises two species, S. bongori and S. enterica, the latter of which is subdivided into six subspecies encompassing over 2,600 serovars, classified by somatic (O) and flagellar (H) antigens (Dos Santos et al., 2019). S. enterica subsp. enterica (subsp. I) includes approximately 1,600 serovars, of which only 32 demonstrate key features of pathogenicity, environmental persistence, and inter-ecosystem transmission, due to food handling and processing (Agbaje et al., 2011;Tack et al., 2020). Twelve serovars, including S. Enteritidis, S. Typhimurium, and S. Monophasic Typhimurium (4, [5],12:i:-), are implicated in ∼90% of foodborne outbreaks (European Food Safety Authority [EFSA] and European Centre for Disease Prevention and Control [ECDC], 2024).
Non-typhoidal Salmonella (NTS) is the second most reported zoonotic infection in the European Union (EU), accounting for the majority of foodborne outbreaks (European Food Safety Authority [EFSA] and European Centre for Disease Prevention and Control [ECDC], 2024). In 2023, 77,486 salmonellosis cases were reported, up from 65,478 in 2022. Eective prevention requires early detection across the food chain, from animal feed and livestock to processing and retail (Centers for Disease Control [CDC], 2013). Notification of NTS is mandatory in most EU countries, supported by surveillance systems such as Enter-Net Italy.
Serotyping is based on the Kaumann-White-Le Minor (KW) scheme, which distinguishes antigenic profiles via slide agglutination using specific antisera targeting 64 O and 114 H antigen variants (Grimont and Weill, 2007). The sero-agglutination test is the standard method. The model relies solely on observable phenotypic traits, which is therefore reflected in a subjective interpretation, which requires correctly trained workers. However, the agglutination method has several limitations, including (i) weaker and non-specific agglutination reactions that can lead to false-positive results; (ii) rough, non-motile, and mucoid strains that exhibit autoagglutination and loss of antigen expression; (iii) the technique necessitates the use of more than 150 highly specific antisera (Schrader et al., 2008).
The introduction of Next Generation Sequencing (NGS) and the decrease of sequencing expenses have made whole genome sequencing (WGS) more aordable, and it now functions as a substantial tool for pathogen subtyping, source tracing, and characterization, including virulence and antimicrobial resistance gene profiling (Deng et al., 2015). The use and application of WGS, including Salmonella serotyping, in routine laboratories is becoming a viable option. WGS oers superior discriminatory power over the gold-standard agglutination method because it enables rapid, high-resolution pathogen characterization and enhances outbreak source tracing, representing a transformative tool in Salmonella epidemiology surveillance (Deng et al., 2015;Koser et al., 2012).
Several studies have evaluated the use of WGS for Salmonella serotyping, assessing its reliability, performance, and applicability as an alternative to traditional phenotypic methods (Cooper et al., 2020;Ibrahim and Morin, 2018;Jibril et al., 2020;Robertson et al., 2018;Uelze et al., 2020;Wu et al., 2021;Xu et al., 2020;Yachison et al., 2017). These works collectively highlight the potential of WGS to provide accurate and high-throughput serotype prediction, streamline laboratory workflows, and support large-scale epidemiological investigations. However, the analysis of genomic data and the accurate identification of isolates are strongly influenced by the choice of bioinformatic tools. Among those most commonly employed, SeqSero2 and SISTR are widely used (Yoshida et al., 2016;Zhang et al., 2019). The extensive use of WGS for the identification of Salmonella serovars has led to the accumulation of Salmonella genomes in comprehensive databases such as NCBI's Pathogen Detection (comprising approximately 765,000 S. enterica genomes) and EnteroBase (with over 704,000 genomes), which are regularly updated to support ongoing surveillance and research (Dyer et al., 2024;The NCBI Pathogen Detection Project, 2016;Robertson et al., 2018).
The aim of this study was to assess the concordance between the gold-standard sero-agglutination method and two alternative approaches, WGS combined with SeqSero2 and SISTR, for the routine serotyping of Salmonella isolates in our hospital.
## Materials and methods
## Sample collection and Kauffmann-White-Le Minor serotyping
A total of 282 Salmonella isolates were collected from 2021 to 2024 at the Microbiology and Virology Unit of Fondazione IRCCS Policlinico San Matteo, a 900-bed hospital in Northern Italy.
Clinical samples (feces, blood, urine, synovial fluid and abscess) were cultured according to laboratory protocols and genus identification was confirmed by matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (Bruker Daltonik GmbH, Bremen, Germany) and Bruker BioTyper database version 3.1. The isolates were serotyped with rabbit antisera (SSI Diagnostica A/S, Denmark) following the traditional KW phenotypic method based on Ryan's scheme (Ryan et al., 2017). The serotype was determined by agglutinating pure colonies with polyvalent and then monovalent antisera to determine the somatic antigen (O) and then to characterize the flagellar antigens (H). This provided the antigenic formula of the isolate associated with the serotype name and subspecies.
## Whole genome sequencing
The genomic DNA extraction was performed by using the QIAsymphony DSP Virus/Pathogen Midi Kit (QIAGEN, Germany), according to the manufacturer's instructions. DNA was quantified by fluorescence, using the Qubit 4.0 fluorometer kit (Invitrogen, Carlsbad, CA, United States). All isolates underwent WGS on the Illumina MiSeq platform (San Diego, California, United States) performed with a 2 × 250 bp paired-end sequencing run. Libraries were prepared according to the manufacturer's protocol using the Nextera XT DNA library preparation kit to ensure optimal fragment size distribution and sequencing quality. The evaluation of raw sequencing data integrity was performed using the FastQC tool (Andrews, 2010). Raw reads with Phred scores below 20 were removed using the fastp tool (Chen, 2023) to include only high-quality reads.
## Serovar prediction from sequencing data
The SeqSero2 and SISTR tools were used to determine the serotype of the Salmonella isolates (Yoshida et al., 2016;Zhang et al., 2019). Paired-end FASTQ reads were used as input for the allele micro-assembly workflow of SeqSero2 (version 1.2.1) (Zhang et al., 2019), which predicts the serovar based on the detection and combination of specific genetic determinants associated with the O (somatic) and H (flagellar) antigens. Consensus sequences of the Salmonella isolates were generated using the Shovill tool (Shovill, 2020; Supplementary Table 1). The resulting FASTA files were subsequently analyzed using the SISTR tool (Yoshida et al., 2016), which similarly infers the serotype through gene presence and sequence analysis of loci involved in antigen biosynthesis and expression. The results from the traditional Salmonella serotyping and from WGS were analyzed for their dierences and similarities using an in-house R script to generate figures (libraries: tidyverse, stringr, data.table, ggplot2, ggnewscale, RColorBrewer, Biostrings, reshape2, plyr) (RStudio Team, 2015).
McNemar's chi-squared test was used to assess the statistical significance of the classification performance between SeqSero2 and SISTR versus KW methods.
## Results
The Salmonella specimen cohort Among the 282 clinical samples analyzed, fecal specimens accounted for 80.5% (n = 227) of the total. Blood and urine samples represented 9.6% (n = 27) and 8.1% (n = 23), respectively. The remaining five samples included abscess material (n = 3), biopsy specimen, and synovial fluid (Supplementary Table 2). The latest ECDC report on salmonellosis reports an NTS invasiveness rate of around 6% (European Centre for Disease Prevention and Control [ECDC], 2022). However, the latest data collected by the Enter-Net Italia system (National Surveillance 2016-2021 of Salmonella, Campylobacter, Shigella, and Yersinia infections) shows that approximately 8.3% of isolated Salmonella strains cause invasive infections (Istituto Superiore di Sanità, 2024). This study highlights a higher rate of invasiveness among NTS strains compared to both European and Italian data.
The average annual number of samples was 70 per year. No monthly peak was observed in the sample collection, as it remained constant throughout the years (Figure 1).
The sero-agglutination of 282 isolates showed 32 dierent Salmonella serovars, among which S. Monophasic Typhimurium was the most prevalent (Figure 2 and Supplementary Table 3). This serovar was identified in 100 of the 282 isolates (35.4%) followed by serovar Enteritidis in 31 (10.9%) isolates. All other serovars were detected at frequencies below 10%. In addition, the monthly distribution of the two most prevalent serovars was analyzed for the years 2021-2024, but no distinct monthly peaks in prevalence were observed (Supplementary Figure 1).
Finally, serotyping was not assessed in 12% of the isolates (n = 34) because of the presence of autoagglutination or nonagglutination isolates or the non-availability of certain specific antisera in our laboratory (Wattiau et al., 2011).
For serovars causing sepsis, no distinct trend was observed, and the overall numerical pattern recorded at our hospital was confirmed. Among the total isolates, six isolates of S. Monophasic Typhimurium, four isolates of S. Enteritidis, and two isolates of S. Napoli were responsible for invasive infections (Supplementary Table 2). Additionally, two cases of typhoid fever were reported, with S. Typhi isolated exclusively from blood cultures.
## Comparison WGS and gold-standard method
Of the 282 samples analyzed, both the gold-standard method and WGS successfully assigned the isolates to a specific serovar or antigenic profile in 248 cases (88%), whereas for the remaining 34 isolates, only WGS was able to achieve the identification (Figure 3, bottom table).
Comparing the predicted Salmonella serotypes derived from the sequencing protocol with respect to the gold-standard, both SeqSero2, SISTR and KW serotyping predicted identical Salmonella serotypes in 236 out of 248 isolates, showing 95% of concordance.
In the remaining 12 cases (5%), both methods (KW and SeqSero2/SISTR) achieved classification and yielded discordant results (Figure 3, upper table).
Eight isolates were classified successfully by KW and SeqSero2 while SISTR was able to recover the antigenic profiles without the serovar identifier (Figure 3, upper table, orange box).
One sample was classified as Typhimurium by KW and SeqSero2 while SISTR identified it as Monophasic Typhimurium (Figure 3, upper table, green box).
Two isolates were phenotypically identified as Salmonella Chincol and Colindale, whereas the Enteritidis and Strathcona classifications were produced by WGS (Figure 3, upper table, yellow box).
The last isolate was classified by KW as Stanley, Monophasic Typhimurium by SISTR, whereas SeqSero2 could only identify the antigen (I 4:d:-) and not the classification (Figure 3, upper table, blue box).
Regarding the classification recovery rate, WGS assigned a specific classification to 34/282 (12%) isolates that were not characterized by KW (Figure 3, bottom table). In one of these cases, SeqSero2 and SISTR were able to recover only the antigenic profile without reaching classification (Figure 3, bottom table, pink box). Moreover, the SeqSero2 tool recovered a classification and an antigenic profile in the case of two isolates not classified by KW and SISTR, revealing high sensibility (Figure 3, bottom table, purple box).
McNemar's chi-squared test revealed a statistically significant dierence in classification performance between the KW and WGS-SeqSero2 method, with the latter outperforming the goldstandard approach (χ 2 = 34, df = 1, p = 5.511 × 10 -9 ).
McNemar's chi-squared test between KW and WGS-SISTR method was also significant (χ 2 = 32, df = 1, p-value = 1.542 × 10 -8 ). Overall, WGS significantly reduced the proportion of untyped isolates compared with KW.
Regarding the comparison between SeqSero2 and SISTR, the test resulted in a non-significant p-value (χ 2 = 2, df = 1, p = 0.1573), indicating that there are no statistically significant dierences between the two methods in the sample identification rate.
## Discussion
In Italy, non-typhoidal Salmonella is the second most prevalent infection linked to diarrhea. S. Monophasic Typhimurium was the most prevalent serotype associated with non-typhoidal salmonellosis in Italy, which is in contrast with the European trend reported by Centers for Disease Control [CDC] (2013). S. Enteritidis is the most prevalent serotype in Europe, while in our hospital, as well as in the rest of Italy, ranks second.
Determining the phenotype of the isolates is a labor-intensive process that requires a significant amount of time, typically taking up to three days of active work of a trained operator to complete, and this applies to only a limited number of isolates to be typed. The turnaround time for WGS can vary depending on the library preparation kit and the sequencing platform, as well as the possibility to automate the library preparation process. A single MiSeq run (500 cycles paired-end) comprising 16 bacterial genomes generally takes 48-72 h (McGann et al., 2016) for Nextera XT protocol library by-hand preparation, sequencing, and data analysis, with 50% of active hands-on work time. Importantly, sequencing allows for a high-throughput, parallel processing of multiple genomes simultaneously, enabling the analysis of many isolates at once and obtaining much more information than sample typing (e.g., species, subspecies, serovar, virulence, pathogenicity and antimicrobial resistance), a scale that is not achievable with conventional phenotypic method (Ibrahim and Morin, 2018).
Certain Salmonella isolates require repeated subculturing in semi-solid media to enhance motility and flagellar antigen expression (Centers for Disease Control [CDC], 2013). Specific genetic mutations, such as single nucleotide polymorphisms can lead to the loss of serotype antigen expression, further limiting the reliability of traditional serotyping methods (Li et al., 2017). Conventional serotyping is also resource-intensive, necessitating the maintenance of a broad and diverse panel of antisera.
Recently, WGS has been applied in the area of Salmonella subtyping and has the potential to be a more dependable and eÿcient method (Ashton et al., 2016;Basso et al., 2025;Brümmer et al., 2024;Han et al., 2024;Morton et al., 2024). Tools like SeqSero2 and SISTR can forecast most Salmonella serotypes using large-scale genome sequencing data from databases containing Salmonella serotype determinants.
We performed the WGS approach and KW phenotypic method for 282 Salmonella isolates. The raw WGS data was analyzed using the two tools that have shown the best performance in literature, SeqSero2 and SISTR (Uelze et al., 2020). For 248 Salmonella isolates, both KW and WGS were able to achieve a classification or a profile antigen identification. WGS successfully predicted 236/248 (95%) isolates that shared the same antigenic structure and serotype classification of KW method. Overall, the discordance rate between the gold-standard method and WGS was 5% (12 isolates, Figure 3, upper table).
The discrepancies observed were mostly attributable to inherent dierences between genotypic and phenotypic approaches, with WGS-based tools relying on the detection of antigen-encoding genes and KW depending on antigen expression and agglutination patterns. In eight samples, SISTR failed to return a serovar classification describing only an antigen profile (Figure 3, upper table, orange box). Rather than being associated with lower sensitivity of SISTR, the discrepancy may arise from both methodological dierences among the tools. SISTR integrates information from antigen gene prediction with core genome multilocus sequence typing (cgMLST) to infer the most likely serovar. SeqSero2 relies solely on the presence and combination of antigen-encoding genes (O-and H-antigen loci), while KW classification is based on phenotypic expression of these antigens.
In the second case (Figure 3, upper table, green box), KW and SeqSero2 may have detected partial sequences or traces of the second phase, classifying the isolate as biphasic. Instead, SISTR may be more stringent and classify it as monophasic if the phase 2 gene is too divergent. However, since SISTR classification relies on a consensus derived from other tools, the divergence could be due to a dierent assembly.
The most likely reason for the dierence in KW phenotypic identification with respect to WGS classification in the third and fourth cases (Figure 3, upper table, yellow box) is that the two serovars share highly similar or identical antigenic factors for one phase but dier in antigens. A weakly expressed or absent flagellar/somatic antigen could result in phenotypic agglutination matching Colindale's known antigen formula, even if it carries Strathcona-specific genes. Moreover, the antisera used in the KW scheme can sometimes cross-react with antigens from closely related serovars, leading to misidentification if the full antigenic profile is not clearly expressed (Figure 3, upper table, yellow box).
In the last case, WGS-based algorithms may have assigned a dierent or more generic genetic profile depending on sequence homology and database composition. Thus, if antigen expression is altered, the phenotype may not correspond to the genotype, and the observed antigenic profile could be more readily matched to Stanley in traditional reference charts (Figure 3,upper table,blue box).
Overall, the disparity between genotypic and phenotypic expression and the principle underlying the two methods can be decisive in identifying and classifying the antigenic structure.
WGS enabled the characterization of 34 isolates out of 282 (12%) that were untypeable by conventional methods (Figure 3, lower table). However, it should be noted that of the two tools, SeqSero2 performed slightly better, recovering one classification and one antigenic profile (34/34, 100%) compared to SISTR (32/34, 94%) (Figure 3, lower table, purple boxes). This suggests that SeqSero2 may be more sensitive in detecting partial or divergent antigen gene sequences, while SISTR's more conservative assignments could be advantageous in contexts where minimizing potential misclassification is a priority. However, the overall performances of SeqSero2 and SISTR were similar to each other as previously demonstrated from dierent studies (Ashton et al., 2016;Uelze et al., 2020;Wu et al., 2021;Zhang et al., 2015Zhang et al., , 2019)). Indeed, in the comparison of several open-source tools for serotyping Salmonella spp., SISTR and SeqSero2 produced the most accurate and reliable results, predicting the serovars of 94% and 87% of all isolates, respectively, as reported by Uelze et al. (2020).
The statistically significant dierences observed in McNemar's chi-squared tests support the superior performance of WGS over the KW method in Salmonella classification for both tools (SeqSero2: p = 5.511 × 10 9 , SISTR: p = 1.542 × 10 8 ). Moreover, the comparison between SeqSero2 and SISTR yielded a nonsignificant p-value (p > 0.05), indicating that there are no statistically significant dierences between the two methods in sample identification rates. Our laboratory consistently employs WGS to predict Salmonella serotypes for each isolate from ongoing investigations, as well as from previously archived isolates dating back to previous years.
Despite the demonstrated accuracy and eÿciency of the WGS tool for Salmonella serotyping, certain serovars remain absent from the SeqSero2 and SISTR databases, leading to occasional discrepancies between genomic and traditional serotyping results. It is anticipated that as these databases are progressively expanded, it will be possible to overcome these limitations. Overall, this research confirmed that genomic serotyping is a valuable support for traditional phenotypic methods.
## Study limitations
A limitation of this study is the lack of demographic data that limited our ability to explore possible associations between serovars and specific patient characteristics, infection sources, or clinical outcomes. Moreover, the small number of non-intestinal isolates in our dataset may have reduced the representativeness of invasive or systemic infections. The incomplete panel of antisera also introduced potential biases in the comparison between phenotypic and genomic serotyping methods, as some discrepancies could not be fully resolved through traditional serological testing. Future studies integrating comprehensive epidemiological metadata, a larger number of isolates from diverse clinical and environmental sources, and complete serological characterization will allow a more accurate assessment of the relationships between genotype, phenotype, and epidemiological patterns.
## Conclusion
In conclusion, this study underscores the substantial advantages of WGS for the serotyping of Salmonella isolates, presenting a robust and eÿcient alternative to conventional phenotypic methods. Phenotypic methods for Salmonella identification exhibit inherent limitations due to their reliance on manual operator interpretation, which introduces subjectivity and potential for error. The requirement for skilled personnel to perform and interpret serological assays contributes to variability in results and can reduce reproducibility across laboratories. Additionally, these methods are labor-intensive and timeconsuming, further delaying timely results. Phenotypic approaches often lack the resolution to discriminate closely related isolates, impeding accurate outbreak investigations and epidemiological surveillance. These challenges underscore the necessity for integrating high-resolution, automated molecular techniques such as WGS to enhance accuracy, standardization, and timeliness in Salmonella surveillance (Dyer et al., 2024). The 12% recovery rate of untyped isolates highlight the method's potential to refine and streamline laboratory workflows emerging as a reliable and scalable approach for Salmonella surveillance. Its integration into routine laboratory operations oers notable benefits, including reduced labor demands, enhanced quality control, and expedited reporting. This work also establishes a solid basis for implementing Salmonella surveillance in our hospital and enables more timely reporting of identified serovars to the EnterNet-Italy platform (Istituto Superiore di Sanità, 2024), which coordinates national surveillance eorts. Finally, this approach contributes to national and global strategies for the tracking, prevention, and control of salmonellosis.
## References
1. Agbaje, Begum, Oyekunle et al. (2011) "Evolution of Salmonella nomenclature: A critical note" *Folia Microbiol*
2. Andrews (2010) "FastQC: A quality control tool for high throughput sequence data"
3. Ashton, Nair, Peters et al. (2016) "Identification and typing of Salmonella for public health surveillance using whole genome sequencing" *PeerJ*
4. Basso, Cerri, Possebon et al. (2025) "Whole-genome sequencing of Salmonella serovars isolated from diarrheic and non-diarrheic foals" *J. Vet. Diagn. Invest*
5. Brümmer, Smith, Modise et al. (2022) "Whole genome sequencing assisted outbreak investigation of Salmonella enteritidis" *Access Microbiol*
6. Carrera, Tolini, Trogu et al. (2013) "An atlas of Salmonella in the United States, 1968-2011: Laboratory-based enteric disease surveillance" *Centers for Disease Control*
7. Cooper, Low, Koziol et al. (2020) "Systematic evaluation of whole genome sequence-based predictions of Salmonella serotype and antimicrobial resistance" *Front. Microbiol*
8. Deng, Shariat, Driebe et al. (2015) "Comparative analysis of subtyping methods against a whole-genome-sequencing standard for Salmonella enterica serotype enteritidis" *J. Clin. Microbiol*
9. Santos, Ferrari, Conte-Junior (2019) "Virulence factors in Salmonella typhimurium: The sagacity of a bacterium" *Curr. Microbiol*
10. Dyer, Päuker, Baxter et al. (2024) "EnteroBase in 2025: Exploring the genomic epidemiology of bacterial pathogens" *bioRxiv*
11. (2022) "Annual epidemiological report for 2022 -Salmonellosis. Sweden: ECDC. European Food Safety Authority [EFSA], and European Centre for Disease Prevention and Control" *EFSA J*
12. Grimont, Weill (2007) "Antigenic formulae of the Salmonella serovars"
13. Han, Yu, Zhang et al. (2024) "Draft whole-genome sequencing and phenotypic analysis of Salmonella from retail aquatic products in weifang" *Foodborne Pathog Dis*
14. Huedo, Gori, Zolin et al. (2010) "Salmonella enterica serotype napoli is the first cause of invasive nontyphoidal salmonellosis in lombardy" *Foodborne Pathog Dis*
15. Ibrahim, Morin (2018) "Salmonella serotyping using whole genome sequencing"
16. Jibril, Okeke, Dalsgaard et al. (2020) "Prevalence and risk factors of Salmonella in commercial poultry farms in Nigeria" *PLoS One*
17. Koser, Ellington, Cartwright et al. (2012) "Routine use of microbial whole genome sequencing in diagnostic and public health microbiology" *PLoS Pathog*
18. Li, Liu, Luo et al. (2017) "O-serotype conversion in Salmonella typhimurium induces protective immune responses against invasive non-typhoidal Salmonella infections" *Front. Immunol*
19. Mcgann, Bunin, Snesrud et al. (2016) "Real time application of whole genome sequencing for outbreak investigation -What is an achievable turnaround time?" *Diagn. Microbiol. Infect. Dis*
20. Morton, Kandar, Kearney et al. (2015) "Transition to whole genome sequencing surveillance: The impact on national outbreak detection and response for Listeria monocytogenes, Salmonella, shiga toxin-producing Escherichia coli, and shigella clusters in Canada" *Foodborne Pathog Dis*
21. Pagani, Parenti, Franzetti et al. (2023) "Invasive and non-invasive human salmonellosis cases admitted between 2015 and 2021 in four suburban hospitals in the metropolitan area of milan (Italy): A multi-center retrospective study" *Pathogens*
22. Petrin, Tiengo, Longo et al. (2024) "Uncommon Salmonella infantis variants with incomplete antigenic formula in the poultry food chain" *Italy. Emerg. Infect. Dis*
23. Robertson, Yoshida, Kruczkiewicz et al. (2015) "Comprehensive assessment of the quality of Salmonella whole genome sequence data available in public sequence databases using the Salmonella in silico Typing Resource (SISTR)" *Microb. Genom*
24. Ryan, O'dwyer, Adley (2017) "Evaluation of the complex nomenclature of the clinically and veterinary significant pathogen Salmonella" *Biomed. Res. Int*
25. Schrader, Fernandez-Castro, Cheung et al. (2008) "Evaluation of commercial antisera for Salmonella serotyping" *J. Clin. Microbiol*
26. Shovill ; Tack, Ray, Griÿn et al. (2020) "Preliminary incidence and trends of infections with pathogens transmitted commonly through food-foodborne diseases active surveillance network" *MMWR Morb. Mortal Wkly. Rep*
27. (2016) "The NCBI Pathogen Detection Project"
28. Uelze, Borowiak, Deneke et al. (2020) "Performance and accuracy of four open-source tools for in silico serotyping of Salmonella spp. based on whole-genome short-read sequencing data" *Appl. Environ. Microbiol*
29. Wattiau, Boland, Bertrand (2011) "Methodologies for Salmonella enterica subsp. enterica subtyping: Gold standards and alternatives" *Appl. Environ. Microbiol*
30. Wu, Luo, Xu et al. (2021) "Evaluation of Salmonella serotype prediction with multiplex nanopore sequencing" *Front. Microbiol*
31. Xu, Ge, Luo et al. (2020) "Evaluation of real-time nanopore sequencing for Salmonella serotype prediction" *Food Microbiol*
32. Yachison, Yoshida, Robertson et al. (2017) "The validation and implications of using whole genome sequencing as a replacement for traditional serotyping for a national Salmonella reference laboratory" *Front. Microbiol*
33. Yoshida, Kruczkiewicz, Laing et al. (2016) "The Salmonella In Silico Typing Resource (SISTR): An open webaccessible tool for rapidly typing and subtyping draft Salmonella genome assemblies" *PLoS One*
34. Zhang, Den Bakker, Li et al. (2019) "SeqSero2: Rapid and improved Salmonella serotype determination using whole-genome sequencing data" *Appl. Environ. Microbiol*
35. Zhang, Yin, Jones et al. (2015) "Salmonella serotype determination utilizing high-throughput genome sequencing data" *J. Clin. Microbiol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12750033&blobtype=pdf
|
# Cryo-EM Analysis in CASP16
Thomas Mulvaney, Andriy Kryshtafovych, Maya Topf
## Abstract
Since CASP13, experimentalists have been encouraged to provide their cryo-EM data along with the derived atomic models to the CASP organizers to aid assessment. In CASP16, 38 cryo-EM datasets were provided for assessment, which represented most cryo-EM targets. The corresponding targets typically comprised a single derived atomic structure; however, that model may be only one of several valid conformations. Flexibility often manifests as low-resolution regions in a cryo-EM reconstruction, particularly in RNA but often also in protein complexes. We show that local resolution in the reconstruction correlates well with the root-mean-square fluctuations (RMSF) of residues of accurate CASP predictions. The correlation between the local resolution and pLDDT was less clear, especially when mobile domains were present. When the resolution allowed, assessment of features such as sidechains, using our variant of SMOC with local fragment alignment, indicated that even high-quality predictions have room for improvement; on the other hand, some predictions fitted the density better in specific regions, indicating modeling discrepancies in the target. In one extreme case, a submitted target had regions of low-resolution that limited unambiguous model building. In such cases, part of the target structure is essentially a prediction itself, with implications for the assessment. Experimental data remain essential for model-free assessment of predictions and offer unique analyses such as comparisons to local resolution and thus flexibility.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
## 1 | Introduction
Cryo-electron microscopy has become an important technique for structure determination, providing structural biologists with insights into molecules that were previously unobtainable through crystallographic methods. The Critical Assessment of Structure Prediction (CASP) community has benefited heavily from a stream of cryo-EM targets that are typically larger than crystallographic targets and oftentimes contain folds that are under-represented in structural databases such as the PDB [1].
Since 2020, the number of cryo-EM targets in CASP has been steadily increasing (Figure 1), mirroring the PDB deposition tendency. In this CASP, 49 of the targets (corresponding to 47 maps; T1234 and T1235 were derived from the same map as the complex, H1236) were obtained from cryo-EM experiments, with X-ray crystallography and NMR providing 48 and two targets respectively. Thus, for the first time in CASP, the number of targets from cryo-EM has caught up with that of X-ray crystallography. Predictions for cryo-EM targets are assessed against the target model, similar to how crystallographic and NMR targets are assessed, using standard metrics of accuracy. Our previous papers in this series [2][3][4] explored the possibility of validation of CASP models versus the cryo-EM data using both local and global goodness-of-fit measures. In CASP15, we showed that the fit-to-map-based ranking correlates well with the CASP assessment scores [4]. Additionally, by performing a local assessment of predicted sidechains, we found that they are sometimes poorly positioned in models.
Structure prediction continues to play an important role in interpreting low-resolution cryo-EM data, especially in cases where the resolution is insufficient to build a complete atomic model de novo. One popular approach to model building with a structural prior is flexible fitting and refinement [5,6], where an initial model (potentially a prediction) is adjusted to better fit the density. In CASP14, we demonstrated that state-of-the-art predictions (notably those from AlphaFold2 but also others) can automatically be refined into very accurate models, reaching the accuracy of the structure provided by the experimentalists [3]. We also investigated variations between different predictions of single chains by different groups and correlated these with the quality of local fit in the map. With the improvement in complex prediction and the introduction of RNA targets, refinement in CASP15 shifted to entire protein complexes [4] rather than individual domains. By combining TEMPy-ReFF [7], RIBFIND [8], and, in the case of RNA models, ERRASER2 [9], this refinement pipeline was often able to produce models of equal or better fit to the experimental maps than the targets provided by experimentalists.
In CASP16, of the 47 cryo-EM determined structures, 38 maps were provided for assessment (Figure 2). For the first time, we requested that experimentalists provide the unprocessed halfmaps, in line with established best practices in the cryo-EM community [10]. The requested half-maps allowed additional types of assessment, which were previously not feasible, such as local resolution estimation. This, in turn, the allowed for new avenues of inquiry, such as whether predictions are able to capture this information. By analyzing multiple submissions, we aimed to capture hints of conformational dynamics or alternative local arrangements indicated by the cryo-EM data, essentially asking whether predictors' model diversity reflects regions of uncertainty or flexibility. This allows us to see if regions of a target that were poorly resolved in the cryo-EM map (and thus presumably more flexible or uncertain) also showed larger disagreement across predictions, suggesting that the prediction ensembles can encode experimental dynamics information beyond what a single target model provides.
Finally, the interplay between prediction and experiment in CASP has directly helped structure determination. In CASP14, several targets that could not be solved in time were ultimately resolved with the aid of high-quality predictions from AlphaFold2 [11,12]. Likewise, in CASP16, there were three cases where clear modeling errors in the initially provided "solved" structures were detected and corrected with the aid of accurate highquality structure predictions (T1210, H1220, T1257o). These errors would otherwise have gone unnoticed had the authors not provided their experimental cryo-EM data and engaged with the prediction community. In this paper, we investigate these and other cases in greater detail, employing a local goodness-of-fit metric to the cryo-EM density to systematically identify and analyze regions where predictions and experimental data diverge.
## 2 | Methods
## 2.1 | Experimental Data Collection and Participation
As mentioned above, experimentalists provided the unprocessed half-maps of all 38 cryoEM maps. Datasets covered a range of target types, including 10 protein complexes, 17 RNA structures, both synthetic and natural [13], and 11 hybrid targets composed of proteins and nucleic acids. Due to the size of the targets or the presence of multiple domains or units of interest, some were further broken down to yield additional targets for assessment, such as H1236, which was split into T1234 and T1235 as evaluation units. Across these different targets, the resolutions ranged significantly (Table S1, Figure 2). Generally, protein structures and complexes had higher resolution than RNA and hybrid targets. Nucleic acid-containing structures have historically suffered from lower-resolutions, and only in recent years have the advances in cryo-EM S1.
technology and methodology enabled resolutions sufficient for deriving atomic models. The lowest resolution CASP16 targets were single-chain RNA structures from SARS-2 (R1255 and R1256). Compared with the previous CASP, which had only two RNA-protein targets [14], this CASP saw a huge increase in hybrid targets, many solved by cryo-EM. Unlike in previous CASP rounds, in this round we requested experimentalists to provide the half-maps for their cryo-EM reconstructions. This was in part a response to reconstructions received in CASP15, which had been post-processed using deep learning-based sharpening tools such as DeepEMhancer. Such sharpening approaches are often useful aids in model building, but they have a tendency to make some features, such as ligands, disappear. Local resolution fluctuations are also flattened. Such information can be distracting for model building but are often indications of heterogeneity and mobility, features which might make structure prediction difficult and interpretation challenging.
## 2.2 | Local Resolution Determination
The Relion 3 software [15] was used to prepare local resolution maps from the provided half-maps, referred to here as "LocRes." In local resolution maps, the value of each voxel corresponds to the estimated resolution at that coordinate. The local resolution was then projected onto the experimental model by taking the The distribution of local resolutions is given for each target. In (B) a close-up of visible details for two targets is shown. On the left, T1214, which had an average resolution of 2.17 Å, with sidechains well resolved throughout the reconstruction. On the right H1272, which had an average resolution of 6.47 Å but varied significantly throughout the structure. In one region, the backbone of beta sheets is nicely resolved, but in other regions, even larger secondary structures were not well resolved.
determined resolution of the voxel closest to each atom. These local resolution projections are presented in Figure 2. The average local resolution for each structure is the average of all-atom local resolutions and is shown beneath the targets in Figure 2 and Table S1. (Note that this is not calculated on the entire map but only on regions occupied by atoms).
The mean local resolution of the reconstructions for 10 protein targets varied from 2.17 Å to 6.47 Å. Examples of the structural details present at these two extremes are shown in Figure 3B. T1214 has visible density for sidechains throughout the molecule. On the other hand, H1272 has a high-resolution core with visible backbone density around beta sheets, but other regions lack backbone visibility. Due to this low-resolution, some parts of this target structure were built by rigidly docking AlphaFold 2 and 3 models into the density. Heterogeneity in the resolution of cryo-EM reconstructions can be explained in part by the conformational flexibility of the molecules being imaged. As is often expected, nucleic acid-containing structures had lower mean resolutions, ranging from 3.05 Å and extending down to 18.51 Å. The average of these mean resolutions for RNA targets was 6.99 Å. Hybrid targets had mean local resolutions between 2.69 Å and 7.32 Å. The average mean resolution was 4.83 Å.
Careful alignment and classification of the molecular images ("particles") can help ensure that the reconstructions come from images of molecules with similar conformations. In some datasets, conformationally distinct classes were captured, as in the case of M1228 and M1239. These classes were reconstructed individually, offering targets in alternative states: M1228 v1 and v2, and M1239 v1 and v2. Still, smaller variations are difficult to separate into individual reconstructions, yielding reconstructions with heterogeneous resolution in part due to the averaging over many conformations. The relative local resolutions are displayed for each target, colored blue to red (Figure 2). The absolute ranges are provided in the distributions (Figure 3A).
## 2.3 | RMSF of Predictions Versus Resolution
For each CASP group, the root mean-square fluctuation (RMSF) for each Cα or C4′ atom was computed from five predictions submitted for a given target. This was performed by first aligning models 2-5 against model 1 using least-square fit in ChimeraX [16] using the align command. The RMSF for each residue was calculated by computing the root mean-square fluctuation of the Cα atoms from the 5 aligned models.
Only groups where all five predictions had lDDT scores greater than 0.7, and the TM and IPS (in the case of multimers) scores were greater than 0.8, were included in our assessment. We then calculated Pearson's correlation (PCC) between the RMSF and the local resolution of all maps for which the predictions met the accuracy criteria.
## 2.4 | Local Accuracy Estimates Versus Resolution
The Local Distance Difference Test (LDDT) [17] is a measure of local agreement between structural models while ignoring the long-range discrepancies that negatively affect other scores such as RMSD and TM. LDDT scores are typically reported for each residue on a range of 0-1.0 (or as a percentage), with values close to 1.0 indicating a similar local environment. The predicted Local Distance Difference Test (pLDDT), which was first used in AlphaFold2 [11] attempts to reproduce the LDDT scores and thus is an estimate of local model accuracy. Although adopted as standard since CASP15, other estimates of model accuracy also exist, such as the RosettaFold estimate of positional error (the smaller the better) [18]. To deal with the directionality of different reported accuracy estimates, we use the absolute Pearson's Correlation in our analysis.
## 2.5 | Local Fit to Density Analysis
In recent CASPs, we have noticed potential modeling errors in experimental models submitted as cryo-EM targets. Such errors are typically small and unsurprising given the sheer size of many of these targets. Previously, such errors have been apparent during visual assessment of the predictions against the experimental data or when comparing the goodness-of-fit of the experimental model against cryo-EM refined predictions. Although such small errors are not expected to impact the overall ranking of predictions, they do highlight the potential usefulness of structure prediction in the model-building process.
In previous CASP rounds, the SMOC score [19] was used to assess the goodness-of-fit of experimental models, predictions, and refined predictions to the 3D cryoEM map. One of the important features of SMOC and similar local fitness scores, such as Qscore [20] is that they indicate which parts of the model fit the experimental data well. However, this is also a weakness when assessing structure predictions, which may be locally well modeled but, due to some deviations in domain or secondary structure element positioning (e.g., orientation or shifts), only partially fit the experimental data. In contrast, scores such as lDDT are designed to be sensitive to local modeling errors and forgiving to global deviations such as the one described.
To this end, a new localized cryo-EM fitness score was developed based on aligning fragments to the target, thus avoiding penalties for poor positioning while enabling assessment of local fit to cryo-EM data. The fragments were computed as sliding windows of 11 residues over each chain of the predictions [19]. SMOC scores of the fragment and the corresponding target residues were then computed. The ΔSMOC score was defined as the difference between the fragment and target SMOC scores. Positive ΔSMOC scores correspond to cases where predicted fragments fit the experimental data better than the target, whereas negative scores indicate the opposite. In the case of high-resolution targets where sidechain density is well resolved, a more sensitive variant of the above was used with fragments of 5 residues in length. To better disambiguate different modeling errors at high resolutions, SMOC scores were computed on the backbone (backbone SMOC) and sidechain (sidechain SMOC) atoms of the central residue in the fragment.
## 3 | Results
## 3.1 | RMSF of Predictions Versus Resolution
As stated in methods (Section 3.3), the root mean square fluctuation (RMSF) was computed on well-predicted targets (LDDT > 0.7 for monomers, > 0.8 and IPS > 0.8 for multimers). Six targets satisfied these accuracy criteria. The Pearson's correlation coefficient (PCC) between the RMSF and local resolution for each of these targets is shown in Figure 4A. Positive correlations were observed, with five targets achieving correlations higher than 0.6 for specific targets (Figure 4B).
The most striking example is of the target R1289, a group I intron precursor RNA, where many groups submitted predictions for which the PCC was close to 1.0, with an average of 0.8. The structure consists of two domains, the intron which makes up the largest mass and is solved at a high resolution, and the tRNA component, which is solved at lower resolution likely due to its moving with respect to the intron (Figure 4C). These resolution differences nicely correlate with the RMSF, reflecting uncertainty in the relative positions of these two regions in predictions. Another impressive example of high correlation between RMSFs and local resolution is the Borna virus replication complex (target H1220) (Figure 3A). This dynamic target is explored in more depth in this issue [21]. The RMSF of predictions for T1214 did not correlate as strongly with local resolution. This may reflect a general lack of flexibility in this target, which had one of the narrowest standard deviations in local resolution (0.08 Å).
## 3.2 | Local Accuracy Estimates Versus Local Resolution
Here we study how well local accuracy estimates (LAE) in models correlate with resolution fluctuations in experimental data. For targets with accurate predictions, we observed lower correlation with the experimental data, compared to that of RMSF (see above). Only two out of the six targets had some predictions with LAE-LocRes PCC values above 0.6 (Figure 5a). T1214 is a TonB-dependent transporter of Pyrroloquinoline quinone (PQQ) from E. coli [22], a large beta-barrel-like membrane protein, whereas R1241 is a group-IIC intron from O. iheyensis.
For T1214 (Figure 5B), which only had five high-correlation predictions, all from group Seder2024hard, the exterior loops that connect the beta-strands are of low-resolution (Figure 5Bi), whereas the rest of the structure (Figure 5Bii) has highresolution and high accuracy estimates.
In the Group-IIC intron case, R1241, five out of six domains have been well studied and have structures in the PDB. These five domains had the highest local resolution in the cryo-EM reconstruction. Domain 6 makes non-canonical contacts with Domain 2, which were only correctly predicted in 4 Models: 3 from Vfold and 1 from GeneSilico, explored in more detail in the CASP16 RNA assessment paper [23]. Intriguingly, these three accurate Vfold predictions had the lowest correlation with local resolution. Reporting local accuracy estimates for RNA was not mandatory in CASP16, and the GeneSilico group did not provide them. The high correlation between the rest of the predictions and local resolution is thus likely related to the uncertainty in the modeling of Domain 6. An example of a prediction with high correlation but incorrect Domain 6 contacts is shown in Figure 5C. R1289, the Group I intron precursor tRNA, which had the highest correlation predictions according to RMSF, fared less well against accuracy estimates. Predictors gave high accuracy estimates for the tRNA T-arm and acceptor stem regions (Figure 5Di), and the group I intron domains (Figure 5Diii), whilst the D-arm was given lower accuracy estimates (Figure 5Dii). The highest resolution region was the Group I intron domain, which made up the bulk of the structure, whilst the tRNA region, particularly the T-arm, was solved at a lower-resolution. Although the pLDDT scores did not correlate well with resolution, the low pLDDT scores of the D-arm reflect its flexible nature.
Low pLDDT can indicate disordered regions that may not be visible in cryo-EM reconstructions. Predictions for T1210, a vitellogenin protein from the honey bee, A. cerana [24], illustrate two such scenarios. In the first case (Figure 5Ei), a loop with pLDDT scores below 60 was not visible in the experimental data. In the second case (Figure 5Eii), a high-confidence C-terminal CTCK domain is predicted but was not visible in the data. This is likely due to its flexible attachment to the bulk of the protein via a disordered region. The information from the pLDDT could explain these discrepancies between prediction and experimental map and model, but did not correlate well with the resolution variation seen at the ends of the model (Figure 5Eiii).
## 3.3 | Modeling Errors in Target Structures
In CASP, the target sequence is sometimes provided before the corresponding structure is finalized. Occasionally, challenging targets are not solved in time, and predictions can expedite the process [12]. In this CASP, a few targets had small errors in the submitted experimental model. These errors were identified by comparing the fit of prediction and experimental model to the experimental data using ΔSMOC scores. Assessment against the experimental data using ΔSMOC produced residue scores which followed the same trends as the lDDT, as exemplified by predictions of the Borna Virus polymerase L-protein (target T1220s1), which is part of the larger Borna virus replication complex (target H1220) (Figure 6A). However, in some cases where the lDDT score dipped, the ΔSMOC increased, indicating that although the prediction did not agree with the target structure in this region, it better agreed with the experimental data. Examples from the PEZY Folding (group 015) are visualized in Figure 6B,C. We want to critically note here that although many predictors were able to model some loops better than in the experimental model, they could not do this systematically for all loops, as shown in Figure 6D,E.
## 3.4 | Side Chain Analysis
In CASP15, one target was of sufficiently high resolution that we could directly assess the fit of sidechains against the experimental data. In this CASP, two such targets were present: T1214 and H1236. Given the resolution, we assessed the fit of sidechains and backbone regions of the experimental model and predictions (Figure 7A,B,respectively). Residues with a local resolution worse than 2.5 Å were excluded from these calculations.
Similar to the last CASP, the backbone was generally well predicted, with SMOC scores of predictions approaching those of the experimental model. For sidechains, there was more
## FIGURE 6 | Modeling errors. (A)
The local prediction quality, as scored against the reference structure using lDDT (gray) and against the experimental data using ΔSMOC (black), was in good agreement, with low lDDT scores corresponding to low ΔSMOC scores. However, there were some cases where lDDT scores dipped, but ΔSMOC increased. Two such regions are shown in (B) and (C) for the target (blue) and a prediction from group 015 (orange). Despite surpassing the target structure in accuracy in these regions, there were many more counterexamples where predictions struggled to produce locally accurate models, as seen in the many dips in lDDT and ΔSMOC scores. Two such examples are given in panels (D) and (E). variability in the SMOC scores, with predictors frequently producing sidechains not fitting the experimental data. The residue R594 was one of many poorly predicted sidechains in T1214. This can be seen in the wide ΔSMOC score distribution for this residue in Figure 7C.
One surprising outlier was a proline residue 299 in target T1234-D1 (an assessment unit from H1236), which frequently had better fit to the backbone density in predictions than in the provided experimental model (Figure 7D). On closer inspection, the proline in the experimental model was modeled in the more common trans conformation, but the predictions, which fitted the density best, were in the cis conformation. Such an error is unlikely to affect overall rankings, but highlights how even at relatively high resolutions, modeling errors can be made, which can potentially be avoided with the use of accurate structural predictions. The perhaps surprising ability for AlphaFold and current state-of-the-art methods to predict cis prolines has been noted elsewhere [25].
## 4 | Discussion and Conclusion
## 4.1 | Discussion
Our analysis underscores the increasingly reciprocal relationship between cryo-EM data and structure prediction in the CASP framework. Although the incorporation of cryo-EM data into CASP already provides a valuable means of evaluating prediction accuracy, our results highlight how predictions can also illuminate the strengths and limitations of experimental reconstructions. This dual perspective is particularly important given the growing reliance on structure prediction for experimental model building and validation. This relationship is particularly pertinent today, with highconfidence models being used to build models into maps by rigid docking. With insufficient structural details from experimental data for further refinement, such models would be problematic for CASP as predictions would be compared against predictions. In this CASP, the target H1272 partly raises this issue. Regions of the reconstruction have sufficient resolution to resolve beta sheets, whilst other regions can only delineate domains. The structure was solved using a combination of model building and rigid docking of AlphaFold2 and AlphaFold3 predictions in low-resolution regions that are able to provide important biological and molecular insights [26]. But many parts of the experimental model have been solved at an insufficient resolution to be considered a reliable ground truth for assessment of predictions.
A central finding of this study, although based only on 6 targets for which the predictions passed a certain accuracy threshold (lDDT > 0.7 and TM and IPS (for complexes) > 0.8), is that the variability observed across multiple predictions, quantified as RMSF, correlates with cryo-EM local resolution. This suggests that ensembles of predictions inherently capture conformational heterogeneity, which cryo-EM maps often reflect in poorly resolved FIGURE 7 | Sidechain and backbone SMOC analysis of high-resolution protein targets. Sidechain and backbone SMOC scores were plotted for all residues of all predictions versus target for T1214 (A) and T1234-D1 (B). Despite backbone SMOC scores of predictions generally matching those of the target, the sidechains generally fitted less well. In (C), R594 was an example of a residue from T1214 which had highly variable sidechain SMOC scores. An example of a predicted sidechain is shown. In (D), an example of a backbone region which was better modeled by many predictors. The proline at 299 was trans in the target, but predictions such as the one shown were modeled as cis.
regions. These results align with earlier work showing that molecular dynamics fluctuations correlate with AlphaFold confidence metrics [27]. However, this extends beyond that, indicating that prediction variability is an informative measure of uncertainty that could complement existing model quality metrics in CASP. It is also worth considering that predictors were not asked to submit five models with an RMSF that would correlate with local resolution. Some groups may have favored submitting diverse models or employed other strategies that would have underperformed in this analysis. The degree of diversity among models, and the sampling of diversity, is also an important factor and something that has been a common theme in AlphaFold era CASP experiments including, for example, MassiveFold in this CASP [28].
pLDDT has been shown to correlate only weakly with B-factor [29] and RMSF from MD simulations [27]. Part of the reason could be that well-predicted domains are assigned high pLDDT scores even if their orientation with respect to one another is not well defined. At the same time, low pLDDT scores have also been shown to be good indicators of flexible or disordered regions, which may be unresolvable in cryo-EM experiments. Although less correlated with local resolution variations than RMSF, it remains a valuable metric in its own right for model building and for assessment of predictions including interfaces [30]. Removing low-confidence regions based on pLDDT values or clustering residues in domains based on PAE matrices is an important step in many downstream docking and refinement procedures [31][32][33]. Approaches to combine structure prediction with experimental data such as Phenix PredictAndBuild [32], which iteratively updates input templates for AlphaFold2 by fitting to experimental data, or ROCKET [34] which biases OpenFold's [35] evolutionary space, also exist. Given the importance of local accuracy metrics to these approaches, we hope to see more such methods, particularly in the RNA prediction space, whether in the form of pLDDT or positional error scores.
These results also have implications for assessment methodology. Because predictors often show greater variability in regions of low cryo-EM resolution, scoring functions that account for resolution-dependent reliability may provide fairer benchmarks. Similarly, ensemble-based assessment strategies could help capture conformational dynamics that are otherwise obscured in single static references. Complementary use of prediction-derived uncertainty measures, such as PAE matrices or distograms [36], may further enrich future evaluations by explicitly linking prediction confidence to underlying conformational flexibility.
Our benchmarking further revealed that high-quality predictions can help identify errors in experimental reference models. In several cases, predictors consistently disagreed with local regions of deposited structures that were later shown to be problematic. Fragment-based analysis with ΔSMOC provided a sensitive method for capturing such discrepancies, outperforming manual inspection in both efficiency and precision. This illustrates the utility of prediction-informed evaluation not only for assessing participants but also for improving experimental models themselves. Incorporating automated ΔSMOC-style pipelines into CASP could therefore support both communities, ensuring that experimental models for targets reflect the highest possible quality.
One important limitation of the local analyses of predictions against experimental data is that it relies on an experimental model for fitting fragments of predictions to the data. The assumption is that the experimental model is of sufficient quality to make accurate fragment alignment possible. In the cases where there were small modeling errors in the experimental model, this approach was able to identify them by comparing the fit-to-density. This may, in part, be due to the relatively wide fragment width of 11 residues. Alternatively, docking using secondary structure or a larger rigid body defined by tools such as Slice'N'Dice [33] or RIBFIND [8] might be an alternative way to find well-fitting regions of predictions.
Looking ahead, our findings highlight the important interplay between experimental and computational methods. Structure predictions are now indispensable starting points for cryo-EM, crystallography, and hybrid approaches, accelerating model building while also enabling the detection of errors in deposited structures. At the same time, the availability of highquality experimental data remains crucial for assessing the limits of prediction algorithms and for ensuring that benchmark targets are biologically informative. Strengthening this reciprocal exchange through continued provision of experimental data, resolution-aware evaluation, and systematic pipelines for reporting discrepancies will allow CASP to remain a unique forum for driving both predictive accuracy and experimental rigor.
Uncertainty information in the form of RMSF, local accuracy estimates, or measures such as PAE or distogram information (not explored in this paper) could in principle act as a dataindependent proxy for map interpretation. This may be useful to (i) prioritize regions for focused refinement/model rebuilding, (ii) weight restraints in real-space refinement, (iii) flag likely alternative conformations and inform reconstruction approaches, and (iv) potentially help in designing cryo-EM experiments for challenging molecules.
## 4.2 | Conclusions
Because structure predictions have become essential starting points for model building in cryo-EM, crystallography, and lower-resolution approaches such as cross-linking mass spectrometry [37], the provision of corresponding experimental data is critical for evaluating and extending their applicability to downstream pipelines. We therefore recommend that structure providers make the experimental data (e.g., half-maps) available whenever possible. High-quality predictions can then be used not only for benchmarking but also for detecting and correcting structural errors. In future CASPs, we envision offering automated reporting facilities based on the ΔSMOC method introduced here, to help ensure that reference models are as accurate and informative as possible.
Our analysis further shows that the RMSF across several targets with high-quality predictions can be an indicator of local resolution variation in the experimental data itself. This indicates that even when accurate models deviate slightly from a single reference, they may still reflect biologically relevant conformational states.
Together, these findings emphasize the reciprocal value of combining experimental data with prediction ensembles, both for assessing predictive accuracy and for improving experimental models.
## References
1. Berman, Bhat, Bourne (2000) "The Protein Data Bank and the Challenge of Structural Genomics" *Nature Structural Biology*
2. Kryshtafovych, Malhotra, Monastyrskyy (2019) "Cryo-Electron Microscopy Targets in CASP13: Overview and Evaluation of Results" *Proteins*
3. Cragnolini, Kryshtafovych, Topf (2021) "Cryo-EM Targets in CASP14" *Proteins*
4. Mulvaney, Kretsch, Elliott (2023) "CASP15 Cryo-EM Protein and RNA Targets: Refinement and Analysis Using Experimental Maps" *Proteins*
5. Topf, Lasker, Webb et al. (2008) "Protein Structure Fitting and Refinement Guided by Cryo-EM Density" *Structure*
6. Orzechowski, Tama (2008) "Flexible Fitting of High-Resolution X-Ray Structures Into Cryoelectron Microscopy Maps Using Biased Molecular Dynamics Simulations" *Biophysical Journal*
7. Beton, Mulvaney, Cragnolini et al. (2024) "Cryo-EM Structure and B-Factor Refinement With Ensemble Representation" *Nature Communications*
8. Malhotra, Mulvaney, Cragnolini (2023) "RIBFIND2: Identifying Rigid Bodies in Protein and Nucleic Acid Structures" *Nucleic Acids Research*
9. Chou, Echols, Terwilliger et al. (2016) "RNA Structure Refinement Using the ERRASER-Phenix Pipeline" *Methods in Molecular Biology*
10. Lawson, Kryshtafovych, Adams (2019) "Cryo-EM Model Validation Recommendations Based on Outcomes"
11. (2021) *Nature Methods*
12. Jumper, Evans, Pritzel (2021) "Highly Accurate Protein Structure Prediction With AlphaFold" *Nature*
13. Kryshtafovych, Moult, Albrecht (2021) "Computational Models in the Service of X-Ray and Cryo-Electron Microscopy Structure Determination" *Proteins*
14. Kretsch, Wu, Shabalina (2025) "Naturally Ornate RNA-Only Complexes Revealed by Cryo-EM" *Nature*
15. Das, Kretsch, Simpkin (2023) "Assessment of Three-Dimensional RNA Structure Prediction in CASP15" *Proteins*
16. Scheres (2020) "Amyloid Structure Determination in RELION-3.1" *Acta Crystallographica Section D: Structural Biology*
17. Meng, Goddard, Pettersen (2023) "UCSF ChimeraX: Tools for Structure Building and Analysis"
18. Mariani, Biasini, Barbato et al. (2013) "lDDT: A Local Superposition-Free Score for Comparing Protein Structures and Models Using Distance Difference Tests" *Bioinformatics*
19. Baek, Dimaio, Anishchenko (2021) "Accurate Prediction of Protein Structures and Interactions Using a Three-Track Neural Network" *Science*
20. Joseph, Malhotra, Burnley (2016) "Refinement of Atomic Models in High Resolution EM Reconstructions Using Flex-EM and Local Assessment" *Methods*
21. Pintilie, Shao, Wang (2025) "Q-Score as a Reliability Measure for Protein, Nucleic Acid and Small-Molecule Atomic Coordinate Models Derived From 3DEM Maps" *Acta Crystallographica Section D: Structural Biology*
22. Alexander, Follonier, Kryshtafovych (2026) "Protein Target Highlights in CASP16: Insights From the Structure Providers" *Proteins: Structure, Function, and Bioinformatics*
23. Munder, Voutsinos, Hantke et al. (2025) "High-Affinity PQQ Import Is Widespread in Gram-Negative Bacteria" *Science Advances*
24. Kretsch, Albrecht, Andersen (2026) "Functional Relevance of CASP16 Nucleic Acid Predictions as Evaluated by Structure Providers" *Proteins: Structure, Function, and Bioinformatics*
25. Montserrat-Canals, Schnelle, Leipart (2025) "Cryo-EM Structure of Native Honey Bee Vitellogenin" *Nature Communications*
26. Herzberg, Moult "More Than Just Pattern Recognition: Prediction of Uncommon Protein Structure Features by AI Methods" *Proceedings of the National Academy of Sciences of the United States of America*
27. Liu, Su, Xia et al. (2025) "Native DGC Structure Rationalizes Muscular Dystrophy-Causing Mutations" *Nature*
28. Guo, Perminov, Bekele (2022) "AlphaFold2 Models Indicate That Protein Sequence Determines Both Structure and Dynamics" *Scientific Reports*
29. Raouraoua, Lensink, Brysbaert (2026) "MassiveFold Data for CASP16-CAPRI: A Systematic Massive Sampling Experiment" *Proteins: Structure, Function, and Bioinformatics*
30. Carugo (2023) "PLDDT Values in AlphaFold2 Protein Models Are Unrelated to Globular Protein Local Flexibility" *Crystals*
31. Genz, Mulvaney, Nair et al. (2023) "PICKLUSTER: A Protein-Interface Clustering and Analysis Plug-In for UCSF Chime-raX" *Bioinformatics*
32. Millán, Mccoy, Terwilliger et al. (2023) "Likelihood-Based Docking of Models Into Cryo-EM Maps" *Acta Crystallographica Section D: Structural Biology*
33. Oeffner, Croll, Millán (2022) "Putting AlphaFold Models to Work With Phenix.process_predicted_model and ISOLDE" *Acta Crystallographica. Section D, Structural Biology*
34. Simpkin, Elliot, Joseph (2025) "Slice'N'Dice: Maximizing the Value of Predicted Models for Structural Biologists" *Acta Crystallographica Section D: Structural Biology*
35. Fadini, Li, Mccoy (2025) "AlphaFold as a Prior: Guiding Protein Structure Prediction Using Experimental Data With ROCKET" *Structural Dynamics*
36. Ahdritz, Bouatta, Floristean (2024) "OpenFold: Retraining AlphaFold2 Yields New Insights Into Its Learning Mechanisms and Capacity for Generalization" *Nature Methods*
37. Savaş, Barlas, Karaca (2025) "Exploring the Potential of AlphaFold Distograms for Flexibility Assignment in Cryo-EM Maps"
38. Manalastas-Cantos, Adoni, Pfeifer (2024) "Modeling Flexible Protein Structure With AlphaFold2 and Crosslinking Mass Spectrometry" *Molecular and Cellular Proteomics*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12797278&blobtype=pdf
|
# IgE Anti-Beta Coronaviruses Serology in Napoleon Soldiers, France
Nor El, Houda Merrouche, | Aboudharam, | Sandrine Thiol, Elodie Terrer, Jacques Fantini, Michel Drancourt, Hamadou Hama
## Abstract
To compare the repertoire of anti-beta-Coronavirus antibodies detected in dental pulp samples (systemic immunity) collected from individuals from the early 19th century previously investigated for dental calculus (local immunity) serological response, We investigated 10 dental pulp samples collected from 10 individuals excavated from a 1810-1813 military site in Charleville-Mézières, France. The samples had previously been investigated for dental calculus serology. Dental pulp serology performed under a mini-blot format, incorporated one positive and one negative control, and conjugated antibodies against the five classes of immunoglobulins. Dental pulp IgE serological response reliability was assessed by in silico analyses. Controls yielded expected results. Anti-Coronavirus antibodies were detected in three individuals, comprising anti-beta Coronavirus IgE in three individuals, IgG in two individuals, and IgA in one individual. IgA and IgG anti-alpha Coronavirus were each detected in one individual. These results agreed with those previously obtained from the same 10 individuals with anti-beta-Coronavirus pooled IgG/IgA/IgM dental calculus paleoserology. Dental pulp paleoserology confirmed Coronavirus exposure in three individuals from the start of the 19th century in France. Translating these data into the modern medical literature, we propose that two centuries ago, some individuals suffered a yet unidentified beta-Coronavirus infection.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
## 1 | Introduction
In 1810-1813, the town of Charleville-Mézières, located in the northeast of France, was a garrison for the army of French Emperor Napoleon the First [1]. Ten soldiers exhumed from that archaeological site had previously been investigated by paleoserology for anti-Coronavirus antibodies, following the extraction of paleosera from dental calculus samples. This investigation yielded the detection of pooled IgA/IgM/IgG antibodies against the beta-Coronavirus SARS-CoV-2 antigen in one individual and against the beta-Coronavirus HCoV OC43 antigen in a second individual. Automated Western blot assays confirmed SARS-CoV-2 nucleocapsid protein antibodies [2]. As dental calculus embeds an average 10-day pre-mortem memory of the oral and nasopharyngeal cavity biology, this data indicate that these individuals certainly developed local oral cavity immunity against an as-yet unknown beta-Coronavirus, testifying to these individuals' exposure to such a beta-Coronavirus. This data however left unknown whether these individuals developed systemic immunity against this beta-Coronavirus in addition to local immunity.
In order to answer this question, we further investigated systemic paleoserological responses in these individuals by testing the paleoserum samples extracted from the dental pulp tissues, an organ recognised as containing blood drops from the time of death and thus indicating systemic immunity, if any [3].
## 2 | Materials and Methods
## 2.1 | Dental Pulp Specimens
Dental pulp samples were collected from the 10 individuals exhumed from the 1810-1813 Charleville-Mézières site in agreement with the laws and regulations in France at the time of the study, as previously reported [2].
## 2.2 | Paleoserology
Paleoserum samples were extracted from ancient dental pulp tissues as previously reported [4]. More specifically, paleoserology incorporated heat-inactivated (65°C for 1 h) SARS-CoV-2 MI2 (Wuhan genotype), SARS-CoV-2 2096 (Marseille 4 genotype), OC43 and 229 E Coronavirus, and peroxidiseconjugated immunoglobulins against human pooled IgA/ IgM/IgG (Jackson Immuno Research, Ely, United Kingdom), IgA, IgG, IgM, IgE, and IgD (SouthernBiothech, Birmingham, USA). Serology reactions incorporated Staphylococcus aureus (S. aureus) as a positive control, non-specifically catching any immunoglobulins, and skimmed milk as a negative control [5] (Figure 1).
## 2.3 | Structural Analyses
All simulations were performed with the Hyperchem suite and visualised with Molegro Molecular Viewer, as described previously [6]. In order to interpret the results obtained, we developed an original approach based on the detection of IgE epitopes in the NC protein of Coronaviruses. For this, we developed our own algorithm, taking into account the following criteria: (i) localisation of the epitope on the surface of the protein, which suggests an over-representation of polar amino acids in the motif area; (ii) the presence of a turn in the motif, induced by one or more Gly and Pro residues; and (iii) a negative surface electrostatic potential, requiring the presence of Asp and/or Glu residues. Our algorithm is based on previously published biochemical characterisation of IgE epitopes and careful analysis of available databases [7,8]; Khatri et al. 2022 [9].
## 3 | Results
Both controls yielded expected results. The negative control (skimmed milk) was negative for all antigens and immunoglobulin classes tested and confirmed the absence of nonspecific reactivity, while the positive control (S. aureus) showed reactivity across all antibody classes, confirming the reliability of the tests. Pooled IgA/IgM/IgG detected in three individuals corresponded to IgG and IgA (in two individuals). Interestingly, the three individuals yielded IgE against two or three of the beta-Coronavirus, but not against alpha-Coronavirus 229E (Figure 1), while IgD were never detected (Table 1). In detail, paleoserum US1300 showed the presence of IgE antibodies reactive against two beta-Coronaviruses: SARS-CoV-2 (Wuhan) and OC43. No reactivity was detected with the other antigens tested (Supporting Information Table 1). Paleoserum US1326 showed reactivity against the three beta-Coronaviruses, SARS-CoV-2 (Marseille 4), SARS-CoV-2 (Wuhan), and OC43. Pooled IgA/IgM/IgG antibodies were detected against the SARS-CoV-2 (Marseille 4) antigen, while pooled IgA/IgM/IgG, IgG, and IgE antibodies were detected against the SARS-CoV-2 (Wuhan) antigen. Only IgE was detected against the OC43 antigen, and IgG were detected against the 229E antigens (Supporting Information Table 2). Paleoserum US1339 showed broad reactivity against all antigens tested. IgG and IgE antibodies were detected in response to the SARS-CoV-2 (Marseille 4) antigen, and pooled IgA/IgM/IgG and IgE were detected against both the beta-Coronavirus SARS-CoV-2 (Wuhan) antigen and OC43 antigen, while reactivity was observed with IgA antibodies against the 229E antigen (Supporting Information Table 3). Paleosera from the other individuals (US1131, US1221, US1255, US1257, US1287, US1297, and US1335) were found to be completely negative for all tested Coronavirus antigens, with no antibodies detected for any of the immunoglobulin classes (Supporting Information Tables 45678910).
Analysis of the SARS-CoV-2 NC protein sequence using a home-designed algorithm made it possible to identify the GPEQTQG motif, meeting the required criteria, further localised in the 3D-structure of the protein (Figure 2A). This motif does not exist in the OC43 and 229E Coronaviruses, which have their own IgE epitopes (Figure 2B). Since there is no available structure for the NC proteins of these two Coronaviruses, we created in silico chimeric proteins expressing these epitopes in the context of the 3D-structure of the SARS-CoV-2 NC protein.
We then reconstructed an antigen-IgE complex for the SARS-CoV-2 NC protein. In this case, the complex appears clearly functional with excellent accessibility of the IgE epitope for the antibody (Figure 2B). Moreover, we found that the IgE epitope of OC43 is fully accessible to the antibody, while the 229E epitope is partially masked by an electropositive region, which might be responsible for an electrostatic repulsion and no fit between the protein and the antibody.
## 4 | Discussion
While confirming the exposure of individuals exhumed from the early 19th century site of Charleville-Mézières to beta-Coronavirus, which we previously reported based on dental calculus serology testifying to a local, pharyngeal exposure to the virus [2], the present data further indicate a systemic serological response indicative of an infection in these individuals. We detected an IgE response against beta-Coronavirus in the dental pulp, whereas IgE was not investigated in the dental calculus in our previous investigation [2]. An IgE systemic response against SFL3 and NFL antigens has been reported in 90% to 100% of patients infected with SARS-CoV-2 [10]. Accordingly, the SARS-CoV-2 nucleocapsid was reported to induce specific IgE, while no specific epitope had been reported [11]. Here, structural analysis prompted by our discovery of an IgE response in ancient individuals suggested that the IgE epitope characterised in the NC protein of the SARS-CoV-2 virus may originate from an ancestral epitope shared, at the structural level, by the ancestors of certain Coronaviruses. This would explain the cross-reaction of antibodies detected in the dental pulp of Napoleonic soldiers on antigen-like samples of SARS-CoV-2 and OC43. In contrast, the absence of a cross-reaction with antigenic samples of the Coronavirus 229E suggests a lower accessibility of IgE epitopes by antibodies generated by old Coronavirus infections. The interpretation of IgE anti-Coronavirus remains, however, controversial: while it was shown that an immune response characterised by the production of IgE could be a biomarker for the severity of COVID-19, including death, as patients with severe COVID-19 had significantly higher levels of IgE targeting the SARS-CoV-2 nucleocapsid [11], other studies did not find such a correlation [12,13]. Therefore, the interpretation of data herein reported in two centuries-old specimens cannot be assessed.
This study reinforces the interest of paleoserology studies coupling dental calculus and dental pulp to directly investigate the panel of five anti-human immunoglobulins, with anti-IgD being currently considered as a negative control, because no IgD response has ever been reported against a systemic pathogen.
## References
1. Hubert (1854) "Histoire de Charleville Depuis Son Origine, Jusqu'en 1854"
2. "A translucent yellow disc was superimposed to better visualise the surface potential of the IgE epitope. (B) The top panel shows a functional IgE-NC complex for SARS-CoV-2. The middle panel shows a similar complex for the IgE epitope of OC43 inserted in the SARS-CoV-2 NC protein context. In both cases, a translucent yellow disc was superimposed to better visualise the surface potential of the IgE epitope and its accessibility for the antibody"
3. Merrouche, Edouard, Oumarou et al. (2024) "Paleoserological Detection of Coronavirus Antigens in Dental Calculus of Human Remains Dating From the Beginning of the 19th Century, French Ardennes" *American Journal of Biological Anthropology*
4. Mai, Drancourt, Aboudharam (2020) "Ancient Dental Pulp: Masterpiece Tissue for Paleomicrobiology" *Molecular Genetics & Genomic Medicine*
5. Oumarou Hama, Barbieri, Guirou (2020) "An Outbreak of Relapsing Fever Unmasked by Microbial Paleoserology, 16th Century, France" *American Journal of Physical Anthropology*
6. Oumarou Hama, Chenal, Pible et al. (2023) "An Ancient Coronavirus From Individuals in France, Circa 16th Century" *International Journal of Infectious Diseases*
7. Matveeva, Lefebvre, Chahinian et al. (2023) "Host Membranes as Drivers of Virus Evolution" *Viruses*
8. Bernstein, Canon, Schein (2025) "Fine Resolution of the N-Terminal IgE-Binding Epitope of Ara h 2: Discovery of Variants With Enhanced IgE Binding" *Journal of Allergy and Clinical Immunology*
9. Dreskin, Koppelman, Andorf (2021) "The Importance of the 2S Albumins for Allergenicity and Cross-Reactivity of Peanuts, Tree Nuts, and Sesame Seeds" *Journal of Allergy and Clinical Immunology*
10. Negi, Schein, Braun (2023) "The Updated Structural Database of Allergenic Proteins (SDAP 2.0) Provides 3D Models for Allergens and Incorporated Bioinformatics Tools" *Journal of Allergy and Clinical Immunology: Global*
11. Plūme, Galvanovskis, Šmite et al. (2022) "Early and Strong Antibody Responses to SARS-CoV-2 Predict Disease Severity in COVID-19 Patients" *Journal of Translational Medicine*
12. Tan, Zheng, Sun (2022) "Hypersensitivity May Be Involved in Severe COVID-19" *Clinical and Experimental Allergy: Journal of the British Society for Allergy and Clinical Immunology*
13. De La Poza, Parés, Aparicio-Calvente (2025) "Frequency of IgE Antibody Response to SARS-CoV-2 RBD Protein Across Different Disease Severity COVID-19 Groups" *Virology Journal*
14. Nagarajan, Kothari, Smith-Norowitz et al. (2023) "Serum Immunoglobulin Levels as Potential Biomarkers of COVID-19 Pneumonia" *Annals of Clinical and Laboratory Science*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772245&blobtype=pdf
|
# Clonal outbreak of an extensively drug-resistant NDM-1 producing Pseudomonas aeruginosa in a local hospital in the Czech Republic
Katerina Chudejova, Tsolaire Sourenian, Marc Finianos, Anna Sramkova, Costas Papagiannitsis, Jaroslav Hrabak, Ibrahim Bitar
## Abstract
A clonal outbreak of 18 ST773 NDM-1 producing Pseudomonas aeruginosa strains has been detected for the first time in the Czech Republic. The strains were extremely drug-resistant (XDR) and resistant to serum killing. SNP-based phylogeny and CRISPR assay typing showed minimal genomic variations among the isolates. The results suggest a high-risk, persistent, virulent clone causing the hospital outbreak, with the possibility of a nationwide outbreak. IMPORTANCE Our research on the novel detection of the NDM-1 gene in carbapenemresistant Pseudomonas aeruginosa ST773 in the Czech Republic is of great significance for public health and infection control. Until now, the emergence of this gene in P. aeruginosa strains was uncommon in this region, as carbapenem resistance was primarily associated with IMP and VIM types of MBLs. This nosocomial outbreak was triggered by an index case patient repatriated from areas with reported NDM-1 producing P. aeruginosa, illustrating how international travel contributes to the spread of such resistant pathogens. The results obtained in this study show that it is necessary to focus on tracing the source of infections to control and prevent nosocomial infections, helping to protect public health in the Czech Republic.
T he worldwide spread of carbapenem-resistant Pseudomonas aeruginosa, which frequently belongs to international epidemic high-risk clones widely disseminated in hospital settings, has become an emerging challenge in the last years (1,2). The most important mechanisms of resistance to carbapenems in P. aeruginosa are the production of carbapenemases (3,4). The most frequently described carbapenemases in P. aeruginosa are metallo-beta-lactamases, such as IMP and VIM, and less often NDM (4). In the Czech Republic in 2015, a comprehensive study on 194 carbapenem-resistant P. aeruginosa from 43 health-care facilities confirmed the highest prevalence of IMP metallo-beta-lactamase, followed by VIM type and GES type (3), but no NDM type has been detected. Up to this time, several cases of P. aeruginosa ST773 NDM-1 (PA-NDM-1) spread were described worldwide. The first case was reported by Kocsis et al. in Hungary (5), followed by reports from the USA (6, 7), India (8), Nepal (8), the Netherlands (9), and most recently from Spain (10). In this study, we describe the first case of an imported extremely drug-resistant (XDR) P. aeruginosa isolate of the high-risk clone ST773 harboring the bla NDM-1 gene and causing a local outbreak in a hemato-oncology department of a Czech university hospital.
The first case of a PA-NDM-1 isolate was reported in May 2022 from a rectal swab of a male patient suffering from acute myeloblastic leukemia. He was repatriated from Ukraine and admitted to the hemato-oncology department of the University Hospital in Pilsen. Prior to the admission, he underwent a bone marrow transplantation in Turkey in the spring of 2021 and was subsequently taken back into care in Ukraine. Since the admission in May 2022, a surge in the number of NDM-producing P. aeruginosa isolates has been recorded (up until January 2023). Another 17 patients with PA-NDM-1 were detected in the same hemato-oncology department. Due to the severity of the primary diseases and the limited options of antibiotic therapy, five patients who developed bacterial sepsis died. During the same period, two more cases were detected in other Czech hospitals. The first one was reported from a patient in the University Hospital in Hradec Kralove. The patient, who has been living long-term in Tunisia, was hospitalized there due to a car accident for 10 days and then transferred to Hradec Kralove. The second case was reported from a patient repatriated from Ukraine in a surgery ambu lance of Hospital Ceske Budejovice (Table S1).
Species identification was done using MALDI-TOF MS. Antibiotic susceptibility testing was performed using broth microdilution assay according to the EUCAST (2025) guidelines, and results were interpreted according to its 2025 breakpoints criteria (http:// www.eucast.org/). Carbapenemase production was confirmed using the MALDI-TOF MS meropenem hydrolysis assay (11), phenotypic detection of carbapenemase type was done using the double-disc synergy test, and the PCR amplification of genes encod ing carbapenemases was performed as described previously (12). All isolates were P. aeruginosa harboring bla NDM . All isolates showed an XDR profile showing resistance against most of the tested antibiotics, such as ampicillin-sulbactam, piperacillin-tazobac tam, ceftazidime, meropenem, and ciprofloxacin, yet were susceptible against aztreo nam, colistin, and cefiderocol (Table S2).
Genomic DNAs from all isolates were sequenced using a short-read sequencing platform on NovaSeq 600 (Illumina, Inc., USA). All reads, assembled using SPAdes v3.14.0 (13), were analyzed using public databases in the Center for Genomic Epidemiology (https://genomicepidemiology.org/) (14) (Table 1). All isolates belonged to sequence type ST773 and harbored genes coding for resistance against ciprofloxacin (qnrVC1), aminoglycoside (rmtB, aph(3')-llb), sulfamethoxazole (sul1), fosfomycin (fosA), tetracycline (tet(G)), amphenicol (catB7), and beta-lactams, including carbapenems (intrinsic bla PAO and most importantly bla NDM-1 ). Furthermore, all isolates harbored genes coding for quorum sensing (lasI, lasR, rhlI, rhlR, hdtS) and adherence (such as pilB, pilD-U, pilZ, and fimV). Moreover, isolates harbored genes coding for anti-phagocytosis (alg44, alg8, algA-X, mucA-E, and mucP). These isolates were further tested through analyzing their in vitro virulence by testing the survival of bacteria in pooled human serum (NHS) or heat-inactivated normal human serum (HI-NHS) as described elsewhere (15). We found that all P. aeruginosa isolates, except isolate CZ75789, which is not part of the outbreak, exhibited resistance to killing by complement (Fig. S1). DNA isolated from the index case strain (CZI1002861) and from another strain (CZI1023419), collected at the end of the collection period (after 6 months), were sequenced on PacBio Sequel I (Pacific Biosciences, USA) to produce complete circular genomes and for the purpose of deep genomic comparison for possible detection of microevolution. Long-read WGS data showed that the bla NDM-1 was inserted into the P. aeruginosa chromosome. The bla NDM-1 gene was located within the integrative and conjugative element (ICE) ICE6600-like, as previously described in the P. aeruginosa strain P-600 (GenBank accession no. CP053917) isolated from South Korea.
Furthermore, the clonality of the isolates of this outbreak was determined by CRISPR array typing as previously described (16). All isolates had the CRISPR/Cas I-E type with an identical CRISPR array (100%).
Finally, SNP-based phylogeny for all P. aeruginosa ST773 in the NCBI database (along with our isolates) was done as described previously (16) (index case was used as a reference). The phylogeny grouped the outbreak isolates into one clade (clade A) (Fig. 1), while the isolate from Hradec Kralove belonged to another clade (clade B) of ST773 isolates. The ST773 isolate from Budejovice wasn't grouped in any clade. The SNPs between the genomes in comparison with the genome of the index case ranged from 0 to 11 among the isolates of the outbreak. While it was higher with the ones from Hradec Kralove and Budejovice (56 and 333, respectively) (Table S3). These results strongly suggest that the outbreak was caused by one strain.
The hospitals implemented immediate infection control measures to control the outbreak through isolation/quarantine of colonized patients separately or in groups, depending on the initial diagnosis of the patient. Moreover, they increased cleaning/san itary intervention methods, staff and patients education, and continuous consultation with local epidemiologists. These measures managed to limit the dissemination of this clone.
In conclusion, in the current study, we describe the emergence of an outbreak due to the spread of ST773 P. aeruginosa isolates producing NDM-1 carbapenemase in Pilsen hospital. The outbreak took place due to the repatriation of a patient previously hospitalized in Ukraine. Additionally, two sporadic cases of ST773 P. aeruginosa isolates producing NDM-1 were identified in different hospitals. These findings, which are in agreement with the increased reports in literature reporting the isolation of ST773 P. aeruginosa isolates producing NDM-1 in several countries, highlight the development of ST773 as a high-risk clone.
## References
1. Oliver, Mulet, López-Causapé et al. (2015) "The increasing threat of Pseudomonas aeruginosa high-risk clones" *Drug Resist Updat*
2. Woodford, Turton, Livermore (2011) "Multiresistant Gramnegative bacteria: the role of high-risk clones in the dissemination of antibiotic resistance" *FEMS Microbiol Rev*
3. Papagiannitsis, Medvecky, Chudejova et al. (2017) "Molecular characterization of carbapenemase-producing Pseudomonas aeruginosa of Czech origin and evidence for clonal spread of extensively resistant sequence type 357 expressing IMP-7 metallo-β-lactamase" *Antimicrob Agents Chemother*
4. Tenover, Nicolau, Gill (2022) "Carbapenemase-producing Pseudomonas aeruginosa-an emerging challenge" *Emerg Microbes Infect*
5. Kocsis, Gulyás, Szabó (2021) "Diversity and distribution of resistance markers in Pseudomonas aeruginosa International high-risk clones" *Microorganisms*
6. Alamarat, Babic, Tran et al. (2020) "Long-term compassionate use of cefiderocol to treat chronic osteomyelitis caused by extensively drug-resistant Pseudomonas aeruginosa and extended-spectrum-βlactamase-producing Klebsiella pneumoniae in a pediatric patient" *Antimicrob Agents Chemother*
7. Singh, Pulusu, Pathak et al. (2021) "Complete genome sequence of an extensively drug-resistant Pseudomonas aeruginosa ST773 clinical isolate from North India" *J Glob Antimicrob Resist*
8. Takahashi, Tada, Shrestha et al. (2021) "Molecular characterisation of carbape nem-resistant Pseudomonas aeruginosa clinical isolates in Nepal" *J Glob Antimicrob Resist*
9. Zwittink, Wielders, Notermans et al. (2022) "Dutch CPE and MRSA Surveillance Study Groups. 2022. Multidrug-resistant organisms in patients from Ukraine in the Netherlands" *Euro Surveill*
10. Hernández-García, Cabello, Ponce-Alonso et al. (2024) "First detection in Spain of NDM-1-producing Pseudomonas aeruginosa in two patients transferred from Ukraine to a university hospital"
11. Rotova, Papagiannitsis, Skalova et al. (2017) "Comparison of imipenem and meropenem antibiotics for the MALDI-TOF MS detection of carbapenemase activity" *J Microbiol Methods*
12. Bitar, Papagiannitsis, Kraftova et al. (2022) "Implication of different replicons in the spread of the VIM-1-encoding integron"
13. Bankevich, Nurk, Antipov et al. (2012) "SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing" *J Comput Biol*
14. Bitar, Papagiannitsis, Kraftova et al. (2020) "Detection of five mcr-carrying Enterobacterales isolates in four Czech hospitals. mSphere"
15. Gagaletsios, Tagkalegkas, Bitar et al. (2025) "Exploring virulence characteristics of Klebsiella pneumoniae isolates recovered from a Greek hospital" *Mol Genet Genomics*
16. Chudejova, Sourenian, Palkovicova et al. (2024) "Genomic characterization of ST38 NDM-5-producing Escherichia coli isolates from an outbreak in the Czech Republic" *Antimicrob Agents Chemother*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12654158&blobtype=pdf
|
# Diversity of the Alongshan Virus in Ixodes Ticks Collected in the Russian Federation in 2023
Mikhail Kartashov, Kirill Svirin, Maria Antonets, Alina Zheleznova, Valentina Kurushina, Alexander Agafonov, Vladimir Ternovoi, Valery Loktev
## Abstract
A novel flavi-like virus with a segmented genome-Alongshan virus (ALSV)-has been isolated from Ixodes ticks in Russia. In this study, 4458 ixodid ticks collected in 22 regions of Russia were tested for genetic markers of ALSV by RT PCR. The highest rates of ALSV infection in ticks were detected in the Republic of Khakassia (3.3%) and in Kemerovo Oblast (2.4%), while low infection rates were more typical in the European part of Russia (0.4-0.7%). Complete four-segment genomes of 20 ALSV isolates derived from 22 PCRpositive Ixodes persulcatus ticks were sequenced using a high-throughput approach. The nucleotide sequences for Asian ALSV isolates have a 94.5-96.5% identity to ALSV isolates previously found in China, with this range for the European isolates being 89-93%. This data, together with phylogenetic analysis, indicates the existence of Asian and European subtypes of ALSV, and these may be associated with I. persulcatus and I ricinus ticks. The obtained results express the spread of ALSV in Russia and also may be useful for the diagnosis, prophylactics, and treatment of this infection.
## 1. Introduction
Tick-borne flaviviruses are widespread throughout the world and pose a serious medical problem, causing a significant number of infectious diseases among people [1]. In Russia, the main tick-borne orthoflaviviruses reported are Powassan virus, Omsk hemorrhagic fever virus, and tick-borne encephalitis virus [2][3][4]. Despite the fairly large species diversity, the genome of all flaviviruses has a typical structure and is a non-segmented ss(+)RNA approximately 11 kb long, encoding one extended open reading frame, at the edges of which are the 5 ′ and 3 ′ untranslated regions [5,6].
However, several novel flavi-like viruses have been isolated in recent decades. These viruses are characterized by a segmented genome, a feature that distinguishes them from classical orthoflaviviruses, and they are classified within a distinct taxonomic group, the Jingmenviruses [7][8][9][10]. Such viruses have a segmented positive single-stranded RNA genome, and only two genes have a certain identity to the RNA-dependent RNA polymerase (NS5) and helicase (NS3) of "classical" orthoflaviviruses. This Jingmenvirus group includes the Alongshan virus (ALSV), Jingmen tick, Yanggou tick, Mogiana tick, and Kindia tick viruses, and a number of other viruses [10,11]. To date, these flavi-like viruses have been detected across Asia, Europe, South America, and Africa.
The genomes of segmented flavi-like viruses include either four segments (typical of viruses isolated from ticks, bats, monkeys, and humans) or five segments for viruses isolated from mosquitoes [11]. Segment 1 encodes the NS5-like nonstructural protein, which is similar to NS5 in orthoflaviviruses, and Segment 3 encodes the NS3 polypeptide. The N-terminal domain of NS3 exhibits protease activity, and the C-terminal domain functions as a helicase. The NS3 protein, along with NS5, plays a central role in virus replication. Proteinase activity is required for polyprotein processing, while the helicase domain is involved in capping and viral RNA synthesis. To date, the structure of the NS3 protein in most unsegmented flaviviruses has been studied, and high homology has been shown not only in terms of structure but also in the mechanisms of ATP hydrolysis, recognition, and the unwinding of RNA. Structural proteins VP1, VP2, and VP3 are encoded in Segments 2 and 4 and have no known homologues either among the Flaviviridae family or among other known viruses. Segment 2 in ALSV encodes putative glycoproteins VP1a and VP1b, as well as a small protein with three transmembrane domains, the function of which is unknown. Proteins VP2 (putative capsid protein) and VP3 (putative viral membrane protein) are encoded in Segment 4 and have partially overlapping translation frames.
In addition, additional genomic segments have recently been described for the Jingmenvirus genome [12]. This discovery reveals the fluidity of this genome and the possibility of combinations of segments packaged in different virus particles. This may provide additional evidence indicating that multipartite virions really do exist.
Following the discovery of the first known flavi-like viruses with segmented (multipartite) genomes in China and Brazil [7,8], the circulation of ALSV was detected in ticks and humans in northeastern China (Inner Mongolia and Heilongjiang Province) [13,14]. Subsequent studies detected ALSV RNA in I. ricinus ticks in Finland, France, Serbia, Germany, and Switzerland [11,[15][16][17][18]. ALSV has also been detected in Russia [19][20][21][22]. The genetic material of ALSV has been found in I. persulcatus, I. ricinus, Dermacentor reticulatus, and D. nuttalli ticks collected in the Kaliningrad, Ulyanovsk, and Chelyabinsk oblasts, as well as in the Russian Republics of Karelia, Tatarstan, Gorny Altai, and Tuva. The pathogenicity of multicomponent flavi-like viruses for domestic animals and humans has now been proven. However, this information is fragmentary and limited. It is possible that ALSV's role in infectious pathology may be more significant than is commonly believed.
The aim of this study is to find novel ALSV isolates from ixodid ticks in different regions of Russia and perform whole-genome molecular genetic characterization on them.
## 2. Materials and Methods
## 2.1. Collection and Processing of Ticks
In this study, 4458 individual samples of adult ticks of the species I. persulcatus (N = 4122) and I. ricinus (N = 336) were analyzed. The ticks were collected in 23 regions in the summer of 2023 by flagging from vegetation. Figure 1 shows the locations of tick collection sites and tick species. The species of tick samples was established by their morphological characteristics. The collected ticks were washed twice with 70% ethanol to remove external contaminants and external microflora, following which they were stored at a temperature of -80 • C for subsequent analysis. Additional taxonomic verification of ALSV-positive tick samples was carried out by determining the nucleotide sequence of the mitochondrial cytochrome oxidase gene. S1 andS2.
## 2.2. Reverse Transcriptase PCR (RT-PCR) and Sequencing of Amplified Products
Adult ticks were homogenized using the TissueLyser II laboratory homogenizer (QIAGEN, Hilden, Germany) in 300 µL 0.9% saline solution. Viral RNA from 150 µL tick suspensions was isolated with ExtractRNA (Evrogen, Moscow, Russia), according to the manufacturer's protocols. Screening of the obtained samples for the presence of ALSV RNA was performed by RT-PCR using screening primers complementary to a fragment of Segment 2: Miass_gly_3F TGGATCAGCTCACACCACAC and Miass_gly_3R TCACCGTCACAGTGGAATGG [19]. PCR was performed on a T1000 amplifier (Bio-Rad, Hercules, CA, USA) in 25 µL of the BioMaster RT-PCR Standard reaction mixture (2×) (Biolabmix, Novosibirsk, Russia) containing 0.4 pM primers under the following conditions: polymerase activation at 95 • C for 5 min and then 38 cycles of 95 • C for 10 s, 53 • C for 20 s, and 68 • C for 30 s. The amplification products (expected length 333 bp) were analyzed by electrophoresis in 2% agarose gel containing ethidium bromide at a concentration of 2 µg/mL and visualized in the UV spectrum using a GelDoc Go Gel Imaging System (Bio-Rad, Hercules, CA, USA). To confirm the specificity of RNA detection, the ALSV PCR product was gel-purified and then sequenced in both directions on the ABI PRISM 3500 (Applied Biosystems, Foster City, CA, USA) sequencer using ABI PRISM ® BigDye™ Terminator v.3.1.
## 2.3. NGS Sequencing and NGS Data Analysis
To enrich the library for high-throughput sequencing, we used targeted PCR with a panel of primers for all 4 segments (Table 1). To perform targeted amplification of ALSV, the amplification method was optimized by experimentally choosing the temperature regime and concentrations of the reaction mixture components. The concentration of purified PCR products was estimated by the fluorescence method on a Qubit 2.0 device (Thermo Fisher Scientific, Waltham, MA, USA) using the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific). Sequencing was performed using the MiSeq Reagent Kit v3 (Illumina, San Diego, CA, USA) for 600 cycles. Cutadapt (version 5.1) and SAMtools (version 1.20) were used to remove the Illumina adaptors and duplicate reads. The contigs were assembled de novo using the MIRA assembler with default parameters (version 4.9.6). High-throughput sequencing data were processed using a BLASTN-based (version 2.17.0+) taxonomic read identification algorithm.
## 2.4. Phylogenetic Analysis
The nucleotide sequences of the genome coding regions of each segment were aligned using ClustalW. Phylogenetic analysis was conducted using the maximum likelihood method and the Tamura-Nei model in MEGA 10/11 with 1000 bootstrap replications [23,24]. The percentage identities of nucleotide and amino acid sequences of ALSV were computed in MEGA 10/11 using the default settings.
## 2.5. Nucleotide Sequence Accession Numbers
Nucleotide sequences determined in the study are available in the GenBank database under accession numbers: PP623704-PP623718 and PP942935-PP942941 for Segment 1; PP623719-PP623733 and PP942942-PP942948 for Segment 2; PP623734-PP623748 and PP942949-PP942955 for Segment 3; PP623749-PP623763 and PP942956-PP942962 for Segment 4.
## 2.6. Biosafety
Experiments with potential infectious material were carried out in accordance with the requirements of the following biosafety rules: "Sanitary and Epidemiological Requirements for the Prevention of Infectious Diseases" N 3.3686-21 dated 28 January 2021.
## 3. Results
## 3.1. Tick Collection and ALSV Detection
The study analyzed 4458 individual samples of adult ticks collected from 22 regions of Russia (Figure 1, Tables S1 andS2). Among them, 22 ticks were found to be ALSV-positive by RT-PCR; the average infection rate of ticks was thus 0.5% (22/4458; 95% CI: 0.3-0.7). ALSV-positive ticks were detected in the Transbaikal Territory, Irkutsk Oblast, Tuva Republic, Republic of Khakassia, Kemerovo Oblast, Udmurt Republic, and Vologda Oblast (Table S2). The highest infection rates were in the Republic of Khakassia (3.3%) and Kemerovo Oblast (2.4%). The lowest infection rate was in Vologda Oblast (0.4%). In this study, all ALSV-positive ticks belonged to the species I. persulcatus. No positive samples were found among the I. ricinus ticks studied from the Bryansk and Smolensk oblasts, although ALSV-positive I. ricinus ticks have been detected in Russia [20].
## 3.2. Analysis of Genome Identity
When compared with other ALSV isolates found in China, the studied isolates from the Asian part of Russia have a nucleotide sequence identity level of about 96.5% for Segments 1, 3, and 4, and 94.5% for Segment 2. The corresponding figures for Russian isolates of the European clade are in the range of 90% for segments 1, 3, and 4, and 91% for Segment 2. The level of difference with the prototype isolate found in Finland in the I. ricinus tick is in the 89-93% range.
The level of nucleotide sequence differences between the studied genetic variants of ALSV is about 5% for Segments 1 and 4 and about 4% for Segments 2 and 3 (Figure 2). Differences in deduced amino acid sequences for proteins encoded by Segments 1, 2, and 3 (NS5, VP2, VP3, VP1a, VP1b, and NS3) were roughly 1%, with the highest variation observed in VP2 and VP3 encoded by Segment 4. The most conserved proteins were nonstructural proteins NS3 and NS5. Heat maps of identity for nucleotide and amino acid sequences for the described and studied ALSV isolates also demonstrate the above-described difference between the compared sequences (Figures S1 andS2). However, it is noteworthy that nucleotide substitutions characteristic of all segments do not always lead to pronounced amino acid substitutions. Moreover, the amino acid sequences of NS3 and VP2 have the greatest conservatism, and the variability is typical for the VP1a polypeptide.
## 3.3. Phylogenetic Analysis
The phylogenetic trees demonstrate that genetic variants of ALSV circulating in I. persulcatus ticks in the southern ranges of Eastern Siberia (the Transbaikal Territory, Irkutsk Oblast, Tuva Republic, and the Republic of Khakassia) and Western Siberia (Kemerovo Oblast) are grouped with sequences found in China across four segments (Figure 3). This Asian subtype (clade) is represented by variants that form the Asian isolates found in I. persulcatus ticks (and in humans). Interestingly, this clade also includes a single isolate from the Udmurt Republic (Europe), which is located on the border between the European and Asian parts of Russia. Most of the ALSV variants from the Udmurt Republic, as well as the isolate from Vologda Oblast, belong to separate clades within the European subtype, together with prototype variants from Chelyabinsk Oblast (Ural Mountains). These isolates were detected in I. persulcatus ticks. Another clade of the European subtype is associated with I. ricinus ticks and is found in Western European regions.
## 4. Discussion
The current epidemiological situation in Russia with regard to tick-borne infections is characterized not only by multiple incidents of already known tick-borne infections, but also by the detection novel tick-borne pathogens such as ALSV. ALSV was first isolated from the blood of patients with fever in northeastern China [13,14]. The virus' RNA was detected in 86 of 384 patients with fever and those with a history of tick bites. Patients infected with ALSV had a history of fever, headaches, and other symptoms that resemble the manifestations of other tick-borne infections. Closely related viruses such as Jingmen tick virus, Mogiana tick virus, and Kindia tick virus have also been detected in primates in Uganda, cattle in Brazil and Guinea, and patients with Crimean-Congo hemorrhagic fever in Kosovo and Russia [8, [25][26][27][28][29].
The ALSV genome is represented by ssRNA of positive polarity and consists of four segments [13,14]. Recently, the two novel putative structural proteins in the duplicated segments have been described [12]. This result highlights the fluid nature of the genomes of Jingmenviruses and their multipartite virions. Different combinations of segments packaged in different virus particles could facilitate the acquisition or loss of genomic segments and segment duplication following genomic drift. Comparison of the nucleotide sequences revealed high intraspecific variability at a level of 4.6-7.7%. Amino acid sequences were more conserved, with 0.5-1.9% variability. The NS5 and NS3 flavi-like proteins, encoded by Segment 1 and 3, respectively, are the most conserved polypeptides. The exceptions are the structural glycoproteins VP1a and VP1b, in which the amino acid variability reaches 7.5% and 4.5%, respectively. The increased variability of the putative structural proteins may be due to the pressure of the host immune response or the need for ALSV to adapt to different hosts. The accumulation of point substitutions in these proteins probably provides ALSV with the ability to replicate in various hosts and in different natural foci. Analysis of complete nucleotide sequences of the four segments of the ALSV genome showed that the identity level of the nucleotide sequences (4-6%) of Asian isolates is closer to ALSV isolates previously found in China. The European ALSV isolates have a greater number of differences that indicate the independent evolution of ALSV in different geographical regions of Eurasia.
Phylogenetic analysis of the four genome segments of ALSV showed that ALSV isolates may be divided into Asian and European subtypes (Figure 3). The Asian subtype is closely related to isolates first isolated in China, close to the Russian-Chinese border [13,14], and these isolates are associated with the novel ALSV variants found in this study in the southern ranges of Eastern and Western Siberia. All of these isolates were found in I. persulcatus ticks only. The ALSV isolates of the European genotype are associated with two species of ticks, I. persulcatus and I. ricinus [19][20][21]. These isolates form two separate phylogenetic branches (subclades). They can provisionally be divided into Western European and Eastern European subtypes of ALSV, with the latter including isolates collected in the Ural Mountains. It may be conjectured that the ecosystems of the south of Eastern Siberia and the north of Mongolia are optimal for the circulation ALSV infection [13,30]. Moreover, the vast territory the southern reaches of Eastern Siberia border is territorially close to the interior regions of China, where the circulation of ALSV was firstly detected [14,31].
Previously, ALSV isolates were divided into the I. ricinus and I. persulcatus groups according to the main vector species. The I. persulcatus group is divided into two subgroups, the European (the republics of Karelia and Altai and Chelyabinsk Oblast) and the Asian (China, the republics of Altai, Tuva, and Karelia, the oblasts of Chelyabinsk and Ulyanovsk, and Altai Krai) [19][20][21]. This assumption was confirmed in the present study. All ALSV isolates circulating in the south of Eastern Siberia and in Western Siberia in I. persulcatus ticks were clearly clustered into the Asian subgroup of the corresponding vector when analyzed for each of the genome segments. Of interest is a territory in the Udmurt Republic, where most ALSV variants are clustered into the European branch, but an isolate attributed to the Asian branch is also encountered.
Russia tends to experience persistent foci of tick-borne infections in urban and suburban areas [5,29,32,33]. Ticks inhabiting city parks and squares are especially dangerous, since city dwellers perceive the urban environment as being free of ticks and do not take any non-specific preventive measures, unlike people visiting natural biotopes. In our work, a number of places where ticks with ALSV RNA was detected can be classified as biotopes with a high anthropogenic load and located within rural settlements (for example, in the Vologda Oblast, Udmurtia, and Kemerovo Oblast) or along busy highways, as in the Irkutsk Oblast. Some places where ticks with ALSV RNA were detected in Udmurtia are located near a children's country camp.
## 5. Conclusions
The study shows the wide distribution of ALSV in Russia. The highest levels of virus detection were in the Asian part of Russia, and ALSV genetic markers were predominantly associated with I. persulcatus ticks. Analysis of the complete nucleotide sequences of the four segments of the viral genome showed that the nucleotide sequences of the Asian isolates exhibited a high identity to ALSV isolates previously found in China. The highest level of differences is observed for the VP2 and VP3 polypeptides (Segment 4). Flavi-like proteins NS5 and NS3, encoded by Segments 1 and 3, respectively, are the most conserved polypeptides. This information, together with phylogenetic analysis for the four genome segments of ALSV indicate the existence of Asian and European subtypes (clades) that may be associated with I. persulcatus and I. ricinus ticks, respectively. These data make evident the need to detect changes at the boundaries of the spread of modern ALSV isolates and other flavi-like viruses that are potentially dangerous to humans, which will make it possible to predict the transmission of these tick-borne infections.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms13112564/s1, Table S1: Characteristics of the studied tick samples and the number of PCR-detected ALSV-positive ticks; Table S2
## References
1. Pierson, Diamond (2020) "The continued threat of emerging flaviviruses" *Nat. Microbiol*
2. Leonova, Kondratov, Ternovoi et al. (2009) "Characterization of Powassan viruses from Far Eastern Russia" *Arch. Virol*
3. Ruzek, Avšič Županc, Borde et al. (2019) "Tick-borne encephalitis in Europe and Russia: Review of pathogenesis, clinical features, therapy, and vaccines" *Antivir. Res*
4. Tyulko, Fadeev, Vasilenko et al. (2024) "Analysis of changes in the genome of the Omsk hemorrhagic fever virus (Flaviviridae: Orthoflavivirus) during laboratory practices for virus preservation" *Probl. Virol*
5. Ternovoi, Gladysheva, Ponomareva et al. (2019) "Variability in the 3 ′ untranslated regions of the genomes of the different tick-borne encephalitis virus subtypes" *Virus Genes*
6. Postler, Beer, Blitvich et al. (2023) "Renaming of the genus Flavivirus to Orthoflavivirus and extension of binomial species names within the family Flaviviridae" *Arch. Virol*
7. Qin, Shi, Tian et al. (2014) "A tick-borne segmented RNA virus contains genome segments derived from unsegmented viral ancestors" *Proc. Natl. Acad. Sci*
8. Villa, Maruyama, De Miranda-Santos et al. (2017) "Complete Coding Genome Sequence for Mogiana Tick Virus, a Jingmenvirus Isolated from Ticks in Brazil" *Genome Announc*
9. Colmant, Charrel, Coutard (2022) "Jingmenviruses: Ubiquitous, understudied, segmented flavi-like viruses" *Front. Microbiol*
10. Ogola, Roy, Wollenberg et al. (2025) "Strange relatives: The enigmatic arbo-jingmenviruses and orthoflaviviruses. npj Viruses"
11. Temmam, Bigot, Chrétien et al. (2019) "Insights into the host range, genetic diversity, and geographical distribution of Jingmenviruses. mSphere"
12. Valle, Parry, Coutard et al. (2025) "Discovery of additional genomic segments reveals the fluidity of jingmenvirus genomic organization" *Virus Evol*
13. Wang, Wang, Wei et al. (2019) "A New Segmented Virus Associated with Human Febrile Illness in China" *N. Engl. J. Med*
14. Wang, Wang, Wang et al. (2019) "Prevalence of the emerging novel Alongshan virus infection in sheep and cattle in Inner Mongolia, northeastern China. Parasites Vectors"
15. Kuivanen, Levanov, Kareinen et al. (2019) "Detection of novel tick-borne pathogen, Alongshan virus"
16. Stanojević, Li, Stamenković et al. (2020) "Depicting the RNA virome of hematophagous arthropods from Belgrade, Serbia. Viruses"
17. Ebert, Söder, Kubinski et al. (2023) "Detection and characterization of Alongshan virus in ticks and tick saliva from Lower Saxony, Germany with serological evidence for viral transmission to game and domestic animals. Microorganisms"
18. Stegmüller, Fraefel, Kubacki (2023) "Genome Sequence of Alongshan virus from Ixodes ricinus ticks collected in Switzerland" *Microbiol. Resour. Announc*
19. Kholodilov, Litov, Klimentov et al. (2020) "Isolation and characterisation of Alongshan virus in Russia" *Viruses*
20. Kholodilov, Belova, Morozkin et al. (2021) "Geographical and tick-dependent distribution of flavi-like Alongshan and Yanggou tick viruses in Russia" *Viruses*
21. Kholodilov, Belova, Ivannikova et al. (2022) "Distribution and characterisation of tick-borne flavi-, flavi-like, and phenuiviruses in the Chelyabinsk Region of Russia" *Viruses*
22. Kartashov, Krivosheina, Kurushina et al. (2024) "Prevalence and genetic diversity of the Alongshan virus (Flaviviridae) circulating in ticks in the south of Eastern Siberia" *Probl. Virol*
23. Tamura, Nei (1993) "Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees" *Mol. Biol. Evol*
24. Tamura, Stecher, Kumar (2021) "MEGA11: Molecular Evolutionary Genetics Analysis Version 11" *Mol. Biol. Evol*
25. Ladner, Wiley, Beitzel et al. (2016) "A Multicomponent Animal Virus Isolated from Mosquitoes" *Cell Host Microbe*
26. Souza, Fumagalli, Torres Carrasco et al. (2018) "Viral diversity of Rhipicephalus microplus parasitizing cattle in southern Brazil" *Sci. Rep*
27. Emmerich, Jakupi, Von Possel et al. (2018) "Viral metagenomics, genetic and evolutionary characteristics of Crimean-Congo hemorrhagic fever orthonairovirus in humans" *Kosovo. Infect. Genet. Evol*
28. Kartashov, Krivosheina, Naidenova et al. (2025) "Simultaneous Detection and Genome Analysis of the Kindia Tick Virus in Cattle and Rhipicephalus Ticks in the Republic of Guinea. Vector Borne Zoonotic Dis"
29. Ternovoi, Gladysheva, Sementsova et al. (2020) "Detection of the RNA for new multicomponent virus in patients with Crimean-Congo hemorrhagic fever in southern Russia" *Ann. Russ. Acad. Med. Sci*
30. Su, Cui, Xing et al. (2024) "Metatranscriptomic analysis reveals the diversity of RNA viruses in ticks in Inner Mongolia" *PLoS Negl. Trop. Dis*
31. Xu, Wang, Li et al. (2024) "Alongshan Virus Infection in Rangifer tarandus Reindeer, Northeastern China" *Emerg. Infect. Dis*
32. Korobitsyn, Moskvitina, Tyutenkov et al. "Detection of tick-borne pathogens in wild birds and their ticks in Western Siberia and high level of their mismatch" *Folia Parasitol*
33. Kartashov, Gladysheva, Shvalov et al. (2023) "Novel Flavi-like virus in ixodid ticks and patients in Russia" *Ticks Tick Borne Dis*
34. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12824294&blobtype=pdf
|
# npj | vaccines Article
Tong Sun, Yanfeng Yao, Chuanwen Tian, Yun Peng, Yingnan Liu, Ge Gao, Zhisheng Li, Hang Liu, Jingyi Han, Miaoyu Chen, Shuqi Xiao, Zhiming Yuan, Shan Chao, Jingyi Liu, Hongjun Chen, Chao Shan
Nipah virus (NiV) is a zoonotic pathogen that causes severe encephalitis and respiratory disease in humans and multiple mammalian species. However, no licensed vaccines or therapeutics are currently available against NiV infection. In this study, we developed three mRNA vaccine candidates using a lipid nanoparticle (LNP) delivery platform: mRNA-F-LNP, comprising mRNA encoding the fusion protein (F); mRNA-G-LNP, containing mRNA encoding the attachment glycoprotein (G); and mRNA-GF-LNP, in which mRNAs encoding both F and G proteins were co-encapsulated at a 1:1 molar ratio. All three mRNA-LNPs induced robust and sustained immune responses in both mice and Syrian hamsters. Sera from immunized Syrian hamster showed high levels of cross-neutralizing antibodies against both NiV-Malaysia (NiV-M) and NiV-Bangladesh (NiV-B) strains. Notably, all three mRNA-LNPs conferred complete protection against a lethal challenge with NiV-M in Syrian hamsters. These findings demonstrate that these mRNA-based vaccines are highly immunogenic and efficacious, highlighting their potential as promising candidates for NiV vaccine development.
Nipah virus disease is a highly virulent zoonotic infection caused by Nipah virus (NiV), which was first identified in 1998 during an outbreak among pig farms in Malaysia 1,2 . Fruit bats of the genus Pteropus are the main natural reservoir of NiV 3 . Human infection can lead to severe disease characterized by encephalitis and respiratory symptoms, with high case fatality rates 4 . Beyond humans, other mammals, including pigs, horses, dogs and cats, are susceptible to NiV infection [5][6][7] . Since its initial emergence, NiV has caused recurrent outbreaks in several countries across Southeast and South Asia, with a notable outbreak reported in Kerala, India, in 2024the sixth in that region since 2018 8 . Two major strains, NiV-Malaysia (NiV-M) and NiV-Bangladesh (NiV-B), have been reported in human outbreaks and are associated with varying clinical severity and mortality rates 9,10 . Due to its high pathogenicity and epidemic potential, NiV is classified as a Risk Group 4 (RG-4) agent and has been designated a priority pathogen by the World Health Organization (WHO) 9 .
As a member of the Paramyxoviridae family 11 , NiV encodes two essential surface glycoproteins: the attachment glycoprotein (G) and the fusion protein (F). The G protein mediates viral entry through binding to highly conserved ephrin B2/B3 receptors 12 , a key step that may explain NiV's broad host range 13,14 . This receptor binding triggers conformational changes in the F protein, leading to its cleavage by host proteases into F1/ F2 subunits, enabling membrane fusion and viral entry 15 . Given their essential roles in the viral life cycle, surface accessibility, and strong immunogenicity, both F and G proteins are recognized as key protective antigens and primary targets for neutralizing antibodies and vaccine development 16 .
Current strategies for vaccine development against NiV include several platforms: recombinant viral vectors, such as those based on vesicular stomatitis virus (rVSV) [17][18][19][20] , canarypox virus (rALVAC) 21 , vaccinia virus 22 , adeno-associated virus 23 , and measles virus 24 ; subunit vaccines based on the soluble G protein (sG) of NiV and Hendra virus (HeV) 25 ; and nucleic acid vaccines and immunoinformatics-designed immunogens. For example, one study developed a structurally stabilized multi-epitope vaccine whose immunogenicity was enhanced by computational prediction of B-cell and T-cell epitopes 26 . While numerous vaccine candidates can induce neutralizing antibodies, only three have progressed to Phase I clinical trials: PHV02 (RVSV-Nipah Virus Vaccine Candidate), HeV-sG-V (Hendra Virus Soluble Glycoprotein Vaccine), and mRNA-1215 (NiV mRNA Vaccine) 27 .
The mRNA vaccine technology has emerged as a promising platform against emerging pathogens, owing to its distinct advantages such as rapid development timelines, flexible antigen design, robust induction of both humoral and cellular immune response, and an excellent safety profile 28 . In this study, we developed three mRNA-lipid nanoparticle (mRNA-LNP) vaccine candidates: mRNA-F-LNP encoding the F protein, mRNA-G-LNP encoding the G protein, and mRNA-GF-LNP co-expressing both F and G proteins. These candidates were evaluated in both mice and Syrian hamsters. All three vaccines induced durable, high-titer antibodies with neutralizing activity against NiV pseudoviruses and live viruses. Notably, all vaccinated Syrian hamsters demonstrated 100% survival following lethal NiV-M challenge.
## Results
## Preparation of mRNA vaccines
To develop an effective NiV vaccine, we optimized the codon usage of the full-length F and G glycoprotein genes derived from the NiV-M strain and cloned them into the pcDNA3.1 vector, generating recombinant plasmids pcDNA3.1-F and pcDNA3.1-G. For mRNA production, the plasmids were amplificated by PCR using a downstream primer containing a defined poly(T) sequence to generate the poly(A) tail. The products were subsequently subjected to in vitro transcription and capping, yielding mRNA-F and mRNA-G, respectively (Fig. 1a).
Following purification, the mRNAs were encapsulated into LNP using a standard formulation (SM-102: DSPC: cholesterol: DMG-PEG2000 at 50:10:38.5:1.5 molar ratio), generating three vaccine candidates: mRNA-F-LNP, mRNA-G-LNP, and mRNA-GF-LNP. The mRNA-GF-LNP formulation was generated by co-encapsulating mRNA-F and mRNA-G at a 1:1 molar ratio within the same LNP system. This co-encapsulation approach ensures synchronized delivery of both antigen-encoding mRNAs to target cells, potentially enabling coordinated F and G protein expression that may better mimic natural viral antigen presentation (Fig. 1b andc). All LNP formulations exhibited optimal physicochemical properties. The mRNA-F-LNP showed a size of 98.86 nm with a polydispersity index (PDI) of 0.094, mRNA-G-LNP showed a size of 96.74 nm with a PDI of 0.143, and mRNA-GF-LNP showed a size of 97.3 nm and a PDI of 0.115. These results confirmed consistent nanoparticle formulation (size range: 80-100 nm, PDI < 0.2) across all three formulations. In vitro characterization in HEK293T cells confirmed successful protein expression and proper membrane localization, as validated by western blotting (WB) and immunofluorescence assays (IFA) (Fig. 1d ande).
## Three mRNA vaccines elicit robust immune responses in mice
To evaluate NiV-specific antibody responses, 6-8-week-old female SPF BALB/c mice (n = 9) were immunized intramuscularly with 5 μg of mRNA-F-LNP, mRNA-G-LNP, or mRNA-GF-LNP. A prime-boost regimen was employed, with a booster dose administered at day 21 post-primary immunization (Fig. 2a). Serum samples were collected weekly from day 7 to day 42 and then biweekly until day 98 for analysis by ELISA and pseudovirus neutralization assays.
ELISA results revealed that G-specific antibodies emerged as early as day 7 in both the mRNA-G-LNP and mRNA-GF-LNP groups, whereas F-specific antibodies became detectable at day 14 in the mRNA-F-LNP and mRNA-GF-LNP groups (Fig. 2b-e). Following booster immunization, antibody titers against both glycoproteins increased significantly, peaking (∼10 5 ) by day 42. The mRNA-GF-LNP group elicited F-and G-specific antibody titers comparable to those induced by the mRNA-F-LNP and mRNA-G-LNP groups, respectively. All vaccine formulations maintained high antibody titers through the end of the observation period (day 98).
Pseudovirus neutralization assays revealed distinct kinetic profiles (Fig. 2f). Neutralizing antibodies were initially detected only in the mRNA-GF-LNP group at day 7. By day 21, all vaccines elicited measurable neutralizing antibodies, with mRNA-GF-LNP showing significantly higher titers than mRNA-F-LNP (P < 0.01) and titers equivalent to mRNA-G-LNP. Peak neutralization titers (NT) (∼10 4 ) were achieved by day 42 across all groups, with no significant differences observed at this time point.
To assess NiV-specific T cell responses, IFN-γ ELISpot assays were performed on splenocytes collected 21 days after the second immunization. All three vaccines induced robust cellular immune responses. Following F protein stimulation, the mRNA-GF-LNP group exhibited significantly higher IFN-γ production than the mRNA-F-LNP group (P < 0.05). In contrast, G protein-induced IFN-γ levels were similar between the mRNA-GF-LNP and mRNA-G-LNP groups (Fig. 2g).
All three mRNA vaccines elicit antigen-specific antibodies in Syrian hamsters Prior to challenge experiments, we evaluated the immunogenicity of the three mRNA vaccines in Syrian hamsters. 5-6-week-old female Syrian hamsters (n = 15) were randomly divided into four groups: the mRNA-F-LNP, mRNA-G-LNP, mRNA-GF-LNP, and control groups. Syrian hamsters received prime immunization (10 μg each) followed by an identical booster dose at day 21 via intramuscularly injection. The control group received PBS (Fig. 3a). Serum samples were collected for antibody analysis.
Serological analysis showed that all vaccinated groups developed detectable F-and G-specific antibodies by day 21 after the primary immunization, with antibody titers reaching approximately 10³. Following boosting immunization, we observed robust antibody responses, with antibody titers increasing approximately 10-fold to ～10⁴ by day 28. These elevated antibody levels persisted throughout the 91-day observation period (Fig. 3b-e). Similar to the immunogenicity patterns in mice, Syrian hamsters exhibited consistent antibody kinetics across all vaccine formulations. All groups showed similar antibody responses against both F and G proteins, with synchronized peak responses following booster immunization.
Pseudovirus neutralization profiles in hamsters shared similar features with those in mice (Fig. 3f), with early responses (day 7) in the mRNA-GF-LNP group and delayed neutralization (day 21) detected in the mRNA-G-LNP and mRNA-F-LNP groups. All groups reached similar neutralizing titers by day 42. In live virus neutralization assays, sera from all vaccinated hamsters effectively neutralized both NiV-M and NiV-B, with peak titers reaching approximately 10 3 . Notably, the mRNA-GF-LNP and mRNA-G-LNP groups induced significantly higher neutralizing antibody titers against NiV-B than the mRNA-F-LNP group (P < 0.01 and P < 0.05, respectively) (Fig. 3g).
Collectively, these results demonstrate that all three mRNA vaccines elicit durable antibody responses, indicating a strong potential for long-term protective immunity against NiV.
## Three mRNA vaccines provide complete protection against lethal NiV challenge in Syrian hamsters
To evaluate vaccine efficacy, we utilized a stringent Syrian hamster model of NiV infection [29][30][31][32] . At 21 days post-booster immunization, Syrian hamsters were challenged intraperitoneally with 1000 LD 50 of NiV-M (Fig. 3a). Clinical outcomes were monitored daily for 21 days post-challenge (d.p.c), with virological analysis performed at 5 d.p.c.
All control animals exhibited progressive weight loss and clinical signs after challenge (Fig. 4a), with mortality onset at 5 d.p.c. (1 deaths by 5 d.p.c., 3 by 6 d.p.c., 1 by 9 d.p.c., and the remaining 1 succumbing at 15 d.p.c). In contrast, all vaccinated Syrian hamsters maintained normal body weight, showed no clinical symptoms, and survived through the 21-day observation period.
To assess the effect of vaccination on viral replication, we analyzed the viral loads in spleen, lung and brain tissues collected from 6 Syrian hamsters per group at 5 d.p.c. Control animals exhibited high viral loads (>1 × 10 6 copies/g) across all examined tissues, whereas no viral RNA was detectable in any vaccinated animals, indicating complete viral clearance. Viral titration assays confirmed these findings, with control tissues yielding titers of 1.4 × 10 4 TCID 50 /g (lung), 2.8 × 10 3 TCID 50 /g (spleen), and 60.25 TCID 50 /g (brain) (Fig. 4b). In contrast, no infectious virus was recovered from vaccinated animals. These results demonstrate that all three mRNA vaccine candidates provided complete protection against lethal NiV-M challenge.
## All three mRNA vaccines prevented pathological damage after NiV infection
To further evaluate the protective efficacy of the mRNA vaccine candidates against NiV-induced tissue damage, we performed histopathological and immunohistochemical (IHC) analyses of lung, spleen, and brain tissues collected at 5 d.p.c. Histopathological analyses showed that the control group displayed characteristic pathological manifestations of NiV infection. Lung exhibited extensive hyperplasia, diffuse hemorrhage, alveolar collapse, and marked thickening of alveolar septa attributable to vascular congestion, and inflammatory infiltration; Brain showed prominent perivascular cuffing, and lymphocytic infiltration; Spleen showed significant reduction in follicular size, lymphocyte depletion in white pulp, and increased macrophage infiltration (Fig. 5a). In contrast, vaccinated groups showed no evident pathological alterations. Subsequent IHC analysis of NiV nucleoprotein (N) antigen confirmed these findings. The control tissues displayed abundant diffuse viral antigen distribution across all examined tissues, whereas vaccinated animals showed no detectable viral antigen (Fig. 5b).
Taken together, these data demonstrate that the mRNA-F-LNP, mRNA-G-LNP, and mRNA-GF-LNP vaccines effectively eliminate viral
## Discussion
NiV represents a significant zoonotic threat with case fatality rates of approximately 75% in human outbreaks across Southeast Asia 4,[33][34][35] . The lack of approved treatments or vaccines highlight the urgent need for effective prevention strategies. mRNA technology provides advantages for rapid response to emerging pathogens, featuring simplified production and improved safety profiles 36,37 . In this study, we designed mRNA vaccines targeting the NiV F and G proteins and investigated their protective efficacy in Syrian hamsters.
First, the full-length F and G protein sequences were cloned into plasmids containing the 5' and 3' UTRs. After in vitro transcription and capping, mRNA-G and mRNA-F were obtained. These mRNAs were encapsulated into LNPs containing SM-102 lipid, known to promote efficient cellular uptake and endosomal escape [38][39][40] . Following LNP encapsulation, mRNA-G-LNP and mRNA-F-LNP were prepared. A previous study has shown that chimeric F/G constructs enhanced neutralizing antibody titers 41 . To potentially mimicking natural viral antigen presentation, mRNA-G and mRNA-F were co-encapsulated into LNP, generating mRNA-GF-LNP. Three mRNA vaccine candidates were expressed in HEK293T cells.
Immunogenicity evaluation showed distinct response patterns between the glycoproteins. All three vaccine formulations induced similar antigenspecific antibody titers in mice and Syrian hamsters, with no immunogenicity advantage for the co-formulated vaccine. This phenomenon may be dose-dependent, as 5 μg vaccination dose falls within the 1-10 μg range where mRNA vaccines often show minimal differences in humoral responses 42,43 . Previous studies suggest that immunogenicity differences become more pronounced at lower doses (<0.1 μg), indicating that dose optimization might reveal advantages of the mRNA-GF-LNP 42,43 . These results align with prior vaccinia virus-vectored vaccine study (VV-NiV.G and VV-NiV.F) where coimmunization failed to enhance antibody responses compared to individual immunization 22 .
Notably, F-specific antibodies (from mRNA-F-LNP and mRNA-GF-LNP) became detectable at 14 d.p.i., reaching approximately 10³ by day 21. In contrast, G-specific antibodies (from mRNA-G-LNP and mRNA-GF-LNP) achieved titers of approximately 10³ as early as 7 d.p.i. Previous research has also shown that prefusion-based mRNA vaccine exhibited higher neutralizing activity than post-fusion versions 42 , suggesting protein stability, pre-fusion conformation maintenance, and neutralizing epitope preservation may critically influence immunogenicity. This delayed F-specific antibodies might be caused by this reason or its relatively lower expression levels. Despite these differences, all formulations ultimately produced strong and sustained antibody titers (~10⁵) throughout the 98-day observation period. Additionally, all three candidates elicited strong antigen-specific T-cell responses in mice. Notably, mRNA-GF-LNP triggered significantly higher IFN-γ production than mRNA-F-LNP upon F protein stimulation, whereas G-specific T-cell responses were comparable between mRNA-GF-LNP and mRNA-G-LNP.
Pseudovirus neutralization assays revealed the advantage of mRNA-GF-LNP, generating detectable neutralization by 7 d.p.i. and significantly higher titers than mRNA-F-LNP at 21 d.p.i., suggesting the synergistic effects between G and F proteins. After boosting, all vaccines achieved similar neutralization titers (~10⁴). In live virus neutralization assays, sera from all immunized Syrian hamsters neutralized NiV-M and NiV-B viruses. Challenge studies showed Syrian hamsters immunized with any of the three mRNA vaccine candidates were completely protected against lethal NiV-M challenge. Control group showed significant weight loss by 5 d.p.c. and ultimately succumbed, all vaccinated Syrian hamsters maintained normal weight and survived. Pathological and IHC analyses revealed no detectable lesions or viral antigen in brain, lung, or spleen tissues from immunized groups.
Although this study demonstrates the protective efficacy of the mRNA-LNP candidates against NiV challenge in hamsters, several limitations should be acknowledged. The protective efficacy of these vaccines remains to be evaluated in other susceptible species, such as pigs and non-human primates, which is essential for assessing their broad applicability. While the vaccine dose used in this study provided complete protection, dose-ranging studies are needed to determine the optimal immunization regimen for future development.
In conclusion, we designed mRNA-F and mRNA-G targeting NiV F and G proteins. Through individual and combined LNP encapsulation, we developed three mRNA vaccines candidates: mRNA-F-LNP, mRNA-G-LNP, and mRNA-GF-LNP. All three vaccines induced high-titer specific and neutralizing antibodies in mice and Syrian hamsters. In challenge experiments, all vaccines provided complete protection against lethal NiV-M infection in Syrian hamsters, demonstrating their potential utility for NiV prevention.
## Methods
## Cells and viruses
HEK293T and Vero cells were cultured in complete DMEM (contain 10% fetal bovine serum (FBS) (PAN-Biotech). Cells were cultured at 37 °C in a 5% CO₂ atmosphere. The NiV-M and NiV-B used in this study was provided and handled by the National Virus Resource Center, Wuhan Institute of Virology, Chinese Academy of Sciences.
## Preparation of NiV mRNA vaccines
The F and G gene sequences of the NiV-M (GenBank: AJ627196.1) were retrieved from NCBI and codon-optimized. Following optimization, the sequences were flanked with 5' and 3' UTRs and cloned into the pcDNA3.1 vector downstream of the T7 promoter (Sangon Biotech, Shanghai). For mRNA production, template DNA was amplified by PCR using 2× KeyPo SE Master Mix (Vazyme) with primers containing T7 promoter sequence (forward) and poly(T) plus partial 5' UTR sequence (reverse). Following gel purification, the amplification product was subjected to in vitro transcription using T7 RNA polymerase, followed by the addition of the Cap1 Analog (APExBIO, USA). mRNA was then purified using the MEGAclear™ Kit (Thermo Fisher), yielding final mRNA-F and mRNA-G products.
## LNP encapsulation of NiV mRNA vaccines
The mRNA-F and mRNA-G were encapsulated into LNPs using the following standardized protocol: the aqueous phase was prepared by diluting mRNA solution with nuclease-free water to 180 μg/mL, followed by addition of citrate buffer (10 mM, pH 3.0) to achieve a final mRNA concentration of 90 μg/mL; and the lipid phase consisted of SM-102 (MedChemExpress), 1,2-DSPC (MedChemExpress), cholesterol, and DMG-PEG (2000) (MedChemExpress) dissolved in ethanol (Sinopharm Chemical Reagent Co., Ltd) at a molar ratio of 50:10:38.5:1.5. The two phases were mixed at a 1:3 using a microfluidic chip device (Micro&Nano Co., Ltd) to generate three formulations: mRNA-F-LNP, mRNA-G-LNP, and mRNA-GF-LNP (co-encapsulating both mRNAs at 1:1 molar ratio). LNP characteristics were assessed by dynamic light scattering (DLS).
Western blot analysis HEK293T cells were cultured in 6-well plates (NEST) and incubated with 2 μg of mRNA-F-LNP, mRNA-G-LNP, or mRNA-GF-LNP. The samples were collected after 48 h, denatured in 5× SDS loading buffer at 70 °C for 10 min, and then resolved by 10% SDS-PAGE, the proteins were transferred to a membrane, which was blocked prior to antibody incubation: anti-HA-Tag (6E2) mouse monoclonal antibody (1:1000) (Cell Signaling Technology, CST) and anti-GAPDH polyclonal antibody (1:5000) (Proteintech) conducted at 4 °C for 16-18 h. The membranes were washed with PBST and incubated with HRP-conjugated secondary antibody (CST) (1:5000, 1 h at room tempurature (RT)). Protein bands were developed using chemiluminescent substrate (Yeasen Biotechnology) and visualized by UVP Chemsolo Auto (Analytik Jena). Uncropped and unprocessed original scans are provided in Supplementary Fig. 1.
## Immunofluorescence analysis
When HEK293T cells achieved 70-80% confluency in 96-well plates, they were transfected with mRNA-F-LNP, mRNA-G-LNP, or mRNA-GF-LNP (100 ng per well). Following 24 h transfection, the cells were fixed with anhydrous ethanol (15 min at RT). Following fixation, the cells were blocked with 5% skim milk (1 h at 37 °C). For mRNA-F-LNP or mRNA-G-LNP expression, the cells were incubated with mouse polyclonal antisera specific to F or G protein (1:500, 37 °C for 1 h). After washing with PBST, the cells were incubated with FITC-conjugated anti-mouse IgG secondary antibody (1:2000) (BioLegend) (37 °C for 1 h). Nuclei were then counterstained with DAPI (1:10,000) (Beyotime) (15 min, RT). For mRNA-GF-LNP expression, after blocking and washing, sequential incubations were performed: first with mouse anti-F polyclonal serum (1:500, 37 °C, 1 h), followed by PBST washes, the cells were incubated with pig anti-G polyclonal serum (1:500, 1 h at 37°). Following a wash step, the cells were probed with FITC-conjugated anti-mouse IgG antibody (1:2000) and AF594-conjugated goat anti-pig IgG (HL) antibody (1:500; Abmart), respectively (1 h at 37 °C). Finally, cells were counterstained with DAPI as described above. Expression of G and F proteins were examined using a fluorescence microscope (ZEISS).
## Animal experiments
Thirty-six female BALB/c mice (aged 6-8 weeks) were randomly divided into four groups (n = 9) (SiPeiFu (Suzhou)). Animals received intramuscular immunization with 5 μg of either mRNA-GF-LNP, mRNA-F-LNP, or mRNA-G-LNP, and the control group were injected with 100 μL PBS. All animals were immunized twice with a 21-day interval between doses. Serum samples were collected weekly starting at day 7 post-primary immunization through day 98 for serum antibody monitoring. Mice were anesthetized using 4% isoflurane (RWD Life Science) before blood sampling. Upon completion of the experimental procedures, all mice were euthanized humanely using carbon dioxide inhalation followed by cervical dislocation.
Carbon dioxide was introduced into a sealed chamber at a flow rate that displaced 30% of the chamber volume per minute.
Four experimental groups (n = 15; a total of sixty) were established by randomly assigning SPF female Syrian hamsters (5-6 weeks old, Beijing Vital River Laboratory Animal Technology Co., Ltd). Three immunization groups received 10 μg of mRNA-GF-LNP, mRNA-F-LNP, or mRNA-G-LNP intramuscularly, while controls received 200 μL of PBS. Following two immunizations (21-day interval), randomly selected animals (n = 12 per group) were transferred to ABSL-4 facilities and challenged with 1000 LD 50 of NiV-M via intraperitoneal route. Six per group were euthanized at 5 d.p.c., lung, spleen, and brain tissues were collected for subsequent viral load quantification and histopathological analysis. The remaining six hamsters per group were monitored for body weight changes through 14 d.p.c. and survival through 21 d.p.c. An additional three hamsters per group were maintained for extended antibody monitoring until day 91. At the end of the experiment, all hamsters were euthanized by carbon dioxide inhalation. All animal studies follow the ARRIVE guidelines 44 .
## ELISA
Antibody levels in mouse and hamster serum samples were detected by an indirect ELISA. Briefly, 96-well microplates (NEST) were coated with either NiV Pre-Fusion glycoprotein (ACRO Biosystems) or NiV G Protein (Novoprotein). The blocked plates were incubated with serially diluted serum samples (starting at 1:100, 1 h at 37 °C). After further PBST washes, species-specific HRP-conjugated secondary antibodies: anti-mouse IgG-HRP (1:5000) (CST) or goat anti-hamster IgG-HRP (1:10,000) (Abcam) were added and incubated at 37 °C for 45 min. Plates were washed again and developed with TMB substrate solution (TIANGEN) (37 °C, 15 min) in the dark. The reactions were stopped with 2 M H₂SO₄, and optical density at 450 nm (OD₄₅₀) was measured using a microplate reader (Biotek).
Enzyme-linked immune absorbent spot (ELISpot) assay Splenocytes were adjusted to appropriate density in serum-free medium and plated at 5 × 10⁵ cells per well in 100 μL onto pre-coated mouse IFN-γ ELISpot plates (Mabtech). The cells were stimulated with 5 μg/mL of either G or F protein. Concanavalin A (ConA, 5 μg/mL) and serum-free medium were used as positive and negative controls, respectively. Following incubation, spot-forming units (SFU) were quantified as counts per million cells using AT-Spot™ ELISpot Image Analysis System (Antai Yongxin Tech (Guangzhou) Co., Ltd).
## Pseudovirus neutralization assay
Neutralizing antibody responses were evaluated using a pseudotyped virus system. Pseudoviruses were generated by co-transfecting HEK293T cells (70% confluency in T75 flasks) with the following plasmids at optimized ratio (6:11:1:2): pLOV-CMV-GFP, pSPAX2, pcDNA3.1-F, and pcDNA3.1-G. The total DNA mixture (35 μg) was incubated with 70 μL PEI transfection reagent for 20 min at RT and then added to cells. Following 48 h transfection, the supernatant was harvested and centrifuged at 300 × g for 5 min. The clarified supernatant, designated as the pseudotyped virus, was aliquoted and stored at -80 °C.
After inactivation (56 °C, 30 min), the serum was first diluted 1:20, and a threefold serial dilution was subsequently performed. Sera were mixed with 200 TCID 50 of pseudovirus and incubated at 37 °C for 1.5 h. The mixtures (100 μL) were then added to HEK293T cells. After 48 h incubation, GFP-positive wells were counted, and neutralizing antibody titers (NT) were calculated using the Reed-Muench method.
## Live virus neutralization assay
Serum from Syrian hamsters collected at 14 days post-second immunization have been inactivated (56 °C, 30 min). Using DMEM supplemented with 2% FBS, the serum was subjected to three-fold serial dilutions starting from 1:20. The diluted serum was mixed with 100 TCID₅₀ of NiV-M or NiV-B and incubated at 37 °C for 1 h. The mixture was inoculated onto Vero E6 cells (four replicates per dilution). After 1 h incubation at 37 °C, the inoculum was replaced with complete DMEM and the cells were cultured for 5 days. Cytopathic effect (CPE) was recorded daily. The neutralizing antibody titer was calculated (the highest serum dilution that provided complete protection to 50% of the cell monolayers).
qRT-PCR qRT-PCR was performed to quantify the viral loads in hamster tissues as previously described 45 . Briefly, lung, spleen, and brain tissues were collected from Syrian hamsters at 5 d.p.c. for RNA extraction. The NiV N gene was amplified using a TaqMan probe-based assay on a CFX96 Real-Time System (Bio-Rad) with the HiScript II One Step qRT-PCR Probe Kit (Vazyme). The used primers and probe were: forward primer: 5'-AACATCAGCAG-GAAGGCAAGA-3', reverse primer: 5'-GCCACTCTGTTCTATAGGT TCTTC-3', and probe: FAM-5'-TTGCTGCAGGAGGTGTGCTC-BHQ1-3'. A standard curve was generated to calculate the N gene copy number.
## Virus titration assay
Tissue homogenates were subjected to10-fold serial dilutions in DMEM supplemented with 2% FBS, and 100 μL of each dilution were then applied to Vero E6 cell monolayers and incubated at 37 °C under 5% CO 2 for 1 h, the culture was removed, and the cells were maintained in DMEM containing 2% FBS for 5 days, and final CPE scoring at day 5 was used to calculate viral titers (log10 TCID 50 /g) using the Reed-Muench method.
## Histopathological and immunohistochemical analysis
Following challenge infection, the lung, brain, and spleen from Syrian hamsters were collected and fixed in 10% paraformaldehyde solution for 7 days. After embedding in paraffin, tissue samples were sectioned at a thickness of 4 μm and subsequently stained with H&E. For IHC analysis, an in-house developed monoclonal antibody specific for NiV N protein was used. The sections were scanned using a Pannoramic MIDI system (ZEISS).
## Statistical analysis
All data were analyzed using GraphPad Prism software (version 10.4) and presented as mean ± SEM. Statistical differences were assessed using oneway analysis of variance (ANOVA) and Student's two-tailed t-test. Statistical significance was set at P < 0.05. ns, not significant, *P < 0.05, **P < 0.01.
## Ethic statement
The immunization studies in mice and hamsters complied with the ethical guidelines approved by the Animal Ethics Committee of the Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences (Approval Nos.: SV-20240607-03 and SV-20241018-01). All NiV challenge experiments were performed under Animal Biosafety Level 4 (ABSL-4) facility at the National Biosafety Laboratory (Wuhan), Chinese Academy of Sciences, with protocols approved by the Institutional Animal Care and Use Committee (Approval No.: WIVA21202402).
## References
1. Paliwal, Shinu, Saha (2024) "An emerging zoonotic disease to be concerned about -a review of the nipah virus" *J. Health Popul. Nutr*
2. Amal (2000) "results from a hospital-based case-control study" *Southeast Asian J. Trop. Med. Public Health*
3. Enserink (2000) "Emerging diseases. Malaysian researchers trace Nipah virus outbreak to bats" *Science*
4. Kumar, Anoop Kumar (2018) "Deadly Nipah Outbreak in Kerala: Lessons Learned for the future" *J. Crit. Care Med*
5. Abubakar (2004) "Isolation and molecular identification of Nipah virus from pigs" *Emerg. Infect. Dis*
6. Mills (1999) "Nipah virus infection in dogs" *Emerg. Infect. Dis*
7. Qiu, Wang, Sha (2024) "Infection and transmission of henipavirus in animals" *Comp. Immunol. Microbiol. Infect. Dis*
8. Thiagarajan (2024) "Nipah virus: Kerala reports second death in four months" *BMJ*
9. Bruno (2022) "Nipah Virus disease: epidemiological, clinical, diagnostic and legislative aspects of this unpredictable emerging zoonosis" *Animals*
10. Satter (2006) "Tackling a global epidemic threat: Nipah surveillance in Bangladesh" *PLoS Negl. Trop. Dis*
11. Eaton, Broder, Middleton et al. (2006) "Hendra and Nipah viruses: different and dangerous" *Nat. Rev. Microbiol*
12. Bowden (2008) "Structural basis of Nipah and Hendra virus attachment to their cell-surface receptor ephrin-B2" *Nat. Struct. Mol. Biol*
13. Pernet, Wang, Lee (2012) "Henipavirus receptor usage and tropism" *Curr. Top. Microbiol. Immunol*
14. Ang (2022) "Generating human artery and vein cells from pluripotent stem cells highlights the arterial tropism of Nipah and Hendra viruses" *Cell*
15. Bossart, Fusco, Broder (2013) "Paramyxovirus entry" *Adv. Exp. Med. Biol*
16. Stone, Vemulapati, Bradel-Tretheway et al. (2016) "Multiple strategies reveal a Bidentate Interaction between the Nipah virus attachment and fusion glycoproteins" *J. Virol*
17. Mire (2013) "Single injection recombinant vesicular stomatitis virus vaccines protect ferrets against lethal Nipah virus disease" *Virol. J*
18. Mire (2019) "Use of single-injection recombinant vesicular stomatitis virus vaccine to protect nonhuman primates against lethal Nipah Virus disease" *Emerg. Infect. Dis*
19. Foster (2022) "A recombinant VSV-vectored vaccine rapidly protects nonhuman primates against lethal Nipah virus disease" *Proc. Natl. Acad. Sci. USA*
20. Debuysscher, Scott, Thomas et al. (2016) "Peri-exposure protection against Nipah virus disease using a singledose recombinant vesicular stomatitis virus-based vaccine" *NPJ Vaccines*
21. Weingartl (2006) "Recombinant nipah virus vaccines protect pigs against challenge" *J. Virol*
22. Guillaume (2004) "Nipah virus: vaccination and passive protection studies in a hamster model" *J. Virol*
23. Ploquin (2013) "Protection against henipavirus infection by use of recombinant adeno-associated virus-vector vaccines" *J. Infect. Dis*
24. Yoneda (2013) "Recombinant measles virus vaccine expressing the Nipah virus glycoprotein protects against lethal Nipah virus challenge" *PloS ONE*
25. Mire (2014) "A recombinant Hendra virus G glycoprotein subunit vaccine protects nonhuman primates against Hendra virus challenge" *J. Virol*
26. Rahman (2022) "An immunoinformatics prediction of novel multiepitope vaccines candidate against surface antigens of Nipah Virus" *Int. J. Pept. Res. Ther*
27. Spengler (2025) "Henipaviruses: epidemiology, ecology, disease, and the development of vaccines and therapeutics" *Clin. Microbiol. Rev*
28. Hassett (2019) "Optimization of lipid nanoparticles for intramuscular administration of mRNA Vaccines" *Mol. Ther. Nucleic Acids*
29. Wong (2003) "A golden hamster model for human acute Nipah virus infection" *Am. J. Pathol*
30. De Wit (2011) "Nipah virus transmission in a hamster model" *PLoS Negl. Trop. Dis*
31. Rockx (2011) "Clinical outcome of henipavirus infection in hamsters is determined by the route and dose of infection" *J. Virol*
32. De Wit (2014) "Foodborne transmission of nipah virus in Syrian hamsters" *PLoS Pathog*
33. Chua (1999) "Fatal encephalitis due to Nipah virus among pigfarmers in Malaysia" *Lancet*
34. (1998) "Outbreak of Hendra-like virus--Malaysia and Singapore" *MMWR Morb Mortal Wkly Rep*
35. Harcourt (2004) "Genetic characterization of Nipah virus" *Emerg. Infect. Dis*
36. Ferraro (2011) "Clinical applications of DNA vaccines: current progress" *Emerg. Infect. Dis*
37. Carralot (2004) "Polarization of immunity induced by direct injection of naked sequence-stabilized mRNA vaccines" *Cell Mol. Life Sci*
38. Akinc (2010) "Targeted delivery of RNAi therapeutics with endogenous and exogenous ligand-based mechanisms" *Mol. Ther*
39. Yan (2005) "The role of apolipoprotein E in the elimination of liposomes from blood by hepatocytes in the mouse" *Biochem. Biophys. Res. Commun*
40. Binici, Rattray, Zinger et al. (2025) "Exploring the impact of commonly used ionizable and pegylated lipids on mRNA-LNPs: a combined in vitro and preclinical perspective" *J. Control Release*
41. Loomis (2020) "Structure-based design of Nipah Virus vaccines: a generalizable approach to paramyxovirus immunogen development" *Front. Immunol*
42. Loomis (2021) "Chimeric Fusion (F) and Attachment (G) glycoprotein antigen delivery by mRNA as a candidate Nipah Vaccine" *Front. Immunol*
43. Corbett (2020) "SARS-CoV-2 mRNA vaccine design enabled by prototype pathogen preparedness" *Nature*
44. Kilkenny (2010) "Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research" *PLoS Biol*
45. Lu (2023) "Both chimpanzee adenovirus-vectored and DNA vaccines induced long-term immunity against Nipah virus infection" *NPJ Vaccines*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754552&blobtype=pdf
|
# Viral eye: Emerging insights into corneal and ocular surface viral infections
Marco Zeppieri, Matteo Capobianco, Alessandro Avitabile, Federico Visalli, Cosimo Mazzotta, Mutali Musa, Rosa Giglio, Daniele Tognetto, Caterina Gagliano, Francesco Cappellani, Shah Pt
## Abstract
Viral infections of the ocular surface significantly contribute to morbidity and visual impairment globally. The herpes simplex virus (HSV), adenovirus, cytomegalovirus (CMV), and human papillomavirus (HPV) are predominant pathogens impacting the cornea and conjunctiva, resulting in recurrent illness, epidemic outbreaks, and virus-associated neoplasia. Progress in virology, immunology, and molecular diagnostics has enhanced comprehension of host-virus interactions and introduced novel therapeutic opportunities. A narrative literature review was performed utilizing PubMed, Scopus, and Web of Science, encompassing papers published from 2000 to 2025, with a specific focus on research from 2020 onwards. Eligible publications were peer-reviewed clinical and experimental investigations, together with reviews that focused on epidemiology, Zeppieri M et al. Ocular viral infections WJV https://www.wjgnet.com 2 December 25, 2025 Volume 14 Issue 4etiology, diagnostic methodologies, and therapeutic alternatives. Research indicates that HSV keratitis is the predominant infectious cause of corneal blindness in high-income nations, although adenovirus persists in instigating epidemics of keratoconjunctivitis in the absence of licensed antiviral treatments. CMV keratitis, previously confined to immunocompromised persons, is now acknowledged in immunocompetent patients as a causative agent of corneal endotheliitis. HPV is associated with ocular surface squamous neoplasia, especially in areas with elevated ultraviolet exposure and high human immunodeficiency virus prevalence. Innovative molecular diagnostics, innovative antiviral agents, immunomodulatory approaches, and immunization initiatives signify significant progress that could enhance preventative and therapeutic results.
## INTRODUCTION
Ocular viral infections are a major global health issue, significantly contributing to morbidity, visual impairment, and socioeconomic hardship. The predominant viral ocular illnesses are conjunctivitis and keratitis, with adenoviruses responsible for 65%-90% of viral conjunctivitis in adults, and herpes simplex virus (HSV) causing roughly 1.7 million new cases of keratitis globally each year [1][2][3][4]. These infections exhibit high contagion rates, resulting in recurrent outbreaks in both community and healthcare environments, hence causing significant job absenteeism, inappropriate use of antibiotics, and increased healthcare costs [3,5]. In addition to their immediate effects, viral ocular infections may lead to irreversible corneal scarring, neurotrophic keratopathy, or ocular surface neoplasia, highlighting their enduring implications for vision and quality of life.
The range of viruses associated with ocular surface and corneal diseases is continually broadening. Adenoviruses are the predominant etiological agents of acute viral conjunctivitis in adults, with epidemic keratoconjunctivitis (EKC) attributed to serotypes including HAdV-8, -37, and -64. These highly transmissible strains can persist on fomites for weeks, precipitating epidemic outbreaks that necessitate stringent infection control measures [3,4]. Corneal involvement can lead to subepithelial infiltrates, chronic keratitis, and enduring vision impairments. Topical corticosteroids offer temporary symptomatic relief during the late inflammatory stages; however, their extended usage has hazards such as increased intraocular pressure and delayed virus clearance. Despite ongoing research into novel antivirals aimed at viral entry and replication, no drug has received regulatory approval for adenoviral keratoconjunctivitis.
HSV and varicella-zoster virus (VZV) are primary contributors to infectious corneal blindness in affluent nations [1,4]. HSV keratitis exhibits a recurring pattern, impacting the epithelial, stromal, and endothelial layers, with successive recurrences heightening the likelihood of irreversible visual impairment. Oral prophylaxis with acyclovir markedly lowers recurrence rates, with treatment customized according to disease phenotype: Topical ganciclovir or oral antivirals for epithelial keratitis, and corticosteroids in conjunction with antivirals for stromal keratitis and endotheliitis. VZV, especially as herpes zoster ophthalmicus, can induce both acute and chronic ocular problems. The recombinant zoster vaccination has proven to be an effective preventive strategy, diminishing both the incidence and severity of ocular zoster disease [6].
Cytomegalovirus (CMV), formerly linked to retinitis in immunocompromised patients, has emerged as a notable etiological factor for corneal endotheliitis and anterior segment pathology. CMV keratitis is now observed in immunocompetent individuals, generally presenting as recurrent endotheliitis or keratouveitis. Diagnosis necessitates PCR confirmation from aqueous humor, and therapy typically entails systemic or topical ganciclovir. Antiviral resistance continues to pose a significant barrier in situations necessitating long-term prophylaxis [7].
Alongside these established infections, some developing and re-emerging viruses have been associated with ocular surface illness. Severe acute respiratory syndrome coronavirus 2, monkeypox virus, Zika virus, and dengue virus have all been linked to conjunctivitis, keratitis, and, in certain instances, posterior segment complications, including retinitis and optic neuritis [8,9]. Their acknowledgment highlights the eye's susceptibility as both a locus of infection and a possible reservoir for viral dissemination. CMV retinitis continues to be a significant cause of blindness in areas with elevated human immunodeficiency virus (HIV) incidence, particularly where access to antiretroviral medication is restricted [10]. The prevalence of herpetic ocular illness is probably underestimated in low-and middle-income countries because of elevated HSV seroprevalence and limited access to specialized medical treatment [2].
The extensive repercussions of these infections are seen in the epidemiology of corneal blindness, which is one of the primary contributors to global vision impairment. An estimated 5.5 million individuals are bilaterally blind or experience moderate-to-severe vision impairment due to corneal opacity, while an additional 6.2 million are unilaterally blind. The greatest impact is noted in low-resource environments, when infectious keratitis, ocular damage, and postoperative sequelae are common [11]. Preventive measures, such as infection control, vaccination, and early detection, are essential for mitigating the long-term effects of these disorders.
Recent advancements in molecular biology and immunology are revolutionizing the comprehension of host-virus interactions at the ocular surface. High-throughput systems biology methodologies, single-cell transcriptomics, and computational modeling have elucidated the intricacies of antiviral immune responses, emphasizing the interaction between innate immunity, adaptive mechanisms, and viral evasion tactics [12][13][14]. These discoveries enhance comprehension of viral persistence and latency while also facilitating the advancement of targeted therapeutics and precision medicine approaches in ophthalmology. Innovative diagnostic instruments, including multiplex PCR and next-generation sequencing, provide the swift and precise identification of many eye infections concurrently, enhancing diagnostic efficacy and informing personalized treatment [15].
This study aims to consolidate existing knowledge on four principal viral and virus-associated ocular surface diseases: HSV keratitis, adenoviral conjunctivitis, CMV keratitis, and human papillomavirus (HPV)-related ocular surface squamous neoplasia (OSSN). Each of these diseases illustrates a distinct facet of viral pathology-recurrent infection, epidemic transmission, immune evasion, and oncogenesis. This study emphasizes the problems and potential in tackling viral eye illness by integrating epidemiology, etiology, clinical characteristics, diagnostic innovations, and therapeutic approaches. Advancements in virology, immunology, and clinical care will be essential in alleviating the global impact of corneal and ocular surface viral infections.
This review serves as a narrative minireview aimed at synthesizing contemporary evidence regarding corneal and ocular surface viral infections, specifically emphasizing HSV keratitis, adenoviral conjunctivitis, CMV keratitis, and HPVassociated ocular OSSN. A thorough literature search was conducted utilizing PubMed/MEDLINE, Scopus, and Web of Science to identify articles published from January 2000 to June 2025, with particular focus on studies from 2020 onward to capture the latest advancements in molecular virology, immunology, diagnostics, and therapeutics.
The search strategy utilized a blend of Medical Subject Headings and free-text keywords, encompassing terms such as "herpes simplex keratitis", "HSV keratitis", "adenoviral conjunctivitis", "epidemic keratoconjunctivitis", "cytomegalovirus keratitis", "CMV endotheliitis", "human papillomavirus", "HPV ocular surface squamous neoplasia", and "viral keratitis". Boolean operators were employed to expand or narrow searches as necessary. Further research were obtained by manually examining the reference lists of significant review articles and original publications.
Eligible studies were deemed acceptable if they were peer-reviewed and included original data, clinical observations, or thorough reviews pertinent to viral infections of the ocular surface. Publications were considered if they pertained to epidemiology, etiology, clinical characteristics, diagnostic methods, or therapeutic approaches. Only articles published in English were included. Studies concentrating solely on posterior segment symptoms of viral infections or unrelated to corneal and ocular surface disorders were omitted.
The titles and abstracts were screened for relevance, and full texts were examined in instances of ambiguity. This review, being narrative in style, did not undergo a formal risk-of-bias evaluation; still, priority was accorded to well conducted clinical studies, randomized trials, systematic reviews, and experimental research published in indexed journals. The literature was meticulously examined and compiled to emphasize existing knowledge, developing trends, and prospective directions in ocular surface virology.
## HSV KERATITIS
HSV keratitis is a predominant cause of infectious corneal blindness in affluent nations, with HSV-1 responsible for the majority of cases, while HSV-2 is infrequently involved unless in neonatal or immunocompromised situations [16]. The global prevalence is significant, with around 1.7 million new cases each year and hundreds of thousands affected by recurring or chronic disease. The repetitive occurrence of HSV keratitis significantly impacts visual impairment, healthcare expenses, and the quality of life for patients [2,17]. The implementation of novel therapeutic strategies has progressed at a sluggish pace, and despite extensive study over several decades, the disease continues to pose a significant public health concern [18].
Subsequent to the first ocular infection, HSV develops latency within the trigeminal ganglion. Reactivation may be induced by stress, ultraviolet radiation, fever, or immunosuppression, leading to anterograde transfer of the virus via sensory nerves to the corneal surface, resulting in recurrent epithelial or stromal pathology [16,19]. Latency and reactivation are pivotal to HSV pathogenesis, with immunological dysregulation and viral evasion mechanisms sustaining chronic illness. Corneal pathology is influenced by both viral replication and an increased host immune response, leading to tissue damage and scarring [20].
The clinical spectrum of HSV keratitis is extensive, including epithelial keratitis, stromal keratitis, endotheliitis, and neurotrophic keratopathy [16][17][18][19][20][21]. Epithelial keratitis generally manifests as dendritic or geographic ulcers, directly resulting from viral replication in corneal epithelial cells. Stromal keratitis is primarily immune-mediated, characterized by recurring inflammation that leads to opacification, thinning, and neovascularization, ultimately impairing vision. Endotheliitis arises from inflammation of the corneal endothelium produced by viral antigens, frequently linked to uveitis and corneal edema. Recurring incidents may result in corneal hypoesthesia and neurotrophic keratopathy, characterized by compromised corneal innervation that causes inadequate healing and an increased susceptibility to ulceration and perforation. Each recurrence elevates the probability of permanent visual impairment and necessitates corneal transplantation [21].
Diagnosis is predominantly clinical, augmented by slit-lamp examination; nevertheless, laboratory confirmation is becoming increasingly vital in unusual or treatment-resistant patients. PCR is the most sensitive and specific diagnostic technique, facilitating quick viral detection from corneal scrapings or aqueous humor samples. Recently, metagenomic sequencing techniques have been investigated, especially for atypical or co-infected individuals; however, these methods are primarily limited to research environments and are not yet broadly accessible in clinical practice [22].
Therapeutic techniques are customized according to the clinical subtype of HSV keratitis. Epithelial keratitis is treated with topical antivirals such as ganciclovir gel or acyclovir ointment, or with oral medications such as acyclovir, valacyclovir, and famciclovir [23]. Stromal and endothelial variants necessitate the use of combination oral antivirals and topical corticosteroids to manage immune-mediated injury while inhibiting viral replication [24]. Prolonged oral prophylaxis with acyclovir has been shown to diminish the likelihood of recurrence, serving as a fundamental aspect of preventive therapy [25]. Novel antivirals with enhanced safety profiles are being studied; nonetheless, resistance continues to pose a therapeutic challenge, particularly in immunocompromised individuals [26]. Despite extensive efforts over several decades, no HSV vaccine has achieved clinical application, although numerous vaccine candidates have demonstrated immunogenicity in preclinical or early clinical trials [27]. Recombinant human nerve growth factor (cenegermin) has been effective in facilitating epithelial repair and vision restoration in patients with neurotrophic keratopathy [28].
Recent studies emphasize the intricate relationship among viral infection, immunological dysfunction, and corneal nerve degeneration. Recent immunobiological research has highlighted the contributions of innate and adaptive immune responses, encompassing dendritic cells, T cells, and pro-inflammatory cytokines, in sustaining stromal keratitis [16,29]. The degenerative impact of HSV on corneal nerves intensifies disease chronicity, since structural and functional changes lead to hypoesthesia and compromised wound healing [30]. Comprehending these pathways is essential for creating tailored therapeutics that can disrupt the cycle of recurrence, immune-mediated fibrosis, and neurotrophic injury. Despite advancements in diagnosis and management, HSV keratitis continues to be a significant cause of visual impairment, necessitating more research to develop innovative antivirals, and immunomodulatory approaches that can modify the progression of this complex illness.
## ADENOVIRAL CONJUNCTIVITIS
Adenoviral conjunctivitis is the predominant etiology of viral conjunctivitis in adults, representing 65%-90% of global occurrences, with an estimated annual prevalence of up to 20 million cases in the United States alone. Outbreaks frequently occur in healthcare facilities, educational institutions, military settings, and various community environments, demonstrating the significant transmissibility of adenoviruses. Transmission occurs through direct contact with contaminated hands, fomites, or medical devices, as well as by respiratory droplets and, infrequently, contaminated water [1,3,31]. The virus can remain on environmental surfaces for weeks, with transmission rates in close-contact settings reported between 10% and 50%, highlighting the difficulty of controlling outbreaks [31][32][33].
The clinical manifestations of adenoviral ocular illness differ based on the viral serotype. EKC is predominantly linked to human adenovirus serotypes 8, 19, and 37 whereas pharyngoconjunctival fever (PCF) is generally induced by serotypes 3, 5, 7 and 11 [33]. EKC outbreaks are notably common in Asia, typically peaking around winter and spring, while PCF exhibits a seasonal preponderance in the summer months in areas such China, Australia, and the United States. Epidemiological data over several decades corroborate the worldwide prevalence of pathogenic serotypes and demonstrate changing trends in epidemic dynamics [34].
Adenoviral conjunctivitis clinically manifests with an abrupt onset of erythema, irritation, serous discharge, and conjunctival hyperemia. EKC is generally characterized by chemosis, conjunctival pseudomembranes, and ipsilateral preauricular lymphadenopathy. Corneal involvement occurs in up to 50% of EKC patients, presenting as subepithelial infiltrates that induce discomfort, photophobia, diminished vision, and enduring corneal opacities that may continue for months [31][32][33][34]. Adenoviral keratoconjunctivitis, although typically self-limiting, may lead to persistent keratitis and prolonged morbidity from corneal scarring.
The pathogenesis of adenoviral conjunctivitis is associated with the virus's affinity for the ocular surface. The entry of viruses into corneal epithelial cells is facilitated by the interaction between the adenoviral fiber knob and host cell receptors, particularly sialic acid-containing glycans and integrins like αVβ1 and α3β1 [35,36]. These molecular interactions enable viral binding, internalization, and infection of the corneal epithelium, explaining the pronounced affinity of specific adenovirus serotypes for ocular tissue. This tropism elucidates the clinical severity of EKC and signifies a potential target for innovative antiviral treatments.
The diagnosis of adenoviral conjunctivitis is typically clinical, although point-of-care immunoassays like the AdenoPlus test offer rapid confirmation with high sensitivity and specificity [37]. Molecular diagnostic techniques, like as PCR, are accessible in reference laboratories but are not commonly employed in clinical practice except in severe cases or outbreak scenarios [15].
Management of adenoviral conjunctivitis is primarily supportive, focusing on symptomatic relief with artificial tears, cold compresses, and strict hygiene measures, as no antiviral therapy has yet been proven effective [38]. Topical antibiotics are not warranted unless bacterial superinfection is anticipated. Corticosteroids can mitigate severe inflammation or chronic subepithelial infiltrates; nevertheless, they carry dangers such as viral persistence and steroid-induced glaucoma, thus should be utilized selectively. Emerging evidence indicates that povidone-iodine (1%-5%) may expedite symptom resolution and diminish viral load, especially when used in conjunction with topical corticosteroids like dexamethasone; however, the evidence is of low certainty, and this strategy is not yet regarded as standard practice [39]. Clinical trials of topical antivirals, such as ganciclovir, trifluridine, and cidofovir, have produced inconsistent outcomes, exhibiting limited efficacy and tolerability concerns [40,41]. As of now, no antiviral treatment has received approval from the United States Food and Drug Administration (FDA) or other regulatory bodies especially for adenoviral keratoconjunctivitis.
Prevention is paramount, particularly in outbreak scenarios. Stringent hand hygiene, disinfection of surrounding surfaces, sterilization of ophthalmic tools, and isolation of infected patients are crucial to mitigate transmission. These measures are especially critical in high-risk environments, like hospitals, long-term care institutions, and ophthalmology clinics, where nosocomial transmission has been extensively recorded.
Future research in adenoviral conjunctivitis is exploring next-generation antiviral strategies and preventive approaches. Tailored antivirals, such as multivalent sialic acid conjugates and glycosaminoglycan (GAG) mimetics, have shown promise in in vitro models by blocking viral attachment to cell-surface receptors [42,43]. Broader-spectrum antiviral agents like cidofovir, ganciclovir, and ribavirin are under investigation, though none are yet approved [44]. Research is also being conducted on gene-editing methods that inhibit viral entrance pathways. Innovative approaches, including CRISPR-Cas9 gene editing targeting viral DNA entry or replication, are being studied as potential therapeutics [45]. These advancements offer potential for alleviating the substantial burden of adenoviral ocular illness worldwide.
## CMV KERATITIS
CMV is acknowledged as an opportunistic infection in immunocompromised persons, typically linked to retinitis in patients with severe HIV/AIDS or those receiving immunosuppressive treatment. In the last twenty years, CMV has been recognized as a specific etiological agent of keratitis and corneal endotheliitis in immunocompetent individuals [46]. This paradigm shift has significant therapeutic ramifications, especially in areas where corneal transplantation is prevalent and where anterior segment pathology may resemble other etiologies of keratitis or graft rejection [47]. Rising findings from Asia and Europe have underscored CMV as an undervalued yet substantial contributor to corneal endothelial dysfunction and anterior uveitis [48].
The viral pathogenesis in CMV keratitis entails the reactivation of latent virus within the anterior chamber, resulting in direct infection of corneal endothelial cells [49]. Viral replication results in endothelial cell depletion, persistent inflammation, and compromise of the blood-aqueous barrier, frequently associated with increased intraocular pressure due to trabecular meshwork involvement. CMV can endure latency in myeloid lineage cells while influencing local immune responses. A crucial aspect of immune evasion is the inhibition of cytotoxic T lymphocyte (CTL) activity; research indicates that pp65-specific CTL responses confer protection and correlate with improved long-term results, while IE1specific CTLs may inadvertently induce harmful inflammation [50,51]. These data indicate that the host immune response has a dual role in influencing disease severity and prognosis.
Clinically, CMV keratitis typically manifests as corneal endotheliitis, distinguished by coin-shaped or linear keratic precipitates, corneal edema, and varying degrees of anterior chamber inflammation [49]. Increased intraocular pressure is prevalent, resulting in secondary glaucoma if unrecognized and untreated. Patients who have undergone keratoplasty may experience CMV endotheliitis, frequently misidentified as graft rejection, complicating therapy and jeopardizing graft survival [47].
Precise diagnosis necessitates a heightened level of suspicion in instances of unexplained corneal edema and keratic precipitates, especially when associated with intraocular hypertension. PCR analysis of aqueous humor is the diagnostic gold standard, offering great sensitivity and specificity for the identification of CMV DNA. Quantitative PCR facilitates the evaluation of viral load, which may be associated with disease activity and prognosis. Metagenomic next-generation sequencing (mNGS) has recently emerged as a promising diagnostic instrument in unusual or ambiguous cases, providing the capability to detect CMV with other ocular viruses with enhanced sensitivity [52,53]. Serologic testing is of restricted value in this context, as systemic CMV seropositivity is prevalent and does not indicate localized ocular infection. Imaging methods, such as specular microscopy, can offer supplementary information by assessing endothelial cell density and tracking disease development [54].
Treatment approaches for CMV keratitis emphasize antiviral inhibition of viral replication. Topical ganciclovir gel, generally available in doses between 0.15% and 2%, is frequently utilized as the primary treatment for anterior segment illness in immunocompetent individuals [46][47][48][49]. Oral valganciclovir is utilized for more severe cases or those with inadequate response to topical treatment, whilst intravenous ganciclovir is often designated for immunocompromised individuals or refractory conditions. Induction therapy succeeded by maintenance dose is essential for attaining viral suppression and averting relapse [54]. In cases of resistance or intolerance, other antivirals such as foscarnet have been tested; nevertheless, their ocular application is restricted and linked to toxicity issues [55].
The long-term prognosis for CMV keratitis is typically positive if the infection is detected and treated immediately. Nonetheless, recurrent illness continues to pose a significant issue, frequently requiring extended or repeated antiviral treatment regimens. In patients experiencing repeated relapses, long-term preventive antivirals may be contemplated, however the danger of resistance must be balanced against the advantages of viral suppression.
Future therapies may encompass immunomodulation aimed at reinstating protective CTL responses. Early clinical experience with adoptive virus-specific T-cells (VSTs) shows feasibility and safety for refractory CMV and could be adapted to anterior-segment disease [56]. Furthermore, next-generation antivirals with improved safety/resistance profiles-letermovir (terminase inhibitor) and maribavir (UL97 kinase inhibitor) are effective systemically and may help address recurrence and resistance [57,58]. Among emerging antiviral candidates, filociclovir represents a promising novel drug currently in early-phase clinical evaluation, although clinical trial data remain limited [59].
## HPV AND OSSN
OSSN includes a range of dysplastic and malignant lesions affecting the conjunctiva and cornea, from conjunctival intraepithelial neoplasia (CIN) to invasive squamous cell carcinoma [59]. The epidemiology exhibits significant geographic diversity, with the highest incidence observed in equatorial Africa, where ultraviolet exposure and HIV prevalence are key cofactors and the correlation between HPV and OSSN seems to differ by geographic area [60]. Recent investigations from South Africa demonstrate that HIV-positive individuals have an elevated likelihood of harboring HPV-positive OSSN, and that co-infection leads to more aggressive disease presentations [61].
The carcinogenic processes of HPV in ocular surface disease resemble those identified in cervical and oropharyngeal carcinogenesis. High-risk HPV genotypes, particularly HPV16, HPV18, and HPV33, integrate into host epithelial cells, where the viral oncoproteins E6 and E7 inactivate the tumor suppressors p53 and retinoblastoma, respectively. This disturbance results in unregulated cell cycle progression, genetic instability, and defective apoptosis, facilitating the malignant transformation of conjunctival epithelial cells [62,63]. HPV-induced OSSN typically exhibits a non-keratinizing phenotype, heightened p16 expression as an indicator of viral oncogenic activity, and active transcription of E6/E7 oncogenes [64,65]. Studies have substantiated the involvement of HPV in conjunctival neoplasia by establishing the presence of high-risk HPV DNA in OSSN lesions and detecting viral transcripts indicative of active viral oncogenesis [66].
Clinically, HPV-associated OSSN encompasses a broad range, from moderate dysplasia to aggressive squamous cancer. HPV positive is more commonly observed in CIN and non-keratinizing invasive lesions than in keratinizing malignancies [59][60][61][62][63][64]. Patients affected are typically younger, especially in HPV-positive instances, and may have more aggressive lesions along with an increased probability of recurrence [64,65].
The diagnosis of HPV-associated OSSN depends on a mix of clinical and analytical methods. In vivo confocal imaging facilitates non-invasive lesion assessment, demonstrating aberrant epithelial architecture and elevated nuclear-tocytoplasmic ratios [67]. Conclusive verification of HPV necessitates molecular tests. PCR is extensively employed for the detection of HPV DNA, whereas immunohistochemistry staining for p16 functions as a surrogate marker for viral carcinogenic activity [68]. Advanced techniques, such as RNA in situ hybridization for E6/E7 transcripts, demonstrate transcriptionally active infection and more precisely identify HPV-driven malignancies [64]. These diagnostic strategies resemble methods used in cervical and oropharyngeal cancer, highlighting the molecular parallels of HPV oncogenesis across mucosal epithelia.
The management of OSSN has significantly advanced, offering many surgical and medical alternatives. Surgical excision with extensive margins is fundamental, typically accompanied by cryotherapy to minimize recurrence risk [62]. Topical pharmacological therapy are increasingly preferred for diffuse, multifocal, or recurring conditions. Interferon alpha-2b, delivered topically or through subconjunctival injection, demonstrates great efficacy and tolerability, rendering it a compelling choice for long-term therapy [62,64]. Other frequently utilized topical treatments comprise mitomycin-C and 5-fluorouracil, which are efficacious but linked to ocular surface toxicity, hence reserved for more severe or refractory lesions [64]. Systemic therapy, including immune checkpoint inhibitors, has been investigated for advanced or metastatic cases, especially in individuals with concurrent HIV-related immunosuppression [69,70].
The significance of HPV vaccination in the prevention of OSSN is an increasingly pertinent topic. The prevalence of high-risk HPV genotypes 16, 18, and 33 in HPV-positive OSSN indicates that existing preventive vaccinations aimed at these genotypes may provide indirect ocular protection. Although direct clinical evidence of decreased OSSN incidence post-HPV immunization is currently lacking, preliminary genetic results offer a compelling basis for additional research. Enhancing immunization initiatives, especially in areas with elevated HIV prevalence and OSSN burden, may serve as an effective long-term approach to diminish the incidence of HPV-related ocular neoplasia.
A concise overview of the major viral ocular surface infections, including their clinical features, diagnostic approaches, management strategies, and prognostic considerations, is summarized in Table 1.
## FUTURE PROSPECTIVE
The future care of viral keratitis and conjunctivitis is increasingly influenced by advancements in precision medicine, encompassing medication research, immunization, artificial intelligence (AI), and global collaboration. The primary objective is to amalgamate molecular insights with clinical practice to enhance personalized therapy, mitigate visionthreatening consequences, and bolster public health readiness against viral ocular infections. A particularly promising area pertains to the advancement of tailored antiviral treatments and immunotherapies. Conventional antivirals like acyclovir and ganciclovir, although efficacious, are constrained by inadequate ocular bioavailability, difficulties in corneal penetration, and the development of resistance. Investigations into sustained-release delivery technologies, such as nanocarriers, liposomes, and in situ gelling formulations, have demonstrated considerable promise to extend corneal residence duration and attain therapeutic drug concentrations while minimizing systemic exposure [70]. In the case of adenoviral keratoconjunctivitis, for which no FDA-approved antiviral is available, innovative candidates including sialic acid analogs, cold atmospheric plasma, GAG mimetics and broader-spectrum antiviral agents are undergoing active preclinical and early clinical evaluation [32,40,42,44]. Immunomodulatory treatments, such as topical cyclosporine and tacrolimus, are being investigated as adjuncts to alleviate the inflammatory damage linked to viral keratitis; nevertheless, the data remains uncertain, and long-term outcomes necessitate more confirmation [38].
Vaccination techniques have significant potential in alleviating the global impact of viral ocular diseases. The recombinant zoster vaccination has shown exceptional effectiveness in preventing herpes zoster ophthalmicus, substantially reducing the risk of keratitis and related sequelae [6]. Currently, there is no licensed vaccination for HSV; nevertheless, recent preclinical studies indicate promising outcomes, with several options targeting viral glycoproteins and latencyassociated genes in development [27]. Ocular-specific vaccinations for adenovirus and HPV are currently nonexistent, however systemic HPV immunization offers indirect protection against high-risk oncogenic genotypes associated with OSSN. The prevalence of vaccine-targeted genotypes (HPV16 and HPV18) in OSSN reinforces the justification for possible ocular advantages [59]. Research on adenovirus vaccines is advancing, especially for respiratory and systemic diseases, with ongoing exploration for potential eye protective adaptations.
AI is becoming a vital instrument for diagnosis and outbreak monitoring. The clinical distinction between viral and bacterial conjunctivitis is notoriously difficult, resulting in numerous misdiagnoses and unwarranted antibiotic prescriptions. Deep learning algorithms utilizing multimodal ocular imaging and slit-lamp pictures demonstrate potential for enhancing diagnostic accuracy, while AI-augmented clinical decision-support systems may aid in patient triage and diminish unnecessary antibiotic prescriptions [71,72]. In addition to individual health care, AI is being incorporated into worldwide monitoring systems. Large-scale programs like the SCORPIO study integrate RNA deep sequencing with AI-driven analytics to monitor pathogen diversity, track viral outbreaks, and uncover new ocular pathogens in real time [73]. Such methodologies aim to bridge the divide between laboratory diagnosis and public health interventions.
Global collaboration networks constitute a significant frontier. Multicenter consortia provide swift data dissemination, standardization of diagnostic procedures, and synchronized therapy trials among varied patient demographics. The coronavirus disease 2019 pandemic highlighted the significance of these networks in real-time pathogen surveillance and clinical trial coordination, with analogous systems being applied for viral ocular illnesses. Collaborative registries and clinical trial networks will be crucial in validating innovative antivirals, standardizing treatment protocols, and improving readiness against re-emerging viral threats.
The future of managing viral ocular surface diseases is progressing towards precision medicine. Progress in sustainedrelease antivirals and innovative immunotherapies, along with the enhancement of immunization methods, offers potential for diminishing acute morbidity and long-term consequences. AI-driven diagnostics and outbreak surveillance are poised to transform the precision and promptness of diagnoses, while global networks will facilitate coordinated international responses. These advancements collectively present a vision for the therapy of viral keratitis and conjunctivitis that is increasingly customized, preventative, and internationally integrated.
Viral infections of the ocular surface continue to be a predominant cause of visual impairment globally, presenting significant clinical and public health challenges. Although herpesviruses, adenoviruses, CMV, and HPV have been acknowledged as significant infections for an extended period, emerging knowledge regarding their biology and clinical implications is continually transforming diagnostic and therapeutic strategies. Progress in molecular diagnostics has significantly enhanced the capacity to accurately and swiftly identify viral infections, minimizing wasteful treatments and facilitating targeted therapies. These advancements underscore the manner in which the incorporation of laboratory innovation into clinical practice can directly improve patient outcomes.
A consistent characteristic in all viral ocular infections is the interaction between viral persistence and the host immune response. Although immune activation is crucial for pathogen elimination, dysregulated inflammation frequently leads to corneal scarring, neovascularization, and chronic conditions. This fragile equilibrium highlights the necessity for medicines that both inhibit virus replication and regulate detrimental immune responses. The acknowledgment of immunopathology as a key factor influencing therapeutic outcomes has created new opportunities for the application of immunomodulatory medicines and preventive methods to avert recurrence. Future management will rely on novel drug delivery systems and next-generation antiviral agents. Conventional topical and systemic treatments are constrained by inadequate ocular absorption and the potential for resistance development. Sustained-release formulations, nanomedicine strategies, and innovative medicines aimed at viral entrance or replication are poised to revolutionize therapy paradigms. Vaccines offer potential for prophylaxis, having shown efficacy in herpes zoster and presenting increasing prospects in herpes simplex and HPV-related diseases. Immunization's capacity to diminish incidence and long-term consequences is among the most effective interventions for global eye health.
The wider context of viral ocular illness underscores the necessity of interdisciplinary collaboration and public health involvement. Emerging viral threats indicate that ocular indications may serve as early indicators of systemic outbreaks, highlighting the necessity for worldwide surveillance and prompt response mechanisms. AI and global networks will increasingly connect epidemiology, diagnostics, and medicinal research. Ultimately, continuous advancement will depend on the integration of virology, immunology, pharmacology, and clinical ophthalmology within a cohesive precision medicine framework.
## CONCLUSION
In summary, viral ocular surface diseases provide persistent difficulties while simultaneously offering unique potential. The integration of modern diagnostics, novel medicines, preventive immunization, and international collaboration provides a means to diminish blindness and enhance global quality of life. By adopting these tactics, the forthcoming phase of treatment for patients with viral keratitis and conjunctivitis will be more focused, preventative, and efficacious than ever before.
## FOOTNOTES
## References
1. Durand, Barshak, Sobrin (2023) "Eye Infections" *N Engl J Med*
2. Mccormick, James, Welton et al. (2022) "Incidence of Herpes Simplex Viral Keratitis and Other Ocular Diseases: A Global Overview and Estimates" *Ophthalmic Epidemiol*
3. Garcia-Zalisnak, Rapuano, Sheppard et al. (2018) "Adenovirus Ocular Infections: Prevalence, Pathology, Pitfalls, and Practical Pointers" *Eye Contact Lens*
4. Muto, Imaizumi, Kamoi (2023) "Viral Conjunctivitis" *Viruses*
5. Tsui, Sella, Tham et al. (2023) "Pathogen Surveillance for Acute Infectious Conjunctivitis" *JAMA Ophthalmol*
6. Lu, Sun, Porco et al. (2021) "Effectiveness of the Recombinant Zoster Vaccine for Herpes Zoster Ophthalmicus in the United States" *Ophthalmology*
7. Hwang, Wu, Kang et al. (2025) "Comprehensive insights into cytomegalovirus anterior segment infections: A narrative review" *Taiwan J Ophthalmol*
8. Li, Jiao, Li et al. (2025) "An Overview of Ophthalmic Complications Associated With Emerging/Re-Emerging Viruses: Focus on ZIKV, DENV, SARS-CoV-2, and MPXV" *Rev Med Virol*
9. Venkatesh, Patel, Goyal et al. "Ocular manifestations of emerging viral diseases" *Eye (Lond)*
10. Putera, Distia Nora, Dewi et al. (2025) "Antiviral therapy for cytomegalovirus retinitis: A systematic review and meta-analysis" *Surv Ophthalmol*
11. Wang, Kong, Wolle et al. (2023) "Global Trends in Blindness and Vision Impairment Resulting from Corneal Opacity 1984-2020: A Meta-analysis" *Ophthalmology*
12. Das, Souza, Gorimanipalli et al. (2022) "Ocular Surface Infection Mediated Molecular Stress Responses: A Review" *Int J Mol Sci*
13. Ahsanuddin, Wu (2023) "Single-cell transcriptomics of the ocular anterior segment: a comprehensive review" *Eye (Lond)*
14. De Paiva, Leger, Caspi (2022) "Mucosal immunology of the ocular surface" *Mucosal Immunol*
15. Sugita, Takase, Nakano (2023) "Role of Recent PCR Tests for Infectious Ocular Diseases: From Laboratory-Based Studies to the Clinic" *Int J Mol Sci*
16. Antony, Kinha, Nowińska et al. (2024) "The immunobiology of corneal HSV-1 infection and herpetic stromal keratitis" *Clin Microbiol Rev*
17. Grubešić, Jurak, Čaljkušić-Mance et al. (2024) "Clinical and Demographic Characteristics of Herpetic Keratitis Patients-Tertiary Centre Experience" *RCA*
18. Chodosh, Ung (2020) "Adoption of Innovation in Herpes Simplex Virus Keratitis" *Cornea*
19. Labetoulle, Boutolleau, Burrel et al. (2023) "Herpes simplex virus, varicella-zoster virus and cytomegalovirus keratitis: Facts for the clinician" *Ocul Surf*
20. Azher, Yin, Tajfirouz et al. (2017) "Herpes simplex keratitis: challenges in diagnosis and clinical management" *Clin Ophthalmol*
21. Valerio, Lin (2019) "Ocular manifestations of herpes simplex virus" *Curr Opin Ophthalmol*
22. Pan, Wang, Xu et al. (2024) "Application of Metagenomic Next-Generation Sequencing in the Diagnosis of Infectious Keratitis" *J Ophthalmol*
23. Wilhelmus (2015) "Antiviral treatment and other therapeutic interventions for herpes simplex virus epithelial keratitis" *Cochrane Database Syst Rev*
24. Sibley, Larkin (2020) "Update on Herpes simplex keratitis management" *Eye (Lond)*
25. Labib, Chigbu (2022) "Clinical Management of Herpes Simplex Virus Keratitis" *Diagnostics (Basel)*
26. Lince, Demario, Yang et al. (2023) "A Systematic Review of Second-Line Treatments in Antiviral Resistant Strains of HSV-1, HSV-2, and VZV" *Cureus*
27. Krishnan, Stuart (2021) "Developments in Vaccination for Herpes Simplex Virus" *Front Microbiol*
28. Roszkowska, Spinella, Calderone et al. (2024) "The use of rh-NGF in the management of neurotrophic keratopathy" *Front Ophthalmol (Lausanne)*
29. Rajasagi, Rouse (2019) "The Role of T Cells in Herpes Stromal Keratitis" *Front Immunol*
30. Hamrah, Cruzat, Dastjerdi et al. (2010) "Corneal sensation and subbasal nerve alterations in patients with herpes simplex keratitis: an in vivo confocal microscopy study" *Ophthalmology*
31. Azari, Barney (2013) "Conjunctivitis: a systematic review of diagnosis and treatment" *JAMA*
32. Labib, Minhas, Chigbu (2020) "Management of Adenoviral Keratoconjunctivitis: Challenges and Solutions" *Clin Ophthalmol*
33. Jhanji, Chan, Li et al. (2015) "Adenoviral keratoconjunctivitis" *Surv Ophthalmol*
34. Zhang, Zhao, Sha et al. (2016) "Virology and epidemiology analyses of global adenovirus-associated conjunctivitis outbreaks, 1953-2013" *Epidemiol Infect*
35. Chandra, Frängsmyr, Imhof et al. (2019) "Sialic Acid-Containing Glycans as Cellular Receptors for Ocular Human Adenoviruses: Implications for Tropism and Treatment" *Viruses*
36. Storm, Persson, Skalman et al. (2017) "Human Adenovirus Type 37 Uses α(V)β(1) and α(3)β(1) Integrins for Infection of Human Corneal Cells" *J Virol*
37. Sambursky, Trattler, Tauber et al. (2013) "Sensitivity and specificity of the AdenoPlus test for diagnosing adenoviral conjunctivitis" *JAMA Ophthalmol*
38. Liu, Hawkins, Ng et al. (2022) "Topical pharmacologic interventions versus placebo for epidemic keratoconjunctivitis" *Cochrane Database Syst Rev*
39. Than, Morettin, Harthan et al. (2021) "Efficacy of a Single Administration of 5% Povidone-Iodine in the Treatment of Adenoviral Conjunctivitis" *Am J Ophthalmol*
40. Imparato, Rosa, Bernardo (2014) "Antiviral Drugs in Adenovirus-Induced Keratoconjunctivitis" *Microorganisms*
41. Skevaki, Galani, Pararas et al. (2011) "Treatment of viral conjunctivitis with antiviral drugs" *Drugs*
42. Johansson, Nilsson, Elofsson et al. (2007) "Multivalent sialic acid conjugates inhibit adenovirus type 37 from binding to and infecting human corneal epithelial cells" *Antiviral Res*
43. Chandra, Frängsmyr, Arnberg (2019) "Decoy Receptor Interactions as Novel Drug Targets against EKC-Causing Human Adenovirus" *Viruses*
44. Macneil, Dodge, Evans et al. (2023) "Adenoviruses in medicine: innocuous pathogen, predator, or partner" *Trends Mol Med*
45. Didara, Reithofer, Zöttl et al. (2023) "Inhibition of adenovirus replication by CRISPR-Cas9-mediated targeting of the viral E1A gene" *Mol Ther Nucleic Acids*
46. Gonzales (2018) "Ocular manifestations of cytomegalovirus in immunocompetent hosts" *Curr Opin Ophthalmol*
47. Tan, Tan (2019) "Cytomegalovirus Corneal Endotheliitis After Descemet Membrane Endothelial Keratoplasty" *Cornea*
48. Kobayashi, Hashida (2024) "Overview of Cytomegalovirus Ocular Diseases: Retinitis, Corneal Endotheliitis, and Iridocyclitis" *Viruses*
49. Koizumi, Suzuki, Uno et al. (2008) "Cytomegalovirus as an etiologic factor in corneal endotheliitis" *Ophthalmology*
50. Uotani, Miyazaki, Shimizu et al. (2022) "Antiviral cytotoxic T lymphocyte responses for long term prognosis of corneal infection by cytomegalovirus in immunocompetent subjects" *Sci Rep*
51. Carmichael (2012) "Cytomegalovirus and the eye" *Eye (Lond)*
52. Kandori, Inoue, Takamatsu et al. (2010) "Prevalence and features of keratitis with quantitative polymerase chain reaction positive for cytomegalovirus" *Ophthalmology*
53. Wu, Jiang, Zhang et al. (2022) "Clinical Metagenomic Next-Generation Sequencing for Diagnosis of Secondary Glaucoma in Patients With Cytomegalovirus-Induced Corneal Endotheliitis" *Front Microbiol*
54. Distia, Putera, Mayasari et al. (2022) "Clinical characteristics and treatment outcomes of cytomegalovirus anterior uveitis and endotheliitis: A systematic review and meta-analysis" *Surv Ophthalmol*
55. Vassallo, Nuzzi, Cattani et al. (2020) "CMV retinitis in a stem cell transplant recipient treated with foscarnet intravitreal injection and CMV specific immunoglobulins" *Ther Adv Hematol*
56. Bandeira, Marti, Rother et al. (2024) "Use of Specific T Lymphocytes in Treating Cytomegalovirus Infection in Hematopoietic Cell Transplant Recipients: A Systematic Review" *Pharmaceutics*
57. Marty, Ljungman, Chemaly et al. (2017) "Letermovir Prophylaxis for Cytomegalovirus in Hematopoietic-Cell Transplantation" *N Engl J Med*
58. Avery, Alexander, Blumberg et al. (2022) "SOLSTICE Trial Investigators. Maribavir for Refractory Cytomegalovirus Infections With or Without Resistance Post-Transplant: Results From a Phase 3 Randomized Clinical Trial" *Clin Infect Dis*
59. Ramberg, Møller-Hansen, Toft et al. (2021) "Human papillomavirus infection plays a role in conjunctival squamous cell carcinoma: a systematic review and meta-analysis of observational studies" *Acta Ophthalmol*
60. Hall, Heal (2025) "A systematic review and meta-analysis of the association of human papilloma virus infections with ocular surface squamous neoplasia" *Cancer Epidemiol*
61. Odendaal, Andreae, Sanderson-November et al. (2024) "The prevalence of human papillomavirus in ocular surface squamous neoplasia in HIV positive and negative patients in a South African population" *Infection*
62. Gichuhi, Ohnuma, Sagoo et al. (2014) "Pathophysiology of ocular surface squamous neoplasia" *Exp Eye Res*
63. Mclaughlin-Drubin, Münger (2009) "Oncogenic activities of human papillomaviruses" *Virus Res*
64. Ramberg (2022) "Human papillomavirus-related neoplasia of the ocular adnexa" *Acta Ophthalmol*
65. Moyer, Roberts, Olsen et al. (2018) "Human Papillomavirus-Driven Squamous Lesions: High-Risk Genotype Found in Conjunctival Papillomas, Dysplasia, and Carcinoma" *Am J Dermatopathol*
66. Hanbazazh, Gyure (2018) "Ocular Human Papillomavirus Infections" *Arch Pathol Lab Med*
67. Parrozzani, Lazzarini, Midena (2011) "In vivo confocal microscopy of ocular surface squamous neoplasia" *Eye (Lond)*
68. Woods, Chow, Heng et al. (2013) "Detecting human papillomavirus in ocular surface diseases" *Invest Ophthalmol Vis Sci*
69. Li, Najdawi, Badla et al. (2025) "Immune Checkpoint Inhibitors in the Treatment of Ocular Surface Cancers: A Review" *Semin Ophthalmol*
70. Pandey, Choudhury, Abdul-Aziz et al. (2020) "Advancement on Sustained Antiviral Ocular Drug Delivery for Herpes Simplex Virus Keratitis: Recent Update on Potential Investigation" *Pharmaceutics*
71. Zhang, Wang, Wang et al. (2022) "Deep learning-based classification of infectious keratitis on slit-lamp images" *Ther Adv Chronic Dis*
72. Nusair, Asadigandomani, Farrokhpour et al. (2025) "Clinical Applications of Artificial Intelligence in Corneal Diseases" *Vision (Basel)*
73. Tran, Hoang, Tran et al. (2023) "Pathogen Profiles of Infectious Conjunctivitis in Ho Chi Minh City" *Cornea Open*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12853410&blobtype=pdf
|
# Cardio-Renal Diseases Are Independent Risk Factors of Severe Human Metapneumovirus Infection Among Patients Without Chronic Airway Diseases
Chun Wang, Kwok, Isaac Sze, Him Leung, Chun Ka, Emmanuel Wong, James Chung, Man Ho, | David, Chi Leung, Mary Sau, Man Ip, | Shuk, Man Ngai, Kelvin Kai, Wang To, | Desmond, Yat Yap, Desmond Yat, Hin Yap
## Abstract
Human metapneumovirus (hMPV) causes mild and self-limiting disease in adults. However, the risk factors for serious adverse outcomes following hMPV infection in adult patients without preexisting chronic airway diseases remain poorly understood. We conducted a territory-wide retrospective study on adult patients (aged ≥ 18 years) without chronic airway diseases hospitalized for hMPV infections between January 1, 2016 and June 30, 2023 in Hong Kong. We assessed the incidence and risk factors for in-patient mortality, severe respiratory failure (SRF), secondary bacterial pneumonia and acute kidney injury (AKI) were assessed. A total of 1552 eligible adult patients without chronic airway diseases hospitalized for hMPV infections were analyzed. Within the index admission, 92 (5.9%) patients died. Ischemic heart disease (IHD) was associated with increased risks of SRF [adjusted odds ratio (aOR) 2.00 (95% CI 1.48-2.71), p < 0.001]. IHD, heart failure (HF), and history of ischemic stroke were significant predictors for AKI [aOR 1.51 (95% CI 1.12-2.04), 2.87 (95% CI 2.14-3.85), and 1.47 (95% CI = 1.12-1.93), p = 0.007, < 0.001, and 0.005, respectively). Patients with end-stage kidney disease (ESKD) requiring renal replacement therapy (RRT) were at increased risk of in-patient mortality , p < 0.001] and SRF ), p < 0.001]. The presence of cardiovascular diseases and ESKD requiring RRT is a strong predictor of severe inhospital outcomes among adult patients without chronic airway diseases who are hospitalized for hMPV infections.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
## 1 | Introduction
Human metapneumovirus (hМРV) was first discovered in 2001 as a viral cause of respiratory tract infections [1]. hΜРV is a member of the family Pneumoviridae, which comprises large enveloped negative-sense RNA viruses [2]. hMPV is an enveloped virus with a nonsegmented negative-sense RNA genome.
The surface glycoproteins of hMPV are crucial for infection. The fusion (F) protein mediates viral entry by promoting membrane fusion, while the G glycoprotein is responsible for initial attachment to host cell receptors. The F protein facilitates the merging of viral and cellular membranes, enabling genome entry, while the G protein helps the virus bind to host cells, aiding infection initiation. Together, these glycoproteins are vital for viral attachment, entry, and subsequent replication within host tissues [3].
hΜPV can cause upper and lower respiratory tract infection in patients of all age groups, but symptomatic disease mostly occurs in young children or older adults [4]. In adult patients, hMPV was detected in about 4% of hospitalized adults with community-acquired pneumоnia [5]. The clinical manifestations of hMPV in adults include cough, nasal congestion, rhinorrhoea, dyspnea, hoarseness, and wheezing [6]. hMPV has been demonstrated to cause lower respiratory tract infections in hospitalized adult patients, especially elderly people with preexisting medical comorbidities, such as chronic respiratory diseases and hypertension [7]. In one study, 30% of the patients had bacterial coinfection while the need for mechanical ventilation and/or the hospital death was observed in almost 20% of the patients [7]. Another French study also suggested that the elderly and patients with chronic conditions were mostly affected by hMPV infections and were responsible for frequent cardiac and pulmonary complications [8]. Of note, hMPV was reported to cause exacerbations of chronic obstructive pulmonary disease (COPD) [9][10][11] and asthma [12][13][14].
Currently, the treatment of hMPV is mainly supportive with very limited evidence for ribavirin treatment based on in vitro studies [15]. In recent years, breakthroughs in identifying the structure of the viral fusion (F) protein may bring about advancements in vaccine development and therapeutics, which could serve as a target for future vaccines and drugs [16][17][18][19][20][21][22][23]. While hMPV infections in adults are mostly mild and selflimiting [24], it is crucial to identify the target patient groups for consideration of future vaccination and treatment once they become available, as massive vaccination and liberal use of novel treatment such as neutralizing antibodies [21] will incur a significant burden on healthcare resources and costs.
## 2 | Methods
This was a territory-wide retrospective study to examine the risk factors for mortality and serious clinical outcomes in adult patients hospitalized for hMPV infections among patients who did not have underlying chronic airway diseases (asthma, COPD, asthma/COPD overlap (ACO) and bronchiectasis). Patients with chronic airway diseases were excluded, as the primary outcomes of the study could be related to hMPV infections as well as the underlying chronic airway diseases. Adult patients admitted to public hospitals in Hong Kong for hMPV infection between January 1, 2016 and June 30, 2023 were included. Patients with coexisting respiratory tract viral infections were excluded. Patients were identified from the Clinical Data Analysis and Reporting System (CDARS) of Hospital Authority (HA) by the principal diagnosis with the International Classification of Diseases, Ninth Revision (ICD-9) codes 0789.89 or 466.19. CDARS is an electronic health record database managed by the HA, a public healthcare service provider that covered > 90% of the Hong Kong population since 1993 [25][26][27]. The study was approved by the Institutional Review Board (IRB) of the University of Hong Kong and the HA Hong Kong West Cluster (UW 24-137). Patient informed consent was waived for this retrospective study by the IRB, as it involved no active patient recruitment and all the data were de-identified. The study was conducted in compliance with the Declaration of Helsinki.
## 2.1 | Outcome Measurements and Statistical Analysis
The co-primary outcomes included: (1) death during hospitalization, (2) severe respiratory failure (SRF) requiring invasive or noninvasive mechanical ventilation, (3) secondary bacterial pneumonia, and (4) acute kidney injury (AKI). AKI was defined according to the RIFLE criteria [28]. Secondary bacterial pneumonia was identified by ICD-9 diagnostic code of 481 or 482, or by the growth of pathogenic organisms in respiratory specimens during the same admission episode. The following covariates were assessed as potential risk factors associated with the outcomes: age and Charlson Comorbidity Index (CCI) as continuous variables; gender; history of malignancy; underlying diabetes mellitus (DM), ischemic heart disease (IHD), ischemic stroke, heart failure (HF), and underlying kidney diseases as categorical variables. For underlying kidney diseases, patients were further subclassified into those with end-stage kidney disease (ESKD) requiring renal replacement therapy (RRT) and those with CKD [defined as estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m 2 ] [29]; baseline eGFR was also assessed as a continuous variable.
Descriptive tables and illustrative figures were created to present the incidence rates of severe in-hospital outcomes. Demographic and clinical data were described as actual frequency or mean ± standard deviation (SD), or median [interquartile range] as appropriate. Continuous variables were compared by independent t-test or Mann-Whitney U tests, as appropriate.
Risk factors for adverse clinical outcomes in patients hospitalized for hMPV infections were first assessed by univariate then multivariable analyses. Factors with a p < 0.1 in the univariate analysis were included in the multivariable model using backward selection.
Data analyses were performed using the 28th version of the SPSS statistical package. For all statistical analyses, statistical significance was assessed at an α level of 0.05. This report followed the STROBE and RECORD reporting guidelines.
## 3 | Results
## 3.1 | Patients' Characteristics
A total of 1931 adult patients were hospitalized for hMPV infections in public hospitals in Hong Kong between January 1, 2016 and June 30, 2023. Of these, 379 patients with asthma, COPD, ACO or bronchiectasis were excluded (Figure 1). Consequently, 1552 patients were included in the final analysis. Among the 1552 included patients, 884 (57.0%) were female, with a mean age of 73.4 ± 18.2 years. The mean CCI was 5.08 ± 2.71. Patients' characteristics are summarized in Table 1. In total, 660 (42.5%) of the patients developed AKI. The risk of AKI was increased in female patients [aOR 1.33 (95% CI 1.07-1.65), p = 0.010], patients age ≥ 65 [aOR 1.51 (95% CI 1.17-1.95), p < 0.001], patients with DM [aOR 1.43 (95% CI 1.13-1.82), p = 0.003], patients with IHD [aOR 1.51 (95% CI 1.12-2.04), p = 0.007], patients with HF [aOR 2.87 (95% CI 2.14-3.85), p < 0.001], and patients with a history of stroke [aOR 1.47 (95% CI 1.12-1.93), p = 0.005] (Figure 4).
## 3.2 | Risk Factors for Severe In-Hospital Outcomes
## 3.3 | Subgroup Analysis
As older age is a risk factor for severe hMPV infections, a subgroup analysis was performed for patients aged < 65 (n = 410) and ≥ 65 (n = 1142). Among patients aged < 65 years of age, 9 (2.2%) patients died in the index admission, 111 (27.1%) developed SRF, 253 (61.7%) developed secondary bacterial pneumonia, and 121 (29.5%) developed AKI.
Among patients aged ≥ 65, 83 (7.3%) patients died in the index admission, 238 (20.8%) developed SRF, 805 (70.5%) had S1.
## 4 | Discussion
In this study, we demonstrated that medical comorbidities, especially cardiovascular and renal diseases, are risk factors for adults without chornic airway diseases hospitalized for severe hMPV infections. These findings provide insight for potential patient selection for future hMPV vaccination and treatment when these agents soon become available.
In this study, we demonstrated that adverse in-hospital outcomes were common among adults hospitalized for hMPV infections. While hMPV infection in adults is often considered mild and self-limiting, this may not apply to older adults with more medical comorbidities who require hospitalization, even in the absence of chronic airway diseases. Various cardiovascular and renal diseases were associated with different adverse outcomes, which aligns with literature suggesting chronic medical conditions as risk factors for severe hMPV infection [7].
Our study findings presented here illustrated this phenomenon better by segregating these chronic medical diseases into specific diseases, which allowed a better understanding on the risk stratification. In this context, ESKD requiring RRT emerged as a strong risk factor for in-patient mortality in adults hospitalized for hMPV infections. Patients with ESKD requiring RRT were demonstrated to have an increased mortality during index admission and SRF. It is well recognized that patients with ESKD are immunocompromised, rendering them highly susceptible to serious infections and complications [30]. A similar phenomenon was observed in the coronavirus disease in 2019 (COVID-19) [31]. A Japanese study suggested that the risk of death and renal function decline was increased among ESKD patients requiring RRT who had COVID-19, but not in nondialysis-dependent CKD patients. The risk of severe outcomes from viral infections such as hMPV in RRT-dependent patients should not be overlooked.
hMPV is a respiratory virus that was identified relatively recently, though scientists believe that it has been causing respiratory tract infections for decades [32,33]. Since its discovery in 2001, substantial progress has been made in vaccination and therapeutic research. However, there is a need for case identification in order to design appropriate vaccination and treatment strategies, given the fact that novel vaccines and therapeutic agents can be costly. Our findings could help identify risk groups that are more likely to benefit from receiving vaccination and viral-specific treatment.
Underlying cardiovascular diseases are common across ethnic groups and have been suggested to be associated with increased risks of complications in patients having various viral respiratory tract infections [34][35][36]. This phenomenon is also observed among adult patients hospitalized for hMPV infections. Another high-risk subgroup that was identified to be associated with severe hMPV infections is those who had ESKD requiring RRT.
Studies on other viral infections, such as COVID-19, suggest that these patients have prolonged viral shedding [37] and more severe disease [38]. Immune dysfunction also predisposes patients with CKD to develop severe infections [39]. However, the number of EKSD patients requiring RRT in our cohort is relatively small. Despite a significant result, this finding should be interpreted with caution, and a validation in a larger cohort is warranted.
The high rate of severe in-hospital outcomes concurs with previous reports that elderly patients and those with medical comorbidities are at risk of severe disease, including bacterial coinfection and respiratory failure. hMPV should not be considered as a mild, self-limiting infection among hospitalized patients. These patients are at risk of developing various severe inhospital outcomes, including mortality. Close monitoring of respiratory status and early initiation of antibiotics are warranted, as secondary bacterial infection is common. Apart from respiratory complications, we also overserved a high incidence of AKI, similar to that seen in COVID-19 and influenza infections [40]. This highlights the importance of monitoring other organ functions in patients hospitalized for hMPV, especially elderly patients with medical comorbidities, who may develop AKI due to sepsis, dehydration from anorexia, and also drug-induced AKI.
Our study has several limitations. First, it was conducted in Hong Kong, the majority of the patients are Chinese, and the generalizability to other populations remains to be determined. Some demographic data, such as ethnicity, were missing and were handled by multiple imputation. We also performed subgroup analyses to ensure our data is robust across different age groups. Second, we did not analyze the granular details of secondary bacterial pneumonia. Third, only hospitalized adults were included; non-hospitalized subjects were not analyzed, which may limit the assessment of hMPV infection as a whole and likely elevates the observed complication rate. Nevertheless, it is important to analyze those hospitalized subjects incur much more healthcare burden than nonhospitalized subjects, warranting dedicated assessment. While these inherent limitations may affect our results, it is important to appreciate that our data are derived from a territory-wide electronic health record system that captures comprehensive clinical information for all adults hospitalized for adenoviruses or seasonal influenza infection during the study period, providing a good representation of real-world data.
In summary, our study presented the incidence and risk factors for severe hMPV infections in adult patients hospitalized without chronic airway diseases. The presence of cardiovascular and renal comorbidities, especially among elderly patients, should warrant timely monitoring and management of severe in-hospital outcomes.
## 5 | Conclusions
The presence of cardiovascular diseases and ESKD requiring RRT was associated with an increased risk of severe in-hospital outcomes among adult patients without chronic airway diseases who are hospitalized for hMPV infections.
## References
1. Van Den Hoogen, De Jong, Groen (2001) "A Newly Discovered Human Pneumovirus Isolated From Young Children With Respiratory Tract Disease" *Nature Medicine*
2. Schuster, Williams (2014) "Human Metapneumovirus" *Microbiology Spectrum*
3. Gao, Lin, Ma (2025) "Human Metapneumovirus: Pathogenesis, Epidemiology, Diagnostic Technologies, and Potential Intervention Strategies" *Virology Journal*
4. Boivin, Abed, Pelletier (2002) "Virological Features and Clinical Manifestations Associated With Human Metapneumovirus: A New Paramyxovirus Responsible for Acute Respiratory-Tract Infections in All Age Groups" *Journal of Infectious Diseases*
5. Howard, Edwards, Zhu (2021) "Clinical Features of Human Metapneumovirus-Associated Community-Acquired Pneumonia Hospitalizations" *Clinical Infectious Diseases*
6. Falsey, Erdman, Anderson et al. (2003) "Human Metapneumovirus Infections in Young and Elderly Adults" *Journal of Infectious Diseases*
7. Philippot, Rammaert, Dauriat (2024) "Human Metapneumovirus Infection Is Associated With a Substantial Morbidity and Mortality Burden in Adult Inpatients" *Heliyon*
8. Loubet, Mathieu, Lenzi (2021) "Characteristics of Human Metapneumovirus Infection in Adults Hospitalized for Community-Acquired Influenza-Like Illness in France, 2012-2018: A Retrospective Observational Study" *Clinical Microbiology and Infection*
9. Vicente, Montes, Cilla et al. (2004) "Human Metapneumovirus and Chronic Obstructive Pulmonary Disease" *Emerging Infectious Diseases*
10. Hamelin, Cotu, Laforge (2005) "Human Metapneumovirus Infection in Adults With Community-Acquired Pneumonia and Exacerbation of Chronic Obstructive Pulmonary Disease" *Clinical Infectious Diseases*
11. Martinello, Esper, Weibel et al. (2006) "Human Metapneumovirus and Exacerbations of Chronic Obstructive Pulmonary Disease" *Journal of Infection*
12. Williams, Crowe, Enriquez (2005) "Human Metapneumovirus Infection Plays an Etiologic Role in Acute Asthma Exacerbations Requiring Hospitalization in Adults" *Journal of Infectious Diseases*
13. Bakakos, Sotiropoulou, Vontetsianos et al. (2023) "Epidemiology and Immunopathogenesis of Virus Associated Asthma Exacerbations" *Journal of Asthma and Allergy*
14. Rudd, Thomas, Zaid (2017) "Role of Human Metapneumovirus and Respiratory Syncytial Virus in Asthma Exacerbations: Where Are We Now?" *Clinical Science*
15. Wyde, Chetty, Jewell et al. (2003) "Comparison of the Inhibition of Human Metapneumovirus and Respiratory Syncytial Virus by Ribavirin and Immune Serum Globulin In Vitro" *Antiviral Research*
16. Hsieh, Rush, Palomo (2022) "Structure-Based Design of Prefusion-Stabilized Human Metapneumovirus Fusion Proteins" *Nature Communications*
17. Phan, Ren, Bao (2015) "Recent Vaccine Development for Human Metapneumovirus" *Journal of General Virology*
18. Palavecino, Cespedes, Lay et al. (2015) "Understanding Lung Immunopathology Caused by the Human Metapneumovirus: Implications for Rational Vaccine Design" *Critical Reviews in Immunology*
19. Whitaker, Sahly, Healy (2023) "mRNA Vaccines Against Respiratory Viruses" *Current Opinion in Infectious Diseases*
20. Ogonczyk-Makowska, Brun, Vacher (2024) "Mucosal Bivalent Live Attenuated Vaccine Protects Against Human Metapneumovirus and Respiratory Syncytial Virus in Mice" *NPJ Vaccines*
21. Guo, Li, Liu et al. (2023) "Neutralising Antibodies Against Human Metapneumovirus" *Lancet Microbe*
22. Van Den, Bergh, Bailly et al. (2022) "Antiviral Strategies Against Human Metapneumovirus: Targeting the Fusion Protein" *Antiviral Research*
23. Yim, Mousa, Blanco et al. (2023) "Human Metapneumovirus (hMPV) Infection and MPV467 Treatment in Immunocompromised Cotton Rats Sigmodon hispidus" *Viruses*
24. Haas, Thijsen, Van Elden et al. (2013) "Human Metapneumovirus in Adults" *Viruses*
25. Kwok, Tam, Sing et al. (2023) "Validation of Diagnostic Coding for Bronchiectasis in an Electronic Health Record System in Hong Kong" *Pharmacoepidemiology and Drug Safety*
26. Kwok, Tam, Sing et al. (2023) "Validation of Diagnostic Coding for Asthma in an Electronic Health Record System in Hong Kong" *Journal of Asthma and Allergy*
27. Gao, Leung, Li (2021) "Linking Cohort-Based Data With Electronic Health Records: A Proof-of-Concept Methodological Study in Hong Kong" *BMJ Open*
28. Bellomo, Ronco, Kellum et al. (2004) "Acute Renal Failure -Definition, Outcome Measures"
29. Levey, Eckardt, Tsukamoto (2005) "Definition and Classification of Chronic Kidney Disease: A Position Statement From Kidney Disease: Improving Global Outcomes (KDIGO)" *Kidney International*
30. Pappas, Mpournaka, Katopodis (2019) "The Effect of Dialysis Modality and Membrane Performance on Native Immunity in Dialysis Patients"
31. Ikenouchi, Takahashi, Mandai (2024) "Impact of COVID-19 Versus Other Pneumonia on In-Hospital Mortality and Functional Decline Among Japanese Dialysis Patients: A Retrospective Cohort Study" *Scientific Reports*
32. Peret, Boivin, Li (2002) "Characterization of Human Metapneumoviruses Isolated From Patients in North America" *Journal of Infectious Diseases*
33. Nissen, Mackay, Withers et al. (2002) "Evidence of Human Metapneumovirus in Australian Children" *Medical Journal of Australia*
34. Nguyen, Yang, Ito et al. (2016) "Seasonal Influenza Infections and Cardiovascular Disease Mortality" *JAMA Cardiology*
35. Ielapi, Licastro, Provenzano et al. (2020) "Cardiovascular Disease as a Biomarker for an Increased Risk of COVID-19 Infection and Related Poor Prognosis" *Biomarkers in Medicine*
36. Müller-Wieland, Marx, Dreher et al. (2022) "COVID-19 and Cardiovascular Comorbidities" *Experimental and Clinical Endocrinology & Diabetes*
37. Bardossy, Korhonen, Schatzman (2021) "Clinical Course of SARS-CoV-2 Infection in Adults With ESKD Receiving Outpatient Hemodialysis" *Kidney*
38. Valeri, Robbins-Juarez, Stevens (2020) "Presentation and Outcomes of Patients With ESKD and COVID-19" *Journal of the American Society of Nephrology*
39. Syed-Ahmed, Narayanan (2019) "Immune Dysfunction and Risk of Infection in Chronic Kidney Disease" *Advances in Chronic Kidney Disease*
40. Bhasin, Veitla, Dawson (2021) "AKI in Hospitalized Patients With COVID-19 and Seasonal Influenza: A Comparative Analysis" *Kidney*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12719622&blobtype=pdf
|
# Global Trends in MPOX Research (2014-2025): A Bibliometric Analysis and Overview
Lulu Xie, Yimin Zhou, Xinyu Zhang, Tian Lan, Wenchao Sun
## Abstract
The 2022 global monkeypox outbreak fuelled a surge in related research. In this study, a bibliometric analysis of 2,076 monkeypox-related articles indexed in the Web of Science Core Collection (from January 2014 to June 2025) was performed, and VOSviewer was used to visualize keyword co-occurrence networks and collaboration patterns among countries and institutions. Unlike existing studies, the latest data for 2024-2025 are incorporated; postoutbreak research trends, collaborative patterns, and PHEICdriven resource aggregation effects are systematically incorporated; and gaps in dynamic and multifaceted analyses of the field are addressed. The results revealed a three-stage global research output: low accumulation (2014-2021), a sharp surge after 2022, and sustained high levels after a peak in 2023. The core research clusters focused on viral transmission dynamics and clinical interventions.Country contributions showed a pyramidal hierarchy, with the US leading, followed by China, India, and Saudi Arabia; government health agencies and academic institutions codominated collaborations. The findings confirm the role of the PHEIC in concentrating scientific resources, emphasizing the importance of strengthening international collaboration to address future epidemic challenges.
## Introduction
Monkeypox (MPOX) infection, caused by the monkeypox virus (MPXV), is a zoonotic disease. 1 MPXV was first isolated in 1958 from crab-eating macaques in a research facility in Denmark. 2 The first human MPOX case was diagnosed in 1970 in a 9-month-old boy in the Democratic Republic of the Congo 3 -a country recognized as one of the core endemic regions for MPXV, alongside other Central and West African nations. Since 1970, human MPX cases have been formally documented in 11 countries in Africa, specifically Benin, Cameroon, the Central African Republic, Cote d' lvoire, the DRC, Gabon, Liberia, Nigeria, the Republic of the Congo, Sierra Leone and South Sudan. 4 Historically, the majority of reported human MPOX cases occurred in these African regions, with the first outbreak outside Africa emerging in the United States in 2003. 5 Following the reporting of the first MPOX case in the UK in May 2022, the MPOX outbreak evolved into a widespread global epidemic across multiple regions. 6 Owing to its rapid geographic expansion and increasing case numbers, the World Health Organization (WHO) declared MPOX a public health emergency of international concern (PHEIC) on 23 July 2022. 7 As of 31 July 2024, global surveillance systems have documented 102,997 human MPXV infections and 223 associated fatalities since the first human case was identified in 1970, reflecting the persistent public health burden posed by this pathogen over five decades of continuous circulation. 8 Notably, MPXV exhibits distinct genetic divergence, with two discrete clades responsible for driving outbreaks across Africa: the Central African (Congo Basin) clade and the West African clade. The Central African clade is associated with heightened virulence, featuring a reported case fatality rate (CFR) of 10%-11%, whereas the West African clade is less pathogenic, with a CFR ranging from 1% to 4%. 4 This clade-specific variation in pathogenicity has substantial implications for outbreak response and clinical management strategies. Although existing studies have advanced our understanding of the biological properties, epidemiological dynamics, and clinical care protocols of MPXV, few bibliometric investigations have focused specifically on MPOX. That said, bibliometric methodologies have been employed by researchers to analyse research progress in the field of analogous zoonotic viruses-most prominently COVID-19-wherein publication trends, interinstitutional collaboration networks, and emerging research hotspots have been explored.
MPXV is a member of the genus Orthopoxvirus within the family Poxviridae. The family Poxviridae comprises two subfamilies, the Entomopoxvirinae and the Chordopoxvirinae. 9 Within Chordopoxvirinae, viruses from four genera are known to cause human disease. Among these, the genera Orthopoxvirus, Parapoxvirus, and Yatapoxvirus have zoonotic potential. 10 The genus Orthopoxvirus notably includes four viruses that are pathogenic to humans, namely, MPXV, the variola virus (VARV, a causative agent of smallpox), the vaccinia virus (VACV, used historically in the smallpox vaccine), and the cowpox virus (CPXV). While generally rare, MPOX is a potentially severe disease. 11 MPXV infection shares similarities with smallpox infection in terms of clinical presentation (including the characteristic rash), lesion distribution, and disease progression. 12 These challenges are further exacerbated by restricted access to advanced diagnostic tools in endemic regions, nonspecific early manifestations-including fever and lymphadenopathy-that mimic other febrile conditions such as chickenpox and malaria, and inadequate laboratory capacity for confirmatory testing (notably polymerase chain reaction) in resource-limited settings. 4 Such constraints impede the timely identification of cases and compromise effective outbreak control. Disease severity is correlated with patient age and comorbidities, with reported case fatality rates reaching 11% and the highest risk observed in young children. 4,11 Currently, there is no specific vaccine approved solely for MPXV. 13,14 Consequently, patients infected with MPXV are treated with therapeutics originally developed for smallpox, although the efficacy of historically used smallpox vaccines against MPXV is limited. 15 The sustained global spread of monkeypox has led to substantial public health concerns and has spurred intensive research efforts worldwide, necessitating the synthesis of the burgeoning body of MPOX-related literature. Notably, bibliometric visualization tools-including VOSviewer-play a pivotal role in guiding future research directions: By converting fragmented literature data into intuitive networks, such as keyword co-occurrence maps or institutional collaboration diagrams, these tools empower researchers to rapidly identify research hotspots, uncover collaborative gaps, and prioritize understudied areas-thereby bridging the gap between raw literature data and evidence-based research planning. Integrating insights related to such tools into the framework of the current study serves to more closely align the research background with its methodology, providing a robust rationale for the adoption of visualization-driven bibliometric analysis.
This study conducted a comprehensive analysis of monkeypox-related research literature published between January 2014 and June 2025. To ensure clarity, the key objectives of the study are explicitly outlined as follows: 1) Analysing publication trends of MPOX-related literature over the study period, including annual publication volume and core source journals; 2) Exploring collaboration patterns among countries, institutions, and authors in the MPOX research field; and 3) Identifying keyword clusters to reveal research hotspots and evolutionary trends in MPOX studies.
By utilizing bibliometric visualization techniques, specifically focusing on keywords and source information, we constructed annual publication timelines and keyword, country, institutional, and author co-occurrence network maps. These visualizations provide an intuitive representation of the research dynamics and collaborative trends in monkeypox research over the past decade (from January 2014 to June 2025).
## Materials and Methods
## Data Source and Search Strategy
This bibliometric analysis utilized the Web of Science (WoS) Core Collection as the primary data source. Owing to its comprehensive interdisciplinary coverage, advanced citation analysis functionalities, and compatibility with bibliometric tools such as VOSviewer and R packages Bibliometrix/Biblioshiny, which enable robust network visualization and trend analysis, the WoS database was selected over alternative platforms, including Scopus and PubMed.
To identify relevant publications in the mpox (monkeypox) research field, a comprehensive search was performed using the following author keywords: "MPOX," "MPXV," "monkeypox," or "monkeypox virus." The search strategy was applied to the WoS Core Collection database, with the time span set from January 2014 to June 2025 (H1). The retrieved records were exported in plain text format with full records and cited references for subsequent analysis.
## Eligibility Criteria
Publications were selected on the basis of the following criteria: 1) Document type: Articles and Review Articles were included. Other document types, such as letters, editorial materials, early access, proceedings papers, and meeting abstracts, were excluded. 2) Language: Only English-language publications were included. 3) Time frame: The publication dates ranged from January 2014 to June 2025 (H1). 4) Thematic focus: The explicit thematic focus was mpox. The requisite thematic focus encompassed aspects such as clinical presentations, diagnostic methods, epidemiological features, viral evolution patterns, pathogenicity mechanisms, and prevention/control measures, including vaccine research. The explicit exclusions included all nonresearch outputs and retracted publications. The eligible records were exported in a plain text file with the full record content archived. The PRISMA flowchart outlining the identification, screening, eligibility, and inclusion process is presented in Figure 1.
## Data Processing and Network Visualization for Literature Metrics
Data processing and analysis in this study were executed primarily through VOSviewer software by using the exported full-record literature data. 16 The annual publication volumes, spanning from January 2014 to June 2025 (H1), were first quantified and visualized via GraphPad Prism 9 software. Following data cleaning and optimization, default software parameters were implemented as follows: a frequency threshold of ≥5 occurrences was applied for high-frequency keywords, while a minimum output threshold of ≥5 publications was set for institutions and countries/regions. On the basis of the standardized dataset, four key networks were constructed and visualized: 1) a keyword co-occurrence network, 2) an international collaboration network, 3) an interinstitutional collaboration network, and 4) a researcher coauthorship network. These networks distinctly demonstrated the distribution patterns of prevalent terminology, current research hotspots, dominant trends in transnational collaboration, and institutional cooperation intensity patterns. Within the network diagrams, the node sizes are positively correlated with the occurrence frequency of represented elements across the literature corpus. Lines connecting nodes denote specific relationships between elements, with contextual interpretation determined by the network type. In keyword co-occurrence networks, lines indicate co-occurrence patterns within research contexts, whereas in national/institutional collaboration networks, lines explicitly represent the academic collaborative strength between entities. 17,18
## Results
## Annual Trends Reveal Epidemic-Driven Research Patterns
A total of 2,076 publications meeting the screening criteria were included for subsequent analysis. The temporal distribution of the publications intuitively maps the trajectory of the evolution of the global epidemic, with research In 2022, a sudden outbreak event triggered a surge in publications to 321, which exceeded the total of the previous eight years. The number of publications peaked at a historically high level of 759 in 2023 and slightly decreased to 599 in 2024. The first half of 2025 still had 342 papers published, which is significantly greater than the preoutbreak baseline (Figure 2). This three-phase "dormantsurged-plateaued" curve demonstrates the siphoning effect of public health emergencies on scientific research resources.
Practically, this insight informs policymakers and funding agencies on resource allocation: During postoutbreak "plateau" phases-such as 2024-2025-sustained investment in MPXV research is critical to prevent a regression to the pre-2022 "dormant" state, particularly for long-term research themes including viral evolution and zoonotic spillover monitoring. For global health authorities, the trend also underscores the necessity of establishing proactive rather than reactive research frameworks to prepare for potential MPXV resurgences.
## Keyword Co-Occurrence Analysis
Keyword co-occurrence network analysis revealed that "monkeypox" (1,005 occurrences), "MPOX" (702 occurrences) and "monkeypox virus" (543 occurrences) dominated the network. Clustering revealed two primary research directions. The first direction focuses on viral transmission and pathogenesis, with high-frequency terms "virus" (235), "infection" (194), "outbreak" (170), and "transmission" (118) indicating studies in epidemiology (96) and evolutionary dynamics. The second direction emphasizes clinical treatment and vaccine development. The strong association of "HIV" (169) highlights the priority for protecting immunocompromised populations, whereas "smallpox" (143) and "vaccinia virus" (94) reflect investigations into MPOX pathogenesis and predictive models. Terms such as "vaccine" (82), "disease" (77), and "tecovirimat" (84) underscore the urgency of vaccine prevention and clinical interventions (Figure 3 and Table 1).
These findings have clear practical implications for multiple stakeholders: For clinicians and public health practitioners, the emphasis on "HIV" and "immunocompromised populations" reinforces the need for targeted screening and treatment protocols tailored to high-risk groups. For pharmaceutical authorities, the prominence of "vaccine" and "tecovirimat" signals sustained demand for advancing these interventions-particularly for viral variants that may diminish the efficacy of existing therapeutics. From a forwards-looking perspective, gaps in keyword frequency-such as limited terms related to paediatric cases or African regional transmission-highlight future research priorities: strengthening investigations into age-specific clinical outcomes and region-adapted prevention strategies to address underrepresented areas.
## Country Co-Occurrence Analysis
The national research output has a three-tier pyramid structure. The United States is the main leader, with 522 publications (25.1% of the total). China (386 publications, 18.6%), India (214 publications), and Saudi Arabia (183 2).
This tiered structure provides actionable guidance for international collaboration: Global health authorities can leverage the leading research output of the US and China to coordinate multicentre initiatives-such as vaccine trials -while investing in capacity building for the "broad participation tier", particularly African nations, including Nigeria (a key MPXV-endemic region) and Egypt. Strengthening regional research in Africa not only addresses the imbalance in the 3). 4).
## Institutional and Author Co-Occurrence Analyses
Practically, this analysis assists stakeholders in identifying key partners for collaborative initiatives: Policymakers aiming to develop evidence-based guidelines can engage leading institutions-such as the US CDC (for epidemiological data) or the Chinese Academy of Sciences (for vaccine research). For early-career researchers, collaborating with top scholars-including Fabrizio Maggi in clinical management or Kuldeep Dhama in viral pathogenesis-can accelerate advancements in understudied areas. Looking forward, fostering partnerships between governmental health systems (notably the US CDC) and academic institutions in low-resource regions can bridge the gap between research and on-theground implementation.
## Discussion
This study employed tools such as VOSviewer to visualize bibliometric data from 2,076 publications. Through quantitative, qualitative, and integrative methodologies, the current status and trends in the field of monkeypox (MPOX) research were systematically evaluated.
## Trends in the Field of Monkeypox Research
The transition of MPXV infection from a regional zoonosis to a global public health threat exemplifies the archetypal evolutionary trajectory of emerging infectious diseases in the era of globalization. The intercontinental spread of clade IIb in 2022, which was facilitated by transmission networks within communities of men who have sex with men (MSM), provided the first definitive evidence that MPXV could achieve sustained human-to-human transmission without the need for animal host intermediaries. 19 Concurrently, the 2023 outbreak of clade IIb in the Democratic Republic of the Congo (DRC) further revealed the evolutionary potential of the virus. 20 This outbreak resulted in a distinct shift in demographic distribution, with more than 50% of cases occurring in females-a notable contrast to the 2022 global epidemic, during which 98% of cases occurred in males. 21 MPXV transmission occurs through multiple pathways, including sexual contact (particularly among MSM), vertical mother-to-child transmission, and close household contact. Furthermore, its zoonotic nature enables spillover from natural reservoirs, such as rodents. Critically, viral genomic mutations (eg,
## 106
APOBEC3-driven mutations) coupled with diverse transmission routes collectively amplify the complexity of epidemic containment and control measures. 22 Unlike most DNA viruses, MPXV uniquely completes its entire replication cycle independently within the cytoplasm, markedly reducing its reliance on host cell organelles. 23 Crucially, its genome exhibits functional compartmentalization, with conserved central genes maintaining core replication functions and terminal genes undergoing high-frequency mutations to adapt to host immune pressures. 24 Mutations mediated by APOBEC3 deaminase, identified in clade IIb, exemplify this adaptive strategy. 22 These mutations may not only increase human-to-human transmissibility but also generate variations, posing a risk of false-negative results with existing diagnostic assays. 25 This evolutionary strategy enables MPXV to overcome ecological niche constraints. Following its cross-species jump from African rodents to humans, the virus established transmission foci in nonendemic regions within only three years. This rate drastically exceeds the predicted evolutionary rate of orthopoxviruses, which is historically estimated at 6-12 accumulated SNPs per year. 26 Significant disparities in susceptibility and clinical outcomes among different population groups necessitate precise stratification of prevention and control strategies. Immunocompromised individuals: HIV-coinfected individuals account for 40-90% of global cases. 27 They frequently present with large ulcerative necrotic lesions and multisystem complications (eg, rectal bleeding and encephalitis), increasing their potential as mobile sources of transmission. Children and older adults: During clade I outbreaks, paediatric case fatality rates reached 4.6%, primarily because of the absence of vaccine protection and secondary bacterial infections. 28 Older individuals, burdened by comorbidities, exhibit heightened susceptibility to severe complications, such as pneumonia and renal failure. 29 Sexually active populations: Within the MSM community, individuals reporting high sexual activity demonstrate a 120-fold higher transmission contribution than those with low activity. Female sex workers emerged as critical transmission hubs during the clade IIb outbreak, facilitating viral introduction from animal reservoirs into community households. 20 Bibliometric evidence indicates that research output on MPXV expanded sharply following the 2022 global outbreak. Earlier studies (2014-2019) were dominated by epidemiological observations and animal reservoir investigations, whereas post-2020 research increasingly emphasized viral evolution, genomic diversity, and vaccine development. This temporal evolution of MPXV research reflects the broader pattern observed in outbreak-driven fields-where the transition from descriptive to molecular and translational studies parallels the global escalation of the disease.
## Status and Quality of Authors, Journals and Research
As the world's second-largest contributor to global monkeypox research (18.6% of the total publications), China faces unique pressure from both local transmission and cross-border importation. Imported cases were identified in Chongqing, followed by local cases in Beijing. The spread of clade IIb to neighbouring countries, such as India and Thailand, is concerning and is compounded by the uncertain seroprotection levels conferred by historical smallpox vaccination among individuals born before 1982 in China. 30 Current smallpox vaccines (JYNNEOS and ACAM2000), which rely on cross-immunity, may exhibit waning efficacy against emerging variants. 31 The Rmix6 mRNA vaccine developed by Yan Jinghua et al demonstrated breakthrough potential. Animal studies revealed 100% survival and significantly elevated IFN-γ-secreting T-cell responses (P<0.05), confirming that multiantigen design can activate synergistic immunity. 32 Notably, while leading contributors to global MPXV research-including Fabrizio Maggi and Kuldeep Dhama-have garnered widespread recognition for their seminal insights into viral evolution and clinical management strategies, the discourse on research contributions would be significantly enriched by greater balance, specifically through the acknowledgement of work originating from underrepresented regions. Foremost among these is Africa, the geographical epicentre of the MPXV. African research institutions and scholars have played an indispensable role in advancing MPXV knowledge: they have spearheaded efforts to document early outbreaks-such as conducting long-term epidemiological surveillance in the Democratic Republic of the Congo (DRC) 19 -and have deepened the understanding of region-specific transmission dynamics, including zoonotic spillover events involving local rodent populations. Integrating these regional contributions not only addresses critical gaps in the global research narrative but also highlights contextspecific challenges-such as restricted access to advanced diagnostic tools in rural African settings-that are paramount to developing equitable, locally adaptable prevention and control strategies. priorities. In parallel with COVID-19 research, MPXV-related publications experienced a rapid surge in volume following the 2022 global epidemic, with a primary focus on unravelling transmission dynamics and accelerating vaccine development. 4 However, a key distinction emerges: unlike research on COVID-19-a respiratory virus characterized by high airborne transmissibility-MPXV research has centred heavily on sexual and zoonotic transmission pathways. This focus has, in turn, driven innovations in targeted interventions for high-risk groups, such as MSM communities. In contrast to Ebola, which is constrained by limited geographic spread but marked by high mortality rates, MPXV research confronts the unique challenge of balancing long-term zoonotic surveillance (to monitor spillover from animal reservoirs) with responsive measures for sustained human-to-human transmission. Such comparisons reveal that MPXV research occupies a critical middle ground, bridging gaps between "spillover-only" pathogens (such as Ebola) and "sustained pandemic" pathogens (such as COVID-19). As a result, it offers unparalleled insights into diseases that exhibit moderate transmission potential but have substantial global public health impacts.
## Research Hot Spots, Frontiers, and Future Trends
Beyond vaccines, critical bottlenecks persist in the global response framework, including diagnostic technology limitations and inadequate interventions targeting high-risk groups. 33 Addressing these challenges requires-as a research priority-the development of broad-spectrum diagnostic assays and community integration by establishing integrated service models that combine sexual health clinics with monkeypox screening. The rapid spread of clade IIb within MSM networks exposed blind spots in traditional epidemiological surveillance. 27,33 Concurrently, the paediatric outbreak in the DRC underscores the urgent need for enhanced school health education and subsidized family screening programs.
The evolution of monkeypox fundamentally reflects a process of mutual adaptation between the virus and human society. Globalization accelerates pathogen dispersal, deforestation increases zoonotic spillover frequency, and shifting sexual behaviours alter transmission dynamics-these converging drivers create a "perfect storm" for emerging infectious diseases. Building a resilient next-generation pandemic defence system demands the deep integration of virology, social and behavioural science, and artificial intelligence-based prediction. This approach is not only essential for combating monkeypox but also constitutes vital preparation for future zoonotic pandemics. This study's bibliometric analysis of MPXV research (2014-2025 H1) yields critical, unique insights to guide future research directions and policymaking. First, regarding research prioritization, the analysis identifies two core clustersviral evolution, such as APOBEC3-driven mutations, and targeted interventions, including MSM-focused preventionand highlights understudied areas, notably regional research in Africa and paediatric case management. This guidance empowers researchers and funding agencies to allocate resources to gaps that directly address real-world challenges, such as enhancing access to diagnostics in rural African settings and refining vaccine strategies for immunocompromised populations. Second, for policymaking, the analysis clarifies the need for stratified responses: identifying hightransmission groups-including high-activity MSM and female sex workers-and the geographic spread of clade IIb to China's neighbouring countries provides evidence for targeted public health measures, such as community-based screening at sexual health clinics and cross-border surveillance collaborations. Additionally, comparing MPXV research with that of Ebola and COVID-19 highlights that MPXV's "middle-ground" transmission characteristics demand policies balancing zoonotic spillover prevention-such as rodent population monitoring-and sustained human-to-human transmission control, including vaccine rollouts for high-risk groups. By synthesizing publication trends, collaborative networks, and research hotspots, this bibliometric analysis serves as a data-driven foundation to align future MPXV research with global health needs and ensure evidence-based, adaptable policies amid the virus's evolving threat.
## Strengths and Limitations
This bibliometric and qualitative synthesis provides a comprehensive overview of MPXV research trends from 2014 to mid-2025, combining publication metrics with thematic and contextual analysis. However, several limitations warrant consideration. First, the analysis relied exclusively on the Web of Science Core Collection and English-language publications, potentially omitting relevant non-English or grey literature. Second, citation-based metrics may overrepresent established authors and journals while undervaluing emerging contributions from developing regions. Third, bibliometric indicators alone cannot fully capture the translational value or societal impact of individual studies. 109 Despite these constraints, the integration of analytical tools such as VOSviewer and GraphPad enhances methodological robustness, enabling the identification of publication dynamics, research collaborations, and thematic frontiers. Collectively, these findings provide a data-driven foundation for evidence-based policymaking and targeted funding strategies in the evolving field of monkeypox research.
## Conclusion
This study systematically applied bibliometric and visualization techniques to characterize global monkeypox (MPXV) research from 2014 to mid-2025, providing the first comprehensive mapping of publication dynamics, collaboration patterns, and thematic evolution. The results demonstrate a rapid transformation of the MPXV field, with a transition from a neglected zoonotic topic to a core focus of emerging infectious disease research following the 2022-2023 outbreaks.
These findings reveal that MPXV has evolved from a regionally confined zoonosis to a globally significant pathogen, following an archetypal trajectory of infectious disease emergence in the era of globalization. The 2022 spread of clade IIb across continents-driven primarily by transmission networks among men who have sex with men (MSM)-provided compelling evidence of sustained human-to-human transmission independent of animal reservoirs. The subsequent 2023 outbreak in the Democratic Republic of the Congo (DRC) highlighted the virus's ongoing adaptive evolution, marked by APOBEC3-driven mutations and demographic shifts in infection patterns. The analysis indicates that research activity mirrors these epidemiological transitions, shifting from descriptive surveillance to molecular, genomic, and immunological studies aimed at elucidating transmission dynamics, host interactions, and therapeutic development.
Keyword co-occurrence and clustering analyses identify two dominant research axes: viral evolution and targeted interventions. Studies increasingly emphasize APOBEC3-mediated mutagenesis, vaccine optimization, and protection of high-risk populations such as MSM and immunocompromised individuals. Comparative bibliometric patterns position MPXV research between Ebola and COVID-19-balancing zoonotic surveillance and sustained human-to-human transmission control. Future frontiers include the development of broad-spectrum diagnostic platforms, improved community screening systems integrated with sexual health services, and refined vaccine formulations that maintain efficacy against emerging viral variants.
This bibliometric and qualitative synthesis provides a robust, data-driven overview of MPXV research trends; however, certain limitations remain. The analysis relied solely on the Web of Science Core Collection and Englishlanguage records, potentially excluding relevant non-English or grey literature. Moreover, citation-based metrics may disproportionately favour established scholars and institutions. Despite these constraints, the integration of VOSviewer and GraphPad software ensures analytical precision and visual clarity, enabling the identification of major publication trends, collaboration patterns, and thematic frontiers. Collectively, these findings provide an evidence-based foundation to guide policymaking, funding allocation, and future global health preparedness strategies.
## References
1. Gong, Wang, Chuai (2022) "Monkeypox virus: a re-emergent threat to humans" *Virol Sin*
2. Von Magnus, Anderson, Petersen (1959) "A pox-like disease in cynomolgus monkeys" *Acta Pathol Microbiol Scand*
3. Breman, Kalisa-Ruti, Steniowski (1980) "Human monkeypox" *Bull World Health Organ*
4. Okonji, Okonji (2022) "Monkeypox during COVID-19 era in Africa: current challenges and recommendations" *Ann Med Surg Lond*
5. Farahat, Sah, El-Sakka (2022) "Human monkeypox disease (MPX)" *Infez Med*
6. Overton, Abbott, Christie (2023) "Nowcasting the 2022 mpox outbreak in England" *PLoS Comput Biol*
7. Titanji, Hazra, Zucker (2024) "Mpox clinical presentation, diagnostic approaches, and treatment strategies: a review" *JAMA*
8. Lian, Yang, Qiu (2024) "Evolutionary analysis and antiviral drug prediction of Mpox virus" *Microorganisms*
9. Babkin, Babkina, Tikunova (2022) "An update of orthopoxvirus molecular evolution" *Viruses*
10. Werden, Rahman, Mcfadden (2008) "Poxvirus host range genes" *Adv Virus Res*
11. Li, Yuan, Jiang (2023) "Animal host range of mpox virus" *J Med Virol*
12. Lum, Torres-Ruesta, Tay (2022) "Monkeypox: disease epidemiology, host immunity and clinical interventions" *Nat Rev Immunol*
13. Khaity, Hasan, Albakri (2022) "Monkeypox from Congo 1970 to Europe 2022; is there a difference?" *Int J Surg*
14. Onyeaghala, Alinnor, Irabor (2025) "Neonatal mpox in Nigeria: a case of transplacental or postnatal transmission" *BMC Infect Dis*
15. Ortiz-Prado, Kyriakidis, López-Cortés (2025) "Current and emerging Mpox vaccine strategies: a comprehensive review" *Vaccine*
16. Van Eck, Waltman (2010) "Software survey: vOSviewer, a computer program for bibliometric mapping" *Scientometrics*
17. Wen, Ma, Li (2023) "Research trends and hotspots in exercise rehabilitation for coronary heart disease: a bibliometric analysis" *Medicine*
18. Jiang, Chen, Zhao (2023) "COVID-19 and chronic kidney disease: a bibliometric analysis" *Ann Med Surg Lond*
19. Marcus, Michel, Lunchenkov (2024) "A seroprevalence study indicates a high proportion of clinically undiagnosed MPXV infections in men who have sex with men in" *BMC Infect Dis*
20. Masirika, Udahemuka, Schuele (2024) "Ongoing mpox outbreak in Kamituga, South Kivu province, associated with monkeypox virus of a novel Clade I sub-lineage, Democratic Republic of the Congo" *Euro Surveill*
21. Schwartz (2024) "High rates of miscarriage and stillbirth among pregnant women with clade I Mpox (Monkeypox) are confirmed during 2023-2024 DR Congo outbreak in South Kivu Province" *Viruses*
22. Vakaniaki, Kacita, Kinganda-Lusamaki (2024) "Sustained human outbreak of a new MPXV clade I lineage in eastern Democratic Republic of the Congo" *Nat Med*
23. Silhan, Klima, Otava (2023) "Discovery and structural characterization of monkeypox virus methyltransferase VP39 inhibitors reveal similarities to SARS-CoV-2 nsp14 methyltransferase" *Nat Commun*
24. Li, Hou, Sun (2023) "Monkeypox virus 2022, gene heterogeneity and protein polymorphism" *Signal Transduct Target Ther*
25. Brüssow (2025) "Monkeypox virus: WHO's second public health emergency of international concern within 2 years" *Microb Biotechnol*
26. Isidro, Borges, Pinto (2022) "Phylogenomic characterization and signs of microevolution in the 2022 multi-country outbreak of monkeypox virus" *Nat Med*
27. Bogacka, Wroczynska, Rymer (2025) "Mpox unveiled: global epidemiology, treatment advances, and prevention strategies" *One Health*
28. Halder, Sultana, Himel (2025) "Monkeypox: origin, transmission, clinical manifestations, prevention, and therapeutic options" *Interdiscip Perspect Infect Dis*
29. Savvidis, Rizzo, Ilias (2025) "Mpox Infection and endocrine health: bridging the knowledge gap" *Medicina*
30. Wattal, Datta (2022) "Monkey Pox arrives in India" *Indian J Med Microbiol*
31. Rastogi, Kumar (2024) "Current status of vaccine development for monkeypox virus" *Adv Exp Med Biol*
32. Zeng, Li, Jiang (2023) "Mpox multi-antigen mRNA vaccine candidates by a simplified manufacturing strategy afford efficient protection against lethal orthopoxvirus challenge" *Emerg Microbes Infect*
33. Amer, Khalil, Elahmady (2023) "Mpox: risks and approaches to prevention" *J Infect Public Health*
34. *Veterinary Medicine: Research and Reports*
35. Xie *TCPDF*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12834354&blobtype=pdf
|
Claudia Denkinger, Jane Cunningham, Verena Faehling, Lukas Brümmer, Kim Hanson, Richard Molenkamp, Melissa Miller, Adrian Marcato, David Price, Sandra Ciesek, Emmanuel Agogo, Joseph Fitchett, Ute Ströher, Pragya Yadav, Isabel Bergeri, Nicki Boddington, & Nira, R Pollock
## Perspectives
For highly transmissible pathogens, accurate diagnostic testing is key to early outbreak identification, outbreak mitigation and pandemic prevention. For a novel pathogen, however, diagnostic tests are not initially available. The coronavirus disease 2019 (COVID-19) pandemic clearly demonstrated the importance, and the challenge, of rapidly developing and implementing reliable pathogen-specific diagnostic tools early in an outbreak of a novel pathogen. The diagnostic response to the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was hindered both by the need to develop new tests on a massive scale and by inefficient approaches to test evaluation and deployment. Initial polymerase chain reaction (PCR) protocols were published within two weeks of sequence availability and the World Health Organization (WHO) was shipping kits to 159 laboratories two weeks later. 1 Despite this accomplishment, the early response to SARS-CoV-2 was largely characterized by scarcity of diagnostic testing and few coordinated efforts to generate representative data on pathogen kinetics, the spectrum of clinical presentations, and both optimal sample types and timing of testing. 2 These gaps hindered early efforts to prevent disease transmission and to effectively guide test utilization.
WHO has prepared a series of protocols, collectively titled the Unity Studies, that promote a harmonized approach to surveillance and rapid evidence generation for pandemic influenza viruses, SARS-CoV-2, Middle East respiratory syndrome-CoV and novel respiratory pathogens with pandemic potential (that is, pathogen X). 3 Notably, the first protocol in the existing series, The first few X cases and contacts (FFX) investigation template protocol for respiratory pathogens with pandemic potential, 4 assumes that a validated diagnostic test is already available to facilitate case detection, and does not consider that processes to ensure and optimize rapid development, evaluation and implementation of such a test need to be defined.
To fill this critical gap, a new Unity Studies protocol, First few X cases and contacts diagnostic test evaluation for respiratory pathogens with pandemic potential: template protocol (known as the FFX-Dx protocol) has been developed to facilitate coordinated future development, evaluation and deployment of diagnostics for a novel respiratory pathogen X. The protocol is intended to be the first of the Unity Studies to be implemented during an outbreak. An initial small group of experts developed the first draft, which was subsequently reviewed by a broad group of international diagnostic experts and additional key stakeholders and decision-makers from academia, public health and regulatory agencies. These stakeholders had led SARS-CoV-2 diagnostic test development, evaluation and implementation efforts nationally and internationally. After several rounds of review, including an in-person workshop at the University of Heidelberg, Germany, 5 the pre-final protocol was further circulated for public consultation before it was finalized and posted on WHO's website. 6 The protocol has also been reviewed and revised based on input from the WHO Research Ethics Review Committee.
The protocol is designed to provide specific guidance on methods for the validation and early use of the first molecular diagnostic test for respiratory pathogen X in a rigorous but focused assessment of the first pathogen X cases and contacts at the location of an outbreak anywhere in the world. This first molecular test would need to be an in-house developed test (also known in some countries as a laboratory-developed test), given that timing of clinical and research use would need to precede receipt of any form of emergency use authorization by the relevant regulatory authority. In the protocol, the in-house developed test is assumed to be a realtime reverse transcription (RT)-PCR or PCR assay. As such, the protocol objectives were designed specifically to guide and expedite early pathogen X-specific molecular test development and deployment for clinical use and public health benefit. Optimally, if implemented early in an outbreak of pathogen X, the As a first step in the event of the emergence of a novel pathogen X, initial pathogen discovery and early characterization (including sequencing and culture to yield a viral or bacterial isolate) would be required and would precede the protocol. This step would be followed by early development and early validation of a molecular in-house developed test (Fig. 1) in a laboratory with staff having sufficient expertise to do this work, regardless of its distance from the outbreak site. Early test development would include work to design and develop the test for pathogen X before analytical and clinical validation, including primers, probes and an initial RT-PCR or PCR assay. This work could be done by the same laboratory involved in the discovery and/or early characterization of pathogen X or by another laboratory or laboratory consortium. The early validation phase would involve rigorous and focused analytical studies performed by the test developer to validate the test in preparation for its use for clinical (and research) testing before emergency use authorization, if required by the local regulatory authority. This early validation phase could be done before availability of clinical samples from infected patients and could be conducted by the same laboratory involved in the early development of the test or by another laboratory or laboratory consortium. Detailed considerations for the early development and early validation of the in-house developed test are presented in Annex 4 of the protocol. 6 Following a fully completed and successful early validation performed by the test developer, a focused end-user validation would be required to put the test into clinical use for case diagnosis at the site of the outbreak (Fig. 1). This end-user validation is the subject of Part A of the protocol and would likely be performed by a nationally designated laboratory with characteristics as defined in Annex 5 of the protocol. 6 This stage includes focused analytical validation to verify the analytical test characteristics established by the developing laboratory and a limited clinical validation, that is, prospective testing, with confirmatory testing to be performed by an independent laboratory. Upon completion of the end-user validation, the test could be made available for clinical testing. Importantly, the protocol recognizes that different jurisdictions may have alternative rules and requirements for test validation for clinical use, and that each country implementing the protocol would have to consider what additional validation might be required for emergency use authorization under its national regulatory framework.
## Diagnosis of the first few cases of a novel respiratory pathogen: the FFX-Dx protocol
With a validated molecular inhouse developed test now available for clinical testing, the next critical steps are to define testing approaches with highest yield, optimize and refine the clinical case definition for disease X, and pre-emptively attempt to identify asymptomatic or presymptomatic infection, with the goals of preventing disease transmission and driving fur-ther test development and deployment. Part B of the FFX-Dx protocol is thus focused on early use of the validated inhouse developed test to quantify pathogen X in matched (or paired) samples, that is, collected from different body sites at the same time point. These samples are collected serially over the course of infection and post-exposure to assess early pathogen X kinetics and optimal sample types for diagnostic testing in cases and contacts, including samples relevant to point-of-care and self-testing. Cumulatively, the study is designed to inform test use case(s), infection prevention and isolation duration, as well as to define optimal sample types for early molecular test deployment for clinical and reference testing and for point-of-care and/or self-test development. The workshop group defined best practice methods and materials for collection of each sample type, using knowledge gained from the COVID-19 pandemic. 2,7 Secondary objectives of the study include preliminary estimation of key epidemiological parameters, to inform and be further studied in subsequent Unity Studies protocols. The protocol is designed to be a template that facilitates rapid deployment of in-house developed tests in the event of an outbreak, but should additionally serve as a tool or resource for national regulatory frameworks. Pandemic preparedness teams should consider early adaptation of the template protocol for in-country or national laboratories and public health authorities that are likely to take the lead in deploying the protocol. The teams should also pass the protocol through local and/or national ethics review committees. Furthermore, preparedness teams should consider how the study's data would be used to inform targeted guidance for early test use and further test development and authorization to aid the public health response. Accordingly, the target audience for the protocol includes public health decision-makers, regulatory agencies and academic and/or research institutions, and in particular, epidemiologists and clinical laboratorians.
The expert group also recognized potential barriers to implementation of the protocol that should be addressed in advance of an emergency. 5 These considerations include: (i) protocol implementation partners, local regulatory context and required ethics preapprovals; (ii) funding of initial phases of test development and validation, management of intellectual property rights and possible profit-seeking behaviour by private entities; (iii) development of sample and/or organism panels (and material transfer agreements for sharing) for cross-reactivity analysis as required for early validation; (iv) navigation of regulations affecting the sharing of in-house developed test reagents and control materials between developing laboratories and end user laboratories; (v) managing supply chain logistics; (vi) facilitating data sharing; and (vii) the potential for inequities in global distribution of diagnostics.
A subgroup of experts who participated in this work are developing another template protocol for evaluation of antigen-detecting rapid diagnostic tests for a novel respiratory pathogen X.
Collectively, these efforts will facilitate the rapid generation and synthesis of high-quality evidence to guide and expedite both test development and testing policy for a novel respiratory pathogen X. Doing so will protect people's health by expediting thoughtful deployment of tests at each stage of an outbreak response, effectively combining science, practicality, speed and equity. ■
## References
1. Sheridan (2020) "Coronavirus and the race to distribute reliable diagnostics" *Nat Biotechnol*
2. Theel, Kirby, Pollock (2024) "Testing for SARS-CoV-2: lessons learned and current use cases" *Clin Microbiol Rev*
3. (2025) "Geneva: World Health Organization"
4. (2023) "The First Few X cases and contacts (FFX) investigation template protocol for respiratory pathogens with pandemic potential"
5. (2024) "Workshop to support coordinated evaluation and early deployment of new diagnostics for pathogen X: meeting report"
6. (2025) "First Few X cases and contacts diagnostic test evaluation for respiratory pathogens with pandemic potential: template protocol. Geneva: World Health Organization"
7. Lee, Herigon, Benedetti et al. (2021) "Performance of saliva, oropharyngeal swabs, and nasal swabs for SARS-CoV-2 molecular detection: a systematic review and meta-analysis" *J Clin Microbiol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12565101&blobtype=pdf
|
# Correction: Scharmann et al. Evaluation of the Loop-Mediated Isothermal Amplification Assay (LAMP) Eazyplex ® Pneumocystis jirovecii. J. Fungi 2025, 11, 300
Ulrike Scharmann, Lisa Kirchhoff, Jan Buer, Franziska Schuler, Annerose Serr, Susann Rößler, Jürgen Held, Tobias Szumlanski, Joerg Steinmann, Peter-Michael Rath
## Text Correction
In the original publication [1], we recognized a mistake in our manuscript which was published. In Section 2 Materials and Methods, we specified 125 µL buffer instead of 500 µL buffer for the preparation of the LAMP. However, this was a formal error. The test was carried out correctly, as the buffer provided by the company and ready for use was used for the preparation. This was not pipetted and the amount in milliliters was not changed.
The main text in Section 2.2 should now read as follows:
"For the LAMP, 25 µL of sample was mixed up with 500 µL of buffer and incubated for 3 min at 99 • C and then added to lyophilized reagents."
With the above corrections, the authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.
## References
1. Scharmann, Kirchhoff, Buer et al. (2025) "Evaluation of the Loop-Mediated Isothermal Amplification Assay (LAMP) Eazyplex ® Pneumocystis jirovecii" *J. Fungi*
2. (2025) "MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content" *J. Fungi*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584193&blobtype=pdf
|
# Highly pathogenic avian influenza A H5N1 virus infection in an immunocompromised domestic cat
Chi Chen, Akhila Naru, Vineetha Mareddy, Saraswathi Lanka, Colleen Olmstead, Vanessa Revindran-Stam, Megan Sherman, Heather Yee, Natara Loose, Martha Delaney, Miranda Vieson, Ying Fang, Vineetha Reddy Mareddy
## Abstract
Background Highly pathogenic avian influenza (HPAI) H5N1 viruses of clade 2.3.4.4b have recently caused widespread outbreaks in mammals, including domestic cats that live closely with humans and other animals. In-depth molecular and pathological characterizations of naturally infected cats are urgently needed for developing better strategies to prevent interspecies transmission and further spreading of these viruses.
Case SummaryIn this case report, we characterized a unique case of HPAI H5N1 virus infection in an immunocompromised domestic cat. The pet animal was a diabetic cat with a history of feline infectious peritonitis (FIP). In early 2025, the cat developed acute fever and rapidly worsening respiratory distress and liver dysfunction despite antibi otic treatment. Due to severe clinical deterioration, the cat was euthanized. Postmor tem examination revealed severe bronchointerstitial pneumonia, hepatic and lymphoid necrosis, bone marrow degeneration, and mild lymphohistiocytic meningitis. H5N1 viral RNA/antigens were specifically detected in the lung, brain, urine, or lymphoid tissues. Whole-genome sequencing and phylogeny analysis identified that the virus belongs to influenza clade 2.3.4.4b (B3.13 subgroup), closely related to HPAI H5N1 strains that are currently circulating in domestic cats and cattle. The source of infection for this particular cat might be linked to a fomite/environmental transmission route.
ConclusionThe lethal HPAI H5N1 virus infection in an immunocompromised cat highlights the need for developing an improved prevention plan for pet animals. Clinicians should consider the possibility of H5N1 virus infection in cats with similar acute respiratory or neurologic signs, particularly in animals with chronic illness. KEYWORDS highly pathogenic avian influenza (HPAI), H5N1, feline, cat, feline infectious peritonitis, diabetes mellitus, whole-genome sequencing H ighly pathogenic avian influenza A H5N1 virus, a subtype originating from the A/goose/Guangdong/1/1996 H5 subtype, continues to circulate among wild and domestic birds and spillover to mammals (1). The Eurasian H5N1 clade 2.3.4.4b was first detected in North America in late 2021 and has since caused spillover infections and deaths in terrestrial and marine mammals across the United States (2-7). The detection of HPAI H5N1 clade 2.3.4.4b in severe human infections raises concern about its pandemic potential (8, 9). The recent detection of HPAI H5N1 virus in dairy cattle presents an unusual transmission route (10,11). Notably, domestic cats have been involved in infections on dairy farms, where cats were exposed to H5N1 virus by ingesting raw milk from infected cows or through contact with contaminated farm environments (7,12,13). Cats were also infected by H5N1 virus through consuming
infected birds (14)(15)(16). As a companion animal, cats may function as an intermediate host to facilitate the cross-species transmission of viruses between animals and humans. Domestic cats appeared to be highly susceptible to HPAI H5N1 virus and developed severe respiratory and neurological symptoms and, in some cases, resulted in the death of the animal (17,18). In this study, we characterized a case of H5N1 virus infection in a diabetic cat with a history of FIP virus (FIPV) infection.
## CASE PRESENTATION
A 1-year-old male, castrated domestic shorthair cat was submitted to the University of Illinois (UIUC) Veterinary Diagnostic Laboratory in February 2025 for postmortem examination. This cat was owned and housed at the submitting veterinary clinic in Brooklyn, New York, and had a previous diagnosis of FIPV infection at 8 weeks of age, which was treated with 10 mg/kg/day GS-441524 (Lucky Cat Veterinary Care), initially with the injectable formulation and then switched to oral administration after a seroma developed at an injection site. After recovery from clinical signs attributed to FIP, the cat developed diabetes mellitus that was continually difficult to regulate, contributing to multiple episodes of diabetic ketoacidosis (DKA). During a period of diabetic regulation, the cat developed a high fever (105.7 o F) with slightly different symptoms from previous DKA episodes, prompting emergency care where antimicrobial therapy (Unasyn and Baytril) and supportive care (IV fluids, ondansetron, insulin, metoclopramide, mirataz, and gabapentin) were initiated. Despite these efforts, the fever persisted, and the cat quickly progressed into respiratory and hepatic decline, followed by humane euthanasia.
Gross lesions were mild and non-specific, including multifocal mottling of the lungs, a few small fibrous adhesions between the omentum, spleen, and ventral internal body wall, and mild pallor and swelling of the liver. Significant histopathological lesions were found (Fig. 1A through C), including necrotizing bronchointerstitial pneumonia and fibrinosuppurative exudative interstitial pneumonia. Furthermore, there was necrotizing hepatitis, splenitis, lymphadenitis, bone marrow degeneration, and very mild multifocal lymphohistiocytic meningitis. To detect H5N1 infection, RNAscope in situ hybridization was performed using a probe for Hemagglutinin (HA) H5 gene region (ACD Diagnostics, Probe-V-InfluenzaA-H5N8-M2M1-C1 and RNAscope 2.5 HD Red Assay), while immuno histochemistry was conducted using monoclonal antibody (mAb #42-100) recognizing viral nucleoprotein (NP). The results showed intense H5 gene-specific signal amplification (Fig. 1D through F) and NP antigen-specific immunolabeling (Fig. 1G through I) within the areas that had significant lesions in the lung, liver, and spleen. HPAI H5N1 infection was confirmed by RT-qPCR using the National Animal Health Laboratory Network approved protocol in UIUC Veterinary Diagnostic Laboratory, which targets Matrix (M) and H5 genes. The results revealed a high level of viral RNA load in the brain (M Ct = 17.13; H5 Ct = 19.16), lung (M Ct = 16.54; H5 Ct = 22.71), spleen (M Ct = 19.48; H5 Ct = 22.41), and urine (M Ct = 22.22; H5 Ct = 25.07). Further testing on cat tissue samples yielded negative results for FIPV, SARS-CoV-2, and Francisella tularensis. The aerobic culture of the lung and liver did not result in any significant pathogens, ruling out bacterial sepsis.
After the final diagnosis, contact tracing by the referring clinic revealed that a few days before the onset of illness, as a resident of the clinic, the cat had interacted with the clinic areas and personnel that had not yet been decontaminated from another feline patient. The patient presented with a painful abdomen, fever (103.9 o F), and consumption of food that was recalled for the potential to contain H5N1 virus (19). There is no confirmation of H5N1 infection from that feline patient, and there were also no other known sources of exposure. No related human cases of H5N1 infection were identified.
Viral whole-genome sequencing for brain and lung tissue samples was performed using Illumina MiSeq (20). The resulting sequences were identified as HPAI H5N1 virus genome segments. The cat-derived virus strain is designated as A/cat/New York/ UIUC25-01/2025 (UIUC25-01). Subsequently, maximum likelihood phylogenetic trees were constructed using IQ-TREE (v2.2.6) (21). HA (Fig. 2) and NA (Fig. 3) phylogenies confirm that the UIUC25-01 virus belongs to 2.3.4.4b clade with other circulating HPAI H5N1 strains. Compared to the emerging H5N1 strain originally obtained from the A/ Bovine/texas/24-029328-01/2024 (H5N1), the UIUC25-01 virus exhibited high genetic similarity across all genome segments (99.52%-99.86% nucleotide identity, 99.13%-100% amino acid identity; Table 1). One distinct mutation at amino acid position 71 of the NA protein (N71S) was identified in the cat virus, which was previously reported as a hallmark of emerging HPAI H5N1 viruses (2).
## DISCUSSION
In this report, we characterized a clinical case of HPAI H5N1 virus infection in a 1-year-old immunocompromised cat. Pathological and molecular analyses showed that the lung appeared to be the primary site of viral replication with variable degrees of systemic tissue damage. Consistent with the respiratory and hepatic decline reported clinically, the most severe pathology was observed in the lung and liver, with virus identified by immunohistochemistry and in situ hybridization, along with the necrotizing inflammation. Lesions, organs affected, and viral distribution in this case are similar to those previously documented in cats infected with HPAI H5N1 viruses (25,26). Multiple cases have similarly shown higher levels of virus in the brain, often with lesions that are equally or more severe than those in the respiratory system, indicating a shift towards neurotrop ism (2,10). Interestingly, in this case, there was minimal multifocal lymphohistiocytic inflammation in the leptomeninges and no neurologic symptoms mentioned in the clinical history. Despite this, using an H5-specific probe, RT-qPCR detected the highest level of viral RNA in brain tissue compared to that detected in the lung and spleen. This disparity may be related to spatial distribution. Assuming viral entry into the CNS through the olfactory nerves or across the cribriform plate, as has been documented in mice (27), more virus would be located in the frontal and olfactory regions of the brain that were submitted for testing but not examined by histopathology. Still, this specific route of viral entry into the brain of cats is not clearly documented (25) and warrants further investigation.
The exact source of infection for this cat has not been confirmed. Given no evidence of dairy or poultry consumption or contact with birds, cattle, or individuals working in poultry or cattle industries, we suspect that this cat may have been infected by fomite/ aerosol transmission in the clinic by another cat with possible H5N1 virus infection. This highlights the importance of hand hygiene/decontamination between patients for healthcare. Another important consideration in this cat is concurrent diabetes mellitus The rest of the sequences were from GISAID (23). Sequences were aligned using MAFFT (v7.526) (24). and prior infection with FIPV, both of which can be related to a weakened immune system that may promote transmission of the H5N1 virus. This indicates that animal health conditions need to be considered when assessing spillover risk. Animal handlers/ healthcare providers should understand the potential risks for immunocompromised animals sharing the same space with other infected animals.
In comparison to the emerging bovine-derived H5N1 virus A/Bovine/Texas/ 24-029328-01/2024, the cat UIUC25-01 virus shared >99% identity at the nucleotide and amino acid levels, indicating the same origin. Phylogenetic analysis further confirmed that this UIUC25-01 virus lies in the North American clade 2.3.4.4b, where it clusters with other recently reported emerging HPAI H5N1 strains from cats and dairy cows. GenBank. The rest of the sequences were from GISAID (23). Sequences were aligned using MAFFT (v7.526) (24).
Several mutations had been acquired by clade 2.3.4.4b viruses to adapt to mammalian hosts, particularly in cattle, marine mammals, and humans in recent outbreaks (2,(28)(29)(30)(31)(32). In comparison to our cat H5N1 virus sequences, the previously reported N71S mutation was found in the NA gene (2). This mutation is also present in several other H5N1 virus isolates from both cattle and cats (2,30). The N71S mutation is located in the stalk region between the catalytic head and the transmembrane domain of NA. Although it is far from the catalytic site, unlikely to directly affect substrate specificity, this mutation may introduce a new phosphorylation or glycosylation site that may increase stalk rigidity and indirectly decrease neuraminidase-mediated virion release (33,34). Functional characterization is warranted to determine how N71S affects NA enzymatic activities and overall viral fitness.
In conclusion, we have identified HPAI H5N1 virus infection in a domestic pet cat with chronic illness and performed molecular and pathological analyses. Our study suggests that the immunocompromised patients and their environmental/fomite contacts in clinical settings should be closely monitored to reduce the risk of cat-to-cat and cat-to-human transmission.
## References
1. Nuñez, Ross (2019) "A review of H5Nx avian influenza viruses" *Ther Adv Vaccines Immunother*
2. Chothe, Srinivas, Misra et al. (2025) "Marked neurotropism and potential adaptation of H5N1 clade 2.3.4.4.b virus in naturally infected domestic cats" *Emerging Microbes & Infections*
3. Neumann, Kawaoka (2024) "Highly pathogenic H5N1 avian influenza virus outbreak in cattle: the knowns and unknowns" *Nat Rev Microbiol*
4. Maas, Tacken, Ruuls et al. (2007) "Avian influenza (H5N1) susceptibility and receptors in dogs" *Emerg Infect Dis*
5. Reperant, Van Amerongen, Van De Bildt et al. (2008) "Highly pathogenic avian influenza virus (H5N1) infection in red foxes fed infected bird carcasses" *Emerg Infect Dis*
6. Nidom, Takano, Yamada et al. (2010) "Influenza A (H5N1) viruses from pigs"
7. Burrough, Magstadt, Petersen et al. (2024) "Highly pathogenic avian influenza A(H5N1) Clade 2.3.4.4b virus infection in domestic dairy cattle and cats"
8. Pulit-Penaloza, Brock, Belser et al. (2024) "Highly pathogenic avian influenza A(H5N1) virus of clade 2.3.4.4b isolated from a human case in Chile causes fatal disease and transmits between co-housed ferrets" *Emerg Microbes Infect*
9. Bruno, Alfaro-Núñez, De Mora et al. (2023) "First case of human infection with highly pathogenic H5 avian Influenza A virus in South America: a new zoonotic pandemic threat for 2023?" *J Travel Med*
10. Caserta, Frye, Butt et al. (2024) "Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle" *Nature*
11. Baker, Arruda, Palmer et al. (2025) "Dairy cows inoculated with highly pathogenic avian influenza virus H5N1" *Nature*
12. Briand, Souchaud, Pierre et al. (2022) "Highly pathogenic avian influenza A (H5N1) clade 2.3. 4.4 b virus in domestic cat" *Emerging Infect Dis*
13. Naraharisetti, Weinberg, Stoddard et al. (2024) "Highly pathogenic avian influenza A(H5N1) virus infection of indoor domestic cats within dairy industry worker households -Michigan" *MMWR Morb Mortal Wkly Rep*
14. Marschall, Hartmann (2008) "Avian influenza A H5N1 infections in cats" *J Feline Med Surg*
15. Kuiken, Rimmelzwaan, Van Riel et al. (2004) "Avian H5N1 influenza in cats"
16. Coleman, Bemis (2025) "Avian influenza virus infections in Felines: a systematic review of two decades of literature" *Open Forum Infect Dis*
17. Songserm, Amonsin, Jam-On et al. (2006) "Avian influenza H5N1 in naturally infected domestic cat" *Emerg Infect Dis*
18. Kang, Heo, An et al. (2024) "Highly pathogenic avian influenza A(H5N1) virus infection in cats" *South Korea Emerg Infect Dis*
19. (2025) "Savage Pet Recalls Savage Cat Food Chicken -Large and Small Boxes Because of Possible Bird Flu Health Risk, on U.S. U.S. Department of Health and Human"
20. Spackman (2014) "Animal influenza virus"
21. Nguyen, Schmidt, Haeseler et al. (2015) "IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies" *Mol Biol Evol*
22. Letunic, Bork (2007) "Interactive tree of life (iTOL): an online tool for phylogenetic tree display and annotation" *Bioinformatics*
23. Shu, Mccauley (2017) "GISAID: Global initiative on sharing all influenza data -from vision to reality" *Euro Surveill*
24. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
25. Rimmelzwaan, Van Riel, Baars et al. (2006) "Influenza A virus (H5N1) infection in cats causes systemic disease with potential novel routes of virus spread within and between hosts" *Am J Pathol*
26. Klopfleisch, Wolf, Uhl et al. (2007) "Distribution of lesions and antigen of highly pathogenic avian influenza virus A" *Vet Pathol*
27. Park, Ishinaka, Takada et al. (2002) "The invasion routes of neurovirulent A/Hong Kong/483/97 (H5N1) influenza virus into the central nervous system after respiratory infection in mice" *Arch Virol*
28. Lin, Zhu, Wang et al. (2024) "A single mutation in bovine influenza H5N1 hemagglutinin switches specificity to human receptors" *Science*
29. Kesavardhana, Nataraj, Ashok et al. (2024) "Decoding non-human mammalian adaptive signatures of 2.3. 4.4 b H5N1 to assess its human adaptive potential" *bioRxiv*
30. Pardo-Roa, Nelson, Ariyama et al. (2025) "Crossspecies and mammal-to-mammal transmission of clade 2.3.4.4b highly pathogenic avian influenza A/H5N1 with PB2 adaptations" *Nat Commun*
31. Misra, Gilbride, Ramasamy et al. (2024) "Enhanced diversifying selection on polymerase genes in H5N1 clade 2.3.4.4b: a key driver of altered species tropism and host range expansion"
32. Rzymski (2023) "Avian influenza outbreaks in domestic cats: another reason to consider slaughter-free cell-cultured poultry?" *Front Microbiol*
33. Creytens, Pascha, Ballegeer et al. (2021) "Influenza neuraminidase characteristics and potential as a vaccine target" *Front Immunol*
34. Mcauley, Gilbertson, Trifkovic et al. (2019) "Influenza virus neuraminidase structure and functions" *Front Microbiol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12697156&blobtype=pdf
|
# The complete genome of Escherichia phage Midge
Paige Ripberger, Guadalupe Valencia-Toxqui, Jolene Ramsey
## Abstract
In this report, we describe the isolation and genomic annotation of Escherichia phage Midge. A P2-like myophage, Midge, has a 29,617 bp genome. By nucleotide similarity, Midge is most closely related to Erwinia phage EtG. KEYWORDS bacteriophages, genomics, Escherichia coli E scherichia coli is a commensal bacterium commonly found in the mammalian gut.
E. coli strain 4s is an equine commensal that was isolated from a longitudinal horse microbiome study (1,2). In this report, we describe the isolation and genomic annotation of the 4s-infecting myophage, which we called Midge.
Midge was isolated from an enrichment culture of 10 mL of 0.2 μm-filtered fresh water from Finfeather Lake, Bryan, TX (30°38′55.3″N 96°22′26.3″W) and 100 µL of E. coli 4s grown aerobically in lysogeny broth at 37°C. The phage was propagated three times by the soft agar overlay method (3). DNA extraction from DNase-treated polyethylene glycol-precipitated lysate used the Norgen Biotek kit (#46800) according to manufacturer recommendations. SeqCoast Genomics (Pittsburgh, PA) performed sequencing using the Illumina DNA Prep tagmentation kit and unique dual indexes on a NextSeq2000 platform (300-cycle flow cell), generating 2×150 bp reads. The 1,785,843 total sequence reads were quality controlled with FastQC (www.bioinformat ics.babraham.ac.uk/projects/fastqc), and assembled into a single contig at 109.8-fold coverage using Shovill-SPAdes v1.1.0 with default parameters (4,5). Contig accuracy and completeness were confirmed by PCR and Sanger sequencing using primers listed in the GenBank record (PV287706.1); no terminal repeats were detected. The Midge genome was re-opened before the portal gene according to convention (6), then annotated using the Phage Galaxy and Apollo interfaces (https://phage.usegalaxy.eu), as described in (7). Structural annotation was performed using GLIMMER v3.02 (8), MetaGeneAnnotator v1.0 (9), and GetORFs v19.1.0.0 (10), while tRNA coding regions were predicted by ARAGORN v19.1.0.0 (11). The function of the called genes was predicted by BLAST v2.10.1 (12), InterProScan v5.59 (13), and TMHMM v2.0 (14). Supporting analysis was performed using the HHpred tool (15). All software used the default settings.
The phage Midge has a double-stranded DNA genome with 29,617 bp and 55% G + C content, similar to the 50% G + C content of its 4s host (Table 1). Analysis of Midge's genome predicts a total of 47 protein-coding genes, of which 41 were assigned functional predictions . The Midge lysis proteins include a class II holin (GenBank acc. XRM24217), signal-arrest-release endolysin (XRM24218), and a pair of inner (XRM24220) and outer (XRM24221) membrane spanins. Midge shares nucleotide similarities of 75.3% and 52.3% with Erwinia phage EtG (Genbank acc. MF276773) and Salmonella phage FSL SP-004 (NC_021774). Phylogenetic analysis with Virus Intergenomic Distance Calculator (VIRIDIC) (16) suggests classification within the Peduoviridae. The presence of structural proteins, specifically the tail sheath protein (XRM24230.1) and the tail tube protein (XRM24231.1) and phylogenetic relationships suggest that Midge is a myophage. Midge is identified as a P2-like phage based on its genomic organization and the structural and functional similarities of 31 proteins, although its genome is smaller than the 33,593 bp of phage P2 (Genbank acc. NC_001895). Although Midge and P2 differ in genome size, their shared features and the adaptability of P2-like phages present opportunities for their use in genome engineering (17).
## References
1. Knirel, Prokhorov, Shashkov et al. (2015) "Variations in O-antigen biosynthesis and O-acetylation associated with altered phage sensitivity in Escherichia coli 4s" *J Bacteriol*
2. Golomidova, Efimov, Kulikov et al. (2021) "O antigen restricts lysogenization of non-O157 Escherichia coli strains by Stx-converting bacteriophage phi24B" *Sci Rep*
3. Adams (1956)
4. Community, Abueg, Afgan et al. (2024) "The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update"
5. Seemann (2017) "Shovill: faster SPAdes assembly of Illumina reads"
6. Casjens, Grose (2016) "Contributions of P2-and P22-like prophages to understanding the enormous diversity and abundance of tailed bacteriophages" *Virology (Auckl)*
7. Ramsey, Rasche, Maughmer et al. (2020) "Galaxy and Apollo as a biologist-friendly interface for high-quality cooperative phage genome annotation" *PLoS Comput Biol*
8. Delcher, Bratke, Powers et al. (2007) "Identifying bacterial genes and endosymbiont DNA with Glimmer" *Bioinformatics*
9. Noguchi, Taniguchi, Itoh (2008) "MetaGeneAnnotator: detecting species-specific patterns of ribosomal binding site for precise gene prediction in anonymous prokaryotic and phage genomes" *DNA Res*
10. Cock, Grüning, Paszkiewicz et al. (2013) "Galaxy tools and workflows for sequence analysis with applications in molecular plant pathology" *PeerJ*
11. Laslett, Canback (2004) "ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences" *Nucleic Acids Res*
12. Cock, Chilton, Grüning et al. (2015) "NCBI BLAST+ integrated into Galaxy"
13. Jones, Binns, Chang et al. (2014) "InterProScan 5: genome-scale protein function classification" *Bioinformatics*
14. Krogh, Larsson, Von Heijne et al. (2001) "Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes" *J Mol Biol*
15. Zimmermann, Stephens, Nam et al. (2018) "A completely reimplemented MPI bioinformatics toolkit with a new hhpred server at its core" *J Mol Biol*
16. Moraru, Varsani, Kropinski (2020) "VIRIDIC-a novel tool to calculate the intergenomic similarities of prokaryote-infecting viruses" *Viruses*
17. Cunliffe, Parker, Jaramillo (2022) "Pseudotyping bacteriophage P2 tail fibers to extend the host range for biomedical applications" *ACS Synth Biol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793147&blobtype=pdf
|
# Abstract citation ID: ofaf695.537 P-318. HIV Pre-Exposure Prophylaxis Eligibility and Feasibility in the Health Frontiers in Tijuana Network
Matthew Cappiello, Casey Gaughan, Samvel Gaboyan, Carlos Antonio, Garcia Tovar, Miguel Prado, B Public Health, Jennifer Veltman, Jose Burgos
Background. The Tijuana-California border is a metroplex that integrates multiple risk factors for HIV transmission across borders. However, awareness and uptake of pre-exposure prophylaxis (PrEP) remains low in the region. Health Frontiers in Tijuana (HFiT), a partnership between United States and Tijuana academic training sites for marginalized populations, may serve as an optimal location for PrEP delivery. 1). Anonymized surveys were conducted from January through April 2025 (n = 23) to assess PrEP delivery capacity among HFiT-affiliated physicians and pharmacists (Tables 2 and3). This survey was derived from prior questionnaires on PrEP awareness in health providers. Results. HIV prevalence in this population (3.5%) was over 18 times the estimated national HIV prevalence in Mexico, and more than four times above Baja California background rates. Female gender, use of recreational drugs, and presence of comorbid sexually transmitted infections (p< 0.001) were associated with increased risk for HIV positive test. This data suggests a high number of eligible patients for PrEP. A substantial number of survey respondees (87.6%) felt that PrEP was of high utility in HFiT's patient population. Both physicians (68.8%) and pharmacists (100%) had an adequate understanding of which patients were eligible for PrEP. Education gaps regarding PrEP eligibility were seen in junior providers such as medical students, but less so in senior providers such as residents and attending physicians. Barriers to PrEP adoption appeared feasible to overcome. Provider perception regarding insurance ineligibility (79%) has been mitigated by recent approval of tenofovir disoproxil fumarate-emtricitabine in Mexican public insurance. Concerns about staffing (35%) and lab barriers (60%) can also be mitigated, through provider education materials and identification of discrete sites in the PrEP care continuum (Table 4).
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12791986&blobtype=pdf
|
# P-282. Leveraging Machine Learning and Electronic Health Record Data to Identify Patients at Risk of HIV Care Lapses: A Statewide Analysis in Maryland
Seyed Shams, Chaitrali-Shiri Kher, Colleen Reilly, Divya Hosangadi, Colleen Ennett, Elana Rosenthal, Kristen Stafford
Background. Retention in HIV care is essential to end the HIV epidemic and improve individual patient outcomes, yet 1/3 of people living with HIV in Maryland are not consistently retained in care. Predictive modeling using electronic health record (EHR) data is a promising strategy to identify patients at risk for lapses in care; however, existing efforts remain limited. Regional analyses using statewide health systems can uniquely inform local strategies by accounting for demographic, clinical, and structural variations. This study aimed to develop predictive models utilizing comprehensive EHR data from the University of Maryland Medical System (UMMS) to identify people living with HIV who are at risk of lapsing in care. Methods. Utilizing EHR data from UMMS including HIV-related prescriptions, laboratory tests and clinical visits from adults receiving antiretroviral therapy from 1/ 2016-6/2024, we identified 8,518 patients with 205,633 encounters. Consolidating multiple encounters within 10 days as one encounter, and including only encounters preceding the last recorded encounter for both lapsed and non-lapsed groups, resulted in 4,735 patients. Of those, 2,667 (56%) had lapses -defined as not having an HIV-related clinical encounter within 12 months. We used a Random Forest classifier with extensive hyperparameter tuning via randomized search cross-validation. Model performance was assessed via the performance matrices and SHAP values for feature importance.
Results. The Random Forest model demonstrated an AUC of 0.91 with 98% recall for lapses, and 81% precision (Figure 1). Feature importance analysis highlighted significant predictors, including time on treatment, age, BMI, alcohol use, employment status, racial demographics, and comorbidities such as hypertension and diabetes (Figure 2).
Conclusion. Predictive modeling using EHR data from a comprehensive statewide health system can effectively identify patients at risk for lapses in HIV care. Clinically actionable predictors, such as treatment duration, demographics, and specific health conditions, provide practical insights that can guide interventions to enhance patient retention and ultimately improve health outcomes for people living with HIV.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793288&blobtype=pdf
|
# Abstract citation ID: ofaf695.094 291. Duration of Antibiotic Therapy for Uncomplicated Gram-Negative Bacteremia in Solid Organ Transplantation
William Cappuccio, Emily Heil, Mandee Booth, Ashley Barnes, Kimberly Claeys, Hyunuk Seung, Sara Lee
Background. We hypothesize that short durations (≤10 days) of antibiotics for uncomplicated gram-negative bacteremia (GNB) would have similar clinical outcomes compared to long durations ( >10 days) in solid organ transplant (SOT) patients. Methods. A retrospective cohort review was conducted on patients admitted to University of Maryland Medical System hospitals from 1/1/2019 to 6/30/2024. Included patients were ≥18 years old with an uncomplicated GNB defined as a confirmed Enterobacterales bacteremia, source control, and clinical stability at 72 hours. The primary outcome was a Desirability of Outcome Ranking (DOOR) comparison between < 10 days versus ≥10 days of treatment. Each patient was assigned a DOOR rank from 1 (survival without adverse events) to 5 (death). Ranks 2-4 included patients who survived but had 1, 2, or 3 undesirable events occur during the 30-day follow-up period, respectively. Undesirable events included microbiologic failure, isolation of an MDR gram-negative in the blood, or C. difficile infection. Confidence intervals for the DOOR probability were calculated using the Halperin et al. ( 1989) method with a 95% confidence level, with a statistically significant two-sided p-value of < 0.05.
## DOOR Probability Forest Plot
Results. A total of 105 patients were included in this study, with 27 patients in < 10 day cohort and 78 patients in the ≥10 day cohort. The source of GNB was significantly different between groups (p=0.007), with the < 10 day cohort having more intrabdominal infections (48.2% vs 19.2%) and the ≥ 10 cohort having more urinary sources of infection (67.9% vs 37.0%). There were more patients with an oral switch in the ≥10 day treatment cohort (66.7% vs 33.3%, p=0.003). The overall DOOR probability of having a more favorable outcome with a treatment duration of < 10 days compared to ≥10 day treatment was not statistically different (44.7%, 95% CI: 36.8%, 52.9%; p = 0.1235), highlighting that longer durations of therapy were not associated with improved outcomes.
Conclusion. In patients with a solid organ transplant and uncomplicated GNB, there was no significant difference in outcomes when treated for < 10 days or ≥ 10 days. Further studies with larger sample sizes are needed to highlight the efficacy and safety of shorter antibiotic durations for uncomplicated GNB in solid organ transplantation.
Disclosures. Emily L. Heil, PharmD, MS, Wolters-Kluwer: Advisor/Consultant
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792738&blobtype=pdf
|
# P-1818. Measles Testing in a Large Commercial Laboratory in the United States, 2023-2025 in an Era of Increasing Measles Infections
Charles Walworth, Laura Gillim, Suzanne Dale, Sharon Martens, David Alfego, Ato Aikins
Background. Within the first 4 months of 2025, more than 800 confirmed cases of measles were reported by 22 states in the US. This is more than double the number of cases reported throughout 2024.
Methods. This is a retrospective analysis of measles antibody IgM (IFA/EIA) and IgG (CLIA), and measles PCR (swab and urine) volumes and positivity rates for tests ordered by clinicians between June 2023 through March 2025. De-identified test results were evaluated from a large commercial lab database in the United States. Testing patterns related to age, sex, geography and clinician specialty also were evaluated.
Results. Test volumes for IgM, IgG, and PCR (swab and urine) were 4,772, 1,814,526 and 2221, respectively. Sharp increases in test orders were noted for all measles-related tests beginning in January 2025, although a steady increase in measles IgM was noted beginning in July 2024. PCR testing was 1.39 times more likely to yield a positive result compared to IgM testing (5.0% vs 3.64%, OR=1.39, 95% CI: 1.09 to 1.77), p < 0.05. From January -March 2025, PCR positivity was up 4.5% compared to the same period in 2024. Combined IgM and IgG testing and combined IgM and PCR testing each constituted < 1% of all testing. IgM positivity was highest in the 5-19-year age group (5.1%), whereas PCR positivity was highest between ages 30-39 (6.6%). Overall IgG immunity was 87.4%, with the lowest observed in the < 5-year age group (77%) and the highest in the 66+ year age group (97%). For women of childbearing age (15-49 years), the rate was 86.6%. There were no significant differences in IgG, IgM or PCR positivity rates between sexes. Positivity rates for each geographic region were as follows (IgM vs PCR): South 3.82 vs 6.9, West 2.64 vs 3.09, Northeast 3.79 vs 3.05 and Midwest 2.34 vs 2.38. Testing was ordered mostly by primary care (FP and IM) specialties.
Conclusion. Increases in measles-related testing aligned with recent outbreaks and corresponded to regional positivity rates. Although IgM testing was ordered more than PCR testing, PCR had higher positivity rates. Ordering patterns did not align with current recommendations for the diagnosis of active measles (PCR and IgM) indicating opportunities for provider education. Of note, those >66 years of age demonstrated the greatest level of immunity.
Disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12845413&blobtype=pdf
|
# About Zika Virus (ZIKV). Reply to Weidmann et al. Zika Virus Pathogenicity Versus Transmissibility. Comment on "Roozitalab et al. Distinct Virologic Properties of African and Epidemic Zika Virus Strains: The Role of the Envelope Protein in Viral Entry, Immune Activation, and Neuropathogenesis. Pathogens 2025, 14, 716"
Ashkan Roozitalab, Chenyu Zhang, Jiantao Zhang, Ge Li, Chengyu Yang, Wangheng Hou, Qiyi Tang, Richard Zhao
## 1. African vs. Asian Lineage: Pathogenicity and Host Adaptation
The authors emphasize on the long-standing codon adaptation of African ZIKV lineages to primate hosts, as this evolutionary framework helps explain several characteristics of African strains. Our observations complement this perspective by suggesting that codon adaptation alone does not fully account for the distinct behaviors we observed in human neural cells. In our study, the African strain MR766 exhibited stronger initial binding and higher replication in SH-SY5Y neural cells, while BR15 induced a more prominent innate immune response. These patterns support the idea that African lineage viruses are highly competent in neural cell infection, whereas the resulting cellular outcomes may depend more on host responses and tissue-specific factors than on replication efficiency alone.
## 2. Role of the E Protein in Viral Entry and Neuropathogenesis
We appreciate the commentary's recognition of our findings that link early viral entry and replication dynamics to the envelope (E) protein [1,2]. Our chimeric virus analyses were designed to isolate these effects, and the results suggest that the E protein plays a central role in regulating viral attachment and replication kinetics, while associated structural components such as prM contribute to downstream permissiveness and cytopathic outcomes [1,3]. Although MR766 lacks the N154 glycosylation motif, our experiments indicate that its E protein still mediates efficient neural cell binding. This observation suggests that multiple structural features, including regions within domains II and III, may contribute to strain-specific interactions with neural cells.
## 3. Transmissibility vs. Pathogenicity in ZIKV Lineages
We agree with the authors that viral transmissibility and pathogenicity are shaped by distinct evolutionary pressures and do not necessarily evolve in parallel. As emphasized in the commentary, ecological, vector-related, and population-level factors play central roles in epidemic spread. Although our study did not directly examine transmissibility, available evidence suggests that the enhanced spread of Asian-lineage ZIKV strains, or epidemic success is driven primarily by improved fitness in Aedes mosquito vectors, transmission dynamics, viral persistence, and modulation of host immune responses, rather than by intrinsic replication efficiency or human codon usage alone. Consistent with these observations, our data show that the epidemic BR15 strain induces stronger innate immune activation in human neural cells than the African MR766 strain, which may limit cytopathic effects while allowing sustained viral presence. Together, these findings support the view that transmission efficiency and viral persistence, rather than cellular cytopathogenicity alone, were key contributors to the widespread human impact of Asianlineage ZIKV strains.
## 4. Relevance of MR766 as a Representative African Strain
We acknowledge the commentary's point that MR766 may not fully represent the genetic and phenotypic diversity of African ZIKV strains. At the same time, MR766 remains among the most extensively characterized African isolates, with well-documented neurotropism in experimental systems. This makes it a useful reference strain for mechanistic studies such as ours. We agree that the use of well-characterized reference isolates including a broader range of African isolates in future work would further enrich comparative analyses.
## 5. Interpreting Neurovirulence in the Context of Human Outbreaks
The commentary raises an important distinction between neurovirulence observed in experimental models and the clinical patterns documented during outbreaks. Our findings align with this distinction. While MR766 displayed strong neurotropic and cytopathic properties in vitro, BR15 elicited stronger antiviral and immunomodulatory responses in human neural cells.
While the stronger innate immune activation induced by the epidemic BR15 strain may mitigate direct cytopathic effects and preserve neurosphere integrity despite lower initial replication efficiency, excessive or dysregulated immune responses can also contribute to immune-mediated neuropathology, as has been observed in many encephalitic viral infections. Taken together, our results suggest that host immune activation is a critical modulator of neuropathogenic outcomes and that clinical manifestations likely arise from a complex interplay among viral genetics, transmission dynamics, and host immune responses [4]. This interpretation aligns with the authors' conclusion that epidemic success is unlikely to be driven solely by intrinsic cellular pathogenicity.
## 6. Conclusions
We appreciate the contextual insights provided by Weidmann et al. [5] and regard their commentary as a valuable complement to our mechanistic findings. Taken together, both perspectives highlight the importance of integrating virological, immunological, and evolutionary considerations in understanding ZIKV biology. Our work suggests that the E protein is a major determinant of neural infection, while lineage-specific immune activation patterns influence downstream neuropathogenic outcomes. We hope that our study contributes useful mechanistic context to ongoing efforts aimed at understanding the diverse behaviors of ZIKV lineages.
## References
1. Li, Bos, Tsetsarkin et al. (2019) "The Roles of prM-E Proteins in Historical and Epidemic Zika Virus-mediated Infection and Neurocytotoxicity" *Viruses*
2. Bos, Viranaicken, Frumence et al. (1444) "The Envelope Residues E152/156/158 of Zika Virus Influence the Early Stages of Virus Infection in Human Cells" *Cells*
3. Roozitalab, Zhang, Zhang et al. "Distinct Virologic Properties of African and Epidemic Zika Virus Strains: The Role of the Envelope Protein in Viral Entry, Immune Activation, and Neuropathogenesis. Pathogens 2025"
4. Roozitalab, Zhang, Zhang et al. (2025) "The Evolving Role of Zika Virus Envelope Protein in Viral Entry and Pathogenesis" *Viruses*
5. Weidmann, Faye, Faye (2026) "Zika Virus Pathogenicity Versus Transmissibility. Comment on Roozitalab et al. Distinct Virologic Properties of African and Epidemic Zika Virus Strains: The Role of the Envelope Protein in Viral Entry, Immune Activation, and Neuropathogenesis. Pathogens 2025"
6. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793009&blobtype=pdf
|
Julia Ramponi, Fabián Herrera, Gustavo Torres, ; Maximiliano, Gabriel Castro, Pablo Bonvehí, Specialist, Diego Torres, Agustina Fiori, Nicolás Lasserre, Marcia Querci, Elena Temporiti
Background. BK virus (BKV) reactivation in urine and plasma occurs in 30-60% and 4.5-27% of kidney transplant (KT) recipients, respectively. Of these, 1-5% will develop BKV-associated nephropathy (BKVAN), which can lead to graft loss. BKVAN is defined as histological changes in the graft secondary to BKV reactivation.
Methods. This was a prospective cohort study. KT recipients with BKV reactivation in urine and plasma confirmed by quantitative PCR at a University Hospital (January 2020 -December 2024) were included. Chi-square tests were used for categorical variables and Student's t-test for differences in means and medians; interquartile range (IQR) was reported for continuous variables.
Results. During the study period, 202 KTs were performed, of which 68 (33.7%) had BKV reactivation in urine, and 27 (13.4%) also in plasma. BKV reactivation occurred more than once in 34.1% of cases. Sixteen recipients (23.5%) developed BKVAN. Of these, 81.3% were male. The median age was 53.9 years (IQR 50.5-57.2). Donor type was deceased (75%), unrelated living (5.9%), and related living (19.1%). Induction immunosuppression included corticosteroids (100%), thymoglobulin (86%), mycophenolate (82.8%), tacrolimus (10.9%), and rituximab (4.7%). 6.8% of KT receptors had concurrent CMV infection. Among BKVAN patients, 81.3% experienced rejection and 14.3% lost the graft. The median time from transplant to BKVAN diagnosis was 143 days (IQR 83-853). Recent organ rejection was the only factor associated with BKVAN (p< 0.0001). Clinical presentation included renal insufficiency (56.8%), dysuria (15.9%), hematuria (4.5%), and fever (2.3%). Treatment involved reduction of immunosuppression (77%), regimen changes (40.9%), and leflunomide administration (37%). No hospitalizations or BKVAN-related deaths were recorded.
Conclusion. The incidence of BKVAN was higher than that reported in the literature. Rejection was the only factor associated with BKVAN, suggesting its role in disease progression.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12749113&blobtype=pdf
|
# Five BSC members selected in new cohort of New Cornerstone Investigators
## F
ive distinguished members of the Biophysical Society of China (BSC) have been named to the third cohort of the New Cornerstone Investigator Program, according to the list unveiled by the New Cornerstone Science Foundation on Nov 24, 2025.
Among the 35 scientists recognized this year, the BSC selectees include Yanyi Huang, vice-president of the Single-cell Multiomics Academic Subgroup and vice-president of the Microfluidic System Academic Subgroup, whose research focuses on single-cell multiomics, microfluidic technology, and high-throughput genomic sequencing; Peilong Lu, a committee member of the Molecular Biophysics Academic Subgroup, specializing in protein design, particularly the de novo design of membrane protein antagonists and functional membrane proteins; Xiangxi Wang, secretary-general of the BSC, working in virology and structural biology, with a focus on the precise observation and design of biomacromolecules; Yanli Wang, a member of the BSC, focusing on the structure and function of RNA interference-related proteins and the mechanisms of CRISPR/Cas systems; and Li Yu, a committee member of the Membrane Biophysics Academic Subgroup, who discovered the biological phenomenon of migrasomes and is developing a migrasome-based drug delivery platform.
Their selection demonstrates the strong scientific leadership and innovative capacity of the biophysics community in China and highlights the BSC's ongoing commitment to fostering excellence in frontier research.
The New Cornerstone Investigator Program is an innovative philanthropic funding initiative designed to support fundamental research with a focus on originality and free exploration. In 2022, Tencent committed 10 billion yuan ($1.4 billion) over 10 years to provide long-term and stable support for outstanding scientists working toward groundbreaking innovations.
The program supports creative scientists engaged in high-risk, exploratory basic research projects, aiming to empower them to raise major scientific questions, push disciplinary boundaries and drive original breakthroughs.
It Recipients may apply for renewal upon completion of their terms.
To date, three cohorts have been selected, with a total of 139 exceptional scientists recognized as New Cornerstone Investigators.
## Yanyi Huang
Peilong Lu Xiangxi Wang Yanli Wang Li Yu
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792260&blobtype=pdf
|
# P-255. Receipt of Guideline Concordant Oncologic Care between Older HIV-Positive and Older HIV-Negative Patients with Cancer
Uma Markan, David Riedel, Habib Omari, Hannah Ashe, Jennie Law
Background. With the success of antiretroviral therapy (ART), the population of people living with HIV (PLWH) is aging. The median age of PLWH on ART is projected to be 52 by 2030. PLWH experience worse cancer outcomes and are less likely to receive standard-of-care treatment compared to HIV-negative patients. We conducted a retrospective analysis to assess whether older PLWH (OPLWH, ≥50 years) receive guideline-concordant oncologic care compared to HIV-negative older patients (HNOP).
Methods. A single institution retrospective review in which OPLWH and HNOP diagnosed with diffuse large B cell lymphoma (DLBCL), non-small cell lung cancer (NSCLC), hepatocellular carcinoma (HCC), and cholangiocarcinoma between the years of 2015-2022 were identified using our institutional cancer registry. Guideline concordance was defined as receipt of care aligned with NCCN guidelines from the year of cancer diagnosis. Chi square test was used to compare group characteristics. Multivariable logistic regression models were used to assess factors associated receipt of care. Adjusted odds ratio (aOR) and 95% CI were used to quantify the effect of these associations.
Results. We evaluated 238 patients: 78 OPLWH and 160 HNOP. Median age was 61 (56-66) for OPLWH and 65 (59-71) for HNOP. 83% of OPLWH were Black, compared to 40% of HNOP. Guideline-concordant care was offered to 91% of OPLWH and 93% of HNOP (p=0.446). OPLWH had lower odds of receiving such care (84% vs 91%, p=0.128), though not statistically significant. Among those treated, 86.4% of OPLWH and 93.8% of HNOP tolerated therapy (p=0.08). Compared to publicly insured patients, uninsured patients were less likely to be offered (77.8% vs 92.8%, p=0.134), receive (66.7% vs 91.9%, p=0.06), and tolerate (66.7% vs 91.9%, p=0.06) treatment, though differences were not statistically significant.
Conclusion. There was no significant difference in the provision of guidelineconcordant care between OPLWH and HNOP with DLBCL, NSCLC, or hepatobiliary cancers. Both groups received and tolerated therapy similarly. To our knowledge, this is the first study examining guideline-concordant care in PLWH. These findings suggest progress in delivering evidence-based treatment to OPLWH, with lack of insurance emerging as a potentially greater barrier than HIV status alone.
Disclosures. All Authors: No reported disclosures
Saint Michael's Medical Center, Newark, NJ, USA, Newark, NJ 2 Saint Michaels Medical Center, Newark, New Jersey 3 NJCRI, Newark, New Jersey 4 Northern Jersey Community Research Initiative NJCRI, Newark, New Jersey 1-2 Excess VAT is associated with multiple comorbidities and metabolic syndrome. [1][2][3] Directly measuring VAT through CT scan or DEXA scan can be costly and impractical. 2-3 A potential solution is to use anthropomorphic measurements to predict excess VAT, as illustrated in the VAMOS Study. [3][4] In the VAMOS study, the authors found that Waist Circumference (WC) and Waist-to-Hip Ratio (WHR) are the best predictors for excessive VAT in men. However, the study only had 13 women, and WC was the best predictor, but not WHR. Our study aims to expand on the VAMOS cross-sectional study with a data pool focused on women with HIV.
Poster Abstracts • OFID 2026:13 (Suppl 1) • S299
$$1 2 2 2 2 3 4 1$$
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12777327&blobtype=pdf
|
# Adaptive mutations in HA of avian H9N2 influenza viruses facilitate their transmission to swine
Jia Wang, Peiwen Chen, Qiwei Liu, Sixia Huang, Maocai Wu, Dan'er Wei, Wenshan Hong, Tommy Tsan, Yuk Lam, Huachen Zhu, Yi Guan
## Abstract
The cross-species transmission of avian H9N2 influenza viruses to swine increases the risk of viral adaptation to mammalian hosts. However, the mechanisms by which these H9 viruses can overcome the barriers posed by swine hosts have not yet been fully elucidated. In previous studies, we identified avian H9 strains exhibiting either infective or noninfective phenotypes in swine. Here, we investigated the role of surface genes in cross-species transmission by replacing the surface genes of noninfective strains with those of their infective counterparts. We demonstrated that the surface genes of the infective strains, particularly the hemagglutinin (HA) gene, restored infectivity in pigs for two previously noninfective strains. Surface genes from infective strains significantly increased viral replication efficiency in both CEF and PK15 cells, and recombinant viruses carrying these genes presented superior thermal stability. Amino acid sequence analysis of HA identified six critical residues (30T, 39A, 327R, 373K, 465K, and 490R) associated with infectivity in pigs. Avian H9 viruses bearing these swine-adaptive molecular signatures began emerging in terrestrial poultry before the 2000s and subsequently achieved dominance through widespread dissemination. These findings suggest that molecular changes in HA accumulated during the adaptation of avian H9 viruses in terrestrial poultry may drive the emergence of swine-infective strains. This study elucidates key molecular determinants that enable avian H9 viruses to infect swine and highlights the public health implications of prolonged H9N2 circulation in terrestrial poultry, which facilitates mammalian adaptation. These insights underscore the need for intensified surveillance of avian-to-swine influenza transmission dynamics.
## Introduction
The first recorded instance of the H9 viruses in domestic poultry dates back to 1966 in turkeys in Wisconsin [1]. Since then, these viruses have been sporadically detected in both poultry and wild birds across North America. Owing to the migration of wild birds, H9 viruses were introduced into Korea and China during the 1990s. They subsequently became endemic in poultry throughout Asia, the Middle East, and North Africa over the following three decades [2][3][4]. Currently, H9 is one of the most widespread subtypes of low pathogenic avian influenza A virus globally, and it has also been identified in various mammalian species. To date, the H9 viruses have been detected in humans, swine, dogs, and minks [5][6][7][8]. In addition to cases exhibiting clinical signs of disease, retrospective serological surveys have revealed a relatively high rate of latent infection in human and swine populations [9][10][11]. The broad host range and mild virus-host interactions associated with H9 enhance the survival of H9 viruses in nature and increase the likelihood of genetic exchange between H9 and other influenza virus subtypes [12,13]. The internal genes of H9 viruses contribute to the emergence of H5N6, H7N9, H10N3, and H3N8 viruses that cause human infections [14][15][16][17].
The widespread prevalence of H9 viruses in domestic poultry increases the risk of interspecies transmission to swine, which are considered as the potential "mixing vessels" for influenza virus genes [18]. Since 1998, sporadic reports of swine infected with H9N2 viruses have been published [19,20]. These repeated interspecies transmissions raise concerns that H9 viruses may adapt to humans through the accumulation of molecular changes in swine [21][22][23]. Additionally, there are concerns regarding the potential for the H9 virus to reassort with mammalian influenza viruses [24], which could generate novel pandemic strains, as coinfections of H9N2 with H1N1 and H3N2 have been identified in swine [13,25,26]. Experimental studies have demonstrated that the surface genes of H9 viruses exhibit high compatibility with those of human H1N1 and H3N2 viruses [27,28]. Furthermore, reassortants containing surface genes from H9N2 viruses and internal genes from either H3N2 or H1N1 viruses have shown increased replication efficiency and transmissibility in pigs and ferrets [29]. These findings suggest that the surface genes of the H9N2 virus may possess the essential elements required for infecting swine; nevertheless, efficient aerosol transmission among mammals would require enhancement from internal genes [30,31].
Surface genes play crucial roles in viral tropism, receptor binding, membrane fusion, and virion release. Amino-acid substitutions in these genes can lead to variants that overcome host restrictions. H9 field isolates exhibit spontaneous amino acid mutations, such as Q226L, T137I, I155T, H183N, and A190V in the HA (H3 mature HA numbering throughout the manuscript) [32][33][34][35][36]. These mutations have been experimentally linked to changes in binding preference with human receptors (SAα2,6-Gal) and an increase in viral replication in mammalian cells [33,35,37]. Amino acid substitutions that may facilitate the adaptation of H9 viruses in mammalian hosts were also investigated by serially passaging H9 viruses in animal models such as mice, guinea pigs, ferrets, and pigs. These studies revealed the potential involvement of HA1-D225G, HA1-Q227P, HA2-D46E, NA-27T, and NA-30T in mammalian adaptation [38][39][40]. However, for swine-an animal with a large population, some of which have access to poultry-the determinants of avian H9 viruses related to infectivity in pigs remain unclear. Phylogenetic analyses of the H9 gene revealed three major clades within the Eurasian lineage: the Y439, G1, and Ck-Bei (or Y280) clades. In a previous study, we reported that viruses from the G1 and Y280 lineages were capable of infecting pigs, whereas those originating from the natural reservoir (Y439 lineage) remained noninfective [41]. This finding suggests that the evolution of the H9 gene in land-based birds may have contributed to the introduction of H9 viruses to swine. In light of this, we investigated the role of surface genes in the interspecies transmission of the H9 virus to pigs by replacing the surface genes of infective and noninfective H9 viruses from aquatic and terrestrial sources, respectively.
## Materials and methods
## Viruses
## Generation of recombinants and mutants via reverse genetics
Recombinants were constructed by replacing the surface genes either between ST2030 and SSP177W (recombinants designated reverse genetics [RG]-A, RG-B, RG-C, and RG-D) or between JX7554 and SSP177W (recombinants designated RG-E, RG-F, and RG-H). The gene constellation of each recombinant is shown in Figure 1. However, attempts to rescue RG-G, which was designed to replace the HA gene of JX7554 with that of SSP177W, were unsuccessful after three attempts. Mutants RG-SSP177W-M6 and RG-ST2030-M6, which carry specific amino acids at designated positions (as shown in Figure 5A), were constructed for the HA genes of SSP177W and ST2030. In brief, viral RNA was extracted via the QIAamp Viral RNA Mini Kit (Qiagen). cDNA was synthesized with the Uni-12 primer via the PrimeScript ™ II 1st Strand cDNA Synthesis Kit (TAKARA). Eight full-length genes were amplified with Pfu Ultra ® II Fusion HS DNA Polymerase (Stratagene) and subsequently inserted into the pHW2000 plasmid (provided by Dr. R.G. Webster of St. Jude Children's Research Hospital). The plasmids were sequenced, and only those with sequences identical to those of the parental virus were used to rescue the recombinants. Mutations in the HA genes of ST2030 and SSP177W were introduced via a site-directed mutagenesis kit from Trans Gene, Inc. Eight plasmids, each at a concentration of 1 μg, were incubated with 18 μL of Trans-LT1 (PAN-VERA, Madison, WI, USA) for 45 min. The plasmids were then transfected into cocultured monolayers of human embryonic kidney (HEK) 293T and MDCK cells. After 48-72 h, the culture supernatant was harvested and propagated from 9-to 10-day-old embryonated chicken eggs. The allantoic fluid containing the virus was collected to determine the median tissue culture infective dose (TCID 50 ) and plaque-forming units (PFU). The identities of the recombinants and mutants were confirmed through sequencing.
## Cells, growth kinetics, and thermostability studies
MDCK, 293T, and pig kidney (PK15) cells were obtained from the American Type Culture Collection (ATCC). Primary chicken embryo fibroblast (CEF) cells were isolated from 8-day-old specific pathogen-free (SPF) chicken embryos. MDCK cells were cultured in Eagle's minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), while 293T, PK15, and CEF cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS. For the growth kinetics study, we applied ST2030, JX7554, SSP177W, and the corresponding recombinants to the CEF and PK15 cells at a multiplicity of infection (MOI) of 0.001. In the CEF and PK15 culture systems, trypsin was incorporated at concentrations of 0.1% and 0.15%, respectively. At 12, 24, 48, and 72 h post-inoculation, the supernatants were collected for TCID 50 titration. The thermostability of ST2030, JX7554, and the corresponding recombinants was assessed by incubating the virus continuously for 24 h at 35 °C, 37 °C, 39 °C, 41 °C, and 43 °C. The virus was then applied to MDCK cells for TCID 50 titration at 0, 8, 16, and 24 h post-incubation. Three independent experiments were performed.
## Animals and experimental infections
In total, 36 domestic piglets aged 6-8 weeks were used in the experiment. Prior to infection, the pigs were confirmed to be seronegative (with hemagglutination inhibition [HI] titers < 1:40) for the H1, H3, and H9 viruses through HI assays. Nasal swabs were collected and subsequently inoculated into MDCK cells and embryonated chicken eggs to confirm the influenza-free status. For each recombinant and mutant strain, four pigs were intranasally inoculated with 10 7 PFU of the virus diluted in 1 mL of phosphate-buffered saline (PBS). Nasal swabs were collected daily from 1 to 8 days post-inoculation (dpi), and nasal virus secretion was determined via plaque assays. Rectal temperature and signs of disease were recorded from 0 to 8 dpi. Serum samples were collected at 14 and 28 dpi for detection of antibodies against H9 viruses via HI and viral microneutralization (MN) assays according to protocols recommended by the World Health Organization (WHO).
## Receptor binding preference studies
Chicken and horse erythrocytes were used for hemagglutination (HA) assays to investigate the receptor-binding preferences of ST2030, JX7554, SSP177W, CA07, G1, ML704, and the recombinants. Chicken red blood cells (CRBCs) were suspended in phosphate-buffered saline (PBS) to a final concentration of 0.55%. Horse erythrocytes (HRBCs) were suspended in PBS supplemented with 0.5% bovine serum albumin to a final concentration of 0.7%. For the hemagglutination assay, viruses were serially twofold diluted in a 96-well plate. Subsequently, 50 µL of each diluted virus was then mixed with an equal volume of the respective erythrocyte suspension, and the mixture was incubated at 4 °C to allow complete hemagglutination. The CRBC assay was conducted in U-bottom microtiter plates, whereas the HRBC assay was performed in V-bottom plates. All tests were performed in duplicate.
## Virus elution assay
The HA assay was performed by using viruses with an HA titer of 1:64. Starting from the HA titer of 1:64, the viruses were subjected to serial twofold dilutions. Subsequently, 50 μL of the diluted virus solution was mixed with an equal volume of 0.55% CRBCs in a U-bottom microtiter plate. The plate was held at 4 °C for 30 min for virus adsorption to the CRBCs and then incubated at 37 °C for 12 h continuously. The HA titer was monitored at 30 min, 1 h, and then every 2 h after incubation. A decrease in the HA titer, which coincided with neuraminidase (NA)-mediated virus elution, was observed. The assay was performed in three independent experiments.
## Sequence and protein structure analysis
## Statistical analysis
The data were analyzed via analysis of variance (ANOVA) and Tukey's comparisons test via GraphPad Prism software (version 9). When compared with ST2030 and JX7554, respectively, the recombinant strains with SSP177W surface genes were considered statistically significant differences if p < 0.05, and highly statistically significant differences if p < 0.01.
## Results
Recombinants carrying surface genes from SSP177W were endowed with infectivity in pigs.
Our prior investigation demonstrated that intranasal administration of SSP177W in swine results in viral shedding within the nasal cavity. The shedding titers were observed to range between 10 2 and 10 4 PFU/mL during days 2-6 post-inoculation. Additionally, H9-specific neutralizing antibodies were identified in three pigs, with titers of 1:64, 1:64, and 1:128, respectively. In contrast, no virus was detected in pigs inoculated with either ST2030 or JX7554 [41]. In this study, we exchanged the surface genes of SSP177W and ST2030 and those of SSP177W and JX7554 to investigate the role of surface genes in the infectivity of the H9 virus in pigs. We found that pigs inoculated with RG-A and RG-E, i.e., the reassortant viruses with both HA and NA genes from SSP177W, exhibited virus excretions in the nasal cavity. All pigs in the RG-A group had nasal shedding, with two pigs (Sw2 and Sw3) shedding continuously for 5 days (Figure 1B). The rectal temperatures of these four pigs increased from a baseline of 39.40 ± 0.72 °C (-3 to 0 dpi) to 41.38 ± 0.34 °C (1 to 5 dpi) and returned to normal levels after 6 dpi. All pigs in the study presented a runny nose, with Sw3 displaying the most severe symptoms. Each pig in the RG-A group seroconverted (HI ≥ 1:40), and two pigs (Sw2 and Sw4) were found to have neutralizing antibodies (MN ≥ 1:40) (Table 1). In the RG-E group, Sw2 and Sw4 shed the virus for 2 days after inoculation. Seroconversion was observed in these two pigs, and Sw4 generated H9 neutralizing antibodies. Additionally, Sw1, Sw2, and Sw3 of the RG-E group presented runny nose symptoms, but their rectal temperature remained normal after virus inoculation. In contrast to the RG-A and RG-E groups, pigs inoculated with RG-B and RG-F, which contained the internal genes of SSP177W and the surface genes of ST2030 and JX7554, respectively, yielded negative results in both nasal swab titration and serum antibody detection. Although runny noses were occasionally observed in pigs from these two groups, their temperature remained normal throughout the experiment.
In the recombinant groups containing either the HA or the NA of SSP177W, we found that the inoculation of RG-C, which possessed the HA gene from SSP177W and the remaining genes from ST2030, resulted in a low level of nasal shedding in Sw3 and Sw4 at 1 dpi (Figure 1B). Two pigs (Sw1 and Sw4) in this group seroconverted, but no neutralizing antibodies were detected (Table 1). In the RG-D and RG-H groups, neither viruses nor antibodies were detected, and all pigs remained in normal physical condition throughout the experiment.
## Table 1 Seroconversion of pigs inoculated with the recombinants
a HI or MN titer < 10
## An antibody titer of 40 or greater is regarded as indicative of seroconversion. dpi:days post-inoculation
## Group
Hemagglutination inhibition (HI) titer Microneutralization (MN) titer 14/21 dpi 14/21 dpi
## RG-A RG-B RG-C RG-D RG-E RG-F RG-H RG-A RG-B RG-C RG-D RG-E RG-F RG-H
## RG
## Recombinants harboring surface genes from SSP177W demonstrated superior replication in CEF and PK15
In CEF cells, SSP177W demonstrated significantly better replication efficiency than ST2030 and JX7554 did (p < 0.01) (Figure 2A,B). RG-A, RG-C, and RG-D, which carry either one or two surface genes from SSP177W, achieved higher replication efficiency than the prototype ST2030 (Figure 2A). At 24 hpi, RG-C with only HA from SSP177W presented a significant growth advantage over SSP177W and at 48 hpi over ST2030 (p < 0.0001). In contrast, RG-B, which has internal genes from SSP177W and surface genes from ST2030, exhibited poor replication. The exchange of HA and/or NA between SSP177W and JX7554 resulted in a similar pattern of change in replication efficiency (Figure 2B). The paired HA-NA from SSP177W provided a significant growth advantage to RG-E, whereas the surface genes from JX7554 decreased the growth of RG-F in CEF. These results suggest that surface genes determine the replication efficiency of the recombinants in chicken cells.
In PK15 cells, SSP177W also demonstrated a growth advantage over ST2030 and JX7554 (p < 0.05). The surface genes from SSP177W were found to confer better growth to recombinant viruses RG-A and RG-E relative to their counterparts with surface genes from ST2030 and JX7554, respectively (Figures 2C,D). Additionally, improved growth was observed in RG-C, RG-D, and RG-H, which contained a single surface gene from SSP177W. Interestingly, RG-B and RG-F, which carry internal genes from SSP177W and surface genes from ST2030 or JX7554, presented better replication efficiency than SSP177W. These findings suggest that the internal genes of SSP177W function well with the surface genes from ST2030 and JX7554 in PK15.
## The surface genes of SSP177W conferred better thermostability to the recombinants
The temperature range in the porcine respiratory tract is typically between 36 °C (nasal cavity) and 42 °C (pulmonary artery) [43]. To test the thermostability of the viruses, we incubated the virus aliquots at temperatures of 35, 37, 39, 41, and 43 °C and then titrated the aliquots that were collected at 0, 8, 16, and 24 h post-incubation. In general, all strains presented a similar decrease in titer when incubated at temperatures below 40 °C (Figures 3 A-C, F-H). The prototype viruses (ST2030 and JX7554) were able to maintain 50-60% infectivity after incubation at 40 °C for 16 h, while incubation at 41 °C and 43 °C had an obvious negative effect on viral infectivity. However, the infectivity of the recombinants (RG-A and RG-E), which possessed the HA and NA genes from SSP177W, was better than that of their prototype viruses at temperatures above 40 °C (Figure 3D,E). At 16 h post-incubation, RG-A and RG-E were 20% more infective than their prototype viruses, and RG-E showed a marked survival advantage at high temperatures.
To further investigate whether the HA gene contributes to survival at relatively high temperatures, we analyzed the thermostability of RG-C, which was constructed with the HA of SSP177W and the remaining genes of ST2030. RG-C had better thermostability than RG-A and ST2030 (Figures 3I,J), suggesting that the HA gene from SSP177W alone is sufficient to improve the thermostability of the reassortant.
## Receptor binding preference of SSP177W
By testing the hemagglutination ability of the virus with chicken and horse erythrocytes, we compared the receptor binding preferences of SSP177W, ST2030, JX7554, and the recombinants with those of a panel of viruses with different host adaptation implications. H9 viruses from lineages that are prevalent in terrestrial poultry, including SSP177W, barely bind to horse red blood cells, which possess only α2,3-sialic-acid-linked receptors (Table 2). However, hemagglutination was well observed with these viruses when CRBCs possessing two types of sialic-acid-linked receptors were used. The hemagglutination response pattern of these H9 viruses was similar to that of CA07, which is capable of infecting humans. In contrast, ST2030 and JX7554 hemagglutinated horse erythrocytes as did ML704, showing a similar binding preference to avian influenza viruses circulating in the natural reservoir. Interestingly, recombinant viruses carrying the SSP177W, ST2030, and JX7554 HA genes
## Table 2 Receptor binding preferences of the viruses
The name of each reassortant virus is followed (after the colon) by its gene composition. In the format "XX(Y)-XX(genes)", Y refers to the number of internal genes from the first parent (e.g., ST2030), while the second parent (e.g., SSP177W) specifies the inserted HA/NA surface genes. exhibit red blood cell binding preferences that align with those observed in the corresponding HA parental viruses.
## Host type
## The NA of SSP177W reduced the efficiency of erythrocyte elution in recombinants containing the ST2030 backbone
In the influenza virus replication cycle, HA is responsible for binding to the receptor on the host cell and initiating infection, while the role of NA is to release the progeny virus by cleaving the bonds between HA and sialic acid on the cell membrane. In the porcine infection experiment, the infectivity of RG-C was markedly lower compared with that of RG-A, whereas RG-D exhibited a complete inability to infect pigs. These findings indicate that the compatibility between the functions of NA and HA plays a role in determining viral infectivity in porcine hosts. Here, we investigated the compatibility between the HA and NA genes by assessing the elution efficiency of the recombinants from the CRBCs. Among all the strains tested, ST2030 was found to have a remarkably high elution efficiency (Figure 4). The HA titer of ST2030 started to decrease at 1 hpi and declined rapidly over the next 12 h. In contrast, SSP177W and JX7554 bound well to the CRBC, and their HA titers remained stable throughout the test. Recombinant viruses RG-A, RG-B, and RG-F, which respectively harbor the paired HA and NA gene segments derived from SSP177W, ST2030, and JX7554, demonstrated red blood cell dissociation patterns that were congruent with those observed in their corresponding parental viruses. When comparing elution efficiency between SSP177W and RG-C, and between ST2030 and RG-D, the NA from ST2030 appeared more efficient than that from SSP177W at releasing virus from CRBCs. The elution efficiencies of JX7554, RG-H, and SSP177W were similar, suggesting that the NA genes of JX7554 and SSP177W have similar enzymatic capacities.
## Screening for molecular markers that may be associated with infectivity in pigs
To search for potential molecular signatures of H9 viruses that exhibit infectivity in pigs, we compared the HA amino acid sequences of ST2030 and JX7554 with those of four avian H9 viruses known to infect pigs [41]. Overall, 16 positions with different amino acids between the viruses capable and incapable of infecting pigs were found (Additional file 2). Further analysis was performed using sequences of swine H9 (n = 42). We found that 9 out of the 16 amino acid sites categorized as group B either possessed amino acids situated in positions that were not easily accessible or presented similar side chain motifs (Table 3). Consequently, substitutions at these sites are likely to have a minor effect on the functionality of HA. In contrast, group A presented distinct amino acid characteristics between H9 viruses with the capacity to infect pigs and those lacking this capability. Notably, the presence of arginine at position 327 in viruses exhibiting infectivity in pigs resulted in the introduction of an extra basic amino acid at the P4 position, which lies immediately upstream of the cleavage site. The presence of multiple basic amino acids within the connecting peptides may alter the virus's sensitivity to trypsin, thereby enhancing its infection and replication efficiency. Furthermore, the positioning of the remaining five amino acids within the HA protein suggests that alterations in their properties could influence the interaction between HA1 and HA2, thereby affecting the process of membrane fusion (Additional file 3).
## Amino acid substitutions at six potential determinant sites reversed the infectivity of ST2030 and SSP177W
As presented in our previous study, pigs inoculated with ST2030 exhibited neither viral shedding nor the presence of microneutralization antibodies (MN-Ab) at titers exceeding 1:40. In contrast, inoculation with SSP177W led to nasal viral shedding in four pigs and elicited the production of H9-specific neutralizing antibodies [41]. To test the effects of the potential determinants, RG-ST2030-M6 and RG-SSP177W-M6 were engineered to carry amino acids differing from each other at positions 30, 39, 327, 373, 465, and 490 in the HA and were intranasally inoculated into pigs (Figure 5A). Remarkably, positive shedding was detected in all pigs inoculated with RG-ST2030-M6. Although the shedding titer was low and persisted for only 1-2 dpi, seroconversion, as reflected by the MN titer, was detected in three pigs at 21 dpi, with titers of 1:40, 1:80, and 1:160 (Figure 5B). In contrast, nasal shedding was barely detectable in pigs inoculated with RG-SSP177W-M6 (Figure 5C), indicating that the amino acid substitutions significantly impaired the infectivity of RG-SSP177W-M6 in pigs.
## Distribution dynamics of six potential determinants of H9 viruses detected in poultry, swine, and humans
To investigate whether the amino acids at these positions were selected during the evolution of H9 viruses in nature, we retrieved H9 sequences from the GISAID Epi-Flu ™ database and grouped them by time (before 1990, from 1991 to 2000, from 2001 to 2019, and after 2020) and by host (aquatic birds, terrestrial birds, swine, and humans). The percentages of amino acids at the above positions in viruses circulating in a given time period and host are summarized in Figure 6. Except at position 327, viruses with the same amino acids as those in ST2030 and JX7554 (gray columns) were predominant in both aquatic and terrestrial birds before 1990. During the 1990s, the proportion of viruses that possessed these amino acids decreased, and the proportions of viruses with 30T, 39A, 373K, 465K, and 490R (blue columns) increased.
The downward trend was more pronounced for terrestrial H9 viruses. During the period of 2001-2019, more than 90% of terrestrial H9 viruses and 50% of aquatic H9 viruses contained 30T, 39A, 373K, 465K, and 490R. In contrast, 30I, 39T, 373E, 465N, and 490Q were conserved in the minority of aquatic H9 viruses, and they were rarely found in terrestrial strains after 2000. At position 327, alanine was one of many variants in the early stages of H9 establishment in poultry. However, all the variants were gradually replaced by arginine, which became predominant in terrestrial H9 viruses after 2001. These findings revealed the selection of amino acids during the evolution of H9 viruses in nature, especially during the adaptation of the virus in terrestrial birds. Importantly, these signatures were also observed in the majority of the human and porcine H9 isolates, suggesting that these
## Discussion
Many subtypes of avian influenza viruses can occasionally spill over to swine but not as frequently as H9 [44,45]. A serological survey of 534 swine farms in China revealed that 10.3% of pigs were positive for H9 antibodies [9]. In humans, the number of infected cases has dramatically increased over the last 4 years, bringing the total number of infected cases since 2015 to 133 [5]. The frequent recurrence of the interspecies transmission of the H9 virus to mammals suggests that there may be some factors on the HA gene that contribute to the disruption of mammalian host restriction. In porcine airways, sialic acid α2,6-galactose (SAα2,6-Gal) receptors are abundantly expressed on the epithelium from the nasal cavity to the lobar bronchus, whereas the expression of SAα2,3-Gal receptors increases gradually along the terminal bronchiole toward the alveoli [46,47]. This biological structural basis creates a bottleneck for interspecies transmission: Viruses with a preference for SAα2,6-Gal receptor binding have a greater chance of adhering to and invading the upper respiratory tract. This explained the detection of nasal shedding in pigs inoculated with recombinants carrying the HA gene from SSP177W, which had a receptor binding preference similar to that of the human pandemic strain CA07. Molecular surveillance data have shown that an increasing proportion of H9N2 viruses with binding affinity for SA α2,6-linked receptors are circulating in terrestrial birds [48][49][50][51], suggesting that more H9 viruses have achieved the potential to recognize and bind to mammalian receptors. As shown in many studies, HA-226L and HA-228G are not the exclusive determinants related to the ability of H9 viruses to bind to SA α2,6-linked receptors [22,33]. In the receptor-binding pocket, A190V and D225G, generated during the serial passage of H9 virus in porcine airway epithelial cells and in pigs, were found to have a broader receptor-binding spectrum and better replication efficiency [21][22][23]. However, in this study, the infective strain (SSP177W) and noninfective strains (ST2030 and JX7554) all possessed 190A and 225G, suggesting that these two positions may not be critical factors for infecting pigs. HA-197, which is located at the terminus of the 190 helix, was found to have different amino acids between groups capable and incapable of infecting pigs. Given the structural similarity between arginine and lysine, the K197R mutation was not generated in the present study, leaving its potential role in porcine infection to be further explored.
In addition to the receptor-binding domain, all potential infection molecular markers identified in this study clustered in the HA stem region. A327R introduced an additional basic amino acid at position P4 before the cleavage site. During the evolutionary history of H9 in nature, nil-, mono-, di-, and even tri-basic amino acid connection motifs have been observed [52]. Previous research has demonstrated that H9 viruses possessing di-basic amino acids continue to exhibit low pathogenicity in chickens [53]. However, in the present study, the HA-327R mutation, in conjunction with five additional mutations, facilitated the ability of RG-ST2030-M6 to infect pigs. Conversely, the reversion of these mutations resulted in the loss of infectivity of RG-SSP177W-M6 in pigs. Similarly, an alanine-to-serine substitution at the P5 cleavage site was found to improve the virulence of the H9 virus in mice [54], suggesting that a change in cleavage efficiency may influence the introduction of the H9 virus into mammals. As the emergence of the furin recognition motif (R-X-R/K-R) is critical for the generation of highly pathogenic avian influenza viruses, the molecular evolution at the cleavage site of the H9 virus should be closely monitored. In addition to A327R, the other five potential determinants could be involved in the interaction between HA1 and HA2. Although no molecular marker in this region has been identified in experiments using porcine in vivo or in vitro models, serial adaptation of an H9N2 virus in guinea pigs revealed that a D46E substitution in HA2 could contribute to adaptation. D46E improved the thermostability of the virus and promoted aerosol transmission when it was combined with HA1-Q227P and NP-E434K [38]. This result is consistent with what we found in this study: Six molecular changes in the stem region of HA altered the infectivity of SSP177W and ST2030, and reassortants with HA from SSP177W presented the best thermostability. This is the first study in which amino acid substitutions in the HA stem region were experimentally linked to the infectivity of H9 viruses in pigs.
The prevalence dynamics of these amino acids over the past three decades have shown that these substitutions have been selected by long-term adaptation in terrestrial birds. To date, the vast majority of terrestrial H9 viruses harbor the amino acid residues 30T, 39A, 327R, 373K, 465K, and 490R; however, the functions associated with these residues remain to be elucidated. The long-term prevalence and adaptation of H9 viruses in terrestrial birds has generated and accumulated substitutions that contribute to their infectivity in swine. Moreover, amino acids at these six positions are also present in more than 90% of H9 human viruses, although their role in human infection remains to be investigated. Our findings highlight the risk of the continued prevalence and evolution of H9 viruses in terrestrial birds and urge molecular surveillance of H9N2 viruses for early pandemic preparedness.
This study investigates the influence of surface genes of H9 subtype avian influenza viruses on their infectivity in swine; however, the contribution of the viruses' internal genes in this process remains unexamined. In replication efficiency assays conducted with PK15 cells, the SSP177W strain demonstrated a distinctive pattern characterized by delayed replication during the early phase (12-24 h) followed by a pronounced replication surge in the late phase (48-72 h). Conversely, the recombinant virus RG-B, which harbors the internal genes of SSP177W combined with the surface genes of ST2030, as well as RG-F, containing the surface genes of JX7445, both exhibited significantly enhanced replication efficiency during the early phase relative to the parental SSP177W strain. This phenomenon may be partially attributed to the necessity for wild-type viruses to undergo a period of adaptation to the cellular microenvironment following their isolation and propagation in chicken embryos. Furthermore, it implies that, beyond the role of surface genes, the compatibility of internal genes with porcine host cells and the synergistic interplay between internal and surface genes also influence the H9 subtype avian influenza virus's ability to infect pigs. This insight offers a valuable framework for future comprehensive investigations.
capable and incapable of infecting pigs are highlighted in yellow (in HA1), purple (in HA2) and red (amino acids with different properties). Fuchsia indicates the position of the connection peptide
## References
1. Homme, Easterday (1970) "Avian influenza virus infections. I. Characteristics of influenza A-turkey-Wisconsin-1966 virus" *Avian Dis*
2. Zhou, Li, Chen et al. (2024) "Origin, spread, and interspecies transmission of a dominant genotype of BJ/94 lineage H9N2 avian influenza viruses with increased threat" *Virus Evol*
3. Hu, Zhou, Huang et al. (2017) "Genetic characteristic and global transmission of influenza A H9N2 virus"
4. Li, Adel, Bohlin et al. (2020) "Phylogeographic dynamics of influenza A"
5. (2025) "Regional office for the Western Pacific"
6. Qu, Chen, Chen et al. (2024) "Risk distribution of human infections with avian influenza A" *Front Public Health*
7. Sun, Xu, Liu et al. (2013) "Evidence of avian-like H9N2 influenza A virus among dogs in Guangxi"
8. Peng, Chen, Feng-Xia et al. (2015) "Molecular characterization of H9N2 influenza virus isolated from mink and its pathogenesis in mink" *Vet Microbiol*
9. Ding, Li, Huang et al. (2021) "Infection and risk factors of human and avian influenza in pigs in south China" *Prev Vet Med*
10. De Bruin, Zhang, Ke et al. (2017) "Serological evidence for exposure to avian influenza viruses within poultry workers in southern China" *Zoonoses Public Health*
11. Chauhan, Gordon (2021) "Deciphering transmission dynamics and spillover of avian influenza viruses from avian species to swine populations globally" *Virus Genes*
12. Huang, Yu, Xu et al. (2024) "Long-term co-circulation of multiple influenza A viruses in pigs" *Emerg Microbes Infect*
13. Sun, Cheng, Lam et al. (2022) "Natural reassortment of Eurasian avian-like swine H1N1 and avian H9N2 influenza viruses in pigs" *Emerg Infect Dis*
14. Yang, Sun, Gao et al. (2022) "Human infection of avian influenza A H3N8 virus and the viral origins: a descriptive study" *Lancet Microbe*
15. He, Liu, Gong et al. (2018) "Genetic characterization of the first detected human case of avian influenza A (H5N6) in Anhui Province" *East China. Sci Rep*
16. Lam, Wang, Shen et al. (2013) "The genesis and source of the H7N9 influenza viruses causing human infections in China"
17. Zhu, Yang, Han et al. (2025) "Origin, pathogenicity, and transmissibility of a human isolated influenza A(H10N3) virus from China" *Emer Microbe Infect*
18. Scholtissek, Burger, Kistner et al. (1985) "The nucleoprotein as a possible major factor in determining host specificity of influenza H3N2 viruses"
19. Cong, Pu, Liu et al. (2007) "Antigenic and genetic characterization of H9N2 swine influenza viruses in China" *J Gen Virol*
20. Yu, Zhou, Li et al. (2011) "Genetic diversity of H9N2 influenza viruses from pigs in China: a potential threat to human health"
21. Yang, Punyadarsaniya, Lambertz et al. (2017) "Mutations during the adaptation of H9N2 avian influenza virus to the respiratory epithelium of pigs enhance sialic acid binding activity and virulence in mice" *J Virol*
22. Yang, Lambertz, Punyadarsaniya et al. (2017) "Increased virulence of a PB2/HA mutant of an avian H9N2 influenza strain after three passages in porcine differentiated airway epithelial cells" *Vet Microbiol*
23. Gracia, Van Hoecke, Saelens et al. (2017) "Effect of serial pig passages on the adaptation of an avian H9N2 influenza virus to swine" *PLoS One*
24. Cui, Che, De Jong et al. (2022) "The PB1 gene from H9N2 avian influenza virus showed high compatibility and increased mutation rate after reassorting with a human H1N1 influenza virus" *Virol J*
25. Peiris, Guan, Markwell et al. (2001) "Cocirculation of avian H9N2 and contemporary "human" H3N2 influenza A viruses in pigs in southeastern China: potential for genetic reassortment?" *J Virol*
26. Jallow, Barry, Fall et al. (2023) "Influenza A virus in pigs in Senegal and risk assessment of avian influenza virus (AIV) emergence and transmission to human" *Microorganisms*
27. Wan, Sorrell, Song et al. (2008) "Replication and transmission of H9N2 influenza viruses in ferrets: evaluation of pandemic potential" *PLoS One*
28. Sun, Qin, Wang et al. (2011) "High genetic compatibility and increased pathogenicity of reassortants derived from avian H9N2 and pandemic H1N1/2009 influenza viruses" *Proc Natl Acad Sci U S A*
29. Gracia, Van Den Hoecke, Richt et al. (2017) "A reassortant H9N2 influenza virus containing 2009 pandemic H1N1 internal-protein genes acquired enhanced pig-to-pig transmission after serial passages in swine" *Sci Rep*
30. Sorrell, Wan, Araya et al. (2009) "Minimal molecular constraints for respiratory droplet transmission of an avian-human H9N2 influenza A virus" *Proc Natl Acad Sci U S A*
31. Zhu, Zeng, He et al. (2024) "Reassortant H9N2 canine influenza viruses containing the pandemic H1N1/2009 ribonucleoprotein complex circulating in pigs acquired enhanced virulence in mice" *Virology*
32. Liu, Lai, Li et al. (2016) "Endemic variation of H9N2 avian influenza virus in China"
33. Teng, Xu, Shen et al. (2016) "A single mutation at position 190 in hemagglutinin enhances binding affinity for human type sialic acid receptor and replication of H9N2 avian influenza virus in mice" *J Virol*
34. Guo, Wang, Zhao et al. (2021) "Molecular characterization, receptor binding property, and replication in chickens and mice of H9N2 avian influenza viruses isolated from chickens, peafowls, and wild birds in eastern China" *Emerg Microbes Infect*
35. Zou, Zhang, Li et al. (2013) "Molecular characterization and receptor binding specificity of H9N2 avian influenza viruses based on poultry-related environmental surveillance in China between" *Virology*
36. Ma, Ren, Shi et al. (2025) "Isoleucine at position 137 of haemagglutinin acts as a mammalian adaptation marker of H9N2 avian influenza virus" *Emerg Microbes Infect*
37. Sun, Belser, Pulit-Penaloza et al. (2024) "Dissecting the role of the HA1-226 leucine residue in the fitness and airborne transmission of an A(H9N2) avian influenza virus" *J Virol*
38. Sang, Wang, Ding et al. (2015) "Adaptation of H9N2 AIV in guinea pigs enables efficient transmission by direct contact and inefficient transmission by respiratory droplets" *Sci Rep*
39. Liu, Chen, Huang et al. (2014) "A nonpathogenic duck-origin H9N2 influenza A virus adapts to high pathogenicity in mice" *Arch Virol*
40. Kimble, Sorrell, Shao et al. (2011) "Compatibility of H9N2 avian influenza surface genes and 2009 pandemic H1N1 internal genes for transmission in the ferret model" *Proc Natl Acad Sci U S A*
41. Wang, Wu, Chen et al. (2016) "Infectivity and transmissibility of avian H9N2 influenza viruses in pigs" *J Virol*
42. Guex, Peitsch, Schwede (2009) "Automated comparative protein structure modeling with SWISS-MODEL and Swiss-PdbViewer: a historical perspective"
43. Hanneman, Jesurum-Urbaitis, Bickel (2004) "Comparison of methods of temperature measurement in swine" *Lab Anim*
44. Abente, Gauger, Walia et al. (2017) "Detection and characterization of an H4N6 avianlineage influenza A virus in pigs in the Midwestern United States" *Virology*
45. Bourret (2018) "Avian influenza viruses in pigs: an overview" *Vet J*
46. Nelli, Kuchipudi, White et al. (2010) "Comparative distribution of human and avian type sialic acid influenza receptors in the pig" *BMC Vet Res*
47. Iwatsuki-Horimoto, Nakajima, Shibata et al. (2017) "The microminipig as an animal model for influenza A virus infection" *J Virol*
48. Matrosovich, Krauss, Webster (2001) "H9N2 influenza A viruses from poultry in Asia have human virus-like receptor specificity" *Virology*
49. Li, Shi, Guo et al. (2014) "Genetics, receptor binding property, and transmissibility in mammals of naturally isolated H9N2 avian influenza viruses"
50. Bi, Li, Li et al. (2020) "Dominant subtype switch in avian influenza viruses during 2016-2019 in China" *Nat Commun*
51. Yang, Li, Sun et al. (2025) "Genetic diversity of H9N2 avian influenza viruses in poultry across China and implications for zoonotic transmission" *Nat Microbiol*
52. Parvin, Schinkoethe, Grund et al. (2020) "Comparison of pathogenicity of subtype H9 avian influenza wild-type viruses from a wide geographic origin expressing mono-, di-, or tri-basic hemagglutinin cleavage sites" *Vet Res*
53. Soda, Asakura, Okamatsu et al. (2011) "H9N2 influenza virus acquires intravenous pathogenicity on the introduction of a pair of di-basic amino acid residues at the cleavage site of the hemagglutinin and consecutive passages in chickens" *Virol J*
54. Sun, Tan, Wei et al. (2013) "Amino acid 316 of hemagglutinin and the neuraminidase stalk length influence virulence of H9N2 influenza virus in chickens and mice" *J Virol*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12823530&blobtype=pdf
|
# EDITED AND REVIEWED BY
Michael Kogut, Inkar Castellanos-Huerta, Victor Petrone-Garcia, Marco Ju Árez-Estrada, Benjamín Fuente
## Abstract
KEYWORDSbovine viral diarrhea virus (BVDV), HPAI H N , One Health (OH), porcine reproductive and respiratory syndrome virus (PRRSV), recombinant vaccines Editorial on the Research Topic High-impact respiratory RNA virus diseases, volume II
Infectious respiratory diseases pose a significant economic and social risk to the veterinary industry, particularly viral diseases. Taken together, these findings underscore the urgent need to strengthen surveillance, improve diagnostic tools, and provide a coordinated One Health response. Their relevance has recently increased due to the emergence of new viral variants of infectious diseases in species such as swine, cattle, and poultry, as well as in small animals. This increase in the prevalence and incidence of RNA respiratory viruses has driven the development of new strategies for study, diagnosis, control measures, and vaccination, opening new areas of research and the implementation of multidisciplinary approaches.
This editorial covers several highly relevant topics, including the evolution of one of the most critical pathogens in cattle production systems, bovine viral diarrhea virus (BVDV). It addresses the impact of massive NS2 gene deletions on viral persistence. Recent analysis of 21 field isolates of the BVDV-2 genotype reveals that this virus generates a considerable number of viral genomes with deletions (DelVGs), especially in the NS2 region, where more than 90% of the deletions are recorded. The most notable finding is that BVDV2c strains can generate DelVGs up to 150 times higher than those of BVDV2a strains. Furthermore, cytopathic strains produce twice as many DelVGs as non-cytopathic strains. These NS2 gene deletions could play a fundamental role in viral persistence, a phenomenon essential for generating persistently infected (PI) animals. These results expand our understanding of BVDV's genetic plasticity and immune evasion mechanisms, demonstrating that the virus creates diversity within the host before establishing a persistent infection (Holthausen et al.). A key issue in the study of viral diseases is the development of more accurate and efficient diagnostic methods, particularly for bovine respiratory syncytial virus (BRSV), another critical area of animal production. Multiplex RT-qPCR for BRSV and BPIV3 is implemented as a diagnostic tool in this case. This assay presents an LOD95 of 164 copies for BRSV and 359 for BPIV3, with high efficiency and specificity. It can detect approximately 2.4 times more BRSV than viral isolation. This offers an ideal solution for high-demand diagnostic laboratories (Zulauf and Pastey).
Regarding viruses affecting the swine industry, one highly relevant respiratory RNA virus stands out: porcine reproductive and respiratory syndrome virus (PRRSV). PRRSV continues to evolve at an alarming rate, with the discovery of a new recombinant HP-PRRSV strain in Jiangxi, China (NC2023). This strain exhibits characteristics such as belonging to lineage 8 of PRRSV-2 and a genome of 15,321 nucleotides, with recombinant regions derived from the JXA1, JXA1-R, and HUN4 strains. This study, particularly regarding the genome, reports 15 unique amino acid mutations and a 395-amino-acid frameshift in the Nsp2 protein, suggesting increased evolutionary complexity of PRRSV and evidence of the simultaneous circulation of multiple viral lineages capable of efficiently exchanging genetic segments (Wang et al.). This underscores the importance of studying this viral type due to its extreme evolutionary capacity and its potential future impact on the swine industry. One of the most relevant topics today is the use of biotechnology to develop recombinant vaccines, which enable vaccination under safe protocols and an efficient immune response. However, as a study presented in this edition shows, the risk of using recombinant vaccines, such as the MLV (L5A) vaccine virus for PRRS, is being redefined. It's recombination between field viruses generated highly virulent strains, such as the GX2024 isolate, which exhibits severe disease and mortality rates of up to 100% in 4-week-old piglets within 14 days (Gao et al.). Recombination between wild-type strains and modified live vaccines to generate extremely pathogenic viruses represents a critical biosecurity issue on pig farms. Among the possible strategies for controlling PRRSV, the use of zinc sulfate in an in vitro model (Yang et al.) is being investigated. This demonstrates a reduction in viral replication in the Marc-145 cell model, attributed to the reduction of reactive oxygen species (ROS) and malondialdehyde (MDA), the increase in superoxide dismutase (SOD) and catalase (CAT) activities, and the modulation of the proinflammatory cytokines IL-6, IL-8, and TNF-α, with a concomitant increase in the anti-inflammatory cytokine IL-10, as well as a reduction in the activation of apoptosis-related proteins. Therefore, these results suggest a potential therapeutic role for zinc sulfate as an immune modulator and antioxidant during PRRSV infection. Finally, epidemiological analysis is explored, demonstrating its relevance to these viral types through fluid tests, primarily saliva samples from piglet mortality. These tests support the use of tongue fluids from stillborn piglets as a rapid and low-cost surveillance tool (Machado et al.).
On the other hand, another highly relevant respiratory RNA virus mentioned in this edition is avian influenza. The emerging threat posed by HPAI H5N1 clade 2.3.4.4b and its transmission to pigs is explored, as well as the recent detection of the B3.13, D1.1, and D1.2 genotypes of the virus in dairy cows, poultry, wild birds, wild mammals, humans, and, recently, pigs, indicating a significant shift in the virus's ecology. Pigs, recognized as "mixing vessels" for influenza, constitute a potential bridge to the human population. However, knowledge gaps persist, including the lack of systematic studies in pigs, uncertainty about their actual susceptibility, the still-unexplored impact on production, and the risk of recombination with endemic swine influenza. The review emphasizes the urgent need to implement active surveillance and integrated control strategies under the One Health framework (Mena Vasquez et al.). In the same context, other authors analyze a new canine influenza H3N2 virus isolate from China, which revealed amino acid substitutions similar to those observed in human viruses. This finding indicates a risk of interspecies transmission and the potential for the virus to adapt progressively to humans. Therefore, continuous surveillance of companion animals is required Genetic (Li et al.).
The study of alternative diagnostic methods has also provided insights to avoid potential errors or failures in these promising new diagnostic tools, such as anti-dsRNA antibodies, which have limited potential as universal viral detection tools. Despite initial promises of universality and independence from the viral genome, studies have revealed significant limitations in the application of specific diagnostic tools (de le Roi et al.). Inconsistent detection, incomplete localization of viral antigens, and detection even in uninfected tissues have been observed, compromising their reliability as a universal diagnostic tool. Consequently, exploring new markers based on interferon cascades is suggested as a more robust alternative. In environmental virology, a study of 600 pregnant women found a correlation between ozone exposure during pregnancy and a 40% reduction in the risk of c infection in the third trimester. Additionally, the quartile with the highest exposure showed a 99% lower probability of disease (Zhang et al.). While these results are promising, the study authors emphasize caution due to ozone toxicity and the uncertainty surrounding the mechanisms underlying this association.
The review of the presented studies reveals a complex landscape in virology. BVDV demonstrates an extraordinary capacity to generate defective genomes, which contributes to its persistence and immune evasion. PRRSV continues to recombine and evolve into more pathogenic forms, posing a constant challenge to disease control. The proximity of H5N1 to pigs reinforces concerns about the possibility of an avian pandemic, given the virus's high pathogenicity and its potential for interspecies transmission. In this context, emerging diagnostic and therapeutic tools offer a glimmer of hope. The development of markers based on interferon cascades, as mentioned previously, represents a significant advance in the early diagnosis of viral infections. Likewise, research into new antiviral therapies and vaccines is essential to mitigate the impact of emerging viral diseases. Environmental factors, such as ozone exposure, can also unexpectedly influence viral susceptibility, underscoring the importance of a holistic approach to viral disease research and control. This approach, known as One Health, integrates human, animal, and environmental health to address global health challenges more effectively.
In conclusion, the findings presented in this review call for action on several fronts. It is imperative to strengthen biosecurity measures to prevent the spread of viral diseases, increase molecular surveillance to detect and characterize new viral strains, and invest in the development of effective vaccines and antiviral therapies. Furthermore, adopting a One Health approach is crucial to comprehensively address global health challenges and prepare production systems for constantly evolving viral threats.
## References
1. Bf "Writing -review & editing, Validation. IC-H: Data curation, Methodology, Visualization, Conceptualization, Project administration, Validation, Investigation, Software, Supervision, Writing -original draft, Funding acquisition, Resources, Writing -review & editing"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793539&blobtype=pdf
|
# P-2199. Impact of Ringer's Lactate versus Other Fluids in the Hemoconcentration of Children and Adults with Dengue: a Single-Arm Meta-Analysis and Systematic Review
Sophia Costa, Leticia Campos, Gisella Carpi, Jose Luis Boene, Thiago Netto, ; Oscar, Hernández Rios, Taniela Bes
Gelafundin vs. RL (n=2): RL showed a favorable ΔHct (MD = 3.72%; 95% CI: 2.71-4.73; p < 0.01) with moderate heterogeneity (I² = 52.9%, p = 0.15). 0.9% Saline vs. RL (n=2): No significant difference was observed (MD = 0.84%; 95% CI: -0.16 to 1.85; p = 0.10; I² = 0%)
Conclusion. RL demonstrated superior efficacy in improving hemoconcentration compared to Dextran (Figure 1) and Gelafundin (Figure 2). Its performance was comparable to 0.9% saline (Figure 3). These findings support continued use of RL as the WHO-recommended first-line fluid for dengue management. Further trials with standardized protocols are warranted to optimize fluid resuscitation strategies in DF.
Disclosures. All Authors: No reported disclosures
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793431&blobtype=pdf
|
Kai Sha, Samuel Jauregui, Christos Petropoulos, Terri Wrin
## Abstract
Background. In early 2024, a high-pathogenicity avian influenza (HPAI) H5N1 began spreading from wild birds to domestic fowl and eventually livestock and wildlife in the United States. Human infections have been limited to individuals exposed to infected animals. Concurrently seasonal viruses (H1N1, H3N2, B) are circulating at high levels causing significant morbidity.A high-throughput neutralizing antibody (nAb) assay platform has been developed that is capable of conducting 6000 assays per week and can be utilized for in-depth and widespread surveillance of natural infections, as well as the evaluation of seasonal and pandemic influenza vaccine responses. This assay employs a well-established pseudovirus assay platform that limits virus replication to a single cycle. As a result, it enables the evaluation of nAb responses to highly contagious and/or pathogenic influenza strains, such as the H1N1, H3N2, B strains and H5N1, H7N9.
Illustration of influenza neutralizing antibody (nAb) assayComprehensive panel of influenza pseudovirus with number of strains per subtype.
Methods. HIV pseudovirus stocks that carry a luciferase reporter gene and express influenza HA and NA proteins are generated and incubated with serial dilutions of sera/plasma, mAb or purified antibodies, followed by inoculation of HEK293 target cells (Fig. 1). Pseudovirus infection that is not neutralized results in luciferase production, while neutralization reduces or prevents luciferase production.
Neutralization of H1N1 and H5N1 pseudoviruses by a panel of randomly selected human sera.
## Neutralization antibody titers from Table 2
Results. A comprehensive panel of pseudovirus stocks has been assembled (Table 1). The panel includes H5N1 vaccine strains, human and avian H5N1 field strains, seasonal influenza strains such as H3N2, H1N1 and influenza B. Testing has been conducted using a broad panel of animal antisera and mAb against multiple H5N1 and H1N1 strains demonstrating strain-specific responses with limited cross-reactivity. Interrogation of a panel of 20 human sera collected from 2020-2024 revealed measurable nAb titers directed against historical and seasonal strains (Table 2 and Fig. 2). However, cross neutralization of 2024 H5N1 human and avian isolates were also observed.
Conclusion. Sera of humans previously exposed to seasonal influenza and/or immunization possess a limited ability to neutralize avian H5N1. The impact of low titer H5N1 responses on zoonotic events leading to widespread human to human transmission is uncertain.
Disclosures.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12828963&blobtype=pdf
|
Václav Hönig
## Abstract
Background The Usutu virus (USUV; Orthoflavivirus, Flaviviridae) is a mosquito-borne pathogen causing fatal neuroinfections in susceptible wild and captive birds, particularly blackbirds, other passerines, and owls. Zoological gardens provide favourable conditions for the circulation of such viruses due to the proximity of diverse species and limited options for prevention.Methods Following the sudden death of several Chilean tinamous kept in the Brno zoological garden, we tested tissues sampled from 22 bird cadavers (from zoos, private owners, and free-living birds) for the presence of USUV and West Nile virus (WNV) RNA using duplex reverse transcription qPCR. Near-complete whole-genome sequences were acquired from positive samples by next-generation sequencing and subjected to phylogenetic analyses. Furthermore, serum samples from additional zoo animals and privately owned birds were screened for anti-flavivirus antibodies using ELISA and subsequently confirmed by the virus neutralization test.
ResultsWe report fatal USUV infections in multiple bird species from three zoological gardens in the Czech Republic. Duplex RT-qPCR targeting USUV and West Nile virus (WNV) detected USUV RNA in tissues from two Boreal owls (Aegolius funereus), one Eurasian pygmy owl (Glaucidium passerinum), two Brahminy starlings (Sturnia pagodarum), and four Chilean tinamous (Nothoprocta perdicaria). Additionally, three randomly found cadavers of free-living blackbirds (Turdus merula) tested positive. Pathological findings ranged from minimal pathological changes to pronounced hepatosplenomegaly with intestinal bleeding. Phylogenetic analysis of near-complete genome sequences assigned all viruses to the Europe 2 genetic lineage, revealing partial geographic clustering among isolates obtained in this study. Serological testing confirmed exposure in additional birds and demonstrated cross-neutralisation between anti-USUV and anti-WNV-positive sera.
ConclusionsIn zoological gardens, flavivirus infections can cause substantial losses, even among rarely bred or endangered species. Given the zoonotic potential of both USUV and WNV, documenting their occurrence in avian hosts is important not only for animal health but also for human disease surveillance from the One Health perspective.
First report of Usutu virus fatal infections in Chilean tinamous (Nothoprocta perdicaria), brahminy starlings (Sturnia pagodarum), and multiple other bird species in zoological gardens and wildlife in the Czech Republic Jan Kamiš 1,2 , Veronika Grymová 3 , Petr Suvorov 4 , Luc Tardy 1 , Petr Vrána 5 , Jan Kirner 6 , Soňa Peková 7 , Vladimír Piaček 8 , Miša Škorič 9 , Jan Pokorný 10 , Natalie Rudenko 11 , Martin Palus 1,12,13 and Václav Hönig 1,12,14*
## Background
The Usutu virus (USUV) is a mosquito-borne flavivirus (Flaviviridae, Orthoflavivirus) first described in 1959 in Aedes neavei mosquitoes in southern Africa [1,2]. The virus was likely introduced to Europe as early as the 1950s, first to Spain and subsequently at least twice more via migratory birds [3,4]. It is currently present in most western, southern, and central European countries including the Czech Republic, with multiple genetic lineages known to (co-)circulate [1,[5][6][7]. Culex pipiens is considered the principal vector of USUV in Europe [8].
The introduction of USUV to new geographical areas is often associated with high mortality in susceptible wild bird species, particularly blackbirds, other passerines, and owls, as well as in captive birds, including exotic species kept in zoological gardens [9][10][11][12]. Fatal flavivirus infections in endangered species or trained falconry birds can result in considerable losses, including financial ones. Prevention options are limited, as no targeted vaccine or treatment is approved for use in birds [13,14].
Beyond its veterinary impact, USUV is of increasing zoonotic concern. By 2022, 112 acute human infections had been reported in Europe, including 30 cases of neuroinvasive disease [15]. However, the true number of symptomatic cases is likely underestimated, as mild cases may remain undiagnosed or be misdiagnosed as West Nile virus (WNV) or tick-borne encephalitis virus (TBEV) infections, due to serological cross-reactivity among these flaviviruses [5,16].
Here, we report a series of fatal USUV infections in wild birds and birds housed in zoological gardens in the Czech Republic, including the first cases in Chilean tinamous (Nothoprocta perdicaria), a species rarely bred in Europe and in Brahminy starlings (Sturnia pagodarum).
## Materials and methods
## Sampled birds
The samples were obtained from randomly found cadavers of wild birds with no other apparent cause of death, or from captive-bred birds (zoological gardens or private owners) between June and September 2024. One additional free-living blackbird was found and sampled in August 2022. The geographic localization of the zoological gardens as well as the locations where the cadavers of free-living birds were found are depicted in Fig. 1. Detailed data of sampled individuals are listed in the Supplementary Table 1.
## Sample processing and screening for USUV and WNV RNA
Bird cadavers were dissected in biohazard boxes under sterile conditions. Each organ was removed using sterile tools to prevent cross-contamination between samples. Collected samples (brain, heart, kidney, liver, lungs, spleen or a mixture of the tissues, when individual organs were unavailable) were homogenized to create 20% (w/v) suspensions in C6/36 cell culture medium (L15, 10% fetal bovine serum, 5% tryptose phosphate broth, 1% l-glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin) using 5 mm sterile stainless steel beads in Tissue Lyzer II (Qiagen, Hilden, Germany). RNA was isolated from 200 µl of the cleared homogenate using MagMAX Viral/ Pathogen Nucleic Acid Isolation Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA) in KingFisher Apex system (Thermo Fisher Scientific, Waltham, Massachusetts, USA). USUV and WNV RNA were detected in 5 µl of extracted RNA using a duplex one-step reverse transcription quantitative polymerase chain reaction (RT-qPCR) using SuperScript III One-Step RT-PCR System with Platinum™ Taq DNA Polymerase (Invitrogen, Carlsbad, California, USA) and targeting a portion of NS5 coding sequences using published sets of primers and probes [17,18] (Supplementary Table 2). Synthetic single-stranded standards for each of the detected viruses were included to generate calibration curves which allowed absolute quantification of the targets.
## Next generation sequencing and phylogenetic analyses
From all RT-qPCR-positive individuals, the sample with the lowest Ct value was selected for next generation sequencing. Tiled amplicon sequencing [19], based on a previously developed USUV-specific primer panel [20] was performed on a MinIon Mk1B device (Oxford Nanopore Technologies, Oxford, United Kingdom). Briefly, viral RNA was reverse transcribed using LunaScript RT SuperMix (New England Biolabs, Ipswich, Massachusetts, USA) with random hexamer priming, followed by multiplex PCR in two pools to produce 32 overlapping amplicons of approximately 500 bp each. The amplicons were pooled, cleaned using AMPure XP beads (Beckman Coulter Life Sciences, Indianapolis, Indiana, USA), and 200 fmol used for sequencing library preparation. Samples were barcoded and adapter-ligated according to the Oxford Nanopore Technologies Ligation sequencing amplicons -Native Barcoding SQK-NBD114.24 protocol (NBA_9168_v114_revR_30Jan2025) and run on a R10.4.1 (FLO-MIN114) flow cell (Oxford Nanopore Technologies, Oxford, UK) with a target yield of 20 K pass-reads per barcoded sample. Basecalling and demultiplexing of sequencing data was conducted in MinKNOW 24.11 using the Dorado Super High Accuracy 4.3 basecalling algorithm, quality control, filtering, and assembly in Linux (Ubuntu v. 22.04.3 LTS) using samtools and FastQC, followed by the ARTIC pipeline for trimming, polishing, reference-guided assembly to the USUV reference genome NC_006551.1. Consensus sequences were generated using Geneious Prime v.2025.1.3 (Dotmatics, Boston, Massachusetts, USA). In addition to the newly detected USUV strains, we sequenced three USUV strains (202TM10, 208TM10, and 264TM10) from the brains of infected blackbirds reported previously [7]. The sequences were submitted to the GenBank database under accession numbers PX210786-PX210800. The sequences were aligned with sequences representing different USUV genetic lineages obtained from the Gen-Bank database and truncated to uniform length (10,936 nt). GTR + R substitution model was selected as best suitable using Smart Model Selection tool [21]. Maximum likelihood [22] and Bayesian inference [23] phylogenetic trees were generated in Geneious Prime 2025.1.3.
## Virus isolation in cell culture and immunofluorescent staining of viral antigens
USUV strains were isolated from the brains of positive individuals using the C6/36 mosquito cell culture, as previously described [7]. Viral titres were determined by plaque assay as described previously [24] using A549 cell culture (human pulmonary epithelial cells, ATCC, CCL-185). The presence of viral antigen in the infected mosquito C6/36 and mammalian A549 cell lines was confirmed by flavivirus-specific immunofluorescence staining [7]. The cell infection rate was estimated in two independent experiments for each of the cell lines (1000 individual cells counted per cell line and replicate).
## Serological analyses
Serum samples taken from Mikado pheasants (Syrmaticus mikado) kept in a cage neighboring to infected Chilean tinamous in Brno Zoo and additional archived serum samples from several Czech zoological gardens kindly provided by Dr. Natalie Rudenko (Laboratory of Molecular Ecology of Vectors and Pathogens, Institute of Parasitology, Biology Centre CAS) (Supplementary Table 3) were screened for anti-flavivirus antibodies using ELISA (ID Screen Flavivirus Competition ELISA, Innovative Diagnostics, Grabels, France), following manufacturer's instructions. Positive and borderline samples were subsequently tested using virus neutralization tests with USUV (strain 277TM10, lineage Europe 2; BCCO 50_0521), WNV (WNV EG-101) [25], and TBEV (European subtype strain Hypr; BCCO 50_0259) from the Biology Centre Collection of Organisms (www.bcco.cz) as described previously [26]. Briefly, the sera were inactivated for 30 min at 56 °C, diluted in a 96-well plate in cultivation media to reach the final dilution (including the volume of the added virus) of 25x, 50x, 100x, 200x, 400x, and 800x for USUV and TBEV and 40x, 80x, 160x, 320x, 640x, and 1280x for WNV. Individual viruses were added 50 PFU per well and incubated 90 min at 37 °C in 5% CO 2 . Then, 50,000 A549 cells per well were added and incubated for 5 days at 37 °C in 5% CO 2 . Subsequently, the plate was washed in sterile phosphate-buffered saline and stained in naphthalene blue. The neutralizing antibody titres were expressed as the reciprocal value of the serum dilution leading to 50% reduction in cytopathic effect (CPE) compared to controls. Sera reacting in dilutions > 40 were considered neutralizing. Four-fold differences between the cross-neutralizing antibody titres were considered proof of specific neutralizing capacity [27,28].
## Results
## Pathological findings
In general, the birds in the zoological gardens exhibited common signs, such as weakness, tremor, and locomotion disorders. Neurological signs developed fully in last few days before death or euthanasia. Necropsy findings were significantly variable, even when comparing different individuals of the same species (N. perdicaria) ranging from slightly enlarged marbled spleen and/or liver, to pronounced pathological changes, including cachexy, hepato-and spleno-megaly with hyperaemia, haemorrhagic enteritis, pathological changes in the circulatory system (hypertrophied left ventricle, dilated cranial part of vena cava), slightly enlarged pancreas with miliary petechial haemorrhages. The severity of necropsy findings generally correlated with the duration of the symptoms. Detailed histopathological examination was done only in one Chilean tinamou and revealed small foci of perivascular lymphoplasmocytic encephalitis, lymphoplasmocytic hepatitis, focal cardiomyocyte dystrophy, pronounced pulmonary hyperaemia, and reactive hyperplasia of the lymphatic tissue of the spleen.
## Detection of USUV RNA in tissue samples
Altogether 22 individual birds were dissected, yielding 82 tissue samples. No WNV RNA was detected, whereas USUV RNA was found in 54.5% (N = 12) individuals and 45.1% (37) of the sampled tissues (Table 1). No statistically significant differences were observed in the number of viral RNA (vRNA) copies per gram of tissue when comparing species of birds or infected tissues (ANOVA). The average USUV target copy number was statistically significantly higher in captive birds (6.1E + 09 per g of tissue) than in free-living blackbirds (5.56E + 08 per g) but
## Plaque assay and virus isolation in cell cultures
Viral titres in RT-qPCR positive brain homogenates were determined using plaque assay reaching an average viral load of 3.9E + 06 PFU/g (Table 2). The paired t-test revealed statistically significant (approximately three logs) difference between the number of USUV target copies and replicating viral particles per g of tissue (p < 0.01).
The virus isolation process was successful in eight out of nine USUV vRNA-positive brain homogenates using the C6/36 mosquito cells. Active virus propagation was confirmed using an A549 plaque assay, achieving average viral titres of 3.78E + 06 PFU/ml six days after infection, without any visible CPE in infected C6/36 cultures. The viral isolates are stored and available in the Biology Centre Collection of Organisms (www.bcco.cz). Immunofluorescence staining of flaviviral E protein revealed that in average 22% of the mammalian A549 cells were infected with the strain 277TM/10, while the C6/36 cells exhibited almost 100% infection rate (Fig. 2).
## Phylogenetic analyses
The nearly full-length genome nucleotide sequences (10,936 nt) were used to reconstruct phylogenetic relationships. Both the maximum likelihood (Figs. 3 and4) and Bayesian inference (Supplementary Fig. 1,2) approaches produced well-resolved phylogenetic trees with almost identical topologies. All USUV sequences obtained from dead birds in this study were assigned to the Europe 2 genetic lineage. In general, the sequences from zoo birds formed three geographically partially supported clusters. One cluster consists of sequences from birds in Brno Zoo (Figs. 3 and4 in red) and an additional sequence from a free-living blackbird (violet) also found in Brno but in 2017. The second cluster contains sequences originating from two Ostrava Zoo birds (blue) that died 7-8/2024 and free-living blackbirds (violet) found in Brno in 2024, and Vienna (Austria) in 2016. The last cluster comprises the sole sequence obtained from Olomouc Zoo (green), the remaining two sequences from Ostrava Zoo (blue), from birds that died 9/2024, and two additional sequences from free-living blackbirds (violet) found in the western part of the country (approximately 280 km from Ostrava, 200 km from Olomouc) (Fig. 4).
In the pairwise comparison the sequences from Brno Zoo differed in 1 (99.9% identity) to 48 nucleotide positions (99.5% identity). USUV sequences from the two Chilean tinamous (283NP10, 284NP10) were almost identical (a single nucleotide mismatch, 100% amino acid sequence identity), whereas 281NP10 differed significantly from all other sequences from Brno Zoo samples (47-48 nucleotide sequence differences, 8-9 amino acid differences). The differences among the nucleotide sequences originating from Ostrava Zoo samples ranged from 9 (99.9% identity) to 97 mismatches (99.1% identity) (Fig. 5). Notably the sequence of the sample 292SP10 contains a 37 nt deletion in the 3´UTR region compared to all other acquired sequences.
## Serological analyses
All samples positive or borderline in ELISA (Supplementary Table 3) were subjected to virus neutralization tests with the three flaviviruses that are known to occur in the area: USUV, WNV, and TBEV. Apart from a wolf serum with low titre of anti-USUV neutralization antibodies (NABs), all USUV or WNV neutralizing sera cross-reacted with both viruses (Table 3). No cross-neutralization was recorded for TBEV. Four-fold difference in NABs titre was reached only for a single serum of a Mikado pheasant housed in the cage neighboring with the Chilean tinamous at Brno Zoo allowing to conclude this serum as anti-USUV specific. Similarly, unequivocally anti-WNV NABs titres were found in a Wood owl (Strix aluco) from a private owner from the South Moravian region (close to Brno). The Wood owl had reported
## Discussion
Zoological gardens have unique conditions conducive to virus transmission [29,30]. Many bird species from different parts of the world with different susceptibility to the infection are caged in proximity of each other, making the transmission of the infection easier and potentially more stable [30]. The application of preventive measures targeting the presence of vectors is limited in the zoos, due to requirements for open enclosures and freely accessible water reservoirs. Vaccination is complicated due to the variable immune responses of different species, lack of approved bird-and virus-specific vaccines, and significant costs associated with vaccinating large numbers of individuals [13,14,31]. Consequently, captive bird populations are at high risk of USUV infections, which are frequently associated with substantial losses.
USUV infections have been reported in numerous zoological gardens across Europe, including neighboring Austria and Germany [11,12,[32][33][34]. Our study was initially prompted by several fatal cases of USUV infection in Chilean tinamous at the Brno zoological garden, which devastated the local breeding population and significantly impacted the European/global zoo breeding program. Of the seven individuals housed together, five succumbed to the infection (one individual was not necropsied). These are the first reported cases of USUV infection in this species worldwide. Serological evidence of exposure was reported only in distantly related greater rheas (Rhea americana), emus (Dromaius novaehollandiae) [33,35], and an ostrich (Struthio camelus) [32]. Also, in the case of Brahminy starlings there are no previous reports of USUV infection in this species. On the other hand, passerine birds as well as owls including Boreal owl and a Eurasian pygmy owl are well known to be sensitive to flavivirus infections [11]. In the zoo settings, substantial diversity has been observed among exposed species, ranging from commonly infected individuals of Strigiformes [11,12,32] to isolated cases of less expected species, such as Humboldt penguins, Egyptian vultures or ratites mentioned above [32,33,35].
In sensitive species of birds, USUV infection is usually multisystemic, targeting the central nervous system and internal organs. Hepatomegaly and splenomegaly are considered major macroscopic pathological findings associated with the infection [11,36,37], which is consistent with our findings. USUV RNA has previously been detected in various internal organs, most frequently in the brain [7,9,37,38]. In the present study, the brain was confirmed as the most suitable tissue for the USUV RNA screening, as detectable vRNA was found in brains of all USUV-positive birds, whereas different other tissues tested negative in some individuals. The amount of viral RNA per gram of brain tissue was in average 3 logs higher than the amount of replicating virus particles as determined by plaque assay. A similar ratio was reported previously for other mosquito-borne flaviviruses [39,40]. The difference might be explained by generation of defective viral particles, release of free viral RNA from lysed cells, and other processes in the flaviviral replication cycle [41]. Brain tissues of captive USUV-positive birds contained statistically significantly higher vRNA copy numbers compared to brain tissues of free-living birds. One possible explanation is, that captive birds are more likely to be found sooner after death and their cadavers transferred to freezers preventing RNA degradation. Nevertheless, there are several important limitations possibly affecting this finding, such as limited number of individuals and comparing several different species of captive birds with blackbirds which is the only free-living bird species positive in our study. USUV RNA was found also in cloacal/choanal swabs of one Boreal owl and one Brahminy starling, and in a cloacal swab of another Brahminy starling. The possibility of non-vector mediated transmission has been suggested previously based on USUV shedding in oral, pharyngeal and cloacal secretions [42][43][44][45][46][47], along with evidence for such transmission in WNV [48]. In both Brno Zoo and Ostrava Zoo, we found USUV infected birds housed either together or in close vicinity of each other. Nevertheless, we are unable to conclude on the route of infection.
In the Czech Republic, USUV has been previously detected in blackbirds [1,7], mosquitoes [7,49], and as an autochthonous human infection [16]. Although the Europe 1, 3, and Africa 3 genetic lineages were detected in blackbirds and mosquitoes in the same (South Moravia) region previously [1,7,49], phylogenetic analyses assigned the sequences of all the newly detected viruses in our study to the Europe 2 lineage. The relatively low variability of detected lineages may be due to the limited number of samples analyzed or to the specific conditions of zoological gardens. Furthermore, the Europe 2 lineage appears to currently dominate in Europe [50]. Notably, it was also reported to be more virulent in a mice model compared to Africa 3 and Europe 3 [51]. This lineage is also frequently associated with cases of human neurological infections, while Europe 3 and Africa 3 have been found in healthy donors [3,15,52]. Although all four Chilean tinamous were housed in the same enclosure and appeared to have been infected around the same time, in one individual a significantly genetically different USUV strain was detected. These findings could indicate either rapid intra-host evolution of the virus [53][54][55] or co-circulation of two different strains in Brno Zoo. We assume the latter possibility is more likely, given that the divergent sequence from the tinamou was more similar to a sequence obtained from a free-living blackbird found in the same city in 2017 than to those from the other three tinamous.
As shown previously using local strains of Europe 1 and 3 lineages [7], also in the case of our Europe 2 isolate, the mosquito C6/36 cells were nearly 100% infected but exhibited no CPE, whereas the mammalian cells were infected in a lower proportion and the infection was associated with CPE. These differences might be at least partly associated with the differences in immune response. While the mosquito cell line is deficient in the RNA interference pathway [56], in the mammalian A549 cells USUV replication might be considerably reduced due to interferon response [57].
Although blood from the Chilean tinamous was unavailable for testing, several serum samples from birds or other animals from zoological gardens, as well as all four serum samples obtained from Mikado pheasants that were caged directly next to the USUV-positive Chilean tinamous in Brno Zoo tested positive for anti-USUV/ anti-WNV antibodies. One of the Mikado pheasants had anti-USUV antibody titres more than four times higher than its anti-WNV NAB titres, which is considered proof of anti-USUV antibody specificity [27,28]. Due to the low probability of a previous WNV infection, it is likely that the other three pheasants also had specific antibodies to USUV rather than to WNV, but did not reach the threshold difference between the NABs titres. Also, we could not determine in this study whether the antibody response resulted from a recent or previous infection, as the class of antibodies was not identified. Of the The titres were determined in previously ELISA-positive or borderline sera using virus neutralization tests. The titres are expressed as the reciprocal value of the serum dilution causing a 50% reduction in cytopathic effect. Four-fold and higher difference in neutralization titre or > 40 combined with negative results for the remaining results was considered a threshold to assign the response to one of the viruses. Titres of neutralizing antibodies are indicated in bold and crossneutralizing titres in italics historical serum samples from zoological gardens, one wolf (C. lupus) serum sample from Plzeň Zoo was found weakly anti-USUV-positive, which would indicate the presence of the virus in the western part of the country as early as 2013. Additionally, a single TBEV-neutralizing serum sample was confirmed from a reindeer (R. tarandus) from Olomouc Zoo, as reported previously by Širmarová et al. [26]. Interestingly, anti-WNV NABs were found in the serum sample of a wood owl from a private owner from the South Moravian region, where WNV circulation has been confirmed [16,49,58,59]. Crossneutralization was observed among USUV-and WNVantibody-positive sera (Table 3), which was demonstrated also previously [32,33,35]. Although no approved avian vaccine is currently available for flavivirus infections, several studies have reported the use of equine anti-WNV vaccines providing a certain level of protection in specific bird species [13,14]. The cross-reactivity described above might thus indicate possible cross-protectivity to USUV [46], although further studies are needed to confirm the efficiency [60].
## Conclusions
We have described cases of fatal USUV infections in free-living and captive birds in the Czech Republic. Chilean tinamous were found highly sensitive to the virus as five of seven individuals housed together succumbed to infection. USUV-positive birds were found in three different zoological gardens in the eastern part of the country, in proximity to areas with known endemic circulation of USUV and WNV [9,12,16,49,58,59,61]. However, USUV-infected blackbirds and mosquitoes have also been detected in the southwestern and central regions [7], indicating that the virus is widely distributed across the Czech Republic. Both USUV and WNV are capable of causing fatal disease in captive endangered, rarely bred, or otherwise valuable avian species. Given the limited preventive options, particularly in zoological gardens, rapid identification of infections, and prompt isolation of susceptible species represent an effective mitigation strategy. Furthermore, as both viruses have the potential to be transmitted to humans [15,16,61], surveillance data from birds are an important contribution to public health monitoring and early warning systems for human disease in the One Health context [30].
## References
1. Hubálek, Rudolf, Čapek et al. (2011) "Usutu virus in Blackbirds (Turdus merula)" *Transbound Emerg Dis*
2. Mcintosh (1985) "Usutu (SA ar 1776), nouvel arbovirus du groupe B. Int Cat Arboviruses"
3. Engel, Jöst, Wink et al. (2016) "Reconstruction of the evolutionary history and dispersal of Usutu virus" *Europe and Africa. mBio*
4. Weissenböck, Bakonyi, Rossi et al. (1996) "Usutu virus" *Emerg Infect Dis*
5. Angeloni, Bertola, Lazzaro et al. (2012) "Epidemiology, surveillance and diagnosis of Usutu virus infection in the EU/EEA" *Euro Eur Centre Disease Prev Control*
6. Clé, Beck, Salinas et al. (2019) "Usutu virus: A new threat" *Epidemiol infect*
7. Hönig, Palus, Kaspar et al. (2016) "Multiple lineages of Usutu virus (Flaviviridae, Flavivirus) in Blackbirds (Turdus merula) and mosquitoes (Culex pipiens, Cx. modestus) in the Czech Republic" *Microorganisms*
8. Simonin (2024) "Circulation of West nile virus and Usutu virus in europe: overview and Challenges. viruses"
9. Chvala, Bakonyi, Bukovsky et al. (2003) "Monitoring of Usutu virus activity and spread by using dead bird surveillance in Austria" *Vet Microbiol*
10. Nikolay (2015) "A review of West nile and Usutu virus co-circulation in europe: how much do transmission cycles overlap?" *Trans R Soc Trop Med Hyg*
11. Steinmetz, Bakonyi, Weissenböck et al. (2011) "Emergence and establishment of Usutu virus infection in wild and captive avian species in and around Zurich, Switzerland-genomic and pathologic comparison to other central European outbreaks" *Vet Microbiol*
12. Weissenböck, Kolodziejek, Url et al. (2002) "Emergence of Usutu virus, an African mosquito-borne flavivirus of the Japanese encephalitis virus group" *Europe. Emerg Infect Dis*
13. Angenvoort, Fischer, Fast et al. (2014) "Limited efficacy of West nile virus vaccines in large Falcons (Falco spp)" *Vet Res*
15. Bergmann, Fischer, Fischer et al. (2023) "Vaccination of zoo birds against West nile Virus-A field study" *Vaccines (Basel)*
16. Cadar, Simonin (2022) "Human Usutu virus infections in europe: A new risk on horizon? Viruses"
17. Zelená, Kleinerová, Šikutová et al. (2018) "First autochthonous West nile lineage 2 and Usutu virus infections in Humans" *Czech Republic. Pathogens*
18. Nikolay, Weidmann, Dupressoir et al. (2014) "Development of a Usutu virus specific real-time reverse transcription PCR assay based on sequenced strains from Africa and Europe" *J Virol Methods*
19. Eiden, Vina-Rodriguez, Hoffmann et al. (2010) "Two new Real-Time quantitative reverse transcription polymerase chain reaction assays with unique target sites for the specific and sensitive detection of lineages 1 and 2 West nile virus strains" *J VET Diagn Invest SAGE Publications Inc*
20. Quick, Grubaugh, Pullan et al. (2017) "Multiplex PCR method for minion and illumina sequencing of Zika and other virus genomes directly from clinical samples" *Nat Protoc Nat Publishing Group*
21. Munnink, Kik, De Bruijn et al. (2019) "Towards high quality real-time whole genome sequencing during outbreaks using Usutu virus as example" *Infect Genet Evol*
23. Lefort, Longueville, Gascuel (2017) "SMS: smart model selection in PhyML" *Mol Biol Evol*
24. Guindon, Dufayard, Lefort et al. (2010) "New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0" *Syst Biol*
25. Huelsenbeck, Ronquist (2001) "MRBAYES: bayesian inference of phylogenetic trees" *Bioinformatics*
26. Sirmarova, Salat, Palus et al. (2018) "Kyasanur forest disease virus infection activates human vascular endothelial cells and monocyte-derived dendritic cells" *Emerg Microbes Infections*
27. Melnick, Paul, Riordan et al. (1951) "Isolation from human Sera in Egypt of a virus apparently identical to West nile virus" *Proc Soc Exp Biol Med*
28. Sirmarová, Tichá, Golovchenko et al. (2014) "Seroprevalence of borrelia burgdorferi sensu Lato and tick-borne encephalitis virus in zoo animal species in the Czech Republic" *Ticks Tick Borne Dis*
29. Atama, Martin, Van Horssen et al. (2025) "West nile virus and Usutu virus neutralizing antibodies found in Dutch rodent Species. Vector-Borne and zoonotic diseases"
30. Laidoudi, Durand, Watier-Grillot et al. (2023) "Evidence of antibodies against the West nile virus and the Usutu virus in dogs and horses from the Southeast of France" *Transbound Emerg Dis*
31. Mcnamara (2007) "The role of zoos in biosurveillance. Int Zoo Yearbook"
32. Van Leeuwen, Falconer, Veitch et al. (2023) "Zoos as sentinels? A Meta-Analysis of Seroprevalence of terrestrial mammalian viruses in zoos" *EcoHealth*
33. Wheeler, Langevin, Woods et al. (2011) "Efficacy of three vaccines in protecting western scrub-jays (Aphelocoma californica) from experimental infection with West Nile virus: implications for vaccination of island scrub-jays (Aphelocoma insularis). vector-borne and zoonotic diseases"
34. Buchebner, Zenker, Wenker et al. (0148) "Low Usutu virus Seroprevalence in four zoological gardens in central Europe" *BMC Vet Res*
35. Cano-Terriza, Guerra, Lecollinet et al. (2015) "Epidemiological survey of zoonotic pathogens in feral pigeons (Columba Livia var. domestica) and sympatric zoo species in Southern Spain" *Comp Immunol Microbiol Infect Dis*
36. Ziegler, Fast, Eiden et al. (2016) "Evidence for an independent third Usutu virus introduction into Germany" *Vet Microbiol*
37. Constant, Bollore, Clé et al. (2020) "Evidence of exposure to USUV and WNV in zoo animals in France" *Pathogens*
38. Agliani, Giglia, Marshall et al. (2023) "Pathological features of West nile and Usutu virus natural infections in wild and domestic animals and in humans: A comparative review" *One Health*
39. Giglia, Münger, Mencattelli et al. (2025) "Usutu virus-disease susceptibility in threatened red-breasted geese (Branta ruficollis) in Italy and the Netherlands" *J Comp Pathol*
40. Becker, Jöst, Ziegler et al. (2012) "Epizootic emergence of Usutu virus in wild and captive birds in Germany" *PLoS ONE*
41. Bae, Nitsche, Teichmann et al. (0129) "Detection of yellow fever virus: a comparison of quantitative real-time PCR and plaque assay" *J Virol Methods*
42. Choy, Ellis, Ellis et al. (2013) "Comparison of the mosquito inoculation technique and quantitative real time polymerase chain reaction to measure dengue virus Concentration. The American journal of tropical medicine and hygiene" *Am Soc Trop Med Hygiene*
43. Saito, Fukasawa, Shirasago et al. (2020) "Comparative characterization of flavivirus production in two cell lines: human hepatoma-derived Huh7.5.1-8 and African green monkey kidney-derived Vero" *PLoS ONE*
44. Benzarti, Rivas, Sarlet et al. (2020) "Experimental Usutu virus infection in domestic Canaries Serinus canaria. viruses"
45. Chvala, Bakonyi, Hackl et al. "Limited pathogenicity of Usutu virus for the domestic chicken"
46. (1080) *Avian Pathol*
47. Höfle, Gamino, De Mera et al. (2013) "Usutu virus in migratory song Thrushes" *Spain. Emerg Infect Dis*
48. Agliani, Visser, Marshall et al. (2025) "Experimental Usutu virus infection in Eurasian Blackbirds (Turdus merula). Npj viruses"
49. Reemtsma, Holicki, Fast et al. (2023) "A prior Usutu virus infection can protect geese from severe West nile Disease" *Pathogens. Multidisciplinary Digit Publishing Inst*
50. Münger, Atama, Van Irsel et al. (1038) "One health approach uncovers emergence and dynamics of Usutu and West nile viruses in the Netherlands" *Nat Commun Nat Publishing Group*
51. Komar, Langevin, Hinten et al. (2003) "Experimental infection of North American birds with the new York 1999 strain of West nile virus" *Emerg Infect Dis*
52. Rudolf, Bakonyi, Šebesta et al. (2015) "Cocirculation of Usutu virus and West nile virus in a Reed bed ecosystem" *Parasit Vectors*
54. Vilibic-Cavlek, Petrovic, Savic et al. (2020) "Epidemiology of Usutu virus: the European scenario" *Pathogens*
55. Duyvestyn, Marshall, Bredenbeek et al. (2025) "Dose and strain dependent lethality of Usutu virus in an Ifnar-/-mouse model. Npj viruses"
56. Clé, Barthelemy, Desmetz et al. (2020) "Study of Usutu virus neuropathogenicity in mice and human cellular models" *PLoS Negl Trop Dis*
57. Oliveira, Durigon, Mendes et al. (2018) "Persistence and Intra-Host genetic evolution of Zika virus infection in symptomatic adults: A special view in the male reproductive system" *Viruses*
58. Ehrbar, Ngo, Campbell et al. (2017) "High levels of local inter-and intra-host genetic variation of West nile virus and evidence of finescale evolutionary pressures" *Infect Genet Evol*
59. Grubaugh, Smith, Brackney et al. (2015) "Experimental evolution of an RNA virus in wild birds: evidence for Host-Dependent impacts on population structure and competitive fitness" *PLOS Pathogens Public Libr Sci*
60. Brackney, Scott, Sagawa et al. (2010) "C6/36 Aedes albopictus cells have a dysfunctional antiviral RNA interference Response" *PLOS neglected tropical diseases. Public Libr Sci*
61. Cacciotti, Caputo, Selvaggi et al. (2015) "Variation in interferon sensitivity and induction between Usutu and West nile (lineages 1 and 2) viruses" *Virology*
62. Hubálek, Juricová, Straková et al. (2017) "Serological survey for West nile virus in wild Artiodactyls, Southern Moravia (Czech Republic)" *Vector Borne Zoonotic Dis*
63. Hubálek, Hm, Halouzka et al. (2000) "West nile virus investigations in South Moravia, Czechland. Viral immunology"
64. Bosco-Lauth, Kooi, Hawks et al. (0363) "Cross-Protection between West nile virus and emerging flaviviruses in wild birds" *Am J Trop Med Hyg*
65. Aberle, Kolodziejek, Jungbauer et al. (2018) "Increase in human West nile and Usutu virus infections" *Euro Surveill*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12335058&blobtype=pdf
|
cap 1 structure, which is critical for enhancing stability and translation activity.
Compared to first-generation cap analogues, such as mCap and ARCA, which predominantly result in the cap 0 structure, the CleanCap approach offers several advantages………. "
## Correction:
The second and more advanced approach is a co-transcriptional method using CleanCap AG technology in combination with T7 RNA polymerase [3]. This method incorporates the cap structure during mRNA synthesis, eliminating the need for any additional enzymatic steps. CleanCap AG technology achieves remarkable efficiency, producing mRNA after IVT reactions with concentrations up to 5 mg/ml and a high proportion (94%) of the desired cap 1 structure, which is critical for enhanced stability and translational activity.
Compared to first-generation cap analogues, such as mCap and ARCA, which predominantly result in the cap 0 structure, the CleanCap approach offers several advantages……….
The original article has been corrected. Following publication of the original article [1], there is a duplication in the text on page 3 of the original article.
The following paragraph appears twice consecutively:
"The second and more advanced approach is a co-transcriptional method using CleanCap AG technology in combination with T7 RNA polymerase [3]. This method incorporates the cap structure during mRNA synthesis, eliminating the need for any additional enzymatic steps. CleanCap AG technology achieves remarkable efficiency, producing mRNA after IVT reactions with concentrations up to 5 mg/ml and a high proportion (94%) of the desired cap 1 structure, which is critical for enhanced stability and translational activity.
The second and more advanced approach is a co-transcriptional method using CleanCap AG technology in combination with T7 RNA polymerase [3]. This method incorporates the cap structure during mRNA synthesis, eliminating the need for any additional enzymatic steps. CleanCap AG technology achieves remarkable efficiency, producing mRNA after IVT reaction with concentrations up to 5 mg/ml and a high proportion (94%) of the desired
## Virology Journal
The original article can be found online at h t t p s : / / d o i . o r g / 1 0 . 1 1 8 6 / s 1 2 9 8 5 -0 2 5 -0 2 6 4 5 -6.
## References
1. Leong, Tham, Poh (1186) "Revolutionizing immunization: a comprehensive review of mRNA vaccine technology and applications" *Virol J*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548395&blobtype=pdf
|
# Prediction of COVID-19 disease progression by multiparametric analysis of circulating extracellular vesicles with flow cytometry
Evelyn Hammer, Charlotte Flynn, Johannes Rößler, Johanna Erder, Rudolf Napieralski, Lisa Fricke, Birgit Campbell, Martin Feuerherd, Felix Esslinger, Albrecht Von Brunn, Timm Weber, Siobhan King, Sisareuth Tan, Alain Brisson, Ulrike Protzer, Gabriele Schricker, Kathrin Gärtner, Gregor Ebert, Allessandra Moretti, Florian Klein, Kevin Knoops, Ron Heeren, Wolfgang Hammerschmidt, Reinhard Zeidler, Olaf Wilhelm, Percy Knolle, Bastian Höchst
## Abstract
Extracellular vesicles (EVs) are released from all cells of the body. They are considered to mirror the state of the cells from which they are released and circulate in the blood, suggesting a possible use of EV analysis for diagnostic purposes. Here, we report that the analysis of single EVs by flow cytometry can detect infection of cells with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) by identifying expression of the SARS-CoV-2 spike (S) protein on the surface of EVs and the cellular origin of EVs by detecting cell-type-specific markers such as troponin (cTNT1) for cardiomyocytes. In coronavirus-associated disease 19 (COVID-19) patients, we detected a direct correlation of the frequencies of circulating S-expressing EVs, but not of cTNT/S-co-expressing EVs, with the subsequent development of a severe disease course. Detection of circulating S-expressing EVs indicates widespread SARS-CoV-2 infection in the body, which may contribute to the immune pathogenesis that triggers tissue and organ damage in COVID-19. Our findings suggest that detecting circulating viral antigen-expressing EVs may provide crucial predictive information on infection-associated disease courses in situations of a future viral pandemic.
IMPORTANCEThe ability to predict which patients infected with the SARS-CoV-2 virus will develop severe disease remains a significant clinical challenge. The present study demonstrates that EVs in the peripheral blood, carrying the SARS-CoV-2 spike protein, can be detected by flow cytometry and serve as early biomarkers of disease progression. In contradistinction to PCR or serology, this method provides insight into systemic viral spread and potential organ involvement. The early identification of spike-positive EVs at the time of hospital admission has the potential to facilitate the timely identification of high-risk patients, thereby enhancing the efficacy of triage and subsequent care. This approach may also be of value in terms of facilitating a more rapid and precise response to future virus pandemics.
Concurrently, dysregulation of the coagulation cascade can lead to immunothrombo sis and the formation of microvascular thrombi, causing organ dysfunction across multiple organ systems (3). Overall, the number of deaths caused by COVID-19 amounts to more than 14 million worldwide (4).
SARS-CoV-2 primarily infects epithelial cells of the upper airways and the lungs (5). However, because of the widespread expression on target cells of its recep tors, angiotensin-converting enzyme 2 (ACE2) and transmembrane protease serine 2 (TMPRSS2), mediating cellular infection (6), SARS-CoV-2 can also infect cells in other organs beyond the lungs. These include cells in the cardiovascular system (cardiomyo cytes and endothelial cells), cells in the nervous system (neurons, astrocytes), cells in the gastrointestinal tract (intestinal epithelial cells, hepatocytes, cholangiocytes, pancreatic cells), the kidney (proximal tubular epithelial cells), and cells of the reproductive tract (7). Infection of cells in these organs may contribute to severe disease courses, such as SARS-CoV-2 infection of the heart, or may be involved in persistent infection, as reported in intestinal epithelial cells in prolonged infection (8). Of note, individuals with pre-existing cardiovascular conditions are at an elevated risk of developing severe manifestations of COVID-19 (9,10), as it has been shown that SARS-CoV-2 infection of cardiomyocytes promotes myocardial injury and triggers cardiac arrhythmias (11). Involvement of the cardiovascular system in COVID-19 patients, determined by increased levels of cardiac troponin T (cTNT) as a sign of cardiac ischemia and injury, is associated with elevated morbidity and mortality (10,(12)(13)(14). However, it is impossible to detect SARS-CoV-2 infection of cardiomyocytes in patients unless a biopsy is taken during an invasive procedure and viral RNA is detected within the tissue.
Extracellular vesicles (EVs) are released from all cells of the body and can contain molecular information from their parental cells (15). EVs circulate in the blood (16) and bear the potential to be used as sentinels for the cells from which they are secreted. Here, we present a flow-cytometry-based detection of EVs from cardiomyocytes infected with SARS-CoV-2. This flow cytometry-based analysis allowed for the detection of EVs expressing troponin and the spike antigen of SARS-CoV-2, both from in vitro SARS-CoV-2infected human cardiomyocytes and in the plasma of COVID-19 patients. The number of circulating SARS-CoV-2 spike antigen-bearing EVs directly correlated with a severe COVID-19 disease course in patients, pointing toward a potential use of circulating EVs to detect viral infection of organs that warrants further investigation.
## RESULTS
## Detection of SARS-CoV-2-spike (S) and cTnT1 expressing EVs
We started by identifying EVs using flow cytometry and analyzing them at the level of single EVs for expression of the SARS-CoV-2 spike antigen (S) as an indicator of ongoing viral infection. For this, we transduced HEK293T cells with the S(B.1) gene of SARS-CoV-2, which resulted in S protein expression (Fig. 1a) and secretion of S-expressing virus-like particles, as previously described (17,18). Nanoparticle tracking analysis (NTA) revealed a similar average size of 130 nm for EVs found in the cell culture supernatant of S-transduced and non-transduced HEK293T cells (Fig. 1b; Fig. S1a). Enrichment of EVs by ultracentrifugation (UC) from the cell culture supernatant of S-transduced HEK293T cells and subsequent Western blot analysis demonstrated S expression (Fig. S1b). Ultrastruc tural imaging by electron microscopy of EVs released from the S-transduced HEK293T cells into the cell culture supernatant showed round spheres enclosed by a double membrane, a characteristic feature of EVs, and the typical corona-like S antigen features on EVs from S-transduced HEK293T cells (Fig. 1c).
To analyze EVs by flow cytometry, we employed the signal from the side scatter (SSC-H) photomultiplier as a trigger for the analysis. Beyond the electronic noise signal derived from photomultipliers associated with this highly sensitive analysis, we detected S-expressing EVs, without prior enrichment by UC, in the supernatant of S-transduced HEK293T cells but not untransduced cells using a fluorochrome-labeled anti-S antibody (Fig. 1d). Side-by-side comparison of flow cytometric analysis of EVs directly from cell culture supernatant or after enrichment through UC or size exclusion chromatography (SEC) demonstrated that enrichment prior to flow cytometric analysis is not required and that direct flow cytometric analysis is even more sensitive in detecting EVs contained in cell culture supernatant (Fig. S1c andd). Serial dilution of the supernatant led to a reduction in the number of EVs detected by flow cytometry, whereas anti-S fluorescence intensity remained unchanged (Fig. 1e andf), as expected for a specific and sensitive analysis. Taken together, this demonstrates the feasibility of detecting S-expressing EVs released from S-transduced HEK293T cells by flow cytometry. To identify S + events in the flow cytometric analysis as EVs, we performed a multiparameter analysis and costained for characteristic EV markers, such as CD9, CD63, and CD81, and analyzed single EVs by high-resolution dSTORM imaging. dSTORM imaging revealed colocalization of S, CD63, and CD81 on the same EV obtained from the supernatant of S-transduced HEK293T cells (Fig. 1g). Quantifying dSTORM high-resolution images of EVs from S-transduced HEK293T cells showed that approximately 75% of the vesicles co-expressed S in combination with CD63 and/or CD81 (Fig. 1h). These results revealed that a multiparametric analysis of EVs by flow cytometry is possible and allowed us to generate quantitative data on the number of EVs with a particular surface phenotype.
We proceeded to analyze cell-derived EVs for expressing an organ-specific marker, such as cTNT, released by cardiomyocytes. We used induced pluripotent stem (iPS) cells (iPSC) differentiated into functionally active human cardiomyocytes (19,20). iPS-derived cardiomyocytes expressed the cardiomyocyte-specific marker troponin (cTNT1) (Fig. 2a). NTA revealed a similar average size of 130 nm for EVs from human cardiomyocytes as for HEK293T cells (Fig. 2b). Ultrastructural imaging of EVs released from iPSC-derived cardiomyocytes into cell culture supernatant revealed an EV characteristic double membrane feature (Fig. 2c). By flow cytometry, we detected cTNT1 expression on EVs in the cell culture supernatant of iPSC-derived cardiomyocytes using a fluorochromelabeled anti-cTNT1 antibody (Fig. 2d), a decreased number of cTNT1 + EVs after serial dilution of the supernatant, and no change in anti-cTNT1 antibody fluorescence (Fig. 2e andf), altogether demonstrating the capacity to detect a cell type-specific marker on EVs by flow cytometry.
Next, we infected iPSC-derived cardiomyocytes with SARS-CoV-2 to evaluate whether S expression on cell-derived EVs can be detected by flow cytometry. We verified infection of human iPSC-derived cardiomyocytes using a recombinant SARS-CoV-2 expressing GFP (21) (Fig. S2a). Infection of iPSC-derived cardiomyocytes with SARS-CoV-2 (B.1.1.7) or the seasonal coronavirus hCoV-NL63 was confirmed by detection of viral RNA by PCR from cell lysates (Fig. S2b). Flow cytometric analysis showed that only EVs released from SARS-CoV-2-infected, but not hCoV-NL63-infected or non-infected, iPSC-derived cardiomyocytes expressed S (Fig. S2c). Importantly, S expression was identified by flow cytometry on cTNT1 + EVs released from SARS-CoV-2-infected cardiomyocytes, and increased frequencies of S + cTNT1 + EVs were detected with increasing numbers of SARS-CoV-2 used for infection of cardiomyocytes (Fig. 2g andh). Taken together, these results provide evidence that multiparametric analysis of individual EVs by flow cytometry allows for the simultaneous detection of cell-type-specific markers and a marker for ongoing SARS-CoV-2 infection in that cell.
## Identification of S + and cTNT1 + EVs in the blood of COVID-19 patients
To demonstrate the feasibility of measuring circulating EVs in the blood, we first investigated plasma from healthy individuals. We analyzed EVs from plasma by flow cytometry for expression of canonical EV markers and found that they expressed CD9, CD63, CD81, and HLA-ABC (Fig. S3a). Based on these initial observations, we next investigated the potential presence of S proteins on the surface of circulating EVs from SARS-CoV-2-infected patients. We performed a clinical study and collected plasma samples from COVID-19 patients admitted to the emergency department of the TUM University hospital between 2 February 2021 and 5 June 2021. Patients did not report prior COVID-19 vaccination and did not receive treatment with anti-SARS-CoV-2 antibodies before or at the time of sampling. A total of 25 patients were included in this analysis. In 19 COVID-19 patients, SARS-CoV-2 infection was detected with the B.1.1.7 variant of concern, 2 patients with the original strain of SARS-CoV-2, and 1 patient with the B.1.351 variant of concern. In contrast, in three patients, no information on the variant of concern could be obtained (details listed in the Table S1).
By flow cytometric analysis of circulating EVs, we detected S protein expression on CD81 + EVs in plasma of 24 out of 25 COVID-19 patients, whereas no S-expressing CD81 + EVs were found in plasma from healthy individuals (Fig. 3a through c). In contrast to the analysis by flow cytometry, S-expressing EVs in plasma were not detected using ELISA or Western blot (Fig. S3b andd). There were substantial differences in the number of circulating S-expressing CD81 + EVs detected across COVID-19 patients investigated here (Fig. 3b andc). However, there was no difference in the overall number of CD81 + EVs found in the plasma of COVID-19 patients compared to healthy individuals (Fig. S3e). Furthermore, there was no correlation between the viral load detected in nasal swabs and the number of circulating S + EVs (Fig. S3f through h).
We continued to determine the expression of cTNT1 on circulating CD81 + EVs and also found an increased number of circulating cTNT1 + EVs in the plasma of COVID-19 patients compared to healthy individuals (Fig. 3d ande). Notably, low numbers of cTNT1expressing EVs were found in two healthy donors (Fig. 3e). To assess SARS-CoV-2 infection of cardiomyocytes in COVID-19 patients, we analyzed co-expression of S and cTNT1 on CD81 + EVs. This revealed a considerable heterogeneity among COVID-19 patients for the number of S + cTNT1 + EVs (Fig. 3f andg). Repeated measurements of EVs from four patients demonstrated the high reproducibility of the detection of S + and cTnT1 + EVs by flow cytometry (Fig. S3i). There was no direct correlation between the number of circulating S + and cTNT1 + EVs in COVID-19 patients (Fig. 3h). Altogether, these results indicated that the detection of circulating S + and S + cTnT1 + EVs is possible in the plasma of COVID-19 patients and that it reflected SARS-CoV-2 infection of tissue cells and, in particular, cTnT1-expressing cardiomyocytes.
We next evaluated whether the detection of circulating S + and cTnT1 + EVs at the time of admission might help to predict the disease course in COVID-19 patients over the next 2-3 weeks. Indeed, the number of circulating S + EVs was higher in COVID-19 patients who later on progressed to develop an ARDS requiring mechanical ventilation (Fig. 4a). Moreover, the number of circulating S + EVs but not cTNT1 + EVs or S + cTnT1 + EVs directly correlated with the COVID-19 severity score and the number of severe clinical complica tions developing over the following days, such as ARDS, myocarditis, heart attack, acute kidney failure, pulmonary embolism, and thrombotic events (Fig. 4b through e). How ever, we did not detect a direct correlation of circulating S + EVs with routine laboratory parameters in COVID-19 patients at the time of analysis (Fig. S4) or with cytokines like IL-1, IL-6, IL-7, IL-8, IL-10, TNF, G-CSF, interferons type I and II or chemokines like CCL2, CCL5, and CXCL10 determined in plasma by bead array analysis (Fig. S5). For the number of circulating cTnT1-expressing EVs, however, we did not detect a correlation with the disease course in COVID-19 patients, routine laboratory parameters, and cytokines/ chemokines (Fig. S4 andS5), but found an inverse correlation with the number of blood leukocytes (Fig. 4f andg). Of note, for the number of circulating S + cTnT1 + EVs, we detected a direct correlation with CK-MB levels (Fig. 4h andi). Thus, our analysis reveals that high numbers of circulating S + EVs are not associated with direct organ damage but herald the development of clinical complications and severe disease courses.
## DISCUSSION
Circulating EVs reflect the state of the cells they are released from and bear the promise to serve as biomarkers for noninvasive diagnosis of viral infection or cancer, treatment monitoring, and disease prognosis (22,23). UC and SEC for enrichment, followed by characterization of EVs by Western blot and bead-based immunoassays, are considered gold standards to characterize EVs (24-26). However, these methods bear limitations as they are not only cost-intensive, but due to their complexity, are prone to errors and investigate enriched EVs as a bulk rather than individual EVs (24)(25)(26). We have estab lished a workflow for flow cytometry, which facilitates the precise multiparametric analysis of individual EVs derived from cell culture supernatant as well as patient plasma. Applied to a clinical context, our approach enabled the detection and characterization of EVs bearing markers of ongoing SARS-CoV-2 infection and derived from cardiomyocytes from the plasma of patients with COVID-19. This provides an advantage over previous reports (24-29) as it does not require initial enrichment by UC and enables the quantification of EVs according to their phenotypic characteristics, which allows for the analysis of small numbers of particular EVs in patients' plasma.
COVID-19 pathogenesis remains insufficiently understood until today (30), but it is known that tissue and organ damage is inflicted by a dysregulated immune response (31,32). In particular, SARS-CoV-2 infection of the cardiovascular system appears to contribute to a variety of complications such as myocarditis, vascular damage, arrhyth mia, and thrombosis (12,14,33), and the degree of ongoing inflammation predicts severe disease courses (34,35). Previous studies reported the presence of circulating EVs bearing SARS-CoV-2 proteins (22), such as S, that correlated directly with disease severity. Here, we address the question of whether the detection of SARS-CoV-2 S-bearing circulating EVs would allow for stratifying COVID-19 patients at the time of admission to the emergency department for the subsequent development of clinical complications and severe disease courses. There was no correlation between the number of circulating S-expressing EVs and any laboratory parameter indicating ongoing organ damage in the heart, liver, or kidney, and the plasma concentrations of immune mediators associated with a CRS, indicating that, at the time of admission, no grave dysregulation of immune responses was present causing tissue or organ damage. However, over the course of the next two weeks, we found that the abundance of circulating SARS-CoV-2 S-expressing EVs at the time of admission directly correlated with later development of inflammation, organ failure, and the COVID-19 severity score, which indicates that analysis of circulat ing EVs yields predictive information on the subsequent disease course of SARS-CoV-2 infected patients.
Notwithstanding this early prognostic information provided by quantifying the number of circulating S-expressing EVs, we did not detect a correlation between the number of cTNT1/S-expressing EVs and severe disease courses. Using iPS-derived human cardiomyocytes, we established that SARS-CoV-2 infection of these cells in vitro can be identified by analysis of their EVs bearing S on their surface. However, we failed to detect a correlation between cTNT1/S-expressing EVs and cardiovascular complications. This may be related to a sensitivity issue of the flow cytometry-based assay to detect very low frequencies of circulating cTNT1/S-expressing EVs, or may be the consequence of the absence of patients in our small study cohort who experienced SARS-CoV-2 infection of the heart. Studies involving a larger number of SARS-CoV-2-infected patients will be required to provide further insights into the question of whether cTNT1/S-expressing EVs predict the development of cardiovascular complications.
Thus, flow cytometry-based detection of circulating EVs and their analysis for evidence of S expression as a marker for ongoing SARS-CoV-2 infection proved to be useful at the time of admission of COVID-19 patients for the prediction of severe organ complications over the next two weeks. Beyond the PCR-based analysis of viral infection, the detection of viral replication in the organism through analysis of circulating viral antigen-expressing EVs may therefore provide important prognostic information in situations of a future viral pandemic.
## MATERIALS AND METHODS
## Generation of S-transduced HEK293T cells and supernatants
To generate EVs with high levels of SARS-CoV-2 spike S FL (D614G), a constitutive S-expressing HEK293 cell line (CID4618, Helmholtz Zentrum München) was generated by transduction with a S(B.1) gene expression plasmid (#7413, Helmholtz Zentrum München). S-transduced and untransduced HEK293T cells were cultured in serum-free medium for 12 h, the cell culture supernatant was collected, centrifuged for 15 min at 2,000 × g, and for 10 min at 10,000 × g at 4°C to remove cells and cell debris, and stored at -80°C until further analysis.
## Anti-S antibody generation and production
Anti-S antibody was generated as reported. One antibody with high-affinity binding to S (43A11) was selected for further studies. Anti-S-antibodies were purified from hybridoma supernatant using a GammaBind Plus Sepharose (17088602, Cytiva) column, washed with PBS, and eluted in citric acid buffer at pH 2.7. Antibody preparations were further purified using a Superdex 200 prep grade (17104301, Cytiva) in PBS to obtain monomers and coupled to a fluorochrome (Alexa488) using the labeling kit A10235 (Thermo Fisher Scientific, Waltham, MA, USA) (17).
## Differentiation of human cardiomyocytes from iPS cells
The iPS cells were treated as previously described to induce the differentiation of functional cardiomyocytes (19). iPS-derived cardiomyocytes were cultured in DMEM containing 10% FCS (10 mM), HEPES (0.55 mM), arginine (0.272 mM), and asparagine. The cells were incubated in a humidified atmosphere with 5% CO 2 at 37°C to reach confluence. Twelve hours before the supernatant was collected, the medium was removed, and the cells were washed with PBS. The cells were cultivated in a serum-free medium for 12 h. The cell culture supernatant was collected and centrifuged for 15 min at 2,000 × g, 20°C and for 10 min at 10,000 × g at 4°C to remove cells and cell debris. Samples were stored at -80°C until further analysis.
## Infection of iPSC-derived cardiomyocytes
iPSC-derived cardiomyocytes were infected with SARS-CoV-2 (B1) at an MOI of 1 in the EB2 medium. The SARS-CoV-2 isolate hCoV-19/Germany/BAV-PL-virotum-nacq/ 2020 (GISAID accession ID: EPI_ISL_582134) was derived from a nasopharyngeal swab, expanded in vitro, and aliquots were stored at -80°C until use. iPSC-derived cardiomyo cytes were infected with SARS-CoV-2 at different multiplicities of infection. The cells were washed with PBS 2 h post-infection and further cultivated in EB2 medium for 22 h before analysis. The seasonal coronavirus hCoV-NL63 (AY567487) (36) was used as a control to infect iPS-derived cardiomyocytes.
## Plasma samples from healthy individuals and from COVID-19 patients
A clinical study (approval 471/20 S from the ethics committee of the Technical University of Munich) was performed at the TUM University Hospital between February and June 2021 to investigate circulating EVs in blood samples from COVID-19 patients. Evidence of SARS-CoV-2 infection was obtained by a positive RT-PCR result. Blood samples from patients were obtained at the time of admission to the emergency department of the TUM University Hospital. Human whole blood was drawn using butterfly syringes and EDTA monovettes and transferred to 15 mL reaction tubes and centrifuged for 15 min at 2,000 × g at 20°C. Platelet-free plasma (PPP) was collected and transferred to 2 mL reaction tubes and centrifuged for 15 min at 10,000 × g at 20°C to remove remaining cells and cellular debris. PFP was collected and transferred into fresh reaction tubes. The samples were aliquoted into fresh reaction tubes and stored at -80°C until further use.
## Flow cytometric analysis of EVs
Flow cytometry analyses were performed using a Sony Spectral Flow Cytometer SA3800 equipped with two lasers (405 and 488 nm), a Sony Spectral Cytometer ID7000 with five lasers (355, 405, 488, 561, and 638 nm) (both from Sony Biotechnology, Japan) or a Beckman Coulter CytoFLEX LX Flow cytometer equipped with four lasers (405, 488, 561, and 638 nm) (BA16019, Beckman Coulter, USA).
The system settings listed in Table 1 were used. Samples were measured in plate-loader mode with 1,000 events per second for 10 min. Flow cytometry data were analyzed with FlowJo Software 10.2 (TreeStar Inc., Ashland, OR, USA).
## Immunophenotyping of EVs from plasma or cell culture supernatant EVs
A total of 10 µL of plasma or cell culture supernatant was incubated with the fol lowing antibodies: anti-CD63 (1 ng/µL, Clone H5C6), anti-CD81 (1 ng/µL, Clone 5A6), anti-CD9 (1 ng/µL, Clone HI9a) (BioLegend, San Diego, CA, USA), anti-HLA-ABC (1 ng/µL Clone W6/32) (Sony Biotechnology, San Jose, CA, USA), anti-S (43A11; Helmholtz), and anti-TroponinT (1 ng/µL, Clone REA400) (Miltenyi Biotech, Bergisch Gladbach, Germany) and incubated for 30 min at 20°C in the dark. The samples were diluted with PBS (Gibco Life Technologies, Thermo Fisher Scientific). As a negative control for antibody back ground, 10 µL of PBS was stained with the appropriate antibody and diluted analogously to plasma or cell culture supernatant. Cell counting beads (Thermo Fisher Scientific) were included in each sample prior to flow cytometry to allow for absolute cell quantification.
## Cryo-electron microscopy (cryo-EM)
The samples were processed for cryo-EM as follows. A 4 µL aliquot was deposited on an EM grid coated with a perforated carbon film; the liquid was blotted from the backside of the grid, and the grid was immersed in liquid ethane using a Leica EMCPC cryo-chamber. The EM grids were stored under liquid nitrogen prior to EM observation. Cryo-EM was performed with a Tecnai F20 (FEI, USA) microscope equipped with a USC1000-SSCCD camera (Gatan, USA).
## Western blot
Concentrated S + EVs or control EVs (S -) were lysed in nonreducing 5× Laemmli buffer and separated by SDS-PAGE. For western blot analysis of Spike-S in plasma from COVID-19 patients or healthy donors, 2-12 µL of plasma was diluted with 5× Laemmli buffer and ddH 2 O to a final volume of 15 µL and separated by SDS-PAGE. Proteins were transferred to nitrocellulose membranes (GE Healthcare Life Sciences) by semidry blotting (Bio-Rad), and the membranes were blocked for 1 h in 5% (wt/vol) nonfat milk in ddH 2 O at RT. The membranes were incubated overnight at 7°C with primary anti-Spike-S antibody (43A11, rat IgG, Helmholtz Zentrum München) at a 1:2,000 dilution in 5% (wt/vol) nonfat milk (Roth) in ddH 2 O, washed three times in TBST (Tris-buffered saline with 0.1% Tween-20) and incubated for 1 at RT with horseradish peroxidase (HRP)-conjugated anti-rat antibody (1112-035-062, goat anti-rat IgG, Jackson ImmunoResearch Europe) at a 1:20,000 dilution in 5% (wt/vol) nonfat milk in PBST (phosphate-buffered saline with 0.1% Tween-20). After three washes in TBST, the blots were incubated with enhanced chemiluminescence reagent (GE Healthcare) and imaged using a Fusion FX (Vilber).
## Enzyme-linked immunosorbent assay (ELISA)
For the quantification of the spike protein in samples from various sources, a sandwich ELISA was developed using two anti-spike antibodies with orthogonal epitopes. The wells of a Nunc MaxiSorp plate (Thermo Fisher Scientific) were coated for 5 h at 20°C with 2 µg mL -1 anti-Spike-S capture antibody (55E10, rat IgG, Helmholtz Zentrum München) or isotype control in PBS. After washing with PBST (PBS + 0.05% Tween-20), free binding sites were blocked for 2 h at 20°C in 5% (wt/vol) nonfat milk (Roth) in PBS. Samples of recombinant Spike protein (S1 + S2 extracellular domains, 40589-V08B1, Sino Biologi cal), in vitro-derived S + EVs, or plasma from COVID-19 patients (Technical University of Munich) were diluted as indicated in PBS containing 10 µg mL -1 isotype antibody (Helmholtz Zentrum München) to eliminate nonspecific plasma-derived background (37) before being incubated for 16 h at 7°C in a coated plate. After washing, HRP-conjuga ted anti-Spike-S detection antibody (43A11, rat IgG, Helmholtz Zentrum München) was added for 2 h at 20°C at a 1:500 dilution in 5% (wt/vol) nonfat milk in PBS, after which the samples were rewashed and incubated at 20°C with 100 µL of TMB substrate reagent (BD555214, Becton Dickinson). The reaction was stopped by adding 50 µL of 1 M H 2 SO 4, and the absorbance was measured at 450 nm in a CLARIOstar Plus (BMG Labtech). Data analysis was performed with GraphPad Prism.
## References
1. Wu, Zhao, Yu et al. (2020) "A new coronavirus associated with human respiratory disease in China" *Nature*
2. Zhou, Yang, Wang et al. (2020) "A pneumonia outbreak associated with a new coronavirus of probable bat origin" *Nature*
3. Bikdeli, Madhavan, Jimenez et al. (2020) "COVID-19 and thrombotic or thromboembolic disease: implications for prevention, antithrombotic therapy, and follow-up: JACC state-of-the-art review" *J Am Coll Cardiol*
4. Msemburi, Karlinsky, Knutson et al. (2023) "The WHO estimates of excess mortality associated with the COVID-19 pandemic" *Nature*
5. Lamers, Haagmans (2022) "SARS-CoV-2 pathogenesis" *Nat Rev Microbiol*
6. Hoffmann, Kleine-Weber, Schroeder et al. (2020) "SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor" *Cell*
7. Wong, Klinkhammer, Djudjaj et al. (2021) "Multisystemic cellular tropism of SARS-CoV-2 in autopsies of COVID-19 patients" *Cells*
8. Machkovech, Hahn, Wang et al. (2024) "Persistent SARS-CoV-2 infection: significance and implications" *Lancet Infect Dis*
9. Blagoeva, Hodzhev, Uchikov et al. (2025) "Clinical course and mortality predictors in adult hospitalized patients with COVID-19 infection-a retrospective cohort study"
10. Huang, Wang, Li et al. (2020) "Clinical features of patients infected with 2019 novel coronavirus in Wuhan" *Lancet*
11. Driggin, Madhavan, Bikdeli et al. (2020) "Cardiovascular considerations for patients, health care workers, and health systems during the COVID-19 pandemic" *J Am Coll Cardiol*
12. Shi, Qin, Shen et al. (2020) "Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan" *JAMA Cardiol*
13. Dmytrenko, Das, Kovacs et al. (2024) "Infiltrating monocytes drive cardiac dysfunction in a cardiomyocyte-restricted mouse model of SARS-CoV-2 infection" *J Virol*
14. Guo, Fan, Chen et al. (2020) "Cardiovascular implications of fatal outcomes of patients with coronavirus disease 2019 (COVID-19)" *JAMA Cardiol*
15. Buzas (2023) "The roles of extracellular vesicles in the immune system" *Nat Rev Immunol*
16. Iannotta, Kijas, Rowan (2024) "Entry and exit of extracellular vesicles to and from the blood circulation" *Nat Nanotechnol*
17. Roessler, Pich, Krähling et al. (2023) "SARS-CoV-2 and Epstein-Barr virus-like particles associate and fuse with extracellular vesicles in virus neutralization tests" *Biomedicines*
18. Roessler, Pich, Albanese et al. (2022) "Quantitation of SARS-CoV-2 neutralizing antibodies with a virus-free, authentic test" *PNAS Nexus*
19. (2025) *Full-Length Text Journal of Virology*
20. Moretti, Laugwitz, Dorn et al. (2013) "Pluripotent stem cell models of human heart disease" *Cold Spring Harb Perspect Med*
21. Moretti, Bellin, Jung et al. (2010) "Mouse and human induced pluripotent stem cells as a source for multipotent Isl1 + cardiovascular progenitors" *FASEB J*
22. Klemm, Ebert, Calleja et al. (2020) "Mechanism and inhibition of the papain-like protease, PLpro, of SARS-CoV-2" *EMBO J*
23. Dangot, Kinaani, Zavaro et al. (2023) "Extracellular vesicles of COVID-19 patients reflect inflammation, thrombogenicity, and disease severity" *Int J Mol Sci*
24. Alberro, Iparraguirre, Fernandes et al. (2021) "Extracellular vesicles in blood: sources, effects, and applications" *Int J Mol Sci*
25. Bano, Ahmad, Mohsin (2021) "A perspective on the isolation and characterization of extracellular vesicles from different biofluids" *RSC Adv*
26. Chen, Li, Zhang et al. (2022) "Review on strategies and technologies for exosome isolation and purification" *Front Bioeng Biotechnol*
27. Gorgzadeh, Nazari, Ismaeel et al. (2024) "A state-of-the-art review of the recent advances in exosome isolation and detection methods in viral infection" *Virol J*
28. Théry, Amigorena, Raposo (2006) "Isolation and characterization of exosomes from cell culture supernatants and biological fluids" *Curr Protoc Cell Biol Chapter*
29. Cocucci, Meldolesi (2015) "Ectosomes and exosomes: shedding the confusion between extracellular vesicles" *Trends Cell Biol*
30. Wang, Li, Wu et al. (2021) "Comparative evaluation of methods for isolating small extracellular vesicles derived from pancreatic cancer cells" *Cell Biosci*
31. Yang (2025) "The immunopathogenesis of SARS-CoV-2 infection: overview of lessons learned in the first 5 years" *J Immunol*
32. Moss (2022) "The T cell immune response against SARS-CoV-2" *Nat Immunol*
33. Mohandas, Jagannathan, Henrich et al. (2023) "Immune mechanisms underlying COVID-19 pathology and post-acute sequelae of SARS-CoV-2 infection (PASC)"
34. Dherange, Lang, Qian et al. (2020) "Arrhythmias and COVID-19: a review" *JACC Clin Electrophysiol*
35. Sharifpour, Rangaraju, Liu et al. "Emory COVID-19 Quality & Clinical Research Collaborative. 2020. C-Reactive protein as a prognostic indicator in hospitalized patients with COVID-19" *PLoS One*
36. Luan, Yin, Yao (2021) "Update advances on C-reactive protein in COVID-19 and other viral infections" *Front Immunol*
37. Van Der Hoek, Pyrc, Jebbink et al. (2004) "Identification of a new human coronavirus" *Nat Med*
38. Degn, Andersen, Jensen et al. (2011) "Assay interference caused by antibodies reacting with rat kappa light-chain in human sera" *J Immunol Methods*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12727862&blobtype=pdf
|
# Correction: N6-methyladenosine modification of circ_0003215 suppresses the pentose phosphate pathway and malignancy of colorectal cancer through the miR-663b/DLG4/G6PD axis
Baoxiang Chen, Yuntian Hong, Rui Gui, Huabin Zheng, Shunhua Tian, Xiang Zhai, Xiaoyu Xie, Quanjiao Chen, Qun Qian, Xianghai Ren, Lifang Fan, Congqing Jiang
The original version of this article contained several errors. In the abstract, "Mechanismly" has been corrected to "Mechanistically."
In the figures, the following corrections were made: Figures 6E andS5E, "expresssion" corrected to "expression"; Figure S1C, "divergent primes" corrected to "divergent primers"; Figure S1E, "liner mRNA" corrected to "linear mRNA"; Figure S3C, "predicated miRNAs" corrected to "predicted miRNAs"; and several errors in Figure 5H, Figure 8E, and Supplementary Figure S6E have been corrected. We have carefully re-examined our original data and corrected these figures. We sincerely apologize for these errors and for any confusion they may have caused. The corrected figures are shown below, and these corrections do not affect the conclusions of the article. This study was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University [No. 2020106]. The corrected original data have been updated in the PDF version of the Supplementary Original Files.
The original article has been corrected.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548445&blobtype=pdf
|
# Retraction for Saeed et al., "Parvulin 14 and Parvulin 17 Bind to HBx and cccDNA and Upregulate Hepatitis B Virus Replication from cccDNA to Virion in an HBx-Dependent Manner"
Umar Saeed, Jumi Kim, Zahid Zahra, Hyeonjoong Piracha, Jaesung Kwon, Yong-Joon Jung, Sun Chwae, Ho-Joon Park, Kyongmin Shin, Kim
## Abstract
The authors hereby retract this article. Upon review, we realized that several immunoblot images and other data were reused within the article.Had this issue been identified earlier, we could have addressed it promptly by referring to the original lab notebooks. However, more than 7 years have passed since the experiments were conducted. As a result, it is now extremely difficult to retrieve or verify the original experimental records.Because of these unresolved concerns, we are retracting the article and regret any inconvenience caused to the readers.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645961&blobtype=pdf
|
# Virion aggregation shapes infection dynamics and evolutionary potential
Meher Sethi, David Vaninsberghe, Bernardo Mainou, Anice Lowen
## Abstract
Viral spread is classically thought to be mediated by single viral particles. However, viruses can also disseminate as aggregates, inside membranous vesicles, and as clusters bound to bacterial or complex surfaces. The implications of collective dispersal for viral infectivity and evolution remain incompletely defined. Here, we used mammalian orthoreovirus to evaluate the impact of aggregation on the propagation of infection and the generation of viral diversity through reassortment. Aggregation of free virions was induced by manipulating pH and ionic conditions. This treatment promoted coordinated delivery of viruses to cells, increasing the number of virions per infected cell and the number of virions per occupied endosome at early times of infection. Likely due to a consolidation of infectious units, aggregation concomitantly reduced the overall infectivity of the viral population and progeny virus yields. When viral popula tions comprised two genetically distinct viruses, aggregation increased the frequency of mixed infection and genetic exchange through reassortment. Thus, the formation of collective infectious units lowers the replicative potential of mammalian orthoreovirus populations but increases viral evolutionary potential by promoting genetic diversifica tion.IMPORTANCE A deeper understanding of the processes shaping viral evolution will advance our ability to anticipate viral emergence, escape from immune responses, and resistance to therapeutics. Although much is known about how genetic variation fuels viral evolution, how modes of viral spread influence the generation and structure of genetic variation remains poorly characterized. Here, we examine how the collective dissemination of viruses modulates early infection dynamics and viral diversity. We find that, although infection in groups reduces the number of independently infected cells, it results in a more genetically diverse progeny population, an outcome that may enhance evolutionary potential.
KEYWORDS reovirus, co-infection, reassortment, collectiveC lassical models of viral infection suggest that viruses disseminate and infect cells as individual particles. Many viruses can be transmitted in groups through a proc ess known as collective dissemination (1). This process includes both within-host and between-host transmission, enabling simultaneous delivery of multiple viral particles to a single cell or a single host, respectively. Mechanisms of collective dissemination include cell-to-cell spread via syncytia or cellular junctions such as plasmodesmata, synapses, and tunneling nanotubes (2, 3); virion aggregation (4-9); transport of virions within extracellular vesicles (10-14); and scaffolding of virions onto bacterial cells (15)(16)(17)(18)(19).Group-based transmission can confer advantages over singular infections (20)(21)(22). These include enhanced infectivity and more rapid establishment of initial infection (10,23,24). For vesicular stomatitis virus (VSV), aggregation increases early-stage reproduc tive success by raising cellular multiplicity of infection, thereby reducing stochastic
infection failure (25). Collective dissemination can also enhance immune evasion by shielding viruses from detection by the host's immune system (26) and boost viral output through sharing of viral resources (24,27). In addition, collective infectious units have elevated stability in extracellular environments compared to single particles (28)(29)(30)(31)(32)(33). Finally, collective dissemination can increase viral evolutionary potential by supporting the diversification of viral populations through recombination (16).
Conversely, collective dissemination may also have negative effects (22,34). These fall into two main categories. First, the merging of multiple virions into a single infectious unit is expected to reduce the number of cells that can potentially be infected by those virions. Thus, particularly if increasing viral input into a cell does not yield a concordant increase in viral output, group-based spread is expected to diminish viral propagation. Second, the replication of multiple genomes within a cell can allow the propagation of deleterious mutations, thereby reducing population fitness. Indeed, aggregation of VSV lowered viral fitness over multiple generations by facilitating accumulation of defective interfering particles, which compete with full-length genomes and impair productive infection (35).
Mammalian orthoreovirus (reovirus), a member of the Reoviridae, is a non-enveloped virus with a segmented double-stranded RNA genome comprising 10 segments (36,37). Reovirus primarily infects epithelial cells of the gastrointestinal and respiratory tracts, but it can also infect the heart and central nervous system in human infants and murine models (38)(39)(40)(41)(42)(43). Reovirus enters cells via receptor-mediated endocytosis, followed by partial disassembly in endosomes to release transcriptionally active cores into the cytoplasm, where replication initiates inside non-membranous structures called inclusion bodies (44,45).
For viruses with segmented genomes, co-infection enables reassortment, a form of genetic exchange in which intact genome segments are shuffled into novel combina tions (46)(47)(48)(49). Reassortment can be a major source of viral diversity: a cell co-infected with two reoviruses that differ in all 10 segments can generate 2 10 (1,024) unique reassortants. In line with this theoretical prediction, we previously found that reovirus reassortment occurs with high efficiency in co-infected cells (50). This high reassort ment capacity (50,51) makes reovirus an ideal model for investigating how collective dissemination and co-infection influence reassortment and viral diversity.
Virus aggregation offers a tractable and physiologically relevant system for study ing collective dissemination. Reovirus engages in collective spread via adhesion to bacteria or dispersal inside extracellular vesicles (12,28,52). Reovirus also aggregates near its isoelectric point, where surface charge is minimized and interparticle attrac tion is favored (53)(54)(55). As an enteric virus, reovirus encounters dramatic pH gradients (~1.5 to ~7.4) during transit through the gastrointestinal tract (56). While the impact of these pH fluctuations on aggregation in vivo remains unclear, acidic conditions have been shown to induce aggregation in vitro (57,58). This aggregation is reversible and can be modulated by adjusting environmental pH, enabling experimental control (7).
Here, we used reovirus aggregation to test our hypothesis that collective disper sal promotes genetic diversification through reassortment. In parallel, we examined whether aggregation limits viral infectious potential by consolidating infectious units. We manipulated viral aggregation state using low pH and assessed outcomes through co-infection assays, genotyping viral progeny, and infectivity measurements. We find that collective dissemination augments viral diversity, but this potential evolutionary benefit comes at a cost of reduced infection efficiency.
## RESULTS
## Reovirus aggregation is pH dependent, reversible, and influenced by buffer composition
To probe the evolutionary implications of collective dissemination, we established a system for modulating the degree of aggregation of mammalian orthoreovirus type 3 Dearing (T3D). We leveraged the work of Floyd and Sharp (7,53,57,59), who showed that virion aggregation is triggered by changes in pH and ionic conditions and that reovirus aggregates around its isoelectric point of pH 3.9. Guided by their findings, we performed a targeted screen of buffers ranging from pH 4 to pH 7.4, analyzing virion aggregation by transmission electron microscopy (TEM) (Fig. 1). We found that, when the virus was suspended in PBS or incubated in acetate pH 5 buffer, most viral objects comprised five or fewer virions with a predominance of single, unaggregated virions. Incubation in tris citrate pH 5 gave an intermediate outcome, and incubation in any of the three pH 4 buffers (tris-citrate, acetate, and citrate) yielded a substantial increase in the frequency of larger aggregates (Fig. 1A andB). Given the prominent aggregation observed in acetate and tris-citrate pH 4 buffers, additional replicates were analyzed with these conditions to confirm reproducibility and identify aggregation conditions suitable for downstream studies (Fig. 1C).
For downstream experiments, it was important to consider the impact of buffer composition on cell health (60). L929 cells showed a significant reduction in viability following exposure to PBS and all low pH buffers tested (Fig. S1A). To mitigate cytotox icity, we diluted these buffers in Opti-MEM. We initially diluted the non-aggregated inoculum (PBS) by 50% and aggregated inocula (acetate, tris-citrate, and citrate) to a composition of 30% low pH buffer and 70% Opti-MEM. However, dilution of the low pH buffers in Opti-MEM reduced the extent of viral aggregation (Fig. S1B). We then tested a broader range of conditions. Except for acetate, dilution to 70% buffer and 30% Opti-MEM improved cell viability (Fig. S2A). Selecting tris-citrate pH 4 for further examination, we evaluated the impact of dilution on aggregation by dynamic light scattering (DLS) (a complementary approach to TEM image analysis that is suitable for more dilute samples) (Fig. S2B). Aggregation levels remained elevated across all tris-citrate formulations compared to PBS, and the 70% formulation resulted in larger aggregates than the 50% condition (Fig. S2B). Finally, we sought to confirm that the treatment conditions used do not impair viral function. We first assessed infectivity in tris citrate vs PBS in the absence of aggregation by omitting the incubation of the virus in the buffer. Viral infectivity and replication were comparable between samples prepared in tris citrate and those prepared in PBS (Fig. S2C andD). We then assessed whether reovirus infectivity is diminished by incubation in PBS or at low pH. Viral titers did not differ significantly between samples prepared in PBS with no incubation and those in PBS with a 4-h incubation (Fig. S2E). To assess the impact of a 4-h incubation at low pH, reovirus was first aggregated using a low pH buffer (tris-citrate, pH 4), and aggregation was then reversed by dilution to a final formulation of 15% tris-citrate in Opti-MEM. Reversal of aggregation was confirmed using DLS (Fig. S3A), and viral infectivity was assessed by plaque assay (Fig. S3B). Aggregated samples showed reduced titers, as expected, but reversal of aggregation restored infectivity to levels comparable to non-aggregated virus prepared in PBS. Thus, the infectivity of T3D reovirus particles was not compromised by incubation in tris-citrate, pH 4 buffer.
For downstream experiments investigating collective dissemination, non-aggregated virus was prepared by suspending the virus in a 1:1 mixture of PBS and Opti-MEM, and aggregated virus was prepared through incubation in tris-citrate pH 4, followed by the addition of Opti-MEM prior to infection, such that tris-citrate made up 50% or 70% of the final inoculum volume.
## Aggregation ensures coordinated delivery of viruses to cells
To evaluate whether viral aggregation enhanced multi-virion delivery to individual cells, we infected cells with virus prepared under non-aggregating and aggregating condi tions. We used TEM to determine if clusters remain intact during viral entry (Fig. 2), if aggregation increases the likelihood of multiple viruses being taken up into endosomes (Fig. 3), and if aggregation results in greater intracellular virion accumulation per cell (Fig. 4).
In cells infected with viral aggregates, large clusters of virions were frequently observed at the cell surface entering as a group (Fig. 2A). In some instances, aggregates appeared to dissipate at the cell membrane, such that virions entered in patches along the cell surface (Fig. 2B). In contrast, cells infected with non-aggregated virus showed individual virions sparsely distributed along the membrane (Fig. 2). Within infected cells, individual endosomes were observed to carry multiple virions more frequently when virus was aggregated compared to when it was not (Fig. 3). Finally, infected cells treated with aggregated virus showed a markedly higher number of intracellular virions per cell compared to those infected with non-aggregated virus (Fig. 4). Together, our TEM analysis demonstrates that viral aggregation facilitates simultaneous entry and uptake of multiple viral particles within individual cells.
## Aggregation promotes co-infection and genetic exchange
To evaluate whether aggregation enhances reassortment, we assessed the frequencies of cellular co-infection and genetic exchange between wild-type (WT) and variant (Var) parental viruses. Var differs from WT by a single synonymous mutation per gene segment, enabling its differentiation from WT but ensuring comparable fitness. The use of WT and Var, therefore, allows quantification of co-infection and reassortment without bias resulting from differing replication kinetics of the two viruses (50,61). We prepared a 1:1 mixture of WT and Var viruses and validated the ratio using two approaches: quantification of WT and Var genome copies (GC) within the bulk mixture and genotyping of 48 plaque isolates derived from the mixture. Both methods showed comparable representation of WT and Var, confirming a 1:1 mixture (Fig. S4A andB).
Using the 1:1 WT-Var inoculum, we measured the impact of aggregation on the frequency of cellular co-infection. To this end, we treated the 1:1 mixture under aggregating and non-aggregating conditions and then isolated 32 plaques from each preparation from three independent experiments. Plaques were genotyped to determine whether the viruses therein were WT-only, Var-only, or mixed. We found that aggre gated samples showed a significantly higher proportion of mixed plaques, indicating that aggregation promotes co-infection (Fig. 5A). Importantly, this result indicates that aggregates typically comprise multiple infectious virions, suggesting that non-infectious virions are not a major component of the viral population.
We next examined how aggregation modulates the frequency of reassortment. Cells were co-infected with WT and Var using the 1:1 viral inoculum at MOIs of 0.1 or 1 GC/cell, and replication was allowed to proceed for 24 h. Levels of reassortment were determined by analyzing progeny from six biological replicates, genotyping 32 plaque isolates per replicate. We observed that both aggregation formulations (50% and 70% tris-citrate) showed increased frequencies of reassortant progeny compared to the non-aggregated control at both MOIs (Fig. 5B). To confirm that the observed effects were not an artifact of buffer composition, we validated PBS as a reliable negative control by comparing virus prepared in PBS to that made in tris-citrate buffer without incubation (Fig. S4C). As expected, when the incubation period that allows virion aggregation was omitted, no significant differences in reassortment frequencies were observed among inocula delivered in PBS and 50% and 70% tris-citrate formulations (Fig. S4C). Together, these findings indicate that viral aggregation increases the likelihood of co-infection, which in turn facilitates genetic exchange through reassortment.
## Aggregation modulates infection dynamics
To test the impact of aggregation on viral infection dynamics, we first evaluated the infectious potential of the inoculum. To this end, we performed plaque assays using virus preparations made under aggregating and non-aggregating conditions (Fig. 6). Both aggregation conditions (50% and 70% tris-citrate) yielded significantly lower titers compared to the non-aggregated virus (PBS) (Fig. 6). Additionally, the 70% tris-cit rate condition, which showed a higher frequency of large aggregates (Fig. S2B), was associated with slightly lower titers compared to the 50% tris-citrate condition. Next, we assessed the percentage of cells infected following inoculation of cell monolayers with aggregated or non-aggregated viral preparations and found that aggregation reduced the percentage of cells infected, with 70% tris-citrate showing a greater effect than 50% tris-citrate (Fig. 7A). Finally, we evaluated viral output at 24 h post-infection to define the impact of aggregation on early rounds of infection (Fig. 7B). Progeny viral titers were significantly reduced in the 70% tris-citrate aggregation condition at both MOIs. In the 50% tris-citrate condition, however, progeny viral titers remained comparable with PBS (Fig. 7B). In conclusion, while aggregation enhances coordinated delivery of multiple viruses to cells and reassortment, it imposes a cost by limiting viral yield from infected cell populations.
## DISCUSSION
We investigated how reovirus aggregation, a form of collective dispersal, shapes viral genetic exchange and infection dynamics. We found that aggregation increases the frequency with which multiple viruses enter the same cell and, under conditions of mixed infection, enhances reassortment. However, high levels of aggregation diminished replication efficiency by reducing the number of cells infected. These findings suggest that collective dispersal can augment viral evolutionary potential but may incur the cost of reduced dissemination.
Several distinct mechanisms of viral clustering can mediate collective dissemination (7,9,12,17). Here, we focus on aggregation governed by direct physicochemical interactions (7,62). Aggregation likely occurs as the surrounding pH approaches the virus's isoelectric point, reducing net surface charge and enabling transient particle-par ticle associations (54). This type of aggregation is dynamic and reversible (7), suggesting that nonspecific electrostatic interactions may modulate infection efficiency in response to environmental pH. Consistent with prior work (53,57,59), we show that exposure of reovirus to low pH near its isoelectric point (~3.9) promoted aggregation, and this aggregation was reversible. Importantly, this mode of aggregation is likely physiologi cally relevant, as reovirus encounters pH ranging from 1.5 to 7.4 during gastrointestinal transit (53,56).
Our data show that virion aggregation facilitates group-based delivery, increasing the number of virions delivered to the same cell and resulting in high intracellular virus accumulation. When multiple viral genotypes are present, we found that aggre gation enhances the frequency of their co-infection and reassortment, leading to the production of genetically diverse progeny. Diversity produced through reassortment may accelerate the emergence of high fitness variants under selective pressure (63)(64)(65)(66)(67)(68)(69). Thus, aggregation may offer a benefit by increasing the adaptability of viral populations. Of note, not all modes of viral clustering are likely to enhance diversity (34). Viral aggregates that form extracellularly can include progeny viruses from distinct foci of infection, which are more likely to differ genetically. In contrast, vesicular ensembles of viruses formed within cells would be more likely to carry closely related viruses replicated from the same parental genome (34). In the latter case, co-infection and reassortment would not increase diversity. Infection of single cells with multiple viruses can be beneficial for viral replication. The extent to which multiple infection modulates per virion burst size has been examined in detail for influenza A virus (70-72) but has not, to our knowledge, been evaluated for reovirus. In the case of influenza A virus, the speed and productivity of replication are often increased under conditions conducive to multiple infection via two mechanisms: genetic complementation and increased efficiency of replication through "strength in numbers" (70,73,74). The former mechanism is necessitated by a high frequency of incomplete viral genomes within singly infected cells, while the latter is more potent under adverse conditions (70,74,75). Importantly, however, diminishing returns are observed at very high MOIs (70,73). While the frequency of incomplete genomes for reovirus is unknown, the production of defective interfering particles that lack portions of essential open reading frames has been documented (76). Collective dissemination would allow complementation of defective interfering genomes, but these genomes may then suppress virus production from the cells that carry them (77,78). The extent to which coordinated infection with multiple, genetically identical, reoviruses may modulate replication efficiency is not clear. Our data offer some insight, but not in the quantitative detail needed to draw strong conclusions. We found that, despite increased intracellular viral density, high levels of reovirus aggregation (70% tris-citrate condition) resulted in reduced progeny output during early rounds of infection. This reduction was likely due to the consolidation of infectious units, leading to fewer infected cells, an effect that was not fully counterbalanced by any potential benefits of multiple infection. Intermediate levels of aggregation (50% tris-citrate condition) did not result in detecta ble replication deficits, however, suggesting that costs and benefits of multiple infection may be balanced in this context. The differing effects seen with high and intermediate levels of reovirus aggregation may reflect diminishing returns of increasing multiplicity, as seen with increasing influenza A virus MOI (70,71). Diminishing returns could result from the competition of co-infecting viruses for the same, finite pool of resources within a cell and/or the strong activation of antiviral responses by the entry of viral aggregates (71).
Some limitations of our study are important to consider. First, to allow controlled manipulation of aggregation conditions, the study was conducted in an in vitro system, which does not capture the complexity of in vivo environments. Aspects of the host environment, such as spatiotemporal variation in pH and the presence of bacteria (7,15,17,28), can modulate viral clustering and may alter the balance of costs and benefits associated with collective dissemination. Second, experimental manipulation of viral aggregation was applied only to the inoculum and did not persist through multiple rounds of viral replication. Although our focus on the first 24 h of infection reduces the contribution of secondary spread, this factor limits our ability to interpret results quantitatively. Third, we focused on a single reovirus strain, T3D, leaving uncertainty in the extent to which our findings apply across other reovirus strains and viral families.
In conclusion, our results support a model in which virion aggregation mediates collective dissemination, enhancing group-based delivery, co-infection, and reassort ment under certain environmental conditions, in turn increasing viral population diversity and evolutionary potential. Aggregation shifts viral dynamics from widespread, low multiplicity infection to localized, high multiplicity infection, increasing genetic exchange but limiting infection potential (Fig. 8). In this way, aggregation may balance short-term replication inefficiency with long-term evolutionary resilience.
## MATERIALS AND METHODS
## Cells
Spinner-adapted L929 (NCTC clone 929 mouse fibroblast) cells (a gift from Dr. Terry Dermody, University of Pittsburgh) were maintained at 37°C in a humidified atmosphere without CO 2 supplementation in complete Joklik's modified Eagle medium (JMEM). For experiments, L929 cells were seeded and incubated at 37°C in a humidified atmos phere containing 5% CO 2 , which induced an adherent phenotype. Complete JMEM was prepared from a powdered formulation (US Biological # M3867). The pH was adjusted to 7.2 using 10 N NaOH. The medium was then supplemented with 5% FBS, 2 mM L-glutamine (Corning), 1% penicillin-streptomycin (PS) (Corning), and 0.25 mg/mL amphotericin B (Sigma-Aldrich). Prior to use, the medium was sterilized by filtration through a 0.22 µm polyether sulfone membrane (PES) filter (Corning). BHKT-7 cells (79) were cultured in Dulbecco's Modified Eagle's Medium (Gibco) supplemented with 5% FBS, 2 mM L-glutamine, PS, and 1 mg/mL G418 (Invivogen) at 37°C in 5% CO 2 . Cells were tested for mycoplasma contamination monthly, and any positive cultures were promptly discarded.
## Viruses
Wild-type T3D reovirus (NCBI accession no. SRX6802327) and variant (PMID 31511390) T3D reovirus (61) were generated by reverse genetics (80,81). T3D Var virus contained the following previously described point mutations: L1 C612T, L2 C853T, L3 G481A, M1 C919T, M2 A650G, M3 T702C, S1 G312A, S2 A438G, S3 T318C, and S4 C383T. Both viruses were produced in BHK-T7 cells and propagated in L929 cells for three serial passages. Viral stocks were subjected to Vertrel-XF extraction, and the virus in the resulting aqueous fraction was purified by ultracentrifugation through a 1.2-1.4 g/mL cesium chloride gradient. The resulting viral band was collected, dialyzed, and stored at 4°C (82). 2), viral yield is largely maintained (5). However, under high aggregation conditions (1) where large aggregates are formed, yield decreases (6). Importantly, reassortment increases ( 8) with aggregation (1) due to enhanced co-infection, resulting in greater genetic diversity among progeny and increasing evolutionary potential.
## Preparation of aggregated and non-aggregated forms of T3D reovirus
To induce aggregation, T3D reovirus was added to low pH buffers at a final concentra tion of 2.7 × 10 4 genome copies/µL. The mixture was then left undisturbed at room temperature for 4 h to allow aggregation. For the non-aggregated control, the virus was re-suspended in 1× PBS and used immediately. Buffers for aggregation were prepared in cell culture-grade water (Corning), filtered through a 0.22 µm PES filter (Corning), and stored at 4°C. Tris-citrate buffers (0.05 M) were prepared using 0.1 M Tris base and 0.04 M citric acid. Acetate buffers (0.05 M) were prepared using glacial acetic acid. Citrate buffer (0.05 M) was prepared using citric acid. In each case, the final pH was adjusted using 10 N NaOH.
To minimize L929 cytotoxicity associated with exposure to low pH buffers and PBS (Corning), Opti-MEM (Gibco) was added to viral samples prior to infection. Non-aggrega ted control virus "PBS (No incubation)" was resuspended in a 1:1 mixture of PBS and Opti-MEM without incubation. Aggregated preparations were combined with Opti-MEM to yield mixtures in which tris-citrate buffer constituted either 50% or 70% of the total volume. All virus preparations were used for infection immediately following dilution with Opti-MEM.
To reverse aggregation, following the 4-h aggregation treatment, Opti-MEM (Gibco) was added to the viral inoculum so that tris-citrate buffer constituted 15% of the total volume, after which it was incubated at 37°C for an additional 1 h.
## Transmission electron microscopy (TEM) to visualize viral aggregates
To analyze virus aggregation, TEM was performed using a Talos 120C microscope. Virus samples of 2.7 × 10 5 GC in 10 µL of specified buffer (final concentration 2.7 × 10 4 GC/µL) were prepared for negative staining as follows. A 0.75% (wt/vol) uranyl formate solution was freshly prepared each day by dissolving 0.0375 g uranyl formate stain in 5 mL of warm deionized water. The solution was vortexed thoroughly to disperse any clumps, followed by the addition of 4 µL of 10 N NaOH and filtered through a 0.22 µm syringe filter. Copper square grids (mesh size 400) with 10 nm carbon film (VWR # 103303-424) were glow discharged to increase hydrophilicity and ensure sample adhesion. For each sample, 3 µL of virus suspension was applied to the grid and allowed to absorb for 1 minute before rinsing twice with a drop of ultra-pure water and blotted dry. The grid was then stained with 0.75% (wt/vol) uranyl formate solution for 1 minute and blot dried before visualization. Micrographs were acquired from multiple randomly selected square openings of the 400-mesh TEM grid. Each field of view encompassed either a section (Fig. 1A) or an entire area (Fig. 1B andC) of one hole in the holey carbon support film.
## Electron microscopy to visualize viral entry
L929 cell monolayers (5 × 10 5 cells/well in a 24-well plate) were infected with aggregated (50% tris-citrate inoculum) or non-aggregated (1:1 PBS-Opti-MEM mixture) forms of T3D reovirus. Following infection for the designated time, cells were fixed in the culture plate without removing the inoculum, with a mixture of 2.5% glutaraldehyde, 1.0% paraformaldehyde, 2.6 mM MgCl 2 , 2.6 mM CaCl 2 , 50 mM KCl, 0.01% picric acid, and 2% sucrose in 0.1 M cacodylate buffer overnight at 4°C. The fixative was then removed, and cells were washed three times with fresh 0.1 M cacodylate buffer at pH 7.4. The cells underwent a second round of fixation with 1.0% osmium tetroxide for 1 h to strengthen the cellular cytoskeleton. The cell monolayers still attached to the culture plate were then washed with water and stained en bloc with 2% aqueous uranyl acetate for 20 minutes at 60°C. Monolayers were then dehydrated in a stepwise manner in 25%, 50%, 70%, and 95% ethanol, followed by three changes in 100% ethanol. Monolayers were then infiltrated with Eponate 12 resin and polymerized for 48 h at 60°C. Polymerized blocks were then removed from the culture plate with the cells flat embedded in the bottom of the block. Ultrathin sections were cut from the bottom of the block with a Leica EM UC6 ultramicrotome, stained with uranyl acetate and lead citrate, and imaged in a Jeol JEM 1400 TEM operated at 80 kV. Electron micrographs were acquired on a 2,048 × 2,048 charge-coupled device camera (UltraScan 1000, Gatan Inc., Pleasanton, CA, USA).
## Cell viability assay to measure cytotoxicity of buffers
L929 cells (1.5 × 10 4 cells/well) were seeded in 96-well, clear bottom plates and incubated overnight. After removing the culture medium, the cell monolayers were treated with 20 µL of buffer mixed with escalating proportions of Opti-MEM for 1 h on a shaker in a 37°C and 5% CO 2 incubator. After 1 h, the bufferOptiMEM mixtures were removed, and 200 µL of complete JMEM was added to each well. Cell viability was tested using the Cell Proliferation Kit (Roche # 11465007001), following the manufacturer's instructions. Briefly, the 96-well culture plates were incubated for an additional 6 h at 37°C and 5% CO 2 . Next, 10 µL of 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide reagent at a final concentration of 0.5 mg/mL was added to each well and mixed thoroughly. After 5 h, 100 µL of solubilization buffer (10% SDS in 0.01 M HCl) was added to each well to dissolve the purple formazan crystals formed by viable cells. The solubilization reaction was allowed to proceed for 14 h at 37°C and 5% CO 2 before absorbance was read at a wavelength of 580 nm using a Synergy H1 plate reader (BioTek).
## Dynamic light scattering to analyze virus aggregation
To assess particle size and aggregation state, DLS was performed using Wyatt Dynapro instrument with Dynamics 8.3.1 software. Each sample contained 3.5 × 10 8 GC/mL of virus, resuspended in PBS or incubated in the designated low pH buffer to induce aggregation; Opti-MEM was included where indicated.
For standard DLS experiments measuring aggregation levels, a 200 µL sample was prepared and run in three 10 µL technical replicates using disposable micro cuvettes. For reversal of aggregation, the full 200 µL sample was imaged by splitting the inoculum across three cuvettes.
DLS detects time-dependent fluctuation in scattered light intensity caused by Brownian motion. These fluctuations were used to calculate the diffusion coefficient, and the hydrodynamic radius (Rh) was derived using the Stokes-Einstein equation by the Dynamics 8.3.1 software. Five measurements, each consisting of 10 acquisitions per sample, were used to generate average size distributions.
To estimate virion counts per object, we assumed an average virion diameter of 80 nm and used equation 1:
(1) Size of objects (number of virions per object) = 2 × Rh 80 nm
## Droplet digital PCR to measure genome copies
To prepare samples for droplet digital PCR (ddPCR), total RNA was extracted from 10 µL of viral stocks (WT, Var, and 1:1 mixture) using the QIAamp Viral RNA Mini Kit (Qiagen), following the manufacturer's protocol (carrier RNA was not added). RNA was eluted in 40 µL of ultra-pure water and stored at -80°C until use. cDNA was synthesized from 12.8 µL of RNA in 20 µL reactions using Maxima reverse transcriptase (Thermo Scientific, Cat # EP0742) with the provided buffer, random hexamers at a final concentration of 0.01 µg/µL (Thermo Scientific, S0142), dNTPs at a final concentration of 0.5 mM, and Ribolock RNase inhibitor (Thermo Scientific, EO0381), following the manufacturer's instructions. DdPCR was performed using the QX200 Droplet Digital PCR system (Bio-Rad). Each reaction included ddPCR Supermix (no dUTP) (Bio-Rad, Cat # 1863024), forward (AGA GTGGCTCAAACGTTGCT) and reverse (TCCAATGCAGTCGGTAGTGA) primers (IDT), probes (IDT) specific for WT (ATT +G+GA GA+a +GT+C T+CT TG) and Var (C+GA TT+G +GAG A+g+G TCT) templates and diluted cDNA. The symbol "+" denotes the sites where LNA was added to raise the Tm of the probe, and the nucleotides represented by a lower case letter in the probe sequences highlight the site of point mutation that distinguishes Var sequence from WT. Final concentration of each primer and probe was 0.5 µM. Droplets were generated according to the manufacturer's instructions and transferred to a 96-well PCR plate. Plates were heat sealed, and targets were amplified using the following thermocycling conditions, with all ramp rates set to 2°C/s: 95°C, 10 minutes (94°C, 30 s; 58°C, 1 min) × 49; 98°C, 10 minutes; and 4°C, hold. Data were analyzed in QuantaSoft software (Bio-Rad, version 1.7.4.0917), and fluorescence thresholds were manually set based on negative controls.
## Plaque assay to determine infectious titers and isolate plaques
Overlay for plaque assays comprised a 1:1 mixture of complete plaque assay medium (2× Medium 199, 5% FBS, 1% L-glutamine, 1% penicillin-streptomycin, and 1% amphotericin B) and 2% Difco-Bacto agar (BD Diagnostics). L929 cells were seeded at a concentration of 2 × 10 6 cells/well in 6-well plates. Cell culture media were removed, and 200 µL of aggregated or non-aggregated forms of T3D reovirus was added (no prior wash step). After a 1-h attachment period at room temperature, the inoculum was removed, and the cells were washed once with DPBS (with calcium and magnesium; Corning) (82). Three milliliters of overlay were added, and plates were incubated at 37°C for 6 days. Cells were supplemented with 2 mL of overlay at 3 days post-infection. On day 6, cells were stained through addition of 1% neutral red (Thermo Scientific) solution prepared in a 1:1 mixture of complete media and agar and incubated at 37°C for 18 h. Plaque isolates were harvested by identifying well-separated viral plaques (at least 1 cm apart) and excising a plug of overlay over each using a sterile 1 mL serological pipette. Each plug was dispensed into 160 µL PBS.
## Genotyping viral plaques using probe-based ddPCR
RNA was extracted from plaque isolates using either the Quick RNA 96 Extraction Kit (Zymo # R1035) or the NucleoMag RNA extraction kit (Macherey-Nagel # 744350.4), following the manufacturer's instructions. NucleoMag RNA extraction was performed with assistance from Eppendorf's epMotion 5075 liquid handler. RNA was eluted in either 40 µL (Zymo) or 60 µL (NucleoMag) of nuclease-free water. A volume of 12.8 µL of RNA was used for cDNA synthesis with Maxima RT, as described above.
To determine whether individual plaques were derived from WT, Var, or a mixture of both viruses, cDNA was diluted 1:20 in nuclease-free water and then 6 µL was used to prepare ddPCR reactions as outlined above.
Genotype assignment was based on probespecific detection of WT or Var alleles according to probespecific fluorescence (FAM for WT and HEX for Var), enabling discrimination of WT, Var, double-positive, and negative droplets. Probe specificity was validated, confirming that each probe bound to its target without minimal offtarget binding. The limit of detection for this assay was set to 787.5 genome copies/µL based on background signal observed for WT probe on Var template and Var probe on WT template.
## Measurement of co-infection and reassortment
L929 cells (5.5 × 10 5 cells/well in 12-well plates) were inoculated with 200 µL of aggrega ted or non-aggregated WT-Var mixtures at MOIs of 1.0 or 0.10 genome copies per cell. After a 1-h attachment period at 37°C, the inoculum was removed, the cells were washed once with DPBS (with calcium and magnesium), 1 mL of complete JMEM was added to each well, and the plates were incubated at 37°C for 24 h.
## Flow cytometry to analyze the proportion of cells infected
At the end of the 24-h incubation, cells were detached from the plate using trypsin (Corning) and moved to a 96-well V-bottom plate for immunostaining. Cells were washed with FACS buffer (2% FBS in 1× PBS) and then resuspended with Zombie NIR dye (BioLegend) to stain dead cells. Next, cells were incubated for 15 minutes with 0.1 µg/mL rat anti-mouse CD16/CD32 Fc block (BD Pharminogen, Clone 2.4G2) at 4°C, followed by fixation and permeabilization according to the BD Cytofix/Cytoperm kit (Bd biosciences # 554714). Cells were then stained with 0.31 mg/mL mouse monoclonal anti-σ3 antibody (clone 10C1) for 45 minutes at 4°C and then washed twice with 1× BD Perm/Wash buffer before adding an AlexaFluor-647 conjugated donkey anti-mouse secondary antibody (Invitrogen A31571) at a 1:1,000 dilution. All samples were analyzed on BD LSRFortessa X-20 instrument at a fixed flow rate of 12 µL/min. Compensation was performed using UltraComp beads (Invitrogen, Cat # 01-2222-42). Data were analyzed using FlowJo 10.9 software.
## Harvesting cell lysates to measure progeny viral yield
At the end of the 24-h incubation, co-infected cell cultures were subjected to three freeze-thaw cycles at -80°C to disrupt cells and release viruses. Plaque assays were then performed using the cell lysates.
## Genotyping viral plaques by high-resolution melt analysis
To distinguish WT and Var segments and thereby quantify reassortment, we applied high-resolution melt analysis to plaque isolates. At the end of the 24-h incubation, cell lysates were collected from co-infected cell cultures. Plaque assays were then performed using these cell lysates, and plaques were isolated. RNA was extracted from plaque isolates and converted to cDNA using Maxima RT as described above. cDNA was diluted 1:3.5 by adding 50 µL of nuclease-free water to 20 µL of cDNA. The diluted cDNA was used as input for qPCR with a 0.4 µM mixture of segmentspecific primers as previously described (61). Five microliter reactions were prepared in 384-well plates using Precision Melt Supermix (Bio-Rad) and run on a CFX384 Touch Real-Time PCR Detection System (Bio-Rad). Amplification was assessed using CFX Manager 3.1 software, and genotypes were assigned using Precision Melt Analysis 1.3 software (Bio-Rad) based on the distinct melting curves of WT and Var templates. Plaques carrying any combination of WT and Var gene segments were defined as reassortant, and percentage of reassortment was calculated using equation 2:
(2) Percent reassortment (%) = Reassortant plaques Total plaques screened × 100
## References
1. Sanjuán (2017) "Collective infectious units in viruses" *Trends Microbiol*
2. Mothes, Sherer, Jin et al. (2010) "Virus cell-to-cell transmis sion" *J Virol*
3. Ganti, Han, Manicassamy et al. (2021) "Rab11a mediates cellcell spread and reassortment of influenza A virus genomes via tunneling nanotubes" *PLoS Pathog*
4. Bald, Briggs (1937) "Aggregation pf virus particles" *Nature*
5. Galasso, Sharp (1962) "Virus particle aggregation and the plaqueforming unit" *J Immunol*
6. Galasso, Sharp, Sharp (1964) "The influence of degree of aggregation and virus quality on the plaque titer of aggregated vaccinia virus" *J Immunol*
7. Floyd, Sharp (1977) "Aggregation of poliovirus and reovirus by dilution in water" *Appl Environ Microbiol*
8. Cuevas, Durán-Moreno, Sanjuán (2017) "Multi-virion infectious units arise from free viral particles in an enveloped virus" *Nat Microbiol*
9. Pradhan, Varsani, Leff et al. (2022) "Viral aggregation: the knowns and unknowns" *Viruses*
10. Chen, Du, Hagemeijer et al. (2015) "Phosphatidylserine vesicles enable efficient en bloc transmission of enteroviruses" *Cell*
11. (2025) *Full-Length Text Journal of Virology*
12. Altan-Bonnet, Perales, Domingo (2019) "Extracellular vesicles: vehicles of en bloc viral transmission" *Virus Res*
13. Smith, Krystofiak, Ogden (2024) "Mammalian orthoreovirus can exit cells in extracellular vesicles" *PLoS Pathog*
14. Williams (1985) "Membrane-associated viral complexes observed in stools and cell culture" *Appl Environ Microbiol*
15. Kadiu, Narayanasamy, Dash et al. (2012) "Biochemical and biologic characterization of exosomes and microvesi cles as facilitators of HIV-1 infection in macrophages" *J Immunol*
16. Berger, Mainou (2018) "Interactions between enteric bacteria and eukaryotic viruses impact the outcome of infection" *Viruses*
17. Erickson, Jesudhasan, Mayer et al. (2018) "Bacteria facilitate enteric virus co-infection of mammalian cells and promote genetic recombination" *Cell Host Microbe*
18. Neu, Mainou (2020) "Virus interactions with bacteria: partners in the infectious dance" *PLoS Pathog*
19. Rowe, Meliopoulos, Iverson et al. (2019) "Direct interactions with influenza promote bacterial adherence during respiratory infections" *Nat Microbiol*
20. Hament, Aerts, Fleer et al. (2005) "Direct binding of respiratory syncytial virus to pneumococci: a phenomenon that enhances both pneumococcal adherence to human epithelial cells and pneumococcal invasiveness in a murine model" *Pediatr Res*
21. Sanjuán (2018) "Collective properties of viral infectivity" *Curr Opin Virol*
22. Sanjuán (2021) "The social life of viruses" *Annu Rev Virol*
23. Sanjuán, Thoulouze (2019) "Why viruses sometimes disperse in groups?" *Virus Evol*
24. Petersen, Lu, Fitzgerald et al. (2022) "Unique aggregation of retroviral particles pseudotyped with the Delta variant SARS-CoV-2 spike protein" *Viruses*
25. Santiana, Ghosh, Ho et al. "Altan-Bonnet N. 2018. Vesicle-cloaked virus clusters are optimal units for inter-organismal viral transmission" *Cell Host Microbe*
26. Andreu-Moreno, Sanjuán (2018) "Collective infection of cells by viral aggregates promotes early viral proliferation and reveals a cellular-level allee effect" *Curr Biol*
27. Feng, Hensley, Mcknight et al. (2013) "A pathogenic picornavirus acquires an envelope by hijacking cellular membranes" *Nature*
28. Shirogane, Watanabe, Yanagi (1235) "Cooperation between different RNA virus genomes produces a new phenotype" *Nat Commun*
29. Berger, Yi, Kearns et al. (2017) "Bacteria and bacterial envelope components enhance mammalian reovirus thermostability" *PLoS Pathog*
30. Aguilera, Pfeiffer (2019) "Strength in numbers: mechanisms of viral co-infection" *Virus Res*
31. Sharp, Floyd, Johnson (1975) "Nature of the surviving plaqueforming unit of reovirus in water containing bromine" *Appl Microbiol*
32. Mattle, Kohn (2012) "Inactivation and tailing during UV254 disinfection of viruses: contributions of viral aggregation, light shielding within viral aggregates, and recombination" *Environ Sci Technol*
33. Galasso, Sharp (1965) "Effect of particle aggregation on the survival of irradiated vaccinia virus" *J Bacteriol*
34. Zhang, Ghosh, Li et al. (2022) "Vesicle-cloaked rotavirus clusters are environmentally persistent and resistant to free chlorine disinfection" *Environ Sci Technol*
35. Bou, Geller, Sanjuán (2019) "Membrane-associated enteroviruses undergo intercellular transmission as pools of sibling viral genomes" *Cell Rep*
36. Andreu-Moreno, Sanjuán (2020) "Collective viral spread mediated by virion aggregates promotes the evolution of defective interfering particles" *mBio*
37. Bellamy, Shapiro, August et al. (1967) "Studies on reovirus RNA. I. Characterization of reovirus genome RNA" *J Mol Biol*
38. Shatkin, Sipe, Loh (1968) "Separation of ten reovirus genome segments by polyacrylamide gel electrophoresis" *J Virol*
39. Konopka-Anstadt, Mainou, Sutherland et al. (2014) "The Nogo receptor NgR1 mediates infection by mammalian reovirus" *Cell Host Microbe*
40. Barton, Forrest, Connolly et al. (2001) "Junction adhesion molecule is a receptor for reovirus" *Cell*
41. Tyler, Mcphee, Fields (1986) "Distinct pathways of viral spread in the host determined by reovirus S1 gene segment" *Science*
42. Wolf, Rubin, Finberg et al. (1981) "Intestinal M cells: a pathway for entry of reovirus into the host" *Science*
43. Excoffon, Guglielmi, Wetzel et al. (2008) "Reovirus preferentially infects the basolateral surface and is released from the apical surface of polarized human respiratory epithelial cells" *J Infect Dis*
44. Sherry, Schoen, Wenske et al. (1989) "Derivation and characterization of an efficiently myocarditic reovirus variant" *J Virol*
45. Dermody, Parker, Sherry (2013) "Orthoreovirus"
46. Roth, Aravamudhan, De Castro et al. (2021) "Ins and outs of reovirus: vesicular trafficking in viral entry and egress" *Trends Microbiol*
47. Cross, Fields (1976) "Use of an aberrant polypeptide as a marker in three-factor crosses: further evidence for independent reassortment as the mechanism of recombination between temperature-sensitive mutants of reovirus type 3" *Virology (Auckl)*
48. Racaniello, Palese (1979) "Isolation of influenza C virus recombinants" *J Virol*
49. Mcdonald, Nelson, Turner et al. (2016) "Reassortment in segmented RNA viruses: mechanisms and outcomes" *Nat Rev Microbiol*
50. Lowen (2018) "It's in the mix: reassortment of segmented viral genomes" *PLoS Pathog*
51. Hockman, Jacobs, Mainou et al. (2022) "Mammalian orthoreovirus reassortment proceeds with little constraint on segment mixing" *J Virol*
52. Thoner, Meloy, Long et al. (2022) "Reovirus efficiently reassorts genome segments during coinfec tion and superinfection" *J Virol*
53. Erickson, Jesudhasan, Mayer et al. (2018) "Bacteria facilitate viral co-infection of mammalian cells and promote genetic recombination" *Cell Host Microbe*
54. (2025) *Full-Length Text Journal of Virology*
55. Floyd, Sharp (1979) "Viral aggregation: buffer effects in the aggregation of poliovirus and reovirus at low and high pH" *Appl Environ Microbiol*
56. Gerba, Betancourt (2017) "Viral aggregation: impact on virus behavior in the environment" *Environ Sci Technol*
57. Michen, Graule (2010) "Isoelectric points of viruses" *J Appl Microbiol*
58. Yamamura, Inoue, Nishino et al. (2023) "Intestinal and fecal pH in human health" *Front Microbiomes*
59. Floyd, Sharp (1978) "Viral aggregation: quantitation and kinetics of the aggregation of poliovirus and reovirus" *Appl Environ Microbiol*
60. (1978) "Viral aggregation: effects of salts on the aggregation of poliovirus and reovirus at low pH"
61. Floyd, Sharp (1978) "Viral aggregation: effects of salts on the aggregation of poliovirus and reovirus at low pH" *Appl Environ Microbiol*
62. Kęska, Rusak, Włostowski et al. (2024) "Low-vacuum SEM imaging and viability test of L929 cells exposed to a Euro 6 diesel exhaust gas mixture in a BAT-CELL chamber in comparison with hydrocarbons emission" *Sci Rep*
63. Hockman, Phipps, Holmes et al. (2020) "A method for the unbiased quantification of reassortment in segmented viruses" *J Virol Methods*
64. Zhang, Li, Prigiobbe (2022) "Population balance modeling of homogeneous viral aggregation" *Chem Eng Sci*
65. Hakim, Gazali, Widyaningsih et al. (2024) "Driving forces of continuing evolution of rotaviruses" *World J Virol*
66. Midthun, Valdesuso, Hoshino et al. (1987) "Analysis by RNA-RNA hybridization assay of intertypic rotaviruses suggests that gene reassortment occurs in vivo" *J Clin Microbiol*
67. Matthijnssens, Ciarlet, Heiman et al. (2008) "Full genome-based classification of rotaviruses reveals a common origin between human Wa-like and porcine rotavirus strains and human DS-1-like and bovine rotavirus strains" *J Virol*
68. Nakagomi, Kaga, Nakagomi (1992) "Human rotavirus strain with unique VP4 neutralization epitopes as a result of natural reassortment between members of the AU-1 and Wa genogroups" *Arch Virol*
69. Palese (2004) "Influenza: old and new threats"
70. Scholtissek, Von Hoyningen, Rott (1978) "Genetic relatedness between the new 1977 epidemic strains (H1N1) of influenza and human influenza strains isolated between 1947 and 1957 (H1N1)" *Virology (Auckl)*
71. Sanjuán, Domingo-Calap (2021) "Genetic diversity and evolution of viral populations" *Encycl Virol*
72. Jacobs, Onuoha, Antia et al. (2019) "Incomplete influenza A virus genomes occur frequently but are readily complemented during localized viral spread" *Nat Commun*
73. Martin, Harris, Sun et al. (2020) "Cellular coinfection can modulate the efficiency of influenza A virus production and shape the interferon response" *PLoS Pathog*
74. Heldt, Kupke, Dorl et al. (2015) "Single-cell analysis and stochastic modelling unveil large cell-to-cell variability in influenza A virus infection" *Nat Commun*
75. Phipps, Ganti, Jacobs et al. (2020) "Collective interactions augment influenza A virus replication in a hostdependent manner" *Nat Microbiol*
76. Shartouny, Lee, Delima et al. (2022) "Beneficial effects of cellular coinfection resolve inefficiency in influenza A virus transcription" *PLoS Pathog*
77. Brooke, Ince, Wrammert et al. (2013) "Most influenza A virions fail to express at least one essential viral protein" *J Virol*
78. Smith, Gribble, Diller et al. (2021) "Reovirus RNA recombination is sequence directed and generates internally deleted defective genome segments during passage" *J Virol*
79. Segredo-Otero, Sanjuán (2019) "The effect of genetic complementa tion on the fitness and diversity of viruses spreading as collective infectious units" *Virus Res*
80. Fonville, Marshall, Tao et al. (2015) "Influenza virus reassortment is enhanced by semi-infectious particles but can be suppressed by defective interfering particles" *PLoS Pathog*
81. Buchholz, Finke, Conzelmann (1999) "Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter" *J Virol*
82. Kobayashi, Antar, Boehme et al. (2007) "A plasmid-based reverse genetics system for animal double-stranded RNA viruses" *Cell Host Microbe*
83. Stewart, Berry, Berger et al. (2019) "Enhanced killing of triple-negative breast cancer cells by reassortant reovirus and topoisomerase inhibitors" *J Virol*
84. Coombs (2023) "Mammalian reoviruses: propagation, quantification, and storage" *Curr Protoc*
85. (2025) *Full-Length Text Journal of Virology*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548455&blobtype=pdf
|
# Retraction for Piracha et al., "Sirtuin 2 Isoform 1 Enhances Hepatitis B Virus RNA Transcription and DNA Synthesis through the AKT/GSK-3β/β-Catenin Signaling Pathway"
Zahra Piracha, Hyeonjoong Kwon, Umar Saeed, Jumi Kim, Jaesung Jung, Yong-Joon Chwae, Sun Park, Ho-Joon Shin, Kyongmin Kim
## Abstract
The authors hereby retract this article. Following a recent request to provide the original data underlying this publication, it was identified that several immunoblot images were reused within the article. This issue was not recognized at the time of submission and publication.As the experiments were conducted more than 7 years ago, it is not possible to recover or verify the original experimental records.Because of these unresolved concerns, we are retracting the article and regret any inconvenience caused to the readers.
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12390207&blobtype=pdf
|
# Evaluation of Allplex™ GI-Parasite Assay-A Multiplex Real Time PCR for the Diagnosis of Intestinal Protozoa: A Multicentric Italian Study
Ester Oliva, Libera Clemente, Nicola Menegotto, Stefania Varani, Antonella Bruno, Raffaele Gargiulo, Luciana Petrullo, Claudio Farina, Annibale Raglio
## Abstract
Background: The microscopic examination of stool samples remains the reference method for the diagnosis of intestinal protozoal infections; however, this technique is time consuming and requires experienced and well-trained operators. Therefore, there is a growing interest in molecular diagnostic techniques, including commercial PCR assays. The aim of this multicentric study was to evaluate a commercial real-time PCR for the detection of intestinal protozoa in fecal samples. Methods: The samples were routinely examined using conventional techniques, such as macro-and microscopic examination after concentration, Giemsa or Trichromic stain, Giardia duodenalis, Entamoeba histolytica/dispar or Cryptosporidium spp. antigens research, and amoebae culture. The samples were frozen by the participating laboratories, retrospectively extracted and examined with one-step real-time PCR multiplex using the Allplex™ GI-Parasite Assay (Seegene Inc., Seoul, Korea). Results: A total of 368 samples were analyzed from 12 Italian laboratories. Compared to traditional techniques, the sensibility and specificity of the real-time PCR kit were as follows: 100% and 100% for Entamoeba histolytica, 100% and 99.2% for Giardia duodenalis, 97.2% and 100% for Dientamoeba fragilis, and 100% and 99.7% for Cryptosporidium spp., respectively. Conclusions: The Allplex™ GI-Parasite Assay exhibited excellent performance in the detection of the most common enteric protozoa.
## 1. Introduction
Intestinal parasitic infections are widespread in low-and middle-income countries, with an estimated 3.5 billion cases annually [1,2]. Enteric protozoan parasites are responsible for a broad spectrum of clinical manifestations, ranging from mild gastrointestinal symptoms to life-threatening watery or hemorrhagic diarrhea, and to extra-intestinal localizations [3].
While Blastocystis hominis is the most common protozoan detected in stool samples [4], giardiasis and dientamoebiasis are the major cause of disease in terms of frequency, and cryptosporidiosis and amoebiasis are, respectively, the third and fourth leading parasitic causes of death worldwide [2]. Regarding the prevalence of these protozoa in Italy, several studies have shown that B. hominis, G. duodenalis, and D. fragilis are the most frequent, often causing co-infections [5]. For example, Clemente at al., has shown that out of 575 enrolled people, 85 (14.8%) were positive for D. fragilis and 37.7% of those had co-infection with B. hominis [6].
However, these infections are often neglected and underreported.
The diagnosis of intestinal protozoan parasites typically relies on the microscopic detection of trophozoites, cysts, and/or oocysts in human fecal samples [7,8]; this is a sharp contrast to other microbiological fields, where innovative modern technology has replaced the more classical diagnostic methods in the last two decades [9]. Molecular techniques are more difficult to apply for the identification of enteric protozoa than other infectious agents, due to the thick wall of parasite (oo)cysts, making DNA extraction difficult, and due to the high density of PCR inhibitors in stool samples [3]. Despite being the reference method, the microscopic examination of stool for the diagnosis of protozoan intestinal infections exhibits several drawbacks, as it is labor-intensive, time consuming and requires a high level of skill for optimal examination, which remains a major challenge for many laboratories in the northern hemisphere due to the low number of positive samples received annually [10]. Moreover, protozoan parasites are difficult to identify, particularly when they are present in low numbers [3]. In addition to the rapid processing of the samples to prevent morphological alterations, iterative stool specimens collected over a few days are usually necessary to increase sensitivity [11]. The sensitivity and specificity of microscopic detection of protozoan parasites are in fact both regarded as scarce, the technique being limited by its poor sensitivity and inability to differentiate between closely related species. For example, it is impossible to differentiate microscopically between the cysts of the pathogenic and the non-pathogenic species of Entamoeba histolytica and E. dispar [12,13].
Diagnosis of D. fragilis relies on direct visualization of the trophozoites in stained fixed fecal smears by light microscopy, as demonstration of the characteristic nuclear structure cannot be achieved in unstained fecal specimens. D. fragilis may be difficult to distinguish from non-pathogenic protozoa [14]. Therefore, alternative diagnostic methods have been developed to overcome the limitations of conventional microscopic techniques. For example, for G. duodenalis and Cryptosporidium spp. identification, direct fluorescent antigen detection by trained microscopists has shown better analytical performances than conventional microscopy. Nevertheless, fluorescent microscopy is still time consuming and requires skilled microscopists and appropriate equipment [15]. Enzymatic immunoassays and immunochromatographic tests are also available, improving the identification of E. histolytica and the analytic turnaround time for G. duodenalis and Cryptosporidium spp. diagnostic. However, analytical performances of these tests vary according to the targeted parasites and manufacturers, with false negative and false positive results still requiring confirmatory tests [16].
Today, nucleic acid amplification tests might offer more sensitive alternatives and allow the distinction between the potentially invasive E. histolytica and the non-pathogenic E. dispar [16][17][18][19] Over the last 10 to 15 years, many clinical microbiology laboratories have been provided with facilities to perform molecular diagnostics and automated approaches based on PCR are becoming increasingly available for detecting intestinal parasites, being less time consuming and demonstrating higher sensitivity and specificity than conventional methods [12,18,20].
The aim of this multicentric study was to evaluate retrospectively the Allplex™ GI-Parasite Assay (Seegene Inc., Seoul, Korea), a multiplex real-time PCR for the detection of intestinal protozoa from fecal samples and to compare the results to conventional methods, such as macro-and microscopic examination after concentration, Giemsa or Trichromic stain, G. duodenalis, E. histolytica/dispar or Cryptosporidium spp. antigens research, and amoebae culture. The DNA was extracted with the Microlab Nimbus IVD system, which automatically performed the nucleic acid processing and PCR setup.
The benefits of using this targeted panel for optimal parasitological diagnosis are discussed considering our results.
## 2. Materials and Methods
## 2.1. Study Design and Sample Collection
We designed a national, multicenter study with the participation of 12 Italian laboratories from northern (n = 10), central (n = 1), and southern (n = 1) Italy, including laboratories in Bergamo, Bologna, Treviso, Pavia, Lecco, Napoli, Legnano, Modena, Verona, Ancona, Pinerolo, and Brunico. The samples were collected during routine parasitological diagnostic procedures from patients suspected of enteric parasitic infection and were examined using traditional techniques according to WHO and CDC guidelines [9,21]. The samples were examined routinely using traditional techniques: macro-and microscopic examination after concentration, Giemsa or Trichrome stain, Giardia duodenalis, Entamoeba histolytica/dispar or Cryptosporidium spp. antigens, and amoebae culture.
A total of 368 samples were collected and stored at -20 or -80 • C in the different laboratories until they were sent to the Unit of Microbiology and Virology of Papa Giovanni XXIII Hospital (Bergamo, Italy) and tested by the Allplex™ GI-Parasite Assay (Seegene Inc., Seoul, Republic of Korea). In case of discrepancies in results, between real-time PCR analysis and traditional investigations, the samples were retested with both real-time PCR and traditional methods (Figure 1).
## 2.2. Real-Time PCR Assay
An amount of 50 to 100 mg of stool specimens was collected and suspended in 1 mL of stool lysis buffer (ASL buffer; Qiagen, Valencia, CA, USA). After pulse vortexing for 1 min and incubation at room temperature for 10 min, the tubes were centrifuged at full speed (14,000 rpm) for 2 min. The supernatant was used for nucleic acid extraction. Nucleic acids were extracted, using the Microlab Nimbus IVD system (Hamilton, Reno, NV, USA). The Microlab Nimbus IVD system automatically performed the nucleic acid processing and PCR setup.
DNA extracts were amplified with one-step real-time PCR multiplex (CFX96™ Realtime PCR, Bio-Rad, California, USA) with CFX Manager 1.6 software using the panel Allplex™ GI-Parasite Assay (Seegene Inc., Seoul, Republic of Korea). Fluorescence was detected at two temperatures (60 • C and 72 • C), and a positive test result was defined as a sharp exponential fluorescence curve that intersected the crossing threshold (Ct) at a value of less than 45 for individual targets. Positive and negative controls were included in each run. The identification panel included Giardia duodenalis, Dientamoeba fragilis, Entamoeba histolytica, Blastocystis hominis, Cyclospora cayetanensis, and Cryptosporidium spp. Results were interpreted using Seegene ® Viewer software (3.28.000 version). The PCR experiment was validated according to the manufacturer's recommendations.
## 2.3. Statistical Assessment
The results were analyzed by descriptive statistics. Sensibility and specificity were evaluated for each pathogen detected by the Allplex™ GI-Parasite Assay. Microscopic examination, G. duodenalis, E. histolytica/dispar or Cryptosporidium spp. antigen detection, and amoebae culture were considered as the reference methods to calculate the sensitivity and specificity of the real-time PCR assay.
The multiplex real-time PCR results were considered true positive (TP) or true negative (TN) when in agreement with the traditional methods. The results were defined as false positive (FP) or false negative (FN) when discrepancies with the reference methods were observed after both the first and second evaluation (Figure 1).
To evaluate the performance between the traditional parasitological examinations carried out by the 12 laboratories and the real-time PCR under evaluation, the Kappa value was calculated. Results were interpreted according to the following Kappa values: (i) 0.01-0.20, slight agreement; (ii) 0.21-0.40, fair agreement; (iii) 0.41-0.60, moderate agreement; (iv) 0.61-0.80, substantial agreement; and (v) 0.81-1.00, perfect agreement.
## 2.4. Ethical Statement
The study protocol received ethical clearance by the Ethics Committee of Papa Giovanni XXIII Hospital, Bergamo, Italy (prot. Nr 172/19 date 19 September 2019).
## 3. Results
This study included 368 stool samples analyzed for the presence of enteric protozoa in 12 Italian laboratories. Specifically, among the 368 samples, microscopic investigation identified 78 negative samples [four samples were positive only for antigens (2 G. duodenalis, 1 E. histolytica, 1 Cryptosporidium spp.) and three fecal samples of patients who were serology positive for anti-E. histolytica antibodies, one positive only by in-house PCR for G. duodenalis used in routine analysis in one of the participant labs]. Microscopic investigations have also allowed us to identify: 18 non-pathogenic protozoa, 30 B. hominis, 52 mixed pathogenic and non-pathogenic protozoa, 87 G. duodenalis, 77 D. fragilis, 14 Cryptosporidium spp., and 12 E. histolytica/dispar (Table 1). Sensibility and specificity for C. cayetanensis detection could not be evaluated because of the lack of positive samples.
Out of the 90 samples positive for G. duodenalis by traditional methods, 90/90 stool samples were also positive by the real-time PCR assay.
In addition, two samples were reported by microscopy as positive for G. duodenalis/Cryptosporidium spp. and G. duodenalis/E. histolytica dispar/E. coli, respectively, and were tested negative for G. duodenalis, Cryptosporidium spp. and E. histolytica by the Allplex™ GI-Parasite Assay; these samples were retested and confirmed to be negative.
The real-time PCR assays yielded two FP results in G. duodenalis detection. The two samples reported by traditional methods as positive, one for Blastocystis and one for Cryptosporidium spp., respectively, were confirmed negative for G. duodenalis both by repeating RT-PCR and traditional methods.
The molecular test under evaluation also found one new positive, confirmed as TP sample for G. duodenalis, which was reported from the labs, positive only for B. hominis and E. coli (Table 2). Compared with the reference methods, the sensibility and specificity of the Allplex™ GI-Parasite Assay for G. duodenalis detection were 100% and 99.2%, respectively. In comparison the traditional methods for G. duodenalis presented sensitivity of 99% and specificity 99.2% (Table 2).
Regarding D. fragilis, out of 96 samples tested positive for this protozoan parasite by conventional methods 64 were confirmed as TP after comparison with real-time PCR results (Table 2). A total of 3 out of 79 samples that tested positive by conventional methods and negative by the molecular test were evaluated as FN; these samples were confirmed as positive by reviewing the previously stained slides. In addition, the real-time PCR test allowed the identification of 42 new positive samples that the conventional methods used by the laboratories participating in the study could not detect, confirmed as TP (Table 2). Following these results, the sensitivity and specificity for the real-time PCR were 97.2% and 100%, respectively (Table 2).
Of the 16 samples testing positive for Cryptosporidium spp. by conventional methods, 13 stool samples tested positive by Allplex™ GI-Parasite Assay, and 3 samples were evaluated as TN reviewing the slides, confirming the real-time PCR results.
One sample testing negative by conventional methods and real-time PCR positive for Cryptosporidium spp. was evaluated as FP; this sample was tested a second time by real-time PCR and was found to be negative. For the first test, the amplification cycles of the sample were too low (9 Ct), probably due to interference in the well.
One additional sample, testing negative by conventional methods and positive at the real-time PCR assay was confirmed as TP by microscopical examination and repetition of the sample (Table 2). As reported in Table 2, the sensitivity and specificity of the Allplex™ GI-Parasite Assay for Cryptosporidium spp. were 100% and 99.7%, respectively.
Regarding E. histolytica, the employment of real-time PCR assay allowed the differentiation between E. histolytica and E. dispar in three samples with negative microscopy, precisely two from patients with positive serology and one with positive antigen test, confirming them as TP. An additional sample tested positive only for B. hominis by microscopy was considered as TP for E. histolytica, in fact we discovered that the patient had positive serology for this parasite (Table 2).
As reported in Table 2, the Allplex™ GI-Parasite Assay exhibited 100% sensitivity and 100% specificity for E. histolytica.
By examining a large samples number, the present study indicates that the Allplex™ GI-Parasite Assay multiplex PCR offers optimal analytical performances, overlapping the routine parasitological investigation procedures performed by microscopists and traditional methods, and shown by the perfect agreement with the traditional methods as seen in Table 3 (Kappa values ranging between 0.96 and 1.0). Table 3. Results of the multiplex PCR assay compared to traditional methods. * (+/+): Positive by both traditional methods and Allplex™ GI-Parasite Assay (i.e., true positive sample, TP). (+/-): Positive by traditional methods and negative by commercial PCR assay (i.e., false negative sample, FN). (-/+): Negative by traditional methods/positive by commercial PCR assays (i.e., false positive sample, FP). (-/-): Negative by both traditional methods and Allplex™ GI-Parasite Assay (i.e., true negative sample, TN).
## Parasites
## 4. Discussion
Despite several limitations, conventional methods, including microscopic examination and antigen detection, still represent the gold standard for diagnosing gastrointestinal parasites [16].
E. histolytica, G. duodenalis, Cryptosporidium spp., and D. fragilis are the four most important and commonly occurring diarrhea causing parasitic protozoa and it is essential that a correct diagnosis is carried out for all four protozoa to reach successful treatment.
The study design, with the collection of stool samples in 12 distinct centers, allowed the inclusion of 368 positive and negatives samples for enteric protozoa. In this multicentric study, the Allplex™ GI-Parasite Assay for the detection of Cryptosporidium spp., D. fragilis, E. histolytica, and G. duodenalis was shown to be an optimal diagnostic tool for rapid, sensitive, and specific detection of these enteric protozoa. This assay combines several advantages such as it being an automated process and this technique could improve the routine diagnosis of protozoan infections by clinical laboratories as it is easier to implement compared to microscopy.
Considering what was published before [4,6,14,16], we observed that by employing the real-time PCR, the sensitivity of D. fragilis identification is higher compared to traditional methods. In this study, we show that traditional methods exhibit scarce specificity for the detection of D. fragilis in stool samples (Table 2), in fact of the 79 samples sent for the study that were positive for this protozoan, 15 were positive only for B. hominis.
Diagnosis of D. fragilis relies on the direct visualization of the trophozoites in stained fixed fecal smears by light microscopy, as demonstration of the characteristic nuclear structure cannot be achieved in unstained fecal specimens [14]. Traditional microscopic assessment encounters significant challenges to identifying D. fragilis trophozoites due to their rapid deterioration outside the intestinal lumen and the fragility of their binucleate structure [22]. Therefore, it may happen that D. fragilis has often been mistaken for B. hominis and vice versa, or both are present and only B. hominis is reported, so it is useful in the real-time PCR panel to have B. hominis to confirm microscopic examination.
Considering the data from the literature (Table 4), there is also increasing evidence that real-time PCR should be preferred over antigen detection for definitive identification of E. histolytica, in fact, we obtained a rapid and accurate identification, allowing the discrimination between the two pathogenic and non-pathogenic amoebas [13,16,20]. [16,23].
The major limits of this study were the poor number of Cryptosporidium spp. and E. histolytica tested, in line with the low prevalence of these intestinal parasites in Italy [5]. Despite this, it was possible to evaluate the usefulness of real-time PCR.
Another potential confounding variable arises from the study's reliance on conventional microscopic examination as the reference standard for diagnosing parasitic infections, despite the well-documented variability in sensibility that is dependent on the microscopist's expertise.
The importance of this study is highlighted by the fact that we were able to define true positive and true negative samples; although the introduction of more sensitive molecular methods to verify the results, such as gene sequencing, would have been interesting.
Sensitive multiplex real-time PCR panels for molecular diagnosis of enteric protozoan parasites have been developed to overcome microscopy-based diagnostic limitations. However, the interpretation and clinical implications of positive real-time PCR results remain a challenge for the treating physician [24], in fact the increasing employment of molecular techniques could cause the loss of the experience of traditional parasitology, which is crucial to confirm indeterminate cases.
The examined panel Allplex™ GI-Parasite assay, could be employed in a diagnostic algorithm for a molecular screening of stool samples for enteric protozoa. Nevertheless, microscopy is necessary to minimize the chance of missing other parasitic pathogens, and to verify the results of the molecular methods, even when the prevalence of nonprotozoal parasites is low, which may be the cause in particular diagnostic settings or patient populations [25]. The use of real-time PCR will allow the detection of Cryptosporidium spp., D. fragilis and E. histolytica as not all laboratories correctly apply all the diagnostic techniques necessary for their identification. The increase in reported cases of Cryptosporidium spp. recently [26] could also be due to the spread of commercial real-time PCR tests in the labs routine.
## 5. Conclusions
In conclusion, the Allplex™ GI-Parasite Assay exhibited excellent performance in the detection of the most common enteric protozoa like G. duodenalis, D. fragilis, E. histolytica and Cryptosporidium spp.
To maintain high quality of care, tailor-made algorithms should identify specific high-risk patient populations who deserve in-depth diagnostic analysis and allow for the identification of rare parasitic diseases. For these patient populations, flexible and highly specialized diagnostic parasitological tools are still essential and should be maintained in designated referral centers [10].
About the implications for future research, our study demonstrates how the collaboration between laboratories is essential for the evaluation of commercial tests for fecal parasitological diagnostics as an important task for microbiologist associations, such as the Parasitology Committee of the Italian Clinical Microbiologists Association (CoSP-AMCLI). While PCR techniques are gaining increased attention in diagnostic laboratories for the development of reliable and cost-effective methods to identify fecal parasites, further studies are needed to standardize procedures for sample collection, storage, and DNA extraction, as these pivotal steps are essential for achieving consistent results.
## References
1. Meurs, Polderman, Melchers et al. "Diagnosing Polyparasitism in a High-Prevalence Setting in Beira, Mozambique: Detection of Intestinal Parasites in Fecal Samples by Microscopy and Real-Time PCR"
2. Autier, Belaz, Razakandrainibe et al. (2018) "Comparison of Three Commercial Multiplex PCR Assays for the Diagnosis of Intestinal Protozoa" *Parasite*
3. Stark, Al-Qassab, Barratt et al. (2011) "Evaluation of Multiplex Tandem Real-Time PCR for Detection of Cryptosporidium spp., Dientamoeba fragilis, Entamoeba histolytica, and Giardia intestinalis in Clinical Stool Samples" *J. Clin. Microbiol*
4. Bartolini, Zorzi, Besutti (2011) "Prevalence of Intestinal Parasitoses Detected in Padua Teaching Hospital" *Infez. Med*
5. Clemente, Pasut, Carlet et al. (2021) "Dientamoeba Fragilis in the North-East of Italy: Prevalence Study and Treatment" *Parasitol. Int*
6. Garcia (2015) "Diagnostic Medical Parasitology"
7. Garcia, Arrowood, Kokoskin et al. (2018) "Practical Guidance for Clinical Microbiology Laboratories: Laboratory Diagnosis of Parasites from the Gastrointestinal Tract" *Clin. Microbiol. Rev*
8. Mchardy, Wu, Shimizu-Cohen et al. (2014) "Detection of Intestinal Protozoa in the Clinical Laboratory" *J. Clin. Microbiol*
9. Laude, Valot, Desoubeaux et al. "Is Real-Time PCR-Based Diagnosis Similar in Performance to Routine Parasitological Examination for the Identification of Giardia Intestinalis, Cryptosporidium Parvum/Cryptosporidium Hominis and Entamoeba Histolytica from Stool Samples? Evaluation of a New Commercial Multiplex PCR Assay and Literature Review" *Clin. Microbiol*
10. Abu-Madi, Boughattas, Behnke et al. (2017) "Coproscopy and Molecular Screening for Detection of Intestinal Protozoa. Parasites Vectors"
11. Mohammad (2013) "Detection Of Human Intestinal Protozoa by Using Multiplex Allele Specific Polymerase Chain Reaction (MAS-PCR) in New Damietta City" *Zagazig Univ. Med. J*
12. Stark, Beebe, Marriott et al. (2006) "Evaluation of Three Diagnostic Methods, Including Real-Time PCR, for Detection of Dientamoeba Fragilis in Stool Specimens" *J. Clin. Microbiol*
13. Verweij, Blangé, Templeton et al. (2004) "Simultaneous Detection of Entamoeba Histolytica, Giardia Lamblia, and Cryptosporidium Parvum in Fecal Samples by Using Multiplex Real-Time PCR" *J. Clin. Microbiol*
14. Argy, Nourrisson, Aboubacar et al. "Selecting a Multiplex PCR Panel for Accurate Molecular Diagnosis of Intestinal Protists: A Comparative Study of Allplex ® (Seegene ® ), G-DiaParaTrio (Diagenode ® ), and RIDA ® GENE (R-Biopharm ® ) Assays and Microscopic Examination"
15. Friesen, Fuhrmann, Kietzmann et al. (2018) "Evaluation of the Roche LightMix Gastro Parasites Multiplex PCR Assay Detecting Giardia Duodenalis, Entamoeba Histolytica, Cryptosporidia, Dientamoeba Fragilis, and Blastocystis Hominis" *Clin. Microbiol. Infect*
16. Verweij (2014) "Application of PCR-Based Methods for Diagnosis of Intestinal Parasitic Infections in the Clinical Laboratory" *Parasitology*
17. Solaymani-Mohammadi, Rezaian, Babaei et al. (2006) "Comparison of a Stool Antigen Detection Kit and PCR for Diagnosis of Entamoeba Histolytica and Entamoeba Dispar Infections in Asymptomatic Cyst Passers in Iran" *J. Clin. Microbiol*
18. Verweij, Stensvold (2014) "Molecular Testing for Clinical Diagnosis and Epidemiological Investigations of Intestinal Parasitic Infections" *Clin. Microbiol. Rev*
19. Specimens (2025)
20. Johnson, Windsor, Clark (2004) "Emerging from Obscurity: Biological, Clinical, and Diagnostic Aspects of Dientamoeba Fragilis" *Clin. Microbiol. Rev*
21. Di Pietra, Gargiulo, Ortalli et al. (2025) "Comparative Analysis of Commercial and "In-House" Molecular Tests for the Detection of Intestinal Protozoa in Stool Samples. Parasites Vectors"
22. Maas, Dorigo-Zetsma, De Groot et al. (2014) "Detection of Intestinal Protozoa in Paediatric Patients with Gastrointestinal Symptoms by Multiplex Real-Time PCR" *Clin. Microbiol. Infect*
23. Bruijnesteijn Van Coppenraet, Wallinga, Ruijs et al. (2009) "Parasitological Diagnosis Combining an Internally Controlled Real-Time PCR Assay for the Detection of Four Protozoa in Stool Samples with a Testing Algorithm for Microscopy" *Clin. Microbiol. Infect*
24. Peake, Inns, Jarvis et al. (2023) "Preliminary Investigation of a Significant National Cryptosporidium Exceedance in the United Kingdom"
25. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12737425&blobtype=pdf
|
# Intensive Therapeutic Plasma Exchange for Severe Yellow Fever: What Is the Evidence? Comment on Ho et al. Intensive Therapeutic Plasma Exchange-New Approach to Treat and Rescue Patients with Severe Form of Yellow Fever. Trop. Med. Infect. Dis. 2025, 10, 39
Till Omansen, Michael Ramharter
Recent outbreaks of yellow fever in Brazil, with hundreds of cases despite available vaccination, have drawn attention to the pressing need for effective therapeutic interventions, with a special focus on the critically ill. The most common manifestation of severe yellow fever is rapidly progressive acute liver failure, requiring specialized intensive care to improve survival rates in the absence of effective antiviral or host-directed therapeutics.
In this context, the study by Ho et al. on the use of therapeutic plasma exchange (TPE) in patients with severe yellow fever provides fascinating insights from a highly challenging outbreak situation, which requires careful consideration [1]. TPE, which involves the removal of plasma and its replacement with donor plasma or other fluids, has been evaluated in various liver failure settings, including sepsis, viral infections, autoimmune diseases, and drug overdoses [2]. TPE has shown promise in alleviating the consequences of liver failure by removing harmful toxic substances, including bilirubin, cytokines, and metabolic waste products. This procedure has been advocated as a strategy to bridge patients to liver transplantation in cases where they experience delays in receiving a transplant [3].
The study by Ho et al. investigated the use of TPE for yellow fever patients in Brazil, contributing to this growing body of evidence for the use of TPE. Their findings suggest that intensive TPE may reduce mortality in severe yellow fever patients, with mortality rates significantly lower in the group receiving intensive TPE (14%) compared to those receiving standard intensive care (85%) or high-volume TPE (82%). While this reduction in mortality seems compelling, it must be interpreted with caution due to the observational nature of the study, lacking a randomization process or parallel control group. This retrospective analysis compared treatment protocols that were implemented sequentially over the course of an outbreak. From a methodological point of view, this comparison using historical controls represents a key limitation of the study, introducing a risk of bias. This may include the possibility that the reported improvements in mortality over time were due to factors unrelated to TPE, such as differences in patient selection, improvements in supportive care over the course of the outbreak, or the accumulation of clinical experience as the outbreak progressed.
These limitations point to a larger issue in clinical research in resource-limited settings: conducting a prospective, randomized trial during an outbreak is highly challenging. On the one hand, withholding a potentially life-saving intervention like TPE from a severely ill patient may not be justifiable from an ethical perspective. On the other hand-and from a methodological point of view-the absence of a well-designed trial makes it difficult to establish a clear cause-and-effect relationship between TPE and improved outcomes. While the results of this observational study are promising, they should be viewed as a first step in our understanding of improved management of yellow fever rather than as definitive evidence. Larger, multicenter, prospective trials with randomized controls are needed to firmly establish the efficacy of TPE in severe yellow fever.
Meanwhile, it is also worth considering other potential therapies for severe liver failure in yellow fever patients. One such alternative is albumin-based dialysis, which has been increasingly used in patients with acute liver failure, particularly in cases where TPE is not available or feasible. Albumin-based dialysis systems, such as ADVOS ® (by Advito, Hannover, Germany) or MARS ® (by Gambro/Baxter, Lund, Sweden), have demonstrated the ability to remove toxins, restore metabolic balance, and support liver function. These therapies also offer the potential to act as a bridge to liver transplantation, similar to TPE [4,5].
The study by Ho et al. offers valuable insight into a novel approach to managing severe yellow fever. The study represents an important contribution to the field, and the authors should be commended for their efforts in advancing our understanding of this life-threatening disease. Further research, including well-designed prospective trials, is necessary to confirm the benefits of TPE and to explore other potential interventions to reduce mortality in severe yellow fever.
## References
1. Ho, Nukui, Villaça et al. (2025) "Intensive Therapeutic Plasma Exchange-New Approach to Treat and Rescue Patients with Severe Form of Yellow Fever" *Trop. Med. Infect. Dis*
2. Kuklin, Sovershaev, Bjerner et al. (2024) "Influence of therapeutic plasma exchange treatment on short-term mortality of critically ill adult patients with sepsis-induced organ dysfunction: A systematic review and meta-analysis" *Crit. Care*
3. Ocak (2023) "A 15-Year Retrospective Study of Supportive Extracorporeal Therapies Including Plasma Exchange and Continuous Venovenous Hemodiafiltration of 114 Adults with Acute Liver Failure Awaiting Liver Transplantation" *Ann. Transplant*
4. Huber, Henschel, Schmid (2017) "First clinical experience in 14 patients treated with ADVOS: A study on feasibility, safety and efficacy of a new type of albumin dialysis" *BMC Gastroenterol*
5. Kantola, Ilmakunnas, Koivusalo et al. (2011) "Bridging therapies and liver transplantation in acute liver failure, 10 years of MARS experience from Finland" *Scand. J. Surg*
6. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods"
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12434107&blobtype=pdf
|
# EDITED AND REVIEWED BY
Michael Ward, Jingqiang Ren, Zhenzhen Zheng, Shen Yang, Xusheng Qiu, Mingxiao Ma
## Introduction
Zoonotic viruses originate from wild animals and livestock. They can circulate between animal reservoir hosts and humans, leading to mild or severe illness in both species (1). In recent decades, with the emergence and re-emergence of zoonotic diseases such as Ebola, SARS-CoV, avian influenza, COVID-19, and Mpox globally (2)(3)(4)(5)(6)(7)(8), people are becoming increasingly aware of the threat these diseases pose to human health. What is astonishing is that these pathogens can be transmitted not only through direct contact but also through indirect contact or via contaminated inanimate objects. To date, no marketed drugs have been proven effective against such diseases. However, with the increasing public awareness of safety and the continuous advancement of scientific research, emergency vaccines for critical situations have been successfully developed (9). This Research Topic aims to delve into the complex mechanisms of the transmission, occurrence, mutation, and dissemination of animal pathogens within and between different populations, hoping to provide information and guidance for disease prevention and treatment.
## Vigilance against cross-species transmission of circoviruses
Circovirus is currently known as the smallest pathogenic DNA virus, traditionally referred to as porcine circovirus because it was initially isolated from pigs (10). However, circoviruses have long since breached the species barrier and have been reported in avian, aquatic, and many mammalian species, including humans (11)(12)(13). Phylogenomic and evolutionary analyses indicate that most of them are related to porcine circovirus (13). Cao et al. in this Research Topic reported a novel canine circovirus, with a code-shifting mutation occurring in the shortened ORF1 region, and a new 13-amino acid sequence emerging at the C-terminus of the replication protein (Rep), suggesting that this mutation may affect the viral replication cycle. Phylogenetic and evolutionary analyses indicate that these isolates belong to canine coronavirus genotype 3, which is more prevalent in the southeastern and southwestern regions of China as well as in neighboring countries (Cao et al.). This epidemiological investigation has broadened our understanding of the genetic diversity of the canine coronaviruses in southwestern China and provides insights into the evolution of the virus.
## Recurrent zoonotic infectious diseases
New or recurrent zoonotic infectious diseases continue to pose a serious threat to public health and the global economy, such as rabies, brucellosis, avian influenza, etc. Das et al. assessed the knowledge, attitude, and practices (KAP) of local residents regarding rabies in Turkana County, the second largest city in Kenya. The results indicated that the level of knowledge, positive attitudes, and behaviors regarding dog vaccination are all below 50% in Turkana region. The main factors influencing dog vaccination include a lack of understanding of rabies, insufficient information about immunization activities, and considerations regarding the cost of vaccination (Das et al.). These findings have significant implications for policy development and the decisionmaking process, emphasizing the necessity of implementing targeted interventions in such situations to enhance rabies awareness and vaccination rates.
Zakharova and Liskova employed spatiotemporal cluster analysis and a negative binomial regression algorithm to investigate the relationship between animal rabies burden and a range of environmental and socio-demographic factors across different administrative districts of Nizhny Novgorod. This approach aimed to evaluate risk factors influencing the transmission and persistence of rabies virus among wild and domestic animal populations, which is critical for developing effective strategies to control and reduce cases. The study found that the density of red foxes and the vaccination rates of wildlife and domestic animals are the most significantly associated risk factors with the severity of rabies in the study area.
## Conclusion and future perspectives
Clearly, the study and understanding of zoonotic diseases extend far beyond the topics discussed above. Although this Research Topic has only gathered a limited number of articles, it does not diminish our commitment to the critical awareness of the importance of zoonotic disease. Future research in this field requires increased investment to better elucidate the pathogenicity and molecular mechanisms of viruses, thereby promoting innovation and development, and ultimately aiding humanity in making greater progress in combating various diseases.
## References
1. Chandra, Kesavardhana (2024) "PANoptosis regulation in reservoir hosts of zoonotic viruses" *Viruses-Basel*
2. Camacho, Kucharski, Funk et al. (2014) "Potential for large outbreaks of Ebola virus disease" *Epidemics*
3. Nakkazi, Sudan (2025) "Ebola virus disease outbreak in Uganda" *Lancet Infect Dis*
4. Peiris, Lai, Poon et al. (2023) "Coronavirus as a possible cause of severe acute respiratory syndrome" *Lancet*
5. Webby, Uyeki (2024) "An update on highly pathogenic avian influenza A(H5N1) virus" *J Infect Dis*
6. Tali, Leblanc, Sadiq et al. (2021) "Tools and techniques for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)/COVID-19 detection" *Clin Microbiol Rev*
7. Hou, Wu, Liu et al. (2025) "Mpox: global epidemic situation and countermeasures" *Virulence*
8. Edward, Gwanafyo, Kimambo et al. (2025) "Mpox outbreak: a call for urgent action and improved response strategies" *Health Sci Rep*
9. Li, Wang, Tian et al. (2022) "COVID-19 vaccine development: milestones, lessons and prospects" *Sig Transd Targeted Ther*
10. Tischer, Rasch, Tochtermann (1974) "Characterization of papovavirus-and picornavirus-like particles in permanent pig kidney cell lines" *Zentralbl Bakteriol Orig A*
11. Da Silva, Akash, De Aquino et al. (2010) "Multiple diverse circoviruses infect farm animals and are commonly found in human and chimpanzee feces" *Int J Surg*
12. Opriessnig, Karuppannan, Castro et al. (2020) "Porcine circoviruses: current status, knowledge gaps and challenges" *Virus Res*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12808268&blobtype=pdf
|
# First evidence of conspecific hyperparasitism in Dermacentor marginatus nymphs feeding on a rabbit under experimental conditions
Lenka Minichová, Ľubomír Vidlička, Mirko Slovák
## Abstract
Hyperparasitism in ticks, particularly in nymphs of the order Ixodidae, is a rare phenomenon. In our laboratory tick colony, female rabbits are used as a blood source for the ticks, housing them under controlled conditions. Feeding Dermacentor marginatus nymphs monitoring was performed daily, and engorged and detached ticks were collected and stored in desiccators at constantly 24 ± 2 °C and 85-90% relative humidity (16 h light/8 hours dark). Nymphs suspected of conspecific hyperparasitism were preserved in ethanol for analysis and imaging. This study presents the first documented case of hyperparasitism in D. marginatus and contributes to the limited literature on hyperparasitism in Ixodidae nymphs. While such cases are observed in controlled tick colonies, their occurrence in the wild, especially in Ixodidae, is extremely rare compared to Argasidae ticks. Furthermore, the frequency of this phenomenon in the wild and its possible eco-epidemiological significance remain poorly understood.
## Introduction
Dermacentor marginatus (Sulzer 1776), also known as ornate sheep tick, infests domestic (sheep, dogs, goats, horses, cattle) and wild hosts including deer, hare, hedgehog, wild boar, and wolf (Accorsi et al. 2022;Estrada-Peña et al. 2017;Garcia-Vozmediano et al. 2020;Rubel et al. 2016;Sgroi et al. 2021). It is adapted to a warmer and drier climate within the belt of 33-51° N latitude and generally inhabits lowlands, steppes (alpine and forest), and semi-desert areas (Rubel et al. 2016;Walter et al. 2016). This three-host tick species has a life cycle of 75-163 days under laboratory conditions, but in nature it usually completes its cycle within 1-2 years (Kahl and Dautel 2013;Nosek 1972). Larvae are commonly found on rodents and small to medium-sized insectivores (Estrada-Peña et al. 2017;Nosek 1972). Adult female ticks have been found on humans, including children, particularly on the scalp (Cazorla et al. 2003;Porta et al. 2008;Raoult et al. 2002;Selmi et al. 2008).
Dermacentor marginatus is associated with the occurrence of numerous tick-borne pathogens (Bonnet et al. 2013;Estrada-Peña et al. 2017;Sonenshine and Roe 2014), some of which are recognised as increasingly important veterinary and public threat (e.g. tickborne encephalitis virus, Rickettsia slovaca, Rickettsia raoultii) (Buczek et al. 2020;Garcia-Vozmediano et al. 2020;Ličková et al. 2025;Nadim et al. 2021;Nosek and Kozuch 1985;Rubel et al. 2016;Špitalská et al. 2012).
Hyperparasitism describes a phenomenon where a tick parasitises another tick feeding on a host. It involves non-fed ticks attaching to fed or feeding ticks of the same (intraspecific, conspecific) (Buczek et al. 2019) or different species/genus (interspecific, heterogeneric) (Uspensky 2024). This behaviour was previously misclassified as cannibalism or kleptoparasitism, but Uspensky later clarified the correct terminology (Uspensky 2023).
The first documented case of hyperparasitism (unfed male of Amblyomma variegatum attached conspecific to a feeding female) was recorded by Barber in 1895 (Barber 1895). Buczek compiled a comprehensive review of hyperparasitism in both soft and hard ticks, including a detailed table on host species, engorgement level, and tick distribution (Buczek et al. 2019). In the most recent review, Uspensky has developed this topic further and provides an in-depth analysis of both the biological and eco-epidemiological aspects of hyperparasitism (Uspensky 2024). Here we report the first case of hyperparasitism in nymphs of D. marginatus under laboratory conditions and provide a detailed description of this phenomenon.
## Materials and methods
In our artificial laboratory breeding of tick colonies, all tick stages are usually fed on female California rabbits obtained from the Research Institute for Animal Production (Nitra, Slovakia). Animals were housed individually in cages in the animal facility of the Institute of Virology, Biomedical Research Center of the Slovak Academy of Sciences under controlled conditions: 15-21 °C, 45-65% relative humidity, and photoperiod 12 h light/12 hours dark cycle. All animals were fed standard pellet diet and given water ad libitum. All procedures followed protocols approved by the Animal Use Protocol approved by the ethical committee of Biomedical Research Center of Slovak Academy of Sciences and the State Veterinary and Food Administration of the Slovak Republic (facility number SK UCH01016, permit number 3954/19-221 and 292/16-221 k).
The rabbit's fur was partially shaved on the back to attach neoprene chambers containing ticks. For more details, please see Slovák et al. (2002). Ticks of each developmental stage were placed separately in individual chambers and on separate animals. Depending on the life stage of ticks, we checked when fully engorged ticks are ready to detach: Larvae were checked twice a day, while nymphs and adult ticks were checked once a day. The collected fully engorged ticks were stored in desiccators at 24 ± 2 °C, 85-90% relative humidity, and a photoperiod of 16 h light and 8 h dark cycle.
Nymphs suspected of being hyperparasitic were placed in 70% ethanol and examined under a Leica M205 C stereomicroscope equipped with a Leica flexacam C3 camera, and the images were processed using Adobe Photoshop CS6.
## Results
The nymphs of D. marginatus (about 500 individuals per chamber) were allowed to feed on a rabbit, and fully engorged nymphs were collected daily. On the sixth day of feeding, we observed two nymphs, one of which was attached to the other. A close examination under the stereomicroscope showed the perfect anchoring of the hypostome of one nymph in the integument of the other, laterally and caudally, near the spiracle (Fig. 1). It can be clearly seen that the nymph has opened its palps, and its hypostome is embedded in the body of the other nymph, which is a case of conspecific hyperparasitism. To our knowledge, this is the first report of conspecific hyperparasitism in D. marginatus. It is also one of the few cases of hyperparasitism in tick nymphs.
## Discussion
As summarized by Buczek et al. (2019) and Uspensky (2024), cases of hyperparasitism are mainly known from tick colonies under laboratory conditions. In argasid ticks, hyperparasitism has been observed primarily during laboratory rearing or in field-collected specimens brought into laboratory conditions. It typically involves unfed individuals stealing blood from engorged individuals, often as a response to starvation or overcrowding. Uspensky reported the occurrence of this phenomenon in nature is extremely rare, and although tickto-tick feeding among argasid ticks has never been observed under field conditions, there is evidence of this behaviour in the form of scars on the cuticle of free-living ticks (Uspensky 2024). In Ixodes ticks, hyperparasitism has been documented both in laboratory settings and among field-collected individuals. In this genus, male hyperparasitism on females is likely associated with mating attempts, whereas in Amblyomminae, it may represent an aberrant feeding behaviour (Buczek et al. 2019;Uspensky 2024). The author even mentions occasional observations of ixodid ticks attaching to horseflies and cites three relevant sources in Russian (Uspensky 2024).
The comprehensive reviews by Buczek et al. (2019) and Rodrigues et al. (2023) show that hyperparasitism in ticks from the families Ixodidae and Argasidae occurs predominantly in the form of males parasitizing females (Buczek et al. 2019;Rodrigues et al. 2023). Cases of nymphal hyperparasitism have been summarized in review of Buczek et al. in the Argasidae, particularly in Ornithodoros turicata, Ornithodoros parkeri and Ornithodoros erraticus, and in the Ixodidae, especially in Hyalomma detritum (Buczek et al. 2019). According to Uspensky, the hyperparasitism of males on unfed or fed females in case of Ixodes ticks is a side effect of mating, whereas in Metastriata ticks, it appears to be a rare feeding aberration (Uspensky 2024).
One potential factor contributing to nymphal hyperparasitism could be overcrowding, i.e. a high number of nymphs per chamber. However, in our observations, a different pattern emerged: despite ample space within the chamber, nymphs consistently clustered in one half, leaving the other half unoccupied. A similar aggregtion behaviour was described by Wang et al. in Rhipicephalus appendiculatus, where ticks gathered in a single area despite the availability of space. This behaviour was linked to the presence of immunoglobulin-binding proteins (IGBPs), particularly the male-specific IGBP-MC, which is secreted in saliva and facilitates female feeding. Males remain near females after mating, likely to enhance their feeding success and thereby increase their own reproductive fitness-a behaviour described as mate guarding (Wang and Nuttall 1998). Hyperparasitism may enable ticks to ingest haemolymph or vertebrate blood from conspecific or heterospecific hosts, potentially influencing intra-tick pathogen transmission. Williamson and Schwan (2018) demonstrated that unfed males of Ornithodoros hermsi frequently parasitize engorged nymphs and are capable of acquiring and transmitting Borrelia hermsii, indicating a possible role of hyperparasitism in the enzootic maintenance of this pathogen (Williamson and Schwan 2018). These findings highlight the need for further investigation into the ecological relevance of hyperparasitism in natural tick populations.
We support the hypothesis proposed by Uspensky that the transmission of pathogens by hyperparasitism is theoretically possible also in hard ticks. Labruna et al. consider this a possible alternative mechanism for the transmission of microorganisms among ticks (Labruna et al. 2007;Uspensky 2024). However, as Uspensky further emphasized, there is no evidence for its occurrence (Uspensky 2024). Therefore, the potential eco-epidemiological relevance of this phenomenon in hard ticks is still uncertain.
## Conclusion
This case illustrates a rare occurrence of conspecific hyperparasitism in hard ticks that is usually reported in controlled laboratory settings. The ecological and evolutionary implications of this behaviour and its possible role in the transmission dynamics of tick-borne diseases are not yet fully understood. Further studies are needed to investigate how such behaviour might influence the transmission of pathogens in hyperparasitism relationships and to expand our knowledge of the eco-epidemiology of tick-borne diseases.
## References
1. Accorsi, Schiavetti, Listorti et al. (2022) "Hard ticks (Ixodidae) from wildlife in Liguria, Northwest italy: tick species diversity and tick-host associations" *Insects*
2. Barber (1895) "The tick pest in the tropics" *Nature*
3. Bonnet, De, Fuente et al. (2013) "Prevalence of tick-borne pathogens in adult Dermacentor spp. Ticks from nine collection sites in France" *Vector-Borne Zoonotic Dis*
4. Buczek, Bartosik, Buczek et al. (2019) "Conspecific hyperparasitism in the Hyalomma excavatum tick and considerations on the biological and epidemiological implications of this phenomenon" *Ann Agric Environ Med*
5. Buczek, Koman-Iżko, Buczek et al. (2020) "Spotted fever group rickettsiae transmitted by Dermacentor ticks and determinants of their spread in Europe" *Ann Agric Environ Med*
6. Cazorla, Enea, Lucht et al. (2003) "First isolation of Rickettsia Slovaca from a patient" *France. Emerg Infect Dis*
7. (2017) "Ticks of Europe and North Africa"
8. Garcia-Vozmediano, Giglio, Ramassa et al. (2020) "Dermacentor marginatus and Dermacentor reticulatus, and their infection by SFG rickettsiae and Francisella-like endosymbionts" *Mountain Periurban Habitats Northwest Italy Veterinary Sci*
9. Kahl, Dautel (2013) "Seasonal life cycle organisation of the Ixodid tick Dermacentor reticulatus in central Europe implications on its vector role and distribution" *Med Kuzbass*
10. Labruna, Ahid, Soares et al. (2007) "Hyperparasitism in Amblyomma rotundatum (Acari: Ixodidae)" *J Parasitol*
11. Ličková, Víchová, Derdáková et al. (2025) "Surveillance of tick-borne encephalitis virus foci in slovakia: A Seroprevalence study in ruminants combined with virus detection in ticks" *Ticks Tick-borne Dis*
12. Nadim, Khanjani, Chegeni et al. (2021) "Identity and microbial agents related to Dermacentor marginatus Sulzer (Acari: Ixodidae) with a new record of rickettsia Slovaca (Rickettsiales: rickettsiaceae) in Iran" *Saa*
13. Nosek (1972) "The ecology and public health importance of Dermacentor marginatus and D. reticulatus ticks in central Europe" *Folia Parasitol (Praha)*
14. Nosek, Kozuch (1985) "Replication of tick-borne encephalitis (TBE) virus in ticks Dermacentor marginatus" *Agnew Parasitol*
15. Porta, Nieto, Creus et al. (2008) "Tick-Borne lymphadenopathy: A new infectious disease in children" *Pediatr Infect Disease J*
16. Raoult, Lakos, Fenollar et al. (2002) "Spotless rickettsiosis caused by Rickettsia Slovaca and associated with Dermacentor ticks" *CLIN INFECT DIS*
17. Rodrigues, Labruna, Ferreira et al. (2023) "First description of conspecific hyperparasitism in Amblyomma sculptum" *Ticks Tick-borne Dis*
18. Rubel, Brugger, Pfeffer et al. (2016) "Geographical distribution of Dermacentor marginatus and Dermacentor reticulatus in Europe" *Ticks Tickborne Dis*
19. Selmi, Bertolotti, Tomassone et al. (2008) "Rickettsia Slovaca in Dermacentor marginatus and tick-borne lymphadenopathy" *Tuscany Italy Emerg Infect Dis*
20. Sgroi, Iatta, Lia et al. (2021) "Spotted fever group rickettsiae in Dermacentor marginatus from wild boars in Italy" *Transbound Emerg Dis*
21. Slovák, Labuda, Marley (2002) "Mass laboratory rearing of Dermacentor reticulatus ticks" *Acarina. Ixodidae)*
22. Sonenshine, Roe, York Špitalská et al. (2012) "Rickettsia Slovaca and Rickettsia raoultii in Dermacentor marginatus and Dermacentor reticulatus ticks from Slovak Republic" *Exp Appl Acarol*
23. Uspensky (2023) "The phenomenon of attachment and feeding of unfed ticks (ixodoidea) on fed and feeding specimens of the same or different species: terminological issues" *Parazitologiâ*
24. Uspensky (2024) "The phenomenon of attachment and feeding of unfed ticks (Ixodoidea) on fed and feeding specimens of the same or different species: biological and epidemiological issues" *Parazitologiâ*
25. Walter, Brugger, Rubel (2016) "The ecological niche of Dermacentor marginatus in Germany" *Parasitol Res*
26. Wang, Nuttall (1998) "Male ticks help their mates to feed" *Nature*
27. Williamson, Schwan (2018) "Conspecific hyperparasitism: an alternative route for Borrelia hermsii transmission by the tick Ornithodoros hermsi" *Ticks Tick-borne Dis*
|
biology
|
europe-pmc
|
https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12265061&blobtype=pdf
|
# Multicenter Cross-sectional Study on the Epidemiology of Human Metapneumovirus in Italy, 2022-2024, With a Focus on Adults Over 50 Years of Age
Alessandro Mancon, Laura Pellegrinelli, Greta Romano, Elisa Vian, Valeria Biscaro, Giulia Piccirilli, Tiziana Lazzarotto, Sara Renteria, Annapaola Callegaro, Elisabetta Pagani, Elisa Masi, Guglielmo Ferrari, Cristina Galli, Francesca Centrone, Maria Chironna, Claudia Tiberio, Erasmo Falco, Valeria Micheli, Federica Novazzi, Nicasio Mancini, Giacomo Tiziano, Allice, Francesco Cerutti, Elena Pomari, Concetta Castilletti, Eleonora Lalle, Fabrizio Maggi, Matteo Fracella, Paolo Ravanini, Giulia Faolotto, Roberta Schiavo, Giuliana Cascio, Carla Acciarri, Stefano Menzo, Fausto Baldanti, Guido Antonelli, Alessandra Pierangeli, Elena Pariani, Antonio Piralla
## Abstract
Background. Human metapneumovirus (hMPV) infections have a significant impact on public health. However, the extent of this burden in Italy remains poorly defined due to a lack of comprehensive data. The aim of this cross-sectional multicenter study was to understand the epidemiology of hMPV in Italy, particularly in older adults.Methods. We analyzed laboratory data from molecular respiratory viral diagnostic tests conducted at 17 centers across Italy from September 2022 to August 2024. Respiratory viruses were tested from outpatients for epidemiologic surveillance and from patients presenting to tertiary hospitals for diagnostic purpose. G gene sequencing was performed on a limited number of circulating strains.Results. Data from 96 460 tests yielded an overall hMPV positivity rate of 3.4%; the hMPV positivity rate was 2.6% in adults aged 50 years and older, a third of whom were aged >80 years. In north-west Italy, hMPV was detected more frequently in outpatients than in hospitalized patients. The temporal distribution of cases showed seasonal peaks in February 2023 and April 2024, which exhibited some geographic variation but overlapped in the general population and in the elderly. Phylogenetic analysis suggested an even distribution of hMPV-A and -B, with a predominance of clades A2c with a 111-nucleotide duplication and B2b, and the possible extinction of previously circulating clades A2c with a 180-nucleotide duplication and B2a.Conclusions. hMPV was shown to be a relevant respiratory pathogen in older adults, who could be more likely to have severe outcomes. These findings may inform hMPV surveillance and the development of prevention strategies.
These range from asymptomatic or mild illness to severe lower respiratory tract disease, in both pediatric and adult populations [3][4][5]. In addition, like other respiratory viruses, hMPV-associated severe disease is not uncommon in older adults and those with comorbidities [6], who represent a priority target population for hMPV vaccination.
hMPV, an enveloped, negative-stranded RNA virus, is classified into 2 antigenically distinct types, A and B, which are further classified into 4 subtypes (or genotypes), A1, A2, B1, and B2 [7], and lineages. The genetic lineages have been classified based on the variable outer G glycoprotein. Because different nomenclatures are used, efforts are being made to establish a unified classification [8]. Over the course of epidemic seasons, the prevalence of different subtypes and lineages can vary widely [8].
As with other respiratory viruses, the coronavirus disease 2019 (COVID-19) pandemic restrictions resulted in the cessation of hMPV circulation for more than a year [9], creating the so-called immune debt, that is, the waning of general population immunity and the formation of groups of hMPV-naive children [9,10]. The immune debt may also be responsible for the changes in seasonality observed following the resurgence of hMPV infection after social distancing restrictions were lifted [10].
As respiratory virus circulation returns to usual patterns, national and regional data on hMPV epidemiology and genetic evolution are needed to assess the impact of potential hMPV vaccines.
The aim of this study was therefore to characterize the postpandemic epidemiologic impact of hMPV in Italy by aggregating multicenter virologic laboratory data from all molecular respiratory viral diagnostic tests from September 2022 to August 2024. In addition, several hMPV strains were characterized by phylogenetic analysis based on sequences of the external glycoprotein (G) gene.
## METHODS
We analyzed data from all molecular respiratory viral diagnostic tests performed between September 2022 and August 2024 in the microbiology and virology laboratories of the Italian Working Group on Respiratory Viral Infections (GLIViRe) throughout Italy. GLIViRe is a network of microbiology laboratories established in November 2018 [11], and includes tertiary and research institutes, as well as academic laboratories (Figure 1).
An email invitation was sent to all GLIViRe laboratories, requesting their participation in a study on hMPV. Data on the total number of respiratory specimens tested and diagnostic results for hMPV were provided by 17 GLIViRe centers throughout Italy (Table 1 and Figure 1). These are academic centers (university hospitals) and tertiary care hospitals, including hospitals accredited as research institutions and large territorial diagnostic centers (Supplementary Table 1).
Laboratories extracted data by week from diagnostic records using the following queries: (1) requests for hMPV detection in respiratory specimens (nasopharyngeal swabs, aspirates, bronchoalveolar lavages) collected from patients with symptoms of influenza-like illness (ILI) or ARI without age restriction; (2) requests for hMPV detection in respiratory specimens from patients ≥50 years with ILI or ARI; and (3) requests submitted between 29 August 2022 (week 35, 2022) and 1 September 2024 (week 34, 2024). Duplicate respiratory samples from the same patient were excluded from the analysis if collected less than 14 days apart.
In the hospital setting, patients from the emergency department or from hospital wards (referred to as inpatients) who had been clinically diagnosed with ILI or ARI were tested for respiratory pathogens at the request of the attending physician. Furthermore, several laboratories also performed molecular diagnosis of hMPV in respiratory samples from patients with ILI symptoms who were not hospitalized (referred to as outpatients) (Table 1). These latter centers participated in the national surveillance of respiratory viruses (RespiVirNet) [12].
Molecular testing for respiratory viruses during all months of the study period was performed in 15 laboratories using commercial platforms (Supplementary 1).
Patients' demographic data were also extracted from diagnostic records and anonymized in accordance with confidentiality requirements. The study was approved by the Ethics Committee of Rome University Hospital (Protocol 0966/2023) and, due to its retrospective nature, informed consent was not required.
## Phylogenetic Analysis
A random selection of residual diagnostic specimens obtained from outpatients and inpatients ≥ 50 years of age at the University of Milan, Pavia and Rome (laboratory Nos. 1, 2, and 15) were subjected to hMPV partial genetic characterization. The entire G gene was amplified [11] and subjected to Sanger sequencing. In addition, 3 Italian hMPV-A 2019 strains identified in a previous study [11], 133 hMPV-A, and 91 hMPV-B sequences from GenBank (https://www.ncbi.nlm.nih.gov/nucleotide/) were added to the datasets (Supplementary Table 2).
## Statistical Analysis
The number of hMPV detections per week and per laboratory were reported. The hMPV detection rate per macrogeographic area and age group was expressed as a crude proportion, with the corresponding 95% confidence interval (CI) with the Wilson interval, assuming a normal distribution. The statistical significance of the observed differences between the proportions in the various groups was determined using Fisher exact test. For continuous variables, Student t test was used. Statistical analysis was conducted using the open-source epidemiological statistics software OpenEpi, version 3.03 [13]. The following definitions were used to describe epidemiologic characteristics of epidemic waves: (1) peak value was defined as the maximum number of hMPV-positive specimens out of the total number of samples tested in a given week during the epidemic period; (2) peak time was defined as the week in which the peak was reached; and (3) hMPV epidemic onset was defined as the time when the number of hMPV-positive samples out of the total number of samples exceeded a certain threshold. The threshold was defined as the average number of hMPV-positive samples observed during the study period plus 2 standard deviations, as originally proposed by the US Centers for Disease Control and Prevention [14], and also described by Tabataba et al [15]. A P value less than .05 was considered statistically significant (2-tailed test).
## RESULTS
## HMPV Epidemiology
The 17 participating laboratories reported results for 96 460 samples collected from individuals with respiratory illness (ILI or ARI). A total of 3280 of 96 460 respiratory samples tested positive for hMPV in 2 consecutive years (from September 2022 to August 2024), resulting in an overall positivity rate of 3.4%. The positivity rate ranged from 2.3% (286/12 419) in southern Italy to 3.9% (1540/39 133) in the north-western macroarea (Table 1).
As shown in Table 1, most centers tested only inpatients attending the hospital or a limited number of outpatients, and 1 center tested only samples from outpatients with ILI in the framework of the RespiVirNet surveillance. A comparison of hMPV positivity rates among inpatients and outpatients was conducted exclusively for the north-west area, where a substantial number of outpatients were tested (Table 1). The hMPV positivity rates were significantly (P < .0001) higher among outpatients (513/10 001; 5.1%) than among inpatients (964/ 27 812; 3.5%).
## Epidemiology of HMPV in the Population Aged 50 Years and Older
The epidemiology of hMPV infection from samples in individuals aged ≥50 years, representing 42.9% (41 365/96 460) of all samples tested for hMPV, was characterized. Overall, 91% of respiratory samples tested were obtained from the upper respiratory tract. The hMPV-positivity rate in adults aged ≥50 years was 2.6% (1068/41 365), of whom 556/1068 (52.3%) were female. hMPV positivity showed a regional variation, ranging from 2.1% (44/2074) in central Italy to 4.4% (35/796) in southern Italy (Table 1). Comparison of hMPV rates among outpatients and inpatients aged 50 years and older tested in the north-west region (Table 1) showed higher positivity rates among outpatients (86/1800, 4.8% vs 286/14 623, 1.9%; P < .0001).
The risk of hMPV positivity was lower in subjects aged ≥50 years (odds ratio [OR], 0.6; 95% CI, .59-.69) than in subjects aged birth to 49 years. Exact age was not available for 5
## Temporal Distribution of hMPV Detections
The positivity rates remained consistent between the 2 time periods (2022-2023 and 2023-2024). The hMPV positivity rate was 3.7% and 4.1% in the north-west in 2022-2023 and 2023-2024, respectively. In the north-east, the hMPV-positivity rate was 3.3% in both periods. The hMPV positivity rate was 3.3% and 2.7% in the central area, and 2.7% and 2.0% in the south, during the 2 periods, respectively. The hMPV positivity rates among adults aged ≥50 years in the Italian macroareas were analyzed in more detail and compared between the study periods, as shown in Table 2. In the north-west, hMPV positivity rates were similar between 2022-2023 and 2023-2024 and during the weeks of the 2 seasonal peaks. In the north-east, hMPV showed trends towards higher positivity rates in 2023-2024 (P = .16 and P = .03, in season and in peak weeks, respectively). In the central macroarea, a decrease in the hMPV rate was observed in 2023-2024 compared to 2022-2023, which was also observed in the south. However, in the south, hMPV cases were more numerous in the 2023-2024 peak weeks compared to the 2022-2023 peak weeks (Table 2).
As shown in Figure 2A, the temporal distribution of hMPV-positive cases in the total population at the national level revealed the presence of 2 peaks. The first wave started in week 2, 2023, peaked in week 6, 2023 and ended in week 12, 2023, with a duration of 11 weeks. The second wave started in week 8, 2024, peaked in week 13, 2024, and ended in week 20, 2024, with a duration of 13 weeks. The overall hMPV positivity rate at peak was 11.9% in 2022-2023 and 12.8% in 2023-2024. Notably, the peak of hMPV detection observed in 2023-2024 was delayed by 7 weeks compared to the peak observed in 2022-2023.
The biweekly distribution of total respiratory specimens and hMPV positivity rate in subjects ≥ 50 years of age, shown in Figure 2B, is comparable to that observed in the overall population (Figure 2A). As in the total population, the hMPV epidemic curve in adults aged ≥50 years peaked in the 2023-2024 season between 6 and 14 weeks after the peak in the previous season (Figure 2B) and varied slightly by region (Table 2 and Figure 3).
## Phylogeny of hMPV Lineages
A phylogenetic analysis was performed on 35 sequences of the entire G glycoprotein gene generated by 3 centers located in the north-west (Milan and Pavia) and central (Rome) Italy. Of these, 18 belonged to subtype A and 17 to subtype B. Phylogenetic analysis showed that only 1 hMPV-A strain was of the A2b genotype, while all others were of the A2c genotype (Figure 4A). Because the A2c strains carrying duplications of 111 and 180 nucleotides (A2c-dup111 and A2c-dup180) have become predominant worldwide [24,25], a duplication calling analysis was performed. All study strains of the A2c genotype showed the 111-nucleotide duplication. Phylogenetic analysis of Italian hMPV-B showed that 1 strain was of the B1 genotype, while 16 belonged to the B2b genotype (Figure 4B).
## DISCUSSION
HMPV is an important pathogen responsible for upper and lower respiratory tract infections in children and adults worldwide [4][5][6]. However, there is a lack of documentation on the seasonal prevalence of hMPV and its impact on ILI and ARI cases in Italy, especially in older adults. In the present study, we report the distribution of hMPV cases over a 2-year period, including the 2 respiratory virus seasons following the complete lifting of pandemic restrictions in Italy. This is based on the results of routine molecular testing of over 96 000 respiratory samples. The data showed that hMPV was initially detected in both the general population and older adults in all study centers in late 2022 and then showed 2 waves during the study period. At the national level, hMPV positivity rates were similar between the 2 periods (September to August 2022-2023 and 2023-2024), with some regional variation. The number of hMPV-positive cases and the positivity rate were higher in the 2023-2024 season than in the previous season in northern Italy, but opposite trends were observed in central and southern Italy. Because cases of ILI or ARI were tested at the request of the attending physician, it is possible that clinical judgment about the need for respiratory multipanel testing differed between hospitals and influenced hMPV positivity rates in the macroareas, particularly where fewer hospitals were included (central and southern Italy). No national or regional algorithms or guidelines have yet been implemented to guide clinicians in respiratory virus testing decisions.
The overall hMPV detection rate of 3.4% is considerably higher than that observed in 2019, in a multicenter study conducted in 8 GLIViRe centers using the same multipanel molecular assays, with the exception of the academic center in Rome [11]. In 2019, hMPV was identified in 2% of samples tested, with rates ranging from 1.2% to 4%; unfortunately, this earlier study did not provide specific hMPV rates in adults aged 50 years and older [11].
Although lower than in younger subjects, the hMPV positivity found in this study in adults ≥50 years of age (2.6%) is noteworthy because in these subjects, the oldest age group was overrepresented, with one-third of cases occurring in subjects over 80 years of age. This observation confirms the results of our previous multicenter study [11], which showed that 30% of all hMPV-positive cases were identified in older adults (≥65 years). Furthermore, the 2.6% hMPV positivity in adults ≥50 years of age is comparable to that found in a recent study on hMPV-positive hospitalized adults (2.8%) [26].
It can be hypothesized that the observed increase in hMPV positivity rates in the postpandemic period, especially in younger age groups, may be due to increased clinician demand for the use of multipathogen panels for the diagnosis of respiratory infections. Another possible factor contributing to increased hMPV circulation in the postpandemic period could be the hMPV-specific immune debt created by pandemic restrictions, that is, a decline in virus-specific population immunity, particularly relevant in children [10].
In our study, the peak of hMPV activity differed between the 2 seasons: in the first, it was recorded in February, whereas in the 2023-2024 season, hMPV showed a peak of cases in April, similar to that observed in north-central Italy before the pandemic [11]. Prior to COVID-19, hMPV epidemics in central Europe alternated between winter and spring-summer every 2 years [27,28]. In addition, a global analysis showed that annual hMPV epidemics in temperate climates occurred in late winter and spring and usually followed the end of RSV epidemics with a delay of 1.7 months [29]. Indeed, perturbations in hMPV seasonality observed in the postpandemic period [10,30,31] have also been attributed to the immune debt that would have favored out-of-season spikes in several viral infections after the end of pandemic restrictions [9]. In this postpandemic context, we observed that the temporal distribution of the hMPV cases showed the earlier seasonal peak in winter when the virus returned to circulation in 2022-2023, followed by a second epidemic with a more typical seasonality in
## A B
Figure Phylogenetic analysis of the G gene of hMPV strains circulating in Italy. A, The phylogenetic tree of hMPV-A includes 18 GLIViRe sequences from study samples (September 2022 to August 2024), 3 GLIViRe sequences from a previous study [11], and 133 reference strains. B, The phylogenetic tree of hMPV-B includes 17 GLIViRe sequences from study samples (September 2022 to August 2024) and 91 reference strains. The hMPV-A dataset (155 sequences) was aligned using the strains NL/00/1 (GenBank accession number AF371337) and NL/00/17 (GenBank accession number FJ168779) as references for genotype A1 and A2, respectively. The hMPV-B dataset (108 sequences) was aligned using NL/99/1 (GenBank accession number AY525843) and NL/94/1 (GenBank accession number FJ168778) as reference strains for sublineages B1 and B2, respectively. The alignment tool used was MAFFT version 7.525 [16] and the visualization was performed using MEGA 11 [17]. A maximum likelihood phylogenetic tree of the dataset was constructed using IQ-TREE multicore version 2.3.3 [18] with a nucleotide substitution model identified by ModelFinder [19]. Branch robustness was assessed using the Shimodaira-Hasegawa approximate likelihood-ratio test [20] and ultrafast bootstrap approximation tests [21]. The phylogenetic tree was visualized using the ggtree package [22] and a custom R script [23]. Abbreviations: G, glycoprotein; GLIViRe, Working Group on Respiratory Viral Infections; hMPV, human metapneumovirus 2023-2024; this distribution is similar to the trend observed in Italy for RSV and influenza [12,32]. In particular, after the relaxation of restrictions in Italy in the summer of 2021, RSV reemerged and caused an intense epidemic in Italy in the fall of 2021, earlier than its historical trend, while RSV circulation was also intense in the fall-winter of 2022-2023, but peaked in January, more similar to the historical seasonality [32]. Influenza A virus returned to abundant circulation in the 2022-2023 season, with the peak of cases in fall-winter, earlier than registered in Italy in most influenza seasons; the 2023-2024 season was also intense, but with the more usual winter peak [12]. In the absence of historical data for hMPV in Italy, it is not possible to say whether the pandemic has altered the hMPV-specific population immunity, resulting in an earlier peak of hMPV cases in the 2022-2023 season, or whether hMPV winter waves have occurred in the past, alternating with spring waves in other seasons. A recent global analysis used routine surveillance data to assess how the timing of several respiratory viruses changed after the SARS-CoV-2 pandemic [33]. Although not statistically significant due to the paucity of national data, the timing of hMPV peak in the Northern Hemisphere appears to be earlier in the 2022-2023 season compared to prepandemic historical data [33].
Extended surveillance for hMPV is needed to understand whether the temporal pattern and peak intensity observed in the postpandemic period may continue to change over time.
Sequencing and phylogenetic analysis of the highly variable G gene was performed on a representative subset of hMPV-positive samples from north-western and central Italy to clarify whether novel strains were circulating after the pandemic period. No strains belonging to the A1 lineage were identified among hMPV-A cases in the present study or in the prepandemic period in Italy [11], consistent with previous reports indicating that no A1 strains have been detected worldwide since 2006 [8,31]. The A2b lineage was represented by a single case in 2023, while the A2c lineage was predominant, with only A2c-dup111 strains identified. The A2c-dup111 and A2c-dup180 variants have been in circulation since 2012, gradually replacing the previously dominant A2a and A2b lineages [8,31]. The duplication in G would confer an evolutionary advantage on both strains with the duplication by masking the conserved antigenic epitopes of the F glycoprotein [31], but A2c-dup111 has dominant compared to A2c-dup180. Indeed, no A2c-dup180 has been detected in the Netherlands since 2019 [31], while in Italy the strain with the longer insertion still accounted for 25% of A2c cases in 2019 [11]. This phylogenetic analysis confirms the prepandemic trend of A2c-dup111 dominance, which may have been accelerated by pandemic restrictions causing the disappearance of A2c-dup180 strain in Italy. However, given the limited number of cases that have been sequenced, it is possible that strains that were not detected after the pandemic are still circulating in areas where surveillance is not in place.
Among hMPV-B cases, 1 genotype B1 was identified during the 2022-2023 season, consistent with low detection rates in Italy in 2019 [11]. In the prepandemic period, genotype B1 represented a minority of hMPV cases worldwide [8,31], with the exception of an outbreak in Asian countries in 2017 [34]. As for the B2 genotype, clade 2a, found in northern Italy in 2019 [11], it was not identified in this study; in contrast, clade B2b was the dominant hMPV-B clade in the postpandemic seasons. The B2b clade is thought to have originated from the B1 genotype and spread mainly between 2012 and 2021 [31], with a notable increase in prevalence between 2012 and 2015 (Figure 4B). Several strains of the B2b clade have undergone an evolutionary process and appear to be dominant in the postpandemic period (Figure 4B). In contrast to B2b, the monophyletic B2a clade, which was predominantly circulating before 2004, was not detected after the onset of the SARS-CoV-2 pandemic (Figure 4B).
The global postpandemic molecular epidemiology of hMPV has been characterized in only a few studies. In the Netherlands, hMPV-A clade A2.2.2 carrying the 111-nucleotide duplication in G (also known as A2c-dup111) was predominant in 2021 [8]. In Taiwan, A2c-dup111, which was dominant in 2021, was not detected in 2022 while the B2 genotype was dominant in 2023 [35]. In Korea, the hMPV 2022 cases were A2.2.1 (the alternative name for A2b) and B2 strains [27]. In Australia, an A2b variant emerged in 2019 and became dominant in 2020-2021; this variant cocirculated with B1 and B2 genotypes in 2021 [10]. In Japan, hMPV-B B1 and B2, but not hMPV-A, were identified following the SARS-CoV-2 pandemic [36]. These data suggest that the pandemic restrictions may have caused distinct local events of predominance and extinction of hMPV genetic lineages, which could potentially lead to further genetic divergence and new evolutionary events. Further surveillance studies using wholegenome sequencing should determine the potential for hMPV genetic variants to spread to distant geographic locations and generate unexpected outbreaks [31,34].
This study has several limitations. First, the study design is limited to 2 years, which may affect the generalizability of the epidemiologic results, and the study is retrospective in nature, which precludes the inclusion of clinical information. In particular, subjects tested in the emergency department are considered here as inpatients because the vast majority were subsequently hospitalized in wards or intensive care units, although the exact number is not known. In addition, it was not possible to obtain detailed age data for all patients tested for hMPV, only for hMPV-positive adults aged 50 years and older, and we could not calculate any age bias in testing. Second, the GLIViRe centers used different molecular methods, laboratory-developed real-time PCR or European union in vitro diagnostic device directive-compliant multiplex assays, which may have different analytical sensitivities. However, 8 and 6 laboratories, respectively, used the same commercial platform for hMPV detection (Allplex Respiratory Panel and Biofire FilmArray), which is likely to reduce analytical variability and increase confidence in estimating hMPV positivity. Third, the number of hMPV-positive samples and the genomic region sequenced are relatively limited. This is due to the limited availability of residual diagnostic specimens and to the different laboratory capacity in terms of sequencing. To address the latter issue, the variability in terms of different sequencing machines was considered in the bioinformatics analysis.
Nevertheless, this study represents a significant contribution to the knowledge of hMPV in Italy, demonstrating the national and regional rates of this infection in the elderly population. Epidemiologic analysis of hMPV circulation in the postpandemic seasons provides insight into potential shifts in hMPV epidemiology. In addition, phylogenetic analysis of circulating strains suggests potential changes in hMPV lineages in the postpandemic era. These findings may prove valuable in the differential surveillance and management of hMPV disease and in the development of prevention strategies.
## References
1. (1990) "Lower Respiratory Infections and Antimicrobial Resistance Collaborators. Global, regional, and national incidence and mortality burden of non-COVID-19 lower respiratory infections and aetiologies"
2. Hansen, Chaves, Demont et al. (1999) "Mortality associated with influenza and respiratory syncytial virus in the US" *JAMA Netw Open*
3. Van Den Hoogen, De Jong, Groen (2001) "A newly discovered human pneumovirus isolated from young children with respiratory tract disease" *Nat Med*
4. Boivin, Abed, Pelletier (2002) "Virological features and clinical manifestations associated with human metapneumovirus: a new paramyxovirus responsible for acute respiratorytract infections in all age groups" *J Infect Dis*
5. Falsey, Erdman, Anderson et al. (2003) "Human metapneumovirus infections in young and elderly adults" *J Infect Dis*
6. Shi, Arnott, Semogas (2020) "The etiological role of common respiratory viruses in acute respiratory infections in older adults: a systematic review and meta-analysis" *J Infect Dis*
7. Van Den Hoogen, Herfst, Sprong (2004) "Antigenic and genetic variability of human metapneumoviruses" *Emerg Infect Dis*
8. Groen, Van Nieuwkoop, Meijer (2023) "Emergence and potential extinction of genetic lineages of human metapneumovirus between 2005 and 2021" *mBio*
9. Chow, Uyeki, Chu (2023) "The effects of the COVID-19 pandemic on community respiratory virus activity" *Nat Rev Microbiol*
10. Foley, Sikazwe, Smith (2022) "An unusual resurgence of human metapneumovirus in western Australia following the reduction of non-pharmaceutical interventions to prevent SARS-CoV-2 transmission" *Viruses*
11. Pierangeli, Piralla, Renteria (2023) "Multicenter epidemiological investigation and genetic characterization of respiratory syncytial virus and metapneumovirus infections in the pre-pandemic 2018-2019 season in northern and central Italy" *Clin Exp Med*
12. Epicentro, Respivirnet (2024)
13. Dean, Sullivan, Soe (2024) "Open source epidemiologic statistics for public health"
14. (2024) "Influenza surveillance: purpose and methods"
15. Tabataba, Chakraborty, Ramakrishnan (2017) "A framework for evaluating epidemic forecasts" *BMC Infect Dis*
16. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
17. Tamura, Stecher, Kumar (2021) "MEGA11: molecular evolutionary genetics analysis version 11" *Mol Biol Evol*
18. Minh, Schmidt, Chernomor (2020) "IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era" *Mol Biol Evol*
19. Kalyaanamoorthy, Minh, Wong et al. (2017) "ModelFinder: fast model selection for accurate phylogenetic estimates" *Nat Methods*
20. Shimodaira, Hasegawa (1999) "Multiple comparisons of log-likelihoods with applications to phylogenetic inference" *Mol Biol Evol*
21. Hoang, Chernomor, Haeseler et al. (2018) "UFBoot2: improving the ultrafast bootstrap approximation" *Mol Biol Evol*
22. Yu, Smith, Zhu et al. (2017) "ggtree: an R package for visualization and annotation of phylogenetic trees with their covariates and other associated data" *Methods Ecol Evol*
23. Team (2015) "RStudio: integrated development environment for R" *PBC*
24. Saikusa, Nao, Kawakami (2019) "Predominant detection of the subgroup A2b human metapneumovirus strain with a 111-nucleotide duplication in the G gene in Yokohama city, Japan in 2018" *Jpn J Infect Dis*
25. Saikusa, Kawakami, Nao (2014) "180-Nucleotide duplication in the G gene of human metapneumovirus A2b subgroup strains circulating in Yokohama city" *Front Microbiol*
26. Falsey, Walsh, House (2024) "Assessment of illness severity in adults hospitalized with acute respiratory tract infection due to influenza, respiratory syncytial virus, or human metapneumovirus" *Influenza Other Respir Viruses*
27. Aberle, Aberle, Redlberger-Fritz et al. (2010) "Human metapneumovirus subgroup changes and seasonality during epidemics" *Pediatr Infect Dis J*
28. Heininger, Kruker, Bonhoeffer et al. (2009) "Human metapneumovirus infections--biannual epidemics and clinical findings in children in the region of Basel, Switzerland" *Eur J Pediatr*
29. Li, Reeves, Wang (2019) "Global patterns in monthly activity of influenza virus, respiratory syncytial virus, parainfluenza virus, and metapneumovirus: a systematic analysis" *Lancet Glob Health*
30. Cho, Kim, Lee (2023) "Re-emergence of HMPV in Gwangju, South Korea, after the COVID-19 pandemic" *Pathogens*
31. Piñana, Vila, Maldonado (2020) "Insights into immune evasion of human metapneumovirus: novel 180-and 111-nucleotide duplications within viral G gene throughout 2014-2017" *J Clin Virol*
32. Pierangeli, Midulla, Piralla (2024) "Sequence analysis of respiratory syncytial virus cases reveals a novel subgroup -B strain circulating in north-central Italy after pandemic restrictions" *J Clin Virol*
33. Del Riccio, Caini, Bonaccorsi (2024) "Global analysis of respiratory viral circulation and timing of epidemics in the pre-COVID-19 and COVID-19 pandemic eras, based on data from the Global Influenza Surveillance and Response System (GISRS)" *Int J Infect Dis*
34. Yi, Zou, Peng (2019) "Epidemiology, evolution and transmission of human metapneumovirus in Guangzhou China" *Sci Rep*
35. Yang, Chiu, Tsai (2024) "Epidemiology of human metapneumovirus in Taiwan from 2013 to 2023" *Arch Virol*
36. Shirato, Suwa, Nao (2024) "Molecular epidemiology of human metapneumovirus in east Japan before and after COVID-19, 2017-2022" *Jpn J Infect Dis*
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.