Mardiyyah/TAPT_data_V2_split · Datasets at Hugging Face

PMCID stringclasses 24 values	Title stringclasses 24 values	Sentences stringlengths 2 40.7k
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Here, by comparing to other methods, we highlight distinct features of MERGE-seq and key biological insights that MERGE-seq can provide.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Early approaches of barcode-based neuronal projection mapping mainly focus on elucidating the projections of individual neurons in a single brain without providing the transcriptional signatures corresponding to those individual neurons (MAPseq; Kebschull et al., 2016).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	We therefore developed MERGE-seq to connect single-neuron transcriptome and projectome with high throughput.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	While there are some conceptual similarities to BARseq or ConnectID (Chen et al., 2019; Klingler et al., 2021), MERGE-seq has its unique features and advantages.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	BARseq can acquire single-neuron transcriptome and projectome, but with only a number of genes due to limited throughout of in situ sequencing.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	An improved version of BARseq can allow tens of genes to be detected, but still with a low throughput compared to scRNA-seq and a high cost in regards to synthesizing RNA probes (Sun et al., 2021).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	ConnectID (scRNA-seq combined with MAPseq) improves the detection of transcriptome using scRNA-seq but has a relatively low recovery rate of cells with transcriptome and projectome simultaneously (~16%, 391 cells with barcode identity in 2450 cells with scRNA-seq; Klingler et al., 2021).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In contrast, MERGE-seq enables transcriptional profiling of thousands of genes per neuron, with valid projectome barcode information recovered from approximately 50% of FAC-sorted cells passing stringent determination criteria.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Another advantage of MERGE-seq is that users only need to sequence one brain region – the source area.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	While, in BARseq or ConnectID, users need to perform numerous tissue homogenization and sequencing for downstream brain regions to query target area barcodes information (projection).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	MERGE-seq is a retro-AAV-based scRNA-seq approach.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Previous research has employed retro-AAV techniques to probe the projection-specific transcriptome or epigenome of individual neurons (Lui et al., 2021; Tasic et al., 2018; Tasic et al., 2016; Yao et al., 2021; Zhang et al., 2021).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Yet, these studies have not developed a multiplexed approach for investigating the complex collateral projection patterns of neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Another retro-AAV based-approach, VECTORseq, was recently developed to associate neuronal projectome and transcriptome (Cheung et al., 2021).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	VECTORseq used several viral transgenes including three recombinases (DreO, Cre, Flpo) and two fluorescent proteins (tdTomato and EGFP) to barcode neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	However, these transgenes are variable in length (DreO, ~1000 bp; Cre, ~1000 bp; FLPo, ~1200 bp; tdTomato, ~1400 bp; EGFP, ~700 bp) and driven by different promoters with different strength (EF1a, hSyn, CAG).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Such an approach will inevitably result in differential expression of these different transgenes in labelled neurons, which in turn leads to different rates of transgene recovery in these neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In addition, viral-mediated overexpression of these recombinase may lead to toxic to the labeled neurons due to non-specific recombination events (Xiao et al., 2012).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Therefore, the transgenes used in VECTORseq method should be carefully selected to avoid any potential interferences with neuronal function or gene expression by these different transgenes.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In contrast, MERGE-seq used 15-nucleotide barcode sequences in the 3’UTR region of EGFP as projection index driven by the same promoter to label different projection neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The expression of these different barcoded EGFP mRNA is comparable, and the number of these barcoded retro-AAV is unlimited.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Therefore, MERGE-seq allows users to examine more populations (theoretically unlimited) in one brain and more extensive analysis of collateralization.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Further, MERGE-seq can reveal projectome of single collateral projection neurons and identify molecular features of these neurons (Figure 5).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	However, the collateral projection patterns of single neurons were not reported in VECTORseq and Retro-seq-based method (Cheung et al., 2021; Lui et al., 2021).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For example, Lui et al., 2021.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	used Retro-seq to investigate the correspondence between transcriptomics and projection patterns of vmPFC neurons, and inferred collateral projection based on the finding that transcriptome-defined neuron subtypes can project to different targets (or neurons projecting to different targets share common transcriptome-defined neuron subtype).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	However, the population level multi-target projections of a transcriptome-defined neuron subtype do not necessarily reflect collateral projection of individual neurons within a subtype.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For instance, individual neurons within a subtype could project to distinct targets (dedicated projection), but their collective projections show multiple targets.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In contrast, in MERGE-seq, individual neurons that were retrogradely labeled multiple projection barcodes are determined as collateral projection neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	By MERGE-seq analysis, we uncovered dedicated and collateral projection patterns of individual vmPFC neurons to the five downstream targets, and revealed molecular features associated with these dedicated or collateral projection neurons (Figures 3—5).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In addition, MERGE-seq strategy can be readily applied to other animal models, which is especially beneficial for research in models (e.g. non-human primate) where genetic manipulation is challenging.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In summary, while Retro-seq methods provide valuable population-level insights, they do not capture the complex collateral projections that MERGE-seq can discern at the single-cell level.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Our findings build upon and extend those of Lui et al. by demonstrating that individual neurons within transcriptome-defined subtypes exhibit a diverse range of projection patterns.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	This contributes a new layer of understanding to the intricate architecture of PFC circuits, emphasizing the nuanced interplay between divergence and convergence in neuronal pathways.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Although MERGE-seq does not offer spatial information of neurons currently, it leverages widely accessible droplet-based scRNA-seq, avoiding specialized equipment.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Meanwhile, the extensive spatially resolved mouse brain atlases available (Allen et al., 2023; Yao et al., 2023; Zhang et al., 2023) allow for easy spatial annotation of cell populations using DEGs identified by scRNA-seq, as we demonstrated by mapping neuronal subtypes with MERFISH data of PFC.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Compared to imaging-based spatial transcriptomics like MERSCOPE with constrained gene numbers, or next-generation sequencing (NGS)-based methods that are lack of true single-cell resolution (e.g. 50 µm 10 x Genomics Visium or 10–50 µm for DBiT-seq-based methods; Deng et al., 2023), we believe our method stands out as a robust solution and offers an advantageous balance between resolution and scope.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	There are several potential concerns and limitations of current study.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	First, a recognized limitation of using retro-AAV-based methods, including MERGE-seq, is the imperfect retrograde labeling efficiency in target regions.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Labeling efficiency could be variable depending on the different source brain regions, projection strength, the distance between source and target brain regions and different AAV serotypes or tropism.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For example, only nine neurons of the L5-Htr2c subtype were recovered with valid barcodes, which may be attributable to technical factors including cell loss during dissociation or AAV2-retro tropism.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Alternatively, this subtype may intrinsically lack projections to the selected target regions examined in this study.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Furthermore, single-cell dissociation for scRNA-seq can result in cell loss, thereby reducing the recovery rate of barcoded neurons.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	All these factors could influence the extent to which the complete range of neuronal projections is captured.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Consequently, the quantitative conclusions drawn here might not fully represent the true extent of neuronal projections.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Second, the robust detection of projection barcodes and its recovery rate in neurons labeled with barcoded AAV-retro viruses is indeed a critical and challenging aspect of our methodology.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	As mentioned above, this challenge is largely due to the differential viral transduction efficiency across neurons, leading to inconsistent barcode expression.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Neurons with low barcode expression may fall beneath the detection threshold of conventional sequencing methods.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	A suboptimal recovery rate can potentially lead to underrepresentation of certain neuron populations or projection patterns in the analyzed data.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	This in turn could impact the interpretation of neuronal connectivity and function, as projections that are less efficiently labeled or harder to detect might be overlooked.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For instance, if a subset of neurons with low barcode expression is systematically missed, it could erroneously suggest that these neurons do not participate in specific projection patterns.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Conversely, overrepresentation of certain barcodes due to higher transduction efficiency could falsely indicate a predominance of certain projections.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	One potential solution to improve barcode detection is to include FAC-sorted EGFP-negative cells as a negative control, which may help to differentiate between true signal and background noise.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Enhancements in sequencing technologies, offering increased read lengths and deeper sequencing, could potentially improve barcode detection sensitivity.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In parallel, applying single-molecule FISH technologies like MERFISH to spatially resolve barcodes offers a robust and direct detection method.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	This technology can provide detailed coverage and resolution of individual RNA molecules within single cells, bypassing additional PCR amplification steps and reducing cell loss during physical isolation.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Furthermore, carefully controlling the viral titer and refining the procedures of single neuron suspension preparation, as performed in this study, is required to control the labeling efficiency and recovery rate.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	DMS is en route from vmPFC to subcortical regions (Shepherd, 2013), thus raising another concern about the transducing ability of AAV2 in axons of passages.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	However, the retrograde transport of AAV has been effectively demonstrated to target projection neurons at axonal terminals, with injections into the DMS exhibiting labeling patterns and efficiencies that match those of synthetic tracers (Tervo et al., 2016).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Further, it has been experimentally verified that AAV2 spread is confined to the vicinity of synaptic terminals and does not affect axon fibers in passages, especially as evidenced by retro-AAV injections in the cervical spinal cord (Wang et al., 2018).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	While these findings are reassuring, additional research is needed to unequivocally eliminate the possibility of transduction along axon fibers of passage.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The five distinct injection sites we chose for our study are spatially disparate, encompassing both cortical and subcortical regions, and span a range from the anterior (Bregma,+2 mm) to the posterior (Bregma, –1.5 mm) brain regions.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	This separation mitigates the potential overlap in labeling when examining spatially proximate nuclei, such as those in the hypothalamus.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Nevertheless, examining such closely situated targets would necessitate meticulous quantification of virus injection volumes to prevent cross-target viral dissemination, ensuring the specificity required for accurate projection mapping.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	In summary, we develop MERGE-seq, a powerful multiplexed projectome and transcriptome analysis platform that will help researchers perform big-data research at low cost.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	This will enable researchers to understand organizing principles and molecular features of neural circuits across modalities, and to construct more comprehensive mesoscale connectomes.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Plasmid pAAV-CAG-tdTomato (Addgene, #59462) was first modified by replacing tdTomato and WPRE with EGFP by T4 DNA Ligase mediated ligation.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	A 15 bp barcode sequence was then inserted after the stop codon of EGFP, linked by EcoRI restriction enzyme recognition site.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Sequences barcode 0 representing the AI target, CTGCACCGACGCATT; barcode 1 (DMS target), GAAGGCACAGACTTT; barcode 2 (MD target), GTTGGCTGCAATCCA; barcode 3 (BLA target), AAGACGCCGTCGCAA; barcode 4 (LH target), TATTCGGAGGACGAC.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Other barcode sequences used for IHC include barcode 10, AGCTATGCACGATCA; barcode 206, GCGTAAGTCTCCTTG; barcode 210, CCTGTATGCGTGGAG.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Engineered viruses were produced by Gene Editing Core Facility, Center for Excellence in Brain Science and Intelligence Technology.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Male adult C57BL/6 mice (8 weeks of age) were anesthetized intraperitoneally using pentobarbital sodium (10 mg/mL, 120 mg/kg b.w.)
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	and unilaterally injected with rAAV2-retro-EGFP-Barcode virus (barcode 0, 1, 2, 3, 4) into five projection targets simultaneously.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Coordinates for these injections are as follows.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Reference from Bregma and dura, AI at two locations (in mm: 2.0 AP, 2.52 ML, –2.0 DV; 1.6 AP, 2.97 ML, –2.2 DV) with rAAV2-retro-EGFP-barcode 0 (250 nl and 200 nl, 2.90×10 VG/ml); DMS at one location (in mm: 0.6 AP, 1.8 ML, –2.2 DV, 8 degree angle), with rAAV2-retro-EGFP-barcode 1 (500 nl, 1.00×10 VG/ml); MD at one location (in mm: –1.25 AP, 1.35 ML, –3.55 DV, 20 degree angle), with rAAV2-retro-EGFP-barcode 2 (300 nl, 1.27×10 VG/ml); BLA at one location (in mm: –1.5 AP, 3.2 ML, –4.2 DV), with rAAV2-retro-EGFP-barcode 3 (300 nl, 2.00×10 VG/ml); LH at one location (in mm: –0.94 AP, 1.2 ML, –4.55 DV), with rAAV2-retro-EGFP-barcode 4 (250 nl, 2.25×10 VG/ml).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Following each injection, the micropipette was left in the tissue for 10 min before being slowly withdrawn to prevent virus spilling and backflow.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Mice were sacrificed 6 weeks after virus injection.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Single-cell suspensions were generated as described in methods below.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For dual-color retrograde virus tracing, two regions were ipsilaterally injected with virus at the same time, one with rAAV2-retro-EGFP-barcode 10 (2.00×10 VG/ml) or rAAV2-retro-EGFPnls-barcode 206 or 210 (3.10×10 VG/ml for barcode 206 and 4.38×10 VG/ml for barcode 210) and one with rAAV2-retro-tdTomato (2.25×10 VG/ml).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	rAAV2-retro-EGFPnls was used to avoid dense fiber staining when performing immunohistochemistry.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	We deposited the virus plasmid constructs to Addgene (pAAV-CAG-EGFP barcode-(0–10)-SV40 polyA, pAAV-CAG-EGFPnls barcode-(206, 210)-SV40 polyA; Addgene ID 190864–190876).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For mice without FAC-sorting (mouse #1, #2, #3), three mice that had been injected with virus were anaesthetized and then subjected to transcranial perfusion with ice-cold oxygenated self-made dissection buffer (in mM: 92 Choline chloride, 2.5 KCl, 1.2 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 Glucose, 5 Sodium ascorbate, 2 Thiourea, 3 Sodium pyruvate, 10 MgSO4.7H2O, 0.5 CaCl2.2H2O, 12 N-Acetyl-L-Cysteine).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The brain was removed, 300 µm vibratome sections were collected, and the PrL and IL regions were microdissected under a stereo microscope with a cooled platform.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Brain slices were incubated in dissection buffer with 10 µM AMPA receptor antagonist CNQX (Abcam, ab120017) and 50 µM NMDA receptor antagonist D-AP5 (Abcam, ab120003) at 33 °C for 30 min.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The pieces were dissociated first using the ice-cold oxygenated dissection buffer added papain (20 units/ml, Worthington, LS003126), 0.067 mM 2-mercaptoethanol (Sigma, M6250), 1.1 mM EDTA (Invitrogen, 15575020), 5.5 mM L-Cysteine hydrochloride monohydrate (Sigma, C7880) and 100 units/ml Deoxyribonuclease I (Sigma, D4527), with 30–40 min enzymatic digestion at 37 °C, followed by 30 min 1 mg/ml protease (Sigma, P5147) and 1 mg/ml dispase (Worthington, LS02106) enzymatic digestion at 25 °C.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Supernatant was removed and digestion was terminated using dissection buffer containing 2% fetal bovine serum (FBS, Bioind, 04-002-1A).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Single-cell suspension was generated by manual trituration using fire-polishing Pasteur pipettes and filtered through a 35 µm DM-equilibrated cell strainer (Falcon, 352052).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Cells were then pelleted at 400 × g for 5 min.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The supernatant was carefully removed and resuspended in 1–2 ml dissection buffer containing 2% FBS.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	The suspension was then subjected to the debris removal step using the Debris Removal Solution (Miltenyi, 130-109-398).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Cell pellets were resuspended and 48,000 cells were loaded into 3 lanes to perform 10x Genomics sequencing.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	For mice with FAC-sorting (mouse #4, #5, #6), PrL and IL regions were microdissected and dissociated as mice without FAC-sorting, cells were sorted to enrich for EGFP-positive rAAV2-retro-EGFP-barcodes labeled cells.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	About 4893 EGFP-positive cells were captured and loaded to perform 10x Genomics sequencing.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Chromium Single Cell 3' Reagent Kits (v3) were used for library preparation (10x Genomics).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Libraries were sequenced on an Illumina Novaseq 6000 system.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Parallel PCR reactions were performed containing 50 ng of post cDNA amplification reaction cleanup material as a template.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	P5-Read1 (AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTC) and P7-index-Read2-EGFP (CAAGCAGAAGACGGCATACGAGATAGGATTCGGTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTGgCATGGACGAGCTGTACAAG) (200 nM each) were used as primers with the NEBNext Ultra II Q5 Master Mix (NEB, M0544L).
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Amplification was performed using the following PCR protocol: (1) 33 °C for 1 min, (2) 98° for 10 s, then 65 °C for 75 s (20–24 cycles), (3) 75 °C for 5 min.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	Reactions were re-pooled during 1 X SPRI selection (Beckman, B23317), which harvested virus projection barcodes library.
PMC10914349	High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.	431–437 bp (with 120 bp adaptors) libraries were sequenced using Illumina HiSeq X Ten.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Here, by comparing to other methods, we highlight distinct features of MERGE-seq and key biological insights that MERGE-seq can provide.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Early approaches of barcode-based neuronal projection mapping mainly focus on elucidating the projections of individual neurons in a single brain without providing the transcriptional signatures corresponding to those individual neurons (MAPseq; Kebschull et al., 2016).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

We therefore developed MERGE-seq to connect single-neuron transcriptome and projectome with high throughput.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

While there are some conceptual similarities to BARseq or ConnectID (Chen et al., 2019; Klingler et al., 2021), MERGE-seq has its unique features and advantages.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

BARseq can acquire single-neuron transcriptome and projectome, but with only a number of genes due to limited throughout of in situ sequencing.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

An improved version of BARseq can allow tens of genes to be detected, but still with a low throughput compared to scRNA-seq and a high cost in regards to synthesizing RNA probes (Sun et al., 2021).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

ConnectID (scRNA-seq combined with MAPseq) improves the detection of transcriptome using scRNA-seq but has a relatively low recovery rate of cells with transcriptome and projectome simultaneously (~16%, 391 cells with barcode identity in 2450 cells with scRNA-seq; Klingler et al., 2021).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In contrast, MERGE-seq enables transcriptional profiling of thousands of genes per neuron, with valid projectome barcode information recovered from approximately 50% of FAC-sorted cells passing stringent determination criteria.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Another advantage of MERGE-seq is that users only need to sequence one brain region – the source area.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

While, in BARseq or ConnectID, users need to perform numerous tissue homogenization and sequencing for downstream brain regions to query target area barcodes information (projection).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

MERGE-seq is a retro-AAV-based scRNA-seq approach.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Previous research has employed retro-AAV techniques to probe the projection-specific transcriptome or epigenome of individual neurons (Lui et al., 2021; Tasic et al., 2018; Tasic et al., 2016; Yao et al., 2021; Zhang et al., 2021).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Yet, these studies have not developed a multiplexed approach for investigating the complex collateral projection patterns of neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Another retro-AAV based-approach, VECTORseq, was recently developed to associate neuronal projectome and transcriptome (Cheung et al., 2021).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

VECTORseq used several viral transgenes including three recombinases (DreO, Cre, Flpo) and two fluorescent proteins (tdTomato and EGFP) to barcode neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

However, these transgenes are variable in length (DreO, ~1000 bp; Cre, ~1000 bp; FLPo, ~1200 bp; tdTomato, ~1400 bp; EGFP, ~700 bp) and driven by different promoters with different strength (EF1a, hSyn, CAG).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Such an approach will inevitably result in differential expression of these different transgenes in labelled neurons, which in turn leads to different rates of transgene recovery in these neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In addition, viral-mediated overexpression of these recombinase may lead to toxic to the labeled neurons due to non-specific recombination events (Xiao et al., 2012).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Therefore, the transgenes used in VECTORseq method should be carefully selected to avoid any potential interferences with neuronal function or gene expression by these different transgenes.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In contrast, MERGE-seq used 15-nucleotide barcode sequences in the 3’UTR region of EGFP as projection index driven by the same promoter to label different projection neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The expression of these different barcoded EGFP mRNA is comparable, and the number of these barcoded retro-AAV is unlimited.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Therefore, MERGE-seq allows users to examine more populations (theoretically unlimited) in one brain and more extensive analysis of collateralization.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Further, MERGE-seq can reveal projectome of single collateral projection neurons and identify molecular features of these neurons (Figure 5).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

However, the collateral projection patterns of single neurons were not reported in VECTORseq and Retro-seq-based method (Cheung et al., 2021; Lui et al., 2021).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For example, Lui et al., 2021.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

used Retro-seq to investigate the correspondence between transcriptomics and projection patterns of vmPFC neurons, and inferred collateral projection based on the finding that transcriptome-defined neuron subtypes can project to different targets (or neurons projecting to different targets share common transcriptome-defined neuron subtype).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

However, the population level multi-target projections of a transcriptome-defined neuron subtype do not necessarily reflect collateral projection of individual neurons within a subtype.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For instance, individual neurons within a subtype could project to distinct targets (dedicated projection), but their collective projections show multiple targets.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In contrast, in MERGE-seq, individual neurons that were retrogradely labeled multiple projection barcodes are determined as collateral projection neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

By MERGE-seq analysis, we uncovered dedicated and collateral projection patterns of individual vmPFC neurons to the five downstream targets, and revealed molecular features associated with these dedicated or collateral projection neurons (Figures 3—5).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In addition, MERGE-seq strategy can be readily applied to other animal models, which is especially beneficial for research in models (e.g. non-human primate) where genetic manipulation is challenging.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In summary, while Retro-seq methods provide valuable population-level insights, they do not capture the complex collateral projections that MERGE-seq can discern at the single-cell level.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Our findings build upon and extend those of Lui et al. by demonstrating that individual neurons within transcriptome-defined subtypes exhibit a diverse range of projection patterns.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

This contributes a new layer of understanding to the intricate architecture of PFC circuits, emphasizing the nuanced interplay between divergence and convergence in neuronal pathways.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Although MERGE-seq does not offer spatial information of neurons currently, it leverages widely accessible droplet-based scRNA-seq, avoiding specialized equipment.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Meanwhile, the extensive spatially resolved mouse brain atlases available (Allen et al., 2023; Yao et al., 2023; Zhang et al., 2023) allow for easy spatial annotation of cell populations using DEGs identified by scRNA-seq, as we demonstrated by mapping neuronal subtypes with MERFISH data of PFC.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Compared to imaging-based spatial transcriptomics like MERSCOPE with constrained gene numbers, or next-generation sequencing (NGS)-based methods that are lack of true single-cell resolution (e.g. 50 µm 10 x Genomics Visium or 10–50 µm for DBiT-seq-based methods; Deng et al., 2023), we believe our method stands out as a robust solution and offers an advantageous balance between resolution and scope.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

There are several potential concerns and limitations of current study.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

First, a recognized limitation of using retro-AAV-based methods, including MERGE-seq, is the imperfect retrograde labeling efficiency in target regions.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Labeling efficiency could be variable depending on the different source brain regions, projection strength, the distance between source and target brain regions and different AAV serotypes or tropism.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For example, only nine neurons of the L5-Htr2c subtype were recovered with valid barcodes, which may be attributable to technical factors including cell loss during dissociation or AAV2-retro tropism.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Alternatively, this subtype may intrinsically lack projections to the selected target regions examined in this study.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Furthermore, single-cell dissociation for scRNA-seq can result in cell loss, thereby reducing the recovery rate of barcoded neurons.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

All these factors could influence the extent to which the complete range of neuronal projections is captured.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Consequently, the quantitative conclusions drawn here might not fully represent the true extent of neuronal projections.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Second, the robust detection of projection barcodes and its recovery rate in neurons labeled with barcoded AAV-retro viruses is indeed a critical and challenging aspect of our methodology.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

As mentioned above, this challenge is largely due to the differential viral transduction efficiency across neurons, leading to inconsistent barcode expression.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Neurons with low barcode expression may fall beneath the detection threshold of conventional sequencing methods.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

A suboptimal recovery rate can potentially lead to underrepresentation of certain neuron populations or projection patterns in the analyzed data.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

This in turn could impact the interpretation of neuronal connectivity and function, as projections that are less efficiently labeled or harder to detect might be overlooked.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For instance, if a subset of neurons with low barcode expression is systematically missed, it could erroneously suggest that these neurons do not participate in specific projection patterns.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Conversely, overrepresentation of certain barcodes due to higher transduction efficiency could falsely indicate a predominance of certain projections.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

One potential solution to improve barcode detection is to include FAC-sorted EGFP-negative cells as a negative control, which may help to differentiate between true signal and background noise.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Enhancements in sequencing technologies, offering increased read lengths and deeper sequencing, could potentially improve barcode detection sensitivity.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In parallel, applying single-molecule FISH technologies like MERFISH to spatially resolve barcodes offers a robust and direct detection method.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

This technology can provide detailed coverage and resolution of individual RNA molecules within single cells, bypassing additional PCR amplification steps and reducing cell loss during physical isolation.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Furthermore, carefully controlling the viral titer and refining the procedures of single neuron suspension preparation, as performed in this study, is required to control the labeling efficiency and recovery rate.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

DMS is en route from vmPFC to subcortical regions (Shepherd, 2013), thus raising another concern about the transducing ability of AAV2 in axons of passages.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

However, the retrograde transport of AAV has been effectively demonstrated to target projection neurons at axonal terminals, with injections into the DMS exhibiting labeling patterns and efficiencies that match those of synthetic tracers (Tervo et al., 2016).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Further, it has been experimentally verified that AAV2 spread is confined to the vicinity of synaptic terminals and does not affect axon fibers in passages, especially as evidenced by retro-AAV injections in the cervical spinal cord (Wang et al., 2018).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

While these findings are reassuring, additional research is needed to unequivocally eliminate the possibility of transduction along axon fibers of passage.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The five distinct injection sites we chose for our study are spatially disparate, encompassing both cortical and subcortical regions, and span a range from the anterior (Bregma,+2 mm) to the posterior (Bregma, –1.5 mm) brain regions.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

This separation mitigates the potential overlap in labeling when examining spatially proximate nuclei, such as those in the hypothalamus.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Nevertheless, examining such closely situated targets would necessitate meticulous quantification of virus injection volumes to prevent cross-target viral dissemination, ensuring the specificity required for accurate projection mapping.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

In summary, we develop MERGE-seq, a powerful multiplexed projectome and transcriptome analysis platform that will help researchers perform big-data research at low cost.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

This will enable researchers to understand organizing principles and molecular features of neural circuits across modalities, and to construct more comprehensive mesoscale connectomes.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Plasmid pAAV-CAG-tdTomato (Addgene, #59462) was first modified by replacing tdTomato and WPRE with EGFP by T4 DNA Ligase mediated ligation.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

A 15 bp barcode sequence was then inserted after the stop codon of EGFP, linked by EcoRI restriction enzyme recognition site.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Sequences barcode 0 representing the AI target, CTGCACCGACGCATT; barcode 1 (DMS target), GAAGGCACAGACTTT; barcode 2 (MD target), GTTGGCTGCAATCCA; barcode 3 (BLA target), AAGACGCCGTCGCAA; barcode 4 (LH target), TATTCGGAGGACGAC.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Other barcode sequences used for IHC include barcode 10, AGCTATGCACGATCA; barcode 206, GCGTAAGTCTCCTTG; barcode 210, CCTGTATGCGTGGAG.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Engineered viruses were produced by Gene Editing Core Facility, Center for Excellence in Brain Science and Intelligence Technology.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Male adult C57BL/6 mice (8 weeks of age) were anesthetized intraperitoneally using pentobarbital sodium (10 mg/mL, 120 mg/kg b.w.)

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

and unilaterally injected with rAAV2-retro-EGFP-Barcode virus (barcode 0, 1, 2, 3, 4) into five projection targets simultaneously.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Coordinates for these injections are as follows.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Reference from Bregma and dura, AI at two locations (in mm: 2.0 AP, 2.52 ML, –2.0 DV; 1.6 AP, 2.97 ML, –2.2 DV) with rAAV2-retro-EGFP-barcode 0 (250 nl and 200 nl, 2.90×10 VG/ml); DMS at one location (in mm: 0.6 AP, 1.8 ML, –2.2 DV, 8 degree angle), with rAAV2-retro-EGFP-barcode 1 (500 nl, 1.00×10 VG/ml); MD at one location (in mm: –1.25 AP, 1.35 ML, –3.55 DV, 20 degree angle), with rAAV2-retro-EGFP-barcode 2 (300 nl, 1.27×10 VG/ml); BLA at one location (in mm: –1.5 AP, 3.2 ML, –4.2 DV), with rAAV2-retro-EGFP-barcode 3 (300 nl, 2.00×10 VG/ml); LH at one location (in mm: –0.94 AP, 1.2 ML, –4.55 DV), with rAAV2-retro-EGFP-barcode 4 (250 nl, 2.25×10 VG/ml).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Following each injection, the micropipette was left in the tissue for 10 min before being slowly withdrawn to prevent virus spilling and backflow.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Mice were sacrificed 6 weeks after virus injection.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Single-cell suspensions were generated as described in methods below.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For dual-color retrograde virus tracing, two regions were ipsilaterally injected with virus at the same time, one with rAAV2-retro-EGFP-barcode 10 (2.00×10 VG/ml) or rAAV2-retro-EGFPnls-barcode 206 or 210 (3.10×10 VG/ml for barcode 206 and 4.38×10 VG/ml for barcode 210) and one with rAAV2-retro-tdTomato (2.25×10 VG/ml).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

rAAV2-retro-EGFPnls was used to avoid dense fiber staining when performing immunohistochemistry.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

We deposited the virus plasmid constructs to Addgene (pAAV-CAG-EGFP barcode-(0–10)-SV40 polyA, pAAV-CAG-EGFPnls barcode-(206, 210)-SV40 polyA; Addgene ID 190864–190876).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For mice without FAC-sorting (mouse #1, #2, #3), three mice that had been injected with virus were anaesthetized and then subjected to transcranial perfusion with ice-cold oxygenated self-made dissection buffer (in mM: 92 Choline chloride, 2.5 KCl, 1.2 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 Glucose, 5 Sodium ascorbate, 2 Thiourea, 3 Sodium pyruvate, 10 MgSO4.7H2O, 0.5 CaCl2.2H2O, 12 N-Acetyl-L-Cysteine).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The brain was removed, 300 µm vibratome sections were collected, and the PrL and IL regions were microdissected under a stereo microscope with a cooled platform.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Brain slices were incubated in dissection buffer with 10 µM AMPA receptor antagonist CNQX (Abcam, ab120017) and 50 µM NMDA receptor antagonist D-AP5 (Abcam, ab120003) at 33 °C for 30 min.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The pieces were dissociated first using the ice-cold oxygenated dissection buffer added papain (20 units/ml, Worthington, LS003126), 0.067 mM 2-mercaptoethanol (Sigma, M6250), 1.1 mM EDTA (Invitrogen, 15575020), 5.5 mM L-Cysteine hydrochloride monohydrate (Sigma, C7880) and 100 units/ml Deoxyribonuclease I (Sigma, D4527), with 30–40 min enzymatic digestion at 37 °C, followed by 30 min 1 mg/ml protease (Sigma, P5147) and 1 mg/ml dispase (Worthington, LS02106) enzymatic digestion at 25 °C.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Supernatant was removed and digestion was terminated using dissection buffer containing 2% fetal bovine serum (FBS, Bioind, 04-002-1A).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Single-cell suspension was generated by manual trituration using fire-polishing Pasteur pipettes and filtered through a 35 µm DM-equilibrated cell strainer (Falcon, 352052).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Cells were then pelleted at 400 × g for 5 min.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The supernatant was carefully removed and resuspended in 1–2 ml dissection buffer containing 2% FBS.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

The suspension was then subjected to the debris removal step using the Debris Removal Solution (Miltenyi, 130-109-398).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Cell pellets were resuspended and 48,000 cells were loaded into 3 lanes to perform 10x Genomics sequencing.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

For mice with FAC-sorting (mouse #4, #5, #6), PrL and IL regions were microdissected and dissociated as mice without FAC-sorting, cells were sorted to enrich for EGFP-positive rAAV2-retro-EGFP-barcodes labeled cells.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

About 4893 EGFP-positive cells were captured and loaded to perform 10x Genomics sequencing.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Chromium Single Cell 3' Reagent Kits (v3) were used for library preparation (10x Genomics).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Libraries were sequenced on an Illumina Novaseq 6000 system.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Parallel PCR reactions were performed containing 50 ng of post cDNA amplification reaction cleanup material as a template.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

P5-Read1 (AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTC) and P7-index-Read2-EGFP (CAAGCAGAAGACGGCATACGAGATAGGATTCGGTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTGgCATGGACGAGCTGTACAAG) (200 nM each) were used as primers with the NEBNext Ultra II Q5 Master Mix (NEB, M0544L).

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Amplification was performed using the following PCR protocol: (1) 33 °C for 1 min, (2) 98° for 10 s, then 65 °C for 75 s (20–24 cycles), (3) 75 °C for 5 min.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

Reactions were re-pooled during 1 X SPRI selection (Beckman, B23317), which harvested virus projection barcodes library.

PMC10914349

High-throughput mapping of single-neuron projection and molecular features by retrograde barcoded labeling.

431–437 bp (with 120 bp adaptors) libraries were sequenced using Illumina HiSeq X Ten.