Add files using upload-large-folder tool

preprint/preprint__97658399979ddce5818a73d2183e854554939b8ade3694c2fef4ee5c223f0f89/images_list.json

@@ -0,0 +1,92 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Fig. 1 The flowchart of EviDTI.",
+    "footnote": [],
+    "bbox": [
+      [
+        163,
+        90,
+        825,
+        660
+      ]
+    ],
+    "page_idx": 9
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Fig 2. The results of two ablation experiments. a Performance comparison on the DrugBank, KIBA and Davis datasets using single-dimensional features and multidimensional feature fusion strategies. b Performance comparison of feature extraction with and without pre-trained models on DrugBank, KIBA, and Davis datasets.",
+    "footnote": [],
+    "bbox": [
+      [
+        177,
+        90,
+        816,
+        460
+      ]
+    ],
+    "page_idx": 15
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Fig. 3 Evidential deep learning provides a favorable measure of uncertainty. a A Mann-Whitney test was performed on the error distribution of uncertainty in samples classified as TP,",
+    "footnote": [],
+    "bbox": [
+      [
+        156,
+        103,
+        835,
+        520
+      ]
+    ],
+    "page_idx": 17
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Fig. 4 Evidential deep learning helps reduce the risk of false predictions in decision-making.",
+    "footnote": [],
+    "bbox": [
+      [
+        155,
+        88,
+        836,
+        308
+      ]
+    ],
+    "page_idx": 22
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Fig. 5 Application of EviDTI in the study of multi-target tyrosine kinase inhibitors. a The",
+    "footnote": [],
+    "bbox": [
+      [
+        171,
+        99,
+        789,
+        562
+      ]
+    ],
+    "page_idx": 25
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Fig. 6 Visualization of attention scores of all the residues in the four randomly selected drug-",
+    "footnote": [],
+    "bbox": [
+      [
+        175,
+        99,
+        848,
+        460
+      ]
+    ],
+    "page_idx": 28
+  }
+]

preprint/preprint__97a3e28d4c90ec91b8f0f9462cfa5ef832e043e9d7b741798d973adb6bc2e4b4/images_list.json

@@ -0,0 +1,212 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Fig. 1. B. thetaiotaomicron PUL transcription reflects target glycan abundance. A. A cartoon depicting prototypical glycan utilization by PUL-encoded gene products. SGBPs (green) specifically bind target glycans in the extracellular milieu, which are transported across the outer membrane by the SusCD complex (blue). Internalized glycans are depolymerized by glycosyl hydrolase (GH) and polysaccharide lyase (PL) activities (purple) that generate ligands that convert PUL sensors (red) to their active form. PUL sensors are deactivated following removal of their glycan ligands from the periplasmin by GH or IM importation activities (orange). B. Ligand-bound sensors direct rapid and dramatic increases in PUL transcription by sequence-specific interactions with PUL promoters. These promoters can be introduced into pBolux to drive glycan-responsive bioluminescence. C. Growth of wild-type Bt (GT23; black) or strains lacking the CS-inducible susC (ΔBT3332; GT2926; blue), 3 CS-specific lyases (ΔBT3324 ΔBT3350 ΔBT4410; GT3086; purple), CS-sensor (ΔBT3334; GT150; red), or a glucuronyl hydrolase (ΔBT3348; VR69; orange) were measured during anaerobic culture in minimal media containing \\(0.1\\%\\) CS as a sole carbon source. Values are the mean of 8 biological replicates, error bars are SEM in color-matched shading. D. The fold increase of the CS-inducible susC (BT3332) mRNA levels were measured by qPCR in either wild-type Bt (GT23; black) or a strain lacking the CS-sensor (GT150; red) following the introduction of CS into the culture media. The fold increase was calculated as the change in transcript levels between cultures before and 2 hours after the introduction of CS. Values are the mean of 6 biological replicates, error bars are SEM. P-values were computed using a two-tailed student's t-test and ** indicates values \\(< 0.01\\) , \\(*< 0.05\\) and \\(ns > 0.05\\) . E. Growth of wild-type Bt (GT23; black) or strains lacking the levan-inducible susC (ΔBT1763; GT3196; blue), 4 levanases (ΔBT1760-1759 ΔBT3082 ΔBT1765; GT3348; purple), fructan sensor (ΔBT1754; GT165; red), or a putative inner membrane fructose importer (ΔBT1758; GT3379; orange) were measured during anaerobic culture in minimal media containing levan as a sole carbon source. Values are the mean of 8 biological replicates, error bars are SEM in color-matched shading. F. The fold increase of levan-inducible susC (BT1763) mRNA levels were measured in either wild-type Bt (GT23; black bars) or a strain lacking the levan sensor (GT165; red bars) following the introduction of \\(0.2\\%\\) levan in the culture media. The fold change was calculated as the change in transcript levels between cultures before and 2 hours after levan introduction. Values are the mean of 6 biological replicates, error bars are SEM. P-values were computed using a two-tailed student's t-test and ** indicates values \\(< 0.01\\) , \\(*< 0.05\\) and \\(ns > 0.05\\) . G. The fold change in BT1763 mRNA levels were measured by qPCR in wild-type Bt (GT23) as described in (D), 120 minutes following the introduction of mixtures of \\(0.2\\%\\) , \\(0.02\\%\\) or \\(0.002\\%\\) levan supplemented with galactose to \\(0.5\\%\\) total carbohydrate. Values are the average of 6 independent measurements, error bars represent SEM, and P-values were calculated by 2-way ANOVA with Tukey's honest significance test and ** represents values \\(< 0.01\\) , \\(*< 0.05\\) , and \\(ns\\) indicates values \\(> 0.05\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        140,
+        52,
+        860,
+        364
+      ]
+    ],
+    "page_idx": 31
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Fig. 2. Construction of a Bacteroides-optimized bioluminescent reporter. A. Schematic depicting the construction of a bioluminescent reporter that encodes the entire PI lux cassette under control of the Bt rpoD promoter and optimized rplL* ribosome binding site (top); a Bacteroides optimized lux cassette with rearranged luxA-E (indicated by the shaded regions) and Bt intergenic regions from a consistently expressed Bt operon (BT1160-1155; middle); or pBolux which has BamHI and SpeI sites positioned upstream of the Bacteroides optimized lux cassette (bottom) in the multi-copy plasmid pLYL01. B. Relative luminescence (solid lines) or growth (dashed lines) from Bt strains harboring an empty vector (GT1846; black) or plasmids containing either the lux operon from P. luminescens (GT3137; blue) or the Bacteroides-optimized lux cassette (GT1541; red) expressed from the Bt rpoD promoter and rplL* RBS were measured during growth in minimal media containing 0.5% galactose as the sole carbon source. C. The relative luminescence (solid lines) or growth (dashed lines) of Bt strains harboring empty pBolux (GT1867; black) or a plasmid with the Bt rpoD promoter cloned into the BamHI and SpeI sites (GT1868; red) during growth in galactose as the sole carbon source. All values in panels B&C are the average of 8 biological replicates and error bars are SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        25,
+        880,
+        175
+      ]
+    ],
+    "page_idx": 32
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Fig. 3. Construction of a glycan-responsive reporter in Bt. A&B. (A) Relative luminescence or (B) growth of wild-type Bt harboring empty pBolux (GT1866; black lines) or a plasmid including the promoter region preceding the CS-inducible susC (P-BT3332; GT1934; pink lines) following introduction of CS (solid lines) or galactose (dashed lines) as the sole carbon source. C. Relative luminescence from strains described in (A) following the introduction of CS normalized by the relative luminescence of identical cultures following the introduction of galactose. For panels A-C, values are the mean of 12 biological replicates and error is SEM in color-matched shading. D. Relative luminescence from wild-type Bt (GT1934, black) or strains lacking a CS-inducible susC (ΔBT3332; GT2939; blue), 3 CS-specific lyases (ΔBT3324 ΔBT3350 ΔBT4410; GT3117; purple), CS-sensor (ΔBT3334; GT2618; red), or a glucuronyl hydrolase (ΔBT3348; GT3102; orange) harboring P-BT3332 following the introduction of an equal mixture of CS and galactose normalized to measurement from identical strains supplied galactose alone. E. Relative luminescence from wild-type Bt (GT1934, black) or strains lacking 3 CS-specific lyases (ΔBT3324 ΔBT3350 ΔBT4410; GT3117; purple) or the CS-sensor (ΔBT3334; GT2618; red) harboring P-BT3332 following the introduction of an equal mixture of unsulfated CS disaccharide (diOS) and galactose normalized to measurements from identical strains supplied galactose alone. F. Relative luminescence from wild-type Bt (GT1934, solid lines) or strains lacking the CS-sensor (GT2618; dashed lines) harboring P-BT3332 following the introduction of a mixture of galactose and either hyaluronic acid (HA, green) or heparan sulfate (HS, blue) and galactose normalized to measurements from identical strains supplied galactose alone. For panels D-F, values are the mean of 8 biological replicates and error is SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        175,
+        46,
+        760,
+        325
+      ]
+    ],
+    "page_idx": 33
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Fig. 4. A levan-responsive reporter reveals multiple levanases coordinate fructan utilization in Bt. A. Relative luminescence from wild-type Bt harboring promoter-less pBolux (black) or a plasmid including the region upstream of the levan inducible susC (P-BT1763, pink) following the introduction of \\(0.5\\%\\) levan as the sole carbon source and normalized by the relative luminescence of identical cultures supplied \\(0.5\\%\\) galactose. Values are the mean of 12 biological replicates, error is SEM in color-matched shading. B&C. Relative luminescence from wild-type Bt (GT1893; black) or strains lacking susC (ΔBT1763; GT3199; blue), 4 levan-specific hydrolases (ΔBT1760-1759 ΔBT3082 ΔBT1765; GT3360; purple), fructan sensor (ΔBT1754; GT2620; red), or a putative inner membrane fructose transporter (ΔBT1758; GT3393; orange) harboring P-BT1763 were measured following the introduction of an equal mixture of galactose and (B) levan or (C) fructose and normalized by the relative luminescence of identical cultures supplied galactose alone. D. Growth of wild-type Bt (GT23; black) or strains lacking the levan-inducible susC (ΔBT1763; GT3196; blue), 4 levan-specific hydrolases (ΔBT1760-1759 ΔBT3082 ΔBT1765; GT3348; purple), fructan sensor (ΔBT1754; GT165; red), or a putative inner membrane fructose transporter (ΔBT1758; GT3379; orange) were measured during anaerobic culture in minimal media containing fructose as a sole carbon source. E. Relative luminescence of wild-type Bt or strains lacking all other levanases except BT1760 (ΔBT1759 ΔBT3082 ΔBT1765; GT3358; pink), BT1759 (ΔBT1760 ΔBT3082 ΔBT1765; GT3356; teal), BT3082 (ΔBT1760-59 ΔBT1765; GT3420; lavender), or BT1765 (ΔBT1760-59 ΔBT3082; GT3345; purple) harboring the levan-responsive reporter following the introduction of a mixture of levan and galactose normalized with measurements from identical cultures supplied galactose alone. F. Growth of wild-type Bt or strains lacking all other levanases except BT1760 (ΔBT1759 ΔBT3082 ΔBT1765; GT3347; pink), BT1759 (ΔBT1760 ΔBT3082 ΔBT1765; GT3346; teal), BT3082 (ΔBT1760-59 ΔBT1765; GT3401; lavender), or BT1765 (ΔBT1760-59 ΔBT3082; GT3308; purple) in levan as a sole carbon source. G. Growth of wild-type Bt or strains lacking either BT1760 (GT3192; pink), BT1759 (GT3246; teal), BT3082 (GT3534; lavender), or BT1765 (GT3282; purple) in levan as a sole carbon source. H. Relative luminescence of wild-type Bt or strains lacking BT1760 (GT3181; pink), BT1759 (GT3226; teal), BT3082 (GT3303; lavender), or BT1765 (GT3299; purple) harboring P-BT1763 following the introduction of a mixture of levan and galactose normalized with measurements from identical cultures supplied galactose alone. For panels B-H, values are an average of 8 biological replicates and error is SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        108,
+        12,
+        856,
+        298
+      ]
+    ],
+    "page_idx": 34
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Fig. 5. Species-specific responses enable PUL reporters to distinguish between compositionally identical yet structurally distinct glycans. A. Growth of wild-type Bt (GT23; gray) and Bo (ATCC 8483; black) or a strain lacking the Bo inulin sensor (ΔBACOVA_04496; GT3183; red) were measured during anaerobic culture in minimal media containing inulin as a sole carbon source. B. Relative luminescence from wild-type Bo harboring empty pBolux (black) or a plasmid that includes the region preceding the inulin-inducible susC (P-BACOVA_04505; pink) were measured following the introduction of 0.5% inulin and normalized by the relative luminescence of identical cultures supplied 0.5% galactose. Values are the mean of 12 biological replicates, error is SEM in color-matched shading. C-E. Relative luminescence from wild-type Bo harboring P-BACOVA_04505 (GT3490; black) or an isogenic strain lacking the Bo inulin sensor (ΔBACOVA_04496; GT3189, red) were measured following the introduction of an equal mixture of galactose and (C) inulin (D) fructose, or (E) levan, normalized by the relative luminescence of identical cultures supplied galactose alone. F. Relative luminescence from wild-type Bt (GT1893; black) or strains lacking the levan-inducible susC (ΔBT1763; GT3199; blue), 4 levan-specific hydrolases (ΔBT1760-1759 ΔBT3082 ΔBT1765; GT3360; purple) or the Bt fructan sensor (ΔBT1754; GT2620; red), harboring P-BT1763 were measured following the introduction of an equal mixture of galactose and inulin normalized by the relative luminescence of identical cultures supplied galactose alone. For panels A and C-F, values are the mean of 8 biological replicates and error is SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        198,
+        15,
+        770,
+        297
+      ]
+    ],
+    "page_idx": 35
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Fig. 6. PUL-reporters reflect dose-dependent transcription. A. Relative luminescence from a wild-type Bt strain harboring P-BT3332 (GT1934) following the introduction of 2-fold serial dilutions of \\(0.4\\%\\) CS containing galactose to a final carbohydrate content of \\(0.5\\%\\) and normalized to identical cultures supplied galactose alone. Values are the mean of 12 biological replicates, error is SEM in color-matched shading. B. The log10 AUC of response curves measured from wild-type Bt (GT1934; black, see panel A) or a mutant lacking the glucuronyl hydrolase \\((\\Delta gh;\\) GT3102; orange, see panel F) harboring P-BT3332 supplied 2x dilutions of CS to each strain and normalized by identical cultures supplied galactose alone. C. The log10 AUC of response curves measured from wild-type Bt (GT1893; black; see Fig. 5SA) or a mutant lacking a putative inner membrane transporter \\((\\Omega\\) BT1758; GT3393; orange; see Fig. S6F) harboring P-BT1763 supplied 2x dilutions of levan to each strain and normalized by identical cultures supplied galactose alone. For panels B&C, values are the mean of 12 biological replicates and error bars are standard deviation. P-values were computed by 2-way ANOVA with Dunnett correction and \\\\*\\\\* indicates values \\(< 0.001\\) \\(^{**}< 0.01\\) \\(^{*}< 0.05\\) , and \\(ns > 0.05\\) . D. The AUC of response curves measured from wild-type Bt strains harboring either P-BT3332 (GT1934; open blue squares) supplied mixtures containing 2-fold serial dilutions of levan or P-BT1763 (GT1893, open red circles) supplied 2-fold serial dilutions dilutions of CS with galactose to \\(0.5\\%\\) total carbohydrate and normalized by identical cultures supplied galactose alone. E. The AUC of response curves measured from wild-type Bt harboring P-BT3332 (GT1934; filled blue squares) supplied a mixture containing 2-fold serial dilutions of levan, constant \\(0.2\\%\\) CS and the balance galactose totaling \\(0.5\\%\\) carbohydrate and normalized by identical cultures supplied galactose alone. The AUC of responses from wild-type Bt harboring a levan-responsive reporter (GT1893; filled red circles) supplied a mixture containing 2-fold serial dilutions of CS, constant \\(0.2\\%\\) levan and the balance galactose totaling \\(0.5\\%\\) carbohydrate normalized by identical cultures supplied galactose alone. For panels D&E, values are the average of 6 biological replicates and error bars represent standard deviation. P-values were computed using 2-way ANOVA with Tukey's honest significance test and \\(^{**}\\) represents values \\(< 0.01\\) and ns indicates values \\(> 0.05\\) . F. Relative luminescence from a BT3348-deficient Bt strain harboring P-BT3332 (GT3102) following the introduction of 2-fold serial dilutions of \\(0.4\\%\\) CS containing galactose to a final carbohydrate content of \\(0.5\\%\\) and normalized to identical cultures supplied galactose alone. Values are the average of 12 biological replicates.",
+    "footnote": [],
+    "bbox": [
+      [
+        198,
+        18,
+        794,
+        295
+      ]
+    ],
+    "page_idx": 36
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_7.jpg",
+    "caption": "Fig. 7. PUL-reporters demonstrate SGBP-mediated target glycan isolation. A-C. Purified BT3330, BSA, or BT1761 was analyzed by affinity-PAGE in the presence of (A) \\(0.1\\%\\) CS, (B) no glycan or (C) \\(0.1\\%\\) levan. Images are representative of 3 independent experiments. D&E. The fold difference in AUC from wild-type Bt strains harboring either P-BT1763 (GT1893, open bars) or P-BT3332 (filled bars, GT1934) supplied fractions eluted from immobilized (D) BT3330 or (E) BT1761 in combination with \\(0.4\\%\\) galactose and normalized against cultures supplied galactose alone. Normalized AUC values were divided by responses from identical strains supplied equivalent fractions eluted from columns supplied control lysates prepared from cells harboring an empty plasmid (pT7-7; see Fig. S7A&B). Values are the average of 4 biological replicates and error bars are SEM. P-values were computed two-tailed Student's t-test and *** indicate values \\(< 0.001\\) , \\(**< 0.01\\) , \\(*< 0.05\\) , and ns \\(> 0.05\\) . F&G. The AUC from wild-type Bt strains harboring either (F) P-BT3332 (GT1934) or (G) P-BT1763 supplied pooled, concentrated elutions fractions in combination with \\(0.4\\%\\) galactose and normalized to identical cultures supplied galactose alone. Measurements were collected alongside identical strains supplied 2-fold serial dilutions of either (F) CS or (G) levan with galactose to a total carbohydrate content of \\(0.5\\%\\) and normalized to identical cultures supplied galactose alone. Linear regression models (gray line) were computed in Prism, and sample concentrations (red) were estimated with the derived equations. Values are the average of 4 technical replicates, error bars are standard deviation, and the dashed gray lines are the \\(95\\%\\) confidence interval computed for each linear regression model in Prism. H&I. The estimated (H) total glycosaminoglycan or (I) fructan content of samples described for panels F and G, respectively, using colorimetric glycan assay kits. Values are the average of 3 technical replicates, error bars are standard deviation, and the dashed gray lines are the \\(95\\%\\) confidence interval computed for each linear regression model in Prism.",
+    "footnote": [],
+    "bbox": [
+      [
+        95,
+        10,
+        870,
+        277
+      ]
+    ],
+    "page_idx": 37
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "Fig. S1. PULs are required for glycan-specific growth conditions and transcriptional responses. A. Growth of wild-type Bt (GT23; black) or strains lacking a CS-inducible susC (ΔBT3332; GT2926; blue), 3 CS-specific lyases (ΔBT3324 ΔBT3350 ΔBT4410; GT3086; purple), CS-sensor (ΔBT3334; GT150; red), or a glucuronyl hydrolase (ΔBT3348; VR69; orange) were measured during anaerobic culture in minimal media containing galactose as a sole carbon source. B. Growth of wild-type Bt (GT23; black) or strains lacking a levan-inducible susC (ΔBT1763; GT3196; blue), 4 levanases (ΔBT1760-1759 ΔBT3082 ΔBT1765; GT3348; purple), fructan-sensor (ΔBT1754; GT165; red), or a putative inner membrane fructose importer (ΔBT1758; GT3379; orange) were measured during anaerobic culture in minimal media containing galactose as a sole carbon source. For panels A&B, values are the mean of 8 biological replicates and error bars are SEM in color-matched shading. C-E. The fold increase of (C&D) BT3332 or (E) BT1763 mRNA levels in wild-type Bt following the introduction of mixtures containing either 0.2%, 0.02%, or 0.02% (C&D) CS or (E) levan supplemented with galactose to 0.5% total carbohydrate. The fold increase was calculated as the change in transcript levels between cultures before and after 2 hours (C) or 1 hour (D&E) following induction of glycan mixtures. Values are the average of 6 independent measurements, error bars represent SEM, and P-values were calculated by 2-way ANOVA with Tukey's honest significance test and *** represents values < 0.001, * < 0.05, and ns indicates values > 0.05.",
+    "footnote": [],
+    "bbox": [
+      [
+        45,
+        35,
+        955,
+        190
+      ]
+    ],
+    "page_idx": 38
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_1.jpg",
+    "caption": "Fig. S2. Bioluminescence during anaerobic growth across Bacteroides species and growth conditions. A. Relative luminescence (solid lines) or growth (dashed lines) from Bt strains harboring an empty vector (GT1866; black) or plasmids containing either the lux operon from P. luminescens (GT3137; blue) or the Bacteroides-optimized lux cassette (GT1541; red) expressed from the Bt rpoD promoter and rplL\\* RBS was measured during growth in minimal media containing \\(0.5\\%\\) glucose. B. The relative luminescence of Bt strains harboring empty pBolux (GT1867; dashed lines) or a plasmid with the corresponding rpoD promoter cloned into the BamHI and SpeI sites (GT1868; solid lines) during growth in glucose (black), fructose (red), arabinose (purple), or xylose (green) as the sole carbon source. C. The relative luminescence (solid lines) or growth (dashed lines) of Bo strains harboring empty pBolux (GT3489; black) or a plasmid with the Bo rpoD promoter cloned into the BamHI and SpeI sites (GT3489; red) during growth in galactose as the sole carbon source. D. The relative luminescence of Bo strains harboring empty pBolux (GT3489; dashed lines) or a plasmid with the corresponding rpoD promoter cloned into the BamHI and SpeI sites (GT3490; solid lines) during growth in glucose (black), fructose (red), arabinose (purple), or xylose (green) as the sole carbon source. For all panels, values are the mean of 8 biological replicates and error is SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        46,
+        42,
+        914,
+        191
+      ]
+    ],
+    "page_idx": 39
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_2.jpg",
+    "caption": "Fig. S3. The BT3332 promoter confers CS and HA-inducible bioluminescence in pBolux. A&B. Relative luminescence from wild-type Bt (GT1934, black lines) or strains lacking the CS-inducible susC (ΔBT3332; GT2939; blue), 3 CS-specific lyases (Δ BT3324 ΔBT3350 ΔBT4410; GT3117; purple), CS-sensor (ΔBT3334; GT2618; red), or a glucuronyl hydrolase (ΔBT3348; GT3102; orange) harboring pBolux including the promoter region preceding BT3332 (P-BT3332) following the introduction of an equal mixture of (A) CS and galactose or (B) galactose alone. C. Relative luminescence from wild-type Bt (GT1934) harboring P-BT3332 following the introduction of galactose alone (black lines) or an equal mixture of galactose and HA (green lines) or HS (blue lines). D&E. Growth of wild-type Bt (GT23; black) or a strain lacking the CS-sensor (ΔBT3334; GT150; red) were measured during anaerobic culture in minimal media containing (D) HS or (E) HA as a sole carbon source. F. Relative luminescence from a CS-sensor deficient strain (ΔBT3334; GT2618) harboring P-BT3332 following the introduction of galactose alone (black lines) or an equal mixture of galactose and HA (green lines) or HS (blue lines). For all panels, values are the mean of 8 biological replicates, error bars are SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        210,
+        58,
+        764,
+        336
+      ]
+    ],
+    "page_idx": 40
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_3.jpg",
+    "caption": "Fig. S4. A fructan-responsive reporter reveal new insights into Bt levan utilization. A. Relative luminescence from wild-type Bt harboring empty pBolux (black) or a plasmid including the region upstream of the levan-inducible susC (P-BT1763, pink) following the introduction of \\(0.5\\%\\) levan (solid lines) or \\(0.5\\%\\) galactose (dashed lines). Values are the mean of 12 biological replicates, error is SEM in color-matched shading. B&C. Growth of wild-type Bt harboring empty pBolux (black) or P-BT1763 (pink) following the introduction of (B) \\(0.5\\%\\) galactose or (C) \\(0.5\\%\\) levan. For panels A-C, values are the mean of 12 biological replicates, error is SEM in color-matched shading. D&F. Growth of wild-type Bt or strains lacking all other levanases except BT1760 (BT1759 & BT3082, BT1765; GT3347; pink), BT1759 (BT1760, BT3082, BT1765; GT3346; teal), BT3082 (BT1760-59, BT1765; GT3401; lavender), or BT1765 (BT1760-59, BT3082; GT3308; purple) in \\(0.1\\%\\) (D) fructose or (F) galactose as a sole carbon source. E&G. Growth of wild-type Bt (GT23, black) or strains lacking either BT1760 (GT3181; pink), BT1759 (GT3226; teal), BT3082 (GT3303; lavender), or BT1765 (GT3282; purple) in \\(0.1\\%\\) (E) fructose or (G) galactose as a sole carbon source. H. Growth of wild-type Bt (GT2111, black) or BT1760-deficient strains (GT3215; pink) harboring empty pNBU2 or a plasmid encoding BT1760 (GT3216, green) in \\(0.1\\%\\) levan as the sole carbon source.",
+    "footnote": [],
+    "bbox": [
+      [
+        95,
+        24,
+        857,
+        310
+      ]
+    ],
+    "page_idx": 41
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_4.jpg",
+    "caption": "Fig. S5. A fructan-responsive PUL in Bo exhibits inulin-inducible activity. A-C. Growth of wild-type Bo (ATCC 8483; black) or a strain lacking the Bo inulin sensor (ΔBACOVA_04496; GT3183; red) were measured during anaerobic culture in minimal media containing (A) levan, (B) fructose or (C) galactose as the sole carbon source. Values are the mean of 8 biological replicates and error bars are SEM in color-matched shading. D. Relative luminescence from wild-type Bo harboring empty pBolux (black) or a plasmid that includes the region preceding an inulin-inducible susC (P-BACOVA_04505; pink) were measured following the introduction of \\(0.5\\%\\) inulin (solid lines) or \\(0.5\\%\\) galactose (dashed lines). Values are the mean of 12 biological replicates and error is SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        110,
+        42,
+        875,
+        188
+      ]
+    ],
+    "page_idx": 42
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_5.jpg",
+    "caption": "Fig. S6. PUL reporters display concentration dependent responses to target glycans. A. Relative luminescence from a wild-type Bt strain harboring a P-BT1763 (GT1893) following the introduction of mixtures containing 2-fold serial dilutions of \\(0.4\\%\\) levan balanced with galactose to a final carbohydrate content of \\(0.5\\%\\) and normalized to identical cultures supplied galactose alone. Values are the mean of 12 biological replicates and error bars are SEM in color-matched shading. B&C. The log10 AUC responses within the linear range from wild-type Bt (black; B: GT1934; C: GT1893) or mutants defective for PUL-sensor deactivation (orange; B:GT3102; C: GT3393) harboring (B) P-BT3332 or (C) P-BT1763 and supplied 2x serial dilutions of \\(0.4\\%\\) (B) CS or (C) levan balanced with galactose to \\(0.5\\%\\) total carbohydrate content and normalized to responses from identical cultures supplied galactose alone. Values are the mean of 12 biological replicates and error is standard deviation. Solid lines represent the simple linear regression models corresponding to responses from each strain and color-matched dashed lines represents the \\(95\\%\\) confidence intervals computed in Prism. D&E. The fold difference between the AUC of responses from wild-type Bt strains harboring either P-BT3332 (blue squares) or P-BT1763 (red circles) supplied glycan mixtures containing 2-fold serial dilutions of \\(0.2\\%\\) CS or levan, respectively, in the presence or absence of constant \\(0.2\\%\\) levan or CS, respectively, and balanced with galactose to \\(0.5\\%\\) total carbohydrate normalized by the AUC of responses from identical cultures supplied galactose alone. E. The AUC of responses from wild-type Bt strains harboring either P-BT3332 (GT1934; blue squares) or P-BT1763 (GT1893; red circles) supplied glycan mixtures containing 2-fold serial dilutions of \\(0.2\\%\\) CS or levan, respectively, in the presence of constant \\(0.2\\%\\) levan or CS, respectively, and balanced with galactose to \\(0.5\\%\\) total carbohydrate normalized by the AUC of responses from identical cultures supplied galactose alone. Values represent the average of 6 biological replicates and error bars are standard deviation. P-values were calculated with 2-way ANOVA with Tukey's honest significance test and *** represents values \\(< 0.001\\) , \\(**< 0.01\\) , \\(*< 0.05\\) , and ns \\(> 0.05\\) . F. Relative luminescence from a BT1758-deficient Bt strain harboring P-BT1763 (GT3393) following the introduction of 2-fold serial dilutions of \\(0.4\\%\\) levan containing galactose to a final carbohydrate content of \\(0.5\\%\\) and normalized to identical cultures supplied galactose alone. Values are the mean of 12 biological replicates and error bars are SEM in color-matched shading.",
+    "footnote": [],
+    "bbox": [
+      [
+        186,
+        54,
+        805,
+        338
+      ]
+    ],
+    "page_idx": 43
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_6.jpg",
+    "caption": "Fig. S7. Glycan-responsive reporter strains can indicate target glycan abundance. A&B. The AUC of responses from wild-type BT strains harboring either a (A) P-BT1332 (GT1934; open bars) or (B) P-BT1763 (GT1893, filled bars) supplied elution fractions from nickel-NTA agarose incubated with E. coli whole cell lysates from strains containing empty pT7-7 vector, or plasmids engineered to over-express BT1761 or BT3330 and pre-incubated with a mixture of \\(0.1\\%\\) of both levan and CS. All elutions fractions were supplemented with \\(0.4\\%\\) galactose. Values represent an average of 4 total replicates from two independent experiments and error bars are SEM. C. Coomassie stained SDS-PAGE gels showing the corresponding protein levels for BT3330 (top 2 gels) or BT1761 (bottom 2 gels) in each elution fraction. D&E. The AUC of responses from wild-type BT strains harboring either (D) P-BT3332 (GT1934; open bars) or (E) P-BT1763 (GT1893, filled bars) supplied galactose alone or concentrated material co-purifying with BT3330 or BT1761 supplemented with \\(0.4\\%\\) galactose. Values are the average of 4 measurements from 2 independent experiments, error is standard deviation and P-values were computed using 1-way ANOVA with Tukey's honest significance test and *** represents values \\(< 0.001\\) and ns indicates values \\(> 0.05\\) . F&G. The AUC of responses from deactivation defective BT strains harboring either (F) P-BT3332 (GT3102; open bars) or (G) P-BT1763 (GT3393, filled bars) supplied galactose alone or concentrated material co-purifying with BT3330 or BT1761 supplemented with \\(0.4\\%\\) galactose. For panels D-G, values are the average of 4 measurements from 2 independent experiments, error is standard deviation and P-values were computed using 1-way ANOVA with Tukey's honest significance test and *** represents values \\(< 0.001\\) , \\(< 0.05\\) , and ns indicates values \\(> 0.05\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        60,
+        25,
+        740,
+        305
+      ]
+    ],
+    "page_idx": 44
+  }
+]

preprint/preprint__97ae97305ce10e362d0efa94157154334e4c670f53e025014566da79d85c660f/images_list.json

@@ -0,0 +1,122 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "Scheme 1. Schematic diagram of the preparation of Cu/CN and diabetic wound healing mechanism of Cu/CN.",
+    "footnote": [],
+    "bbox": [
+      [
+        115,
+        90,
+        875,
+        633
+      ]
+    ],
+    "page_idx": 6
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1. Composition characterization of CN. XRD patterns (a), C 1s XPS spectra (b), N 1s XPS spectra (c), UV-vis spectra (d), EPR spectra (e) and FT-IR spectra (f) of CN with different nitrogen vacancy concentrations.",
+    "footnote": [],
+    "bbox": [
+      [
+        115,
+        153,
+        884,
+        494
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2. Photoelectrochemical behaviours of modified CN. (a) Photocatalytic properties of different catalysts for glucose consumption. (b) Transient photocurrent responses. (c) Electrochemical impedance spectra (EIS). In the simulated electrical equivalent-circuit model (inset), RS, R1, and CPE represent as solution resistance, charge transfer resistance, and double layer capacitance, respectively. (d) The plots of \\((\\mathrm{Ahv})^2\\) and \\(hv\\) of different catalysts. (e) Mott-Schottky curves of different catalysts. (f) Schematic illustration of band structure for the different catalysts.",
+    "footnote": [],
+    "bbox": [
+      [
+        114,
+        232,
+        884,
+        574
+      ]
+    ],
+    "page_idx": 10
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3. Structural characterizations of Cu/CN. (a) UV-vis spectra of Cu/CN. (b) FT-IR spectra of Cu/CN. (c) XRD patterns of Cu/CN. (d) Cu 2p XPS spectra of Cu/CN. (e) XANES spectra at Cu K-edge of Cu/CN. (f) Standard curve of Cu valence state. (g) EXAFS spectra at Cu K-edge of Cu/CN. (h) EXAFS fitting result of Cu/CN at R space. (i) Wavelet transform of Cu foil, \\(\\mathrm{Cu}_2\\mathrm{O}\\) , \\(\\mathrm{CuO}\\) and \\(\\mathrm{Cu/CN}\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        398,
+        884,
+        770
+      ]
+    ],
+    "page_idx": 12
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4. DFT of Cu/CN. (a) Adsorption energy of glucose at non vacancy and nitrogen vacancy sites on CN surface. (b) Adsorption energy of \\(\\mathrm{H}_2\\mathrm{O}_2\\) at nitrogen vacancy, non-vacancy, and Cu single-atom sites on Cu/CN surface.",
+    "footnote": [],
+    "bbox": [
+      [
+        120,
+        88,
+        875,
+        505
+      ]
+    ],
+    "page_idx": 14
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5. Evaluation of antibacterial capacity. Photographs of bacterial colonies formed by (a) ESBL E. coli and (c) MRSA after treatments with Cu/CN. The relative bacterial viability of (b) ESBL E. coli and (d) MRSA. Morphologies of (e) ESBL E. coli and MRSA treated or untreated with the Cu/CN group. (f) CLSM photos of biofilm treated with Cu/CN. (g) Schematic diagram of photocatalytic antibacterial control by photoswitch.",
+    "footnote": [],
+    "bbox": [
+      [
+        120,
+        88,
+        848,
+        710
+      ]
+    ],
+    "page_idx": 16
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Figure 6. In vitro biocompatible evaluation of \\(\\mathrm{Cu / CN}\\) . (a) Fluorescence images of 3T3 fibroblast cells cultured with hydrogel groups after 24 hours. (b) Cell viability of the \\(\\mathrm{CN}_{700}\\) and \\(\\mathrm{Cu / CN}\\) . (c) Hemolysis ratio of the \\(\\mathrm{CN}_{700}\\) and \\(\\mathrm{Cu / CN}\\) . (d) Cell migration of HUVEC cells in the control and the different groups at 0, 24, and 48 hours. (e) The cellular migration after various treatments for different time periods was identified by drawing the green shadow at the edge of cells in (d), and calculate the area of cell migration (\\*p<0.5; \\*\\*p<0.01; \\*\\*\\*p<0.001; \\*\\*\\*\\*p<0.0001).",
+    "footnote": [],
+    "bbox": [
+      [
+        128,
+        81,
+        870,
+        562
+      ]
+    ],
+    "page_idx": 18
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_7.jpg",
+    "caption": "Figure 7. Evaluation of wound healing in diabetic mouse models. (a). Photographs of the wounds of wound closure treat with light \\((\\lambda >420 \\mathrm{nm})\\) . (b) Wound size ratios with different treatments at days 0, 3, 7, and 14. (c) Typical agar plate photos for the remaining MRSA colonies in the ulcer after different treatment on day 14. (d) Immunohistochemical studies of diabetic wound collected at day 14. Scale bar: \\(1000 \\mu \\mathrm{m}\\) for H&E and Masson, \\(100 \\mu \\mathrm{m}\\) for TNF- \\(\\alpha\\) , IL-10 and CD31. (e) Overall performance of control, GOx, CN, and Cu/CN in the in vivo diabetic wound treatment. (f) Schematic diagram of photocatalytic therapy control by photoswitch.",
+    "footnote": [],
+    "bbox": [
+      [
+        135,
+        80,
+        860,
+        731
+      ]
+    ],
+    "page_idx": 21
+  }
+]

preprint/preprint__97ae97305ce10e362d0efa94157154334e4c670f53e025014566da79d85c660f/preprint__97ae97305ce10e362d0efa94157154334e4c670f53e025014566da79d85c660f.mmd

@@ -0,0 +1,332 @@
+
+# Cu Single-Atom Embedded g-C3N4 nanosheets Rehabilitate Multidrug-Resistant Bacteria Infected Diabetic Wounds via Photoswitchable Cascade Reaction  
+
+Peng Li  
+
+iampli@nwpu.edu.cn  
+
+Northwestern Polytechnical University https://orcid.org/0000- 0002- 5876- 2177  
+
+Xichen Sun Northwestern Polytechnical University  
+
+Pengqi Zhu Shanxi Bethune Hospital  
+
+Liuyan Tang Northwestern Polytechnical University  
+
+Pengfei Wang Northwestern Polytechnical University  
+
+Ningning Li Northwestern Polytechnical University  
+
+Qing Wang Northwestern Polytechnical University  
+
+Yan-Ru Lou University of Helsinki https://orcid.org/0000- 0001- 7717- 6010  
+
+Yuezhou Zhang Northwestern Polytechnical University https://orcid.org/0000- 0002- 8560- 5716  
+
+## Article  
+
+Keywords: nitrogen vacancy, graphitic carbon nitride, photocatalytic antibacterial, chronic wound infections, antibiofilm  
+
+Posted Date: February 19th, 2025  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 5916168/v1
+
+<--- Page Split --->
+
+License: © ① This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: There is NO Competing Interest.  
+
+Version of Record: A version of this preprint was published at Nature Communications on October 16th, 2025. See the published version at https://doi.org/10.1038/s41467-025-64242- z.
+
+<--- Page Split --->
+
+## Cu Single-Atom Embedded g- \(\mathbf{C}_3\mathbf{N}_4\) nanosheets Rehabilitate Multidrug-Resistant Bacteria Infected Diabetic Wounds via Photoswitchable Cascade Reaction  
+
+Xichen Sun \(^{1,2\dagger}\) , Pengqi Zhu \(^{3,4\dagger}\) , Liuyan Tang \(^{1,2}\) , Pengfei Wang \(^{1,2}\) , Ningning Li \(^{1,2}\) , Qing Wang \(^{1,2}\) , Yan- Ru Lou \(^{5}\) , Yuezhou Zhang \(^{1,2*}\) , and Peng Li \(^{1,2,6*}\)  
+
+\(^{1}\) State Key Laboratory of Flexible Electronics (LOFE) & Institute of Flexible Electronics (IFE), Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 710072, China \(^{2}\) Ningbo Institute of Northwestern Polytechnical University, Frontiers Science Center for Flexible Electronics (FSCFE), Key laboratory of Flexible Electronics of Zhejiang Province, Ningbo Institute of Northwestern Polytechnical University, 218 Qingyi Road, Ningbo, 315103, China  
+
+\(^{3}\) Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, 99 Longcheng Street, Taiyuan, 030032, China \(^{4}\) Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430030, China  
+
+\(^{5}\) Faculty of Pharmacy, University of Helsinki, P.O. Box 56, FI- 00014 Helsinki, Finland \(^{6}\) School of Flexible Electronics (SoFE) and Henan Institute of Flexible Electronics (HIFE), Henan University, 379 Mingli Road, Zhengzhou, 450046, China \(^{\dagger}\) These authors contributed equally to this work.  
+
+E- mail: iamyzzhang@nwpu.edu.cn; iampli@nwpu.edu.cn  
+
+Keywords: nitrogen vacancy, graphitic carbon nitride, photocatalytic antibacterial, chronic wound infections, antibiofilm
+
+<--- Page Split --->
+
+To tackle elevated blood glucose, multidrug- resistant (MDR) bacterial infections, and persistent inflammation in diabetic wounds, we present a therapeutic strategy that employs a photoswitch- controlled catalytic cascade reaction, utilizing a photocatalytic material engineered through the synergistic regulation of nitrogen vacancies and single- atom embedding. The nitrogen vacancies in \(\mathrm{g - C_3N_4}\) promise the photocatalytic glucose oxidation to \(\mathrm{H_2O_2}\) and facilitate its subsequent conversion into hydroxyl radicals (•OH) through a photocatalytic cascade reaction with Cu single- atom embedded \(\mathrm{g - C_3N_4}\) nanosheets (Cu/CN). Concurrently, the •OH and superoxide anions (• \(\mathrm{O_2}^-\) ) are obtained by photocatalytic water splitting over Cu/CN. Over \(99.9\%\) antibacterial activity and effective biofilm inhibition are achieved via photocatalytic cascade reaction. In the dark, excess ROS are scavenged by Cu/CN, reducing inflammation of wounds and promoting polarization of M2 macrophages. This photoswitchable cascade reaction effectively treated MDR bacterial- infected diabetic wounds, highlighting its potential for antibiotic- free diabetic wound therapy and its promising prospects for clinical applications.  
+
+## 1. Introduction  
+
+Diabetes, along with its complications- particularly non- healing chronic wounds prone to bacterial infections- has emerged as a critical threat to public health, imposing a heavy burden on patients' life quality and social economy. \(^{1, 2}\) The diabetic wounds are characterized by excessive presence of glucose, persistent bacterial colonization, chronic inflammation, increased levels of oxidative stress, and abnormal angiogenesis, etc. \(^{3, 4}\) The rise of bacterial resistance has diminished the efficacy of conventional antibiotics, making bacterial infected diabetic wounds even harder to heal. \(^{5, 6}\) Staphylococcus aureus (Gram- positive) and Escherichia coli (Gram- negative) are the most frequent bacteria isolated from diabetic wounds, the fast emergence and spread of their multidrug- resistant (MDR) strains, such as methicillin- resistant S. aureus (MRSA) and extended- spectrum \(\beta\) - lactamases producing E. coli (ESBL E. coli), have intensified the demand for new antibacterial strategies. \(^{7}\) For example, a case study revealed that MRSA accounted for more than half of S. aureus isolated from diabetic foot ulcers. \(^{8}\) Novel strategies capable of effectively combating MDR bacteria, reducing glucose levels, controlling oxidative stress,
+
+<--- Page Split --->
+
+regulating inflammation, and promoting angiogenesis are highly desired for the treatment of diabetic wounds.  
+
+Glucose oxidase (GOx) has been employed to treat diabetic complications by catalyzing the oxidative depletion of glucose- an essential nutrient for bacterial growth- and generating \(\mathrm{H}_2\mathrm{O}_2\) , which adversely affects bacterial viability. \(^{9,10,11}\) Recently, a number of glucose oxidase- based platforms that effectively promoting diabetic wound healing have been developed. \(^{12,13,14,15,16}\) In addition, nanozymes with the similar activity and wider applicability have been developed to overcome some limitations of natural enzymes. For example, metal- based nanozymes (MnO \(_2\) , Au, TiO \(_2\) , Cu \(_2\) O, and CeO \(_2\) , etc.) have been constructed to mimic GOx activity. \(^{17,18,19}\) However, a challenge with these nanozymes is the lack of precise control over their reaction process, which typically results in non- stop catalysis and the production of excessive ROS at the wound site, thus exacerbating the inflammation and disturbing the wound healing process. \(^{20,21}\) A potential solution strategy involves the use of switchable processing methods that can regulate enzyme activity under specific stimuli. Among various stimulation methods, light stands out due to its spatiotemporal selectivity, enabling precise regulation of the activity of photocatalysis, thus playing a key role in the controllable treatments. For example, He and coworkers used visible light as a switch to activate photocatalytic titanium dioxide, which consumed glucose at the wound site to produce hydrogen and promote wound healing. \(^{22}\)  
+
+Graphitic carbon nitrides (g- C \(_3\) N \(_4\) ) have demonstrated exceptional and controllable photocatalytic performance, making them widely used in energy, environmental, and biomedical fields. \(^{23,24,25}\) g- C \(_3\) N \(_4\) exhibits the ability to photocatalytically oxidize glucose to generate \(\mathrm{H}_2\mathrm{O}_2\) , and has been employed as a GOx- mimicking photocatalyst for glucose detection. \(^{26,27}\) Defect engineering techniques \(^{28,29,30,31}\) such as carbon and nitrogen vacancies \(^{32,33,34,35}\) , as well as oxygen doping \(^{36,37}\) in g- C \(_3\) N \(_4\) , can effectively improve its photocatalytic performance by altering the electronic band structure \(^{38}\) , optimizing carrier transfer \(^{39}\) , and enhancing surface active sites \(^{40}\) . g- C \(_3\) N \(_4\) typically exhibits a low adsorption rate for \(\mathrm{H}_2\mathrm{O}_2\) and is unable to further consume the \(\mathrm{H}_2\mathrm{O}_2\) produced in the previous step through cascade reaction. By loading metal single- atoms such as Ru \(^{41}\) , Pt \(^{42}\) , and Cu \(^{43}\) onto g- C \(_3\) N \(_4\) , the resulting single- atom embedded g- C \(_3\) N \(_4\) demonstrates enhanced photocatalytic cascade reaction performance,
+
+<--- Page Split --->
+
+further consuming \(\mathrm{H}_2\mathrm{O}_2\) to produce ROS that are more effective at killing bacteria. However, few studies have been conducted that combine defect engineering with single- atom embedding to modulate the photocatalytic performance of \(\mathrm{g - C_3N_4}\) for the enhanced treatment of diabetic infected wounds.  
+
+Herein, we present a Cu single- atom embedded nitrogen vacancy- rich \(\mathrm{g - C_3N_4}\) photocatalyst (Cu/CN), which utilizes a cascade reaction controlled via photoswitchable treatment methods to effectively regulate glucose levels and antibacterial activity while preventing excessive ROS, thereby minimizes the risk of overtreatment and reduces inflammation at the wound site, ultimately enhancing chronic wound healing (Scheme 1). The nitrogen vacancies enhance photocatalytic glucose oxidation, effectively reducing blood glucose at the wound site. Meanwhile, the \(\bullet \mathrm{OH}\) produced from the photocatalytic decomposition of \(\mathrm{H}_2\mathrm{O}_2\) in the preceding step, combined with the \(\bullet \mathrm{OH}\) and \(\bullet \mathrm{O}_2^-\) produced through Cu single- atom- mediated photocatalytic water splitting, effectively eliminate MDR bacteria. In addition, Cu/CN scavenge excess ROS in dark reaction to alleviate inflammation at the wound site, thereby promoting the transition of macrophages from pro- inflammatory M1 phenotype to anti- inflammatory M2 phenotype. This photoswitchable cascade reaction of \(\mathrm{Cu / CN}\) thus represents a promising and innovative approach for treatment of chronic diabetic wounds.
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+<center>Scheme 1. Schematic diagram of the preparation of Cu/CN and diabetic wound healing mechanism of Cu/CN. </center>  
+
+## 2. Results and discussion  
+
+### 2.1 Morphology, size, and composition characterization of CN  
+
+As shown in the scheme 1, CN with graphite- phase structure was prepared by thermal polymerization using melamine as the precursor, and the above materials were improved. To introduce nitrogen vacancies, the material was calcined at different temperatures, decomposing some CN under air atmosphere. \(^{44}\) The SEM images (Figure S1) and TEM images (Figure S2) reveal that the surface of CN becomes increasingly wrinkled and
+
+<--- Page Split --->
+
+porous as the calcination temperature rise. Nitrogen adsorption experiments further confirm that higher calcination temperatures increase in surface area and pore size, potentially enhancing the material's capability to adsorb the reaction substrate (Figure S3 and Table S1). XRD results (Figure 1a) indicate that with increasing heat treatment temperature, the peak intensity of the material decreased at \(13^{\circ}\) (100) crystal plane (stacking of in-plane structural units) and \(27^{\circ}\) (002) crystal plane (stacking of interlayer structures), indicating a reduction in the crystallinity of carbon nitride. The XPS was employed to examine the chemical states of the samples, using the binding energy of the C1s line at \(284.8 \mathrm{eV}\) from alkyl or adventitious carbon as a reference. A comparison of the C 1s and N 1s spectra of \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) and \(\mathrm{CN}_{750}\) is shown in Figure 1bc. Both \(\mathrm{CN}_{550}\) and \(\mathrm{CN}_{700}\) exhibited characteristic binding configurations of carbon and nitrogen that are typical of graphitic carbon nitride. Specifically, in addition to the peak observed at \(284.8 \mathrm{eV}\) , the C1s spectrum for CN featured two additional peaks at \(286.1 \mathrm{eV}\) and \(288.05 \mathrm{eV}\) , which correspond to \(\mathrm{C - NH}_{2}\) and \(\mathrm{C(N)}_{3}\) moieties within the graphitic carbon nitride structure, respectively. \(^{45}\)  
+
+After heat treatment, the ultraviolet- visible absorption strength of the material was enhanced (Figure. 1d). The strong electron paramagnetic resonance (EPR) signal peak with a g value of 2.004 was clearly observed (Figure. 1e). As the heat treatment temperature increased, the N vacancies in the material first increase, peaking at \(\mathrm{CN}_{700}\) , and then decrease. FTIR provided further insights into these defects' states. The spectra obtained from the samples (Figure. 1f) displayed distinct peaks characteristic of graphitic carbon nitride. A peak around \(810 \mathrm{cm}^{- 1}\) was observed, indicating out- of- plane bending vibrations in heptazine ring systems. Additionally, peaks in the range of 900 \(\mathrm{cm}^{- 1}\) to \(1800 \mathrm{cm}^{- 1}\) were noted, which are related to \(\mathrm{C - N(- C) - C}\) vibrations within the heterocycles or bridging \(\mathrm{C - NH - C}\) units. Broad peaks spanning from \(3000 \mathrm{cm}^{- 1}\) to \(3500 \mathrm{cm}^{- 1}\) were attributed to the \(\mathrm{NH / NH}_{2}\) group. However, when the CN samples were subjected to rapid thermal treatment, marked changes were evident in the FTIR signals. A prominent peak appeared at \(2175 \mathrm{cm}^{- 1}\) , corresponding to \(- \mathrm{C} \equiv \mathrm{N}\) (cyano groups), and its intensity increased with higher treatment temperature increased. The emergence of these cyano groups disrupts the hydrogen bonds between the polymeric melon strands,
+
+<--- Page Split --->
+
+causing layers fluctuation and the collapse of the periodic stacking structure, as supported by the XRD findings.  
+
+![](images/Figure_1.jpg)
+
+<center>Figure 1. Composition characterization of CN. XRD patterns (a), C 1s XPS spectra (b), N 1s XPS spectra (c), UV-vis spectra (d), EPR spectra (e) and FT-IR spectra (f) of CN with different nitrogen vacancy concentrations. </center>  
+
+### 2.2 VIS-photocatalytic performances  
+
+CN with varying concentrations of N vacancies was used to investigate the photocatalytic activity of glucose consumption, and it was found that \(\mathrm{CN}_{700}\) , with the highest concentration of N vacancies, exhibited the best photocatalytic activity (Figure 2a). Therefore, \(\mathrm{CN}_{700}\) was selected for loading Cu in subsequent experiments. Additionally, to assess the photocatalyst's capability, measurements of the transient photocurrent responses were conducted. As shown in Figure 2b, all samples exhibit reproducible and stable photocurrent signals, with photocurrent density directly proportional to the nitrogen vacancy density. \(\mathrm{CN}_{700}\) showed the highest photocurrent intensity, indicating a lower electron-hole pairs and improved migration efficiency. This is likely due to N vacancies acting as capture centers for photogenerated holes, which
+
+<--- Page Split --->
+
+inhibit electron- hole recombination and promote separation, thus extending charge carrier lifetime. \(^{46, 47}\) Therefore, more photogenerated electrons and holes in \(\mathrm{CN}_{700}\) contribute to the photocatalytic consumption of glucose. Electron- transfer resistance was further investigated using electrochemical impedance spectroscopy (EIS). Typically, a smaller arc radius on the EIS Nyquist plot signifies reduced interfacial contact resistance between the electrode and the electrolyte. \(^{45}\) As shown in Figure 2c, the radius is inversely proportional to the N vacancy density, with \(\mathrm{CN}_{700}\) exhibiting the smallest radius of circularity, suggesting that its lower electron- transfer resistance facilitates carriers separation. According to \((\mathrm{Ahv})^2\) and the tangent intercept of the curve of light energy, the bandgap energy (Eg) was calculated. The Eg values of \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) were 2.8, 2.8, 2.77, 2.73, and 2.78 eV, respectively, indicating that the introduction of nitrogen vacancies slightly reduced the band gap of \(\mathrm{CN}_{700}\) and moderately increased its absorbance, consistent with UV- visible spectrum results. To further verify that the N vacancy modified samples enhance carrier separation, the charge carrier density \((\mathrm{N_d})\) was evaluated using Mott- Schottky (M- S) curves. Figure 2e reveals that the M- S curve has a positive slope, which is characteristic of n- type semiconductors like \(\mathrm{CN}_{700}\) , allowing \(\mathrm{N_d}\) values to be calculated by the Equation (1). The \(\mathrm{N_d}\) values for the \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) were estimated to be \(1.8522 \times 10^{21}\) , \(2.3153 \times 10^{21}\) , \(2.3453 \times 10^{21}\) , \(3.2008 \times 10^{21}\) , and \(2.0185 \times 10^{21}\) cm\(^{- 3}\) , respectively (Table S2). These \(\mathrm{N_d}\) values correlated with N vacancy density, with \(\mathrm{CN}_{700}\) having the highest \(\mathrm{N_d}\) , indicating that N vacancies in the catalyst can increase the carrier density.  
+
+Moreover, the catalyst's potential relative to the standard hydrogen electrode (SHE), was calculated from the Nernst equation (2) and the Mott- Schottky curve, allowing determination of the conduction band (CB) position. The intersection of the tangent with the Mott- Schottky curve's x- axis represents the catalyst's potential relative to the Ag/AgCl electrode; for n- type semiconductors, the CB position is 0.2 V lower than this potential. The CB positions of catalysts \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) (relative to the SHE: \(- 0.17\) , \(- 0.19\) , \(- 0.26\) , \(- 0.19\) and \(- 0.37\) V, respectively) are shown in Table S3. Combining these \(\mathrm{E_g}\) values with the corresponding catalysts' Tauc curves (Figure
+
+<--- Page Split --->
+
+2f), the valence band potentials are determined to be 2.63, 2.61, 2.50, 2.54, and 2.40 V, respectively. Thus, as N vacancies increase, the bandgap width narrows, consistent with the PBE calculation results (Figure S4). This narrowing allows \(\mathrm{CN}_{700}\) to utilize more low-energy photons, facilitating the photocatalytic consumption of glucose, which aligns with it observed the photocatalytic activity.  
+
+![](images/Figure_2.jpg)
+
+<center>Figure 2. Photoelectrochemical behaviours of modified CN. (a) Photocatalytic properties of different catalysts for glucose consumption. (b) Transient photocurrent responses. (c) Electrochemical impedance spectra (EIS). In the simulated electrical equivalent-circuit model (inset), RS, R1, and CPE represent as solution resistance, charge transfer resistance, and double layer capacitance, respectively. (d) The plots of \((\mathrm{Ahv})^2\) and \(hv\) of different catalysts. (e) Mott-Schottky curves of different catalysts. (f) Schematic illustration of band structure for the different catalysts. </center>  
+
+### 2.3 Morphology, size, and composition characterization of Cu/CN  
+
+CN rich in nitrogen vacancies was prepared by thermal etching, and single- atom Cu- dispersed Cu/CN was synthesized using impregnation method. From the kinetics of \(\mathrm{H}_2\mathrm{O}_2\) consumption by Cu/CN photocatalysis, it was found that \(7.4 \mu \mathrm{g} / \mathrm{L}\) of \(\mathrm{H}_2\mathrm{O}_2\) was consumed in 10 minutes (Figure S4). The visible light absorption intensity of Cu/CN in the UV- vis spectrum is stronger than \(\mathrm{CN}_{700}\) (Figure 3a), indicating that Cu/CN is more
+
+<--- Page Split --->
+
+efficient in utilizing visible light for photocatalytic reaction, consistent with the PBE calculation results (Figure S5). The FT- IR spectra of Cu/CN show a reduced intensity in C- N vibration within the heterocycle, possibly due to the influence of Cu single- atom on the vibration modes (Figure 3b). The XRD pattern of Cu/CN shows no Cu crystal peaks (Figure 3c), indicating that Cu single- atom does not disrupt the material's crystal structure. The Cu 2p XPS spectra of Cu/CN confirm the presence of both +1 and +2 oxidation states (Figure 3d).<sup>48</sup> To further investigate the oxidation state of the Cu single atom in Cu/CN, we analyzed X- ray absorption energy near- edge structure (XANES) of Cu foil, Cu<sub>2</sub>O, CuO, and Cu/CN (Figure 3e). The findings suggest that the XANES spectrum of Cu/CN falls between the spectra of Cu<sub>2</sub>O and CuO, indicating that the oxidation state of the single copper atom is between +1 and +2. From the slope and valence plot, the valency of Cu was estimated to be +1.75 (Figure 3f). Extended X- ray absorption fine structure (EXAFS) was investigated to examine the coordination environment of the Cu sites. Quantitative EXAFS curve fitting analysis in R spaces was employed to characterize the coordination structure of the Cu atoms (Figure 3h and Figure S6). The first coordination shell, observed at 1.5 Å in Figure 3g, suggests the presence of Cu- N or/and Cu- O coordination. No significant Cu- Cu scattering was detected at 2.2 Å in Cu/CN samples, further confirming the single- atom dispersion of Cu. Wavelet transform of the EXAFS data distinguished backscattering atoms (Figure 3i), revealing maximum intensities at 7.2 and 10.6 Å<sup>-1</sup> for Cu foil and CuO, respectively, which correspond to the Cu- Cu configuration. In contrast, Cu/CN displayed a maximum intensity at 4.5 Å<sup>-1</sup>, attributed to Cu- N or/and Cu- O coordination. The N species produced from melamine during pyrolysis help fill vacancies, stabilizing the atomic Cu. EXAFS fitting results show the coordination number (CN) of Cu- N was approximately 4. Determining the specific CN for nitrogen and oxygen, however, is challenging due to their similar bond lengths to Cu. Based on previous studies, both N and O back- scattering paths were included for optimal fit; comparison with fixed CN values for N and O were also made (Table S4). By fitting the EXAFS spectra to different structural models that were optimized using DFT (Figure 4), we found that the optimal fit was achieved when the Cu was coordinated with 4 N atoms, as shown in the atomic Cu structure model for
+
+<--- Page Split --->
+
+Cu/CN in the inset of Scheme 1. We also measured the EPR spectra to analyze the in- situ generation of \(\bullet \mathrm{OH}\) and \(\bullet \mathrm{O}_2^-\) by Cu/CN with and without light exposure (Figure S7). Under light, Cu/CN exhibited superior performance in photocatalytic decomposition of \(\mathrm{H}_2\mathrm{O}_2\) to generate \(\bullet \mathrm{OH}\) compared to \(\mathrm{CN}_{700}\) . Additionally, Cu/CN can directly photocatalyze water splitting to produce \(\bullet \mathrm{OH}\) and oxidize \(\mathrm{O}_2\) to produce \(\bullet \mathrm{O}_2^-\) , making it suitable for ROS generation for antibacterial purposes. However, excessive ROS at the wound site can cause inflammation, hindering the healing process. Utilizing DPPH (Figure S8), ABTS (Figure S9) assay, as well as RAW264.7 intracellular ROS scavenging experiments (Figure S10), it has been demonstrated that Cu/CN can effectively scavenge excessive ROS during dark reaction, independent of light exposure. \(^{49,50}\)  
+
+![](images/Figure_3.jpg)
+
+<center>Figure 3. Structural characterizations of Cu/CN. (a) UV-vis spectra of Cu/CN. (b) FT-IR spectra of Cu/CN. (c) XRD patterns of Cu/CN. (d) Cu 2p XPS spectra of Cu/CN. (e) XANES spectra at Cu K-edge of Cu/CN. (f) Standard curve of Cu valence state. (g) EXAFS spectra at Cu K-edge of Cu/CN. (h) EXAFS fitting result of Cu/CN at R space. (i) Wavelet transform of Cu foil, \(\mathrm{Cu}_2\mathrm{O}\) , \(\mathrm{CuO}\) and \(\mathrm{Cu/CN}\) . </center>  
+
+### 2.4 DFT of Cu/CN
+
+<--- Page Split --->
+
+Density functional theory (DFT) calculations were conducted to investigate the effects of nitrogen- containing vacancies in CN and Cu single- atom in the adsorption energies of glucose and \(\mathrm{H}_2\mathrm{O}_2\) on Cu/CN. For glucose, there are two types of adsorption sites on the CN surface: i) glucose adsorbs at non- vacancy sites with an adsorption energy of \(- 0.15\) eV, and ii) glucose adsorbs at nitrogen vacancy sites with an adsorption energy of \(- 0.42\) eV (Figure 4a). For \(\mathrm{H}_2\mathrm{O}_2\) , there are three types of adsorption sites on the Cu/CN surface of: i) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs at nitrogen vacancies, ii) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs between nitrogen vacancies and Cu single- atom sites, and iii) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs at Cu single- atom, with corresponding adsorption energies of \(- 0.26\) eV, \(- 0.74\) eV and \(- 1.39\) eV, respectively (Figure 4b). The more negative the adsorption energy, the more stable the adsorption, indicating that the presence of nitrogen vacancies enhance the catalyst's glucose adsorption capacity. Furthermore, charge transfer between CN and glucose was visualized using the Charge Density Difference (CDD) analysis. Figures S11 and S12 show electrons accumulation (yellow) at non- nitrogen vacancies site and electrons depletion (blue) at nitrogen vacancies, indicating that the presence of nitrogen vacancies are more effective in oxidizing glucose.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4. DFT of Cu/CN. (a) Adsorption energy of glucose at non vacancy and nitrogen vacancy sites on CN surface. (b) Adsorption energy of \(\mathrm{H}_2\mathrm{O}_2\) at nitrogen vacancy, non-vacancy, and Cu single-atom sites on Cu/CN surface. </center>  
+
+### 2.5 In vitro antibacterial property of Cu/CN  
+
+Motivated by the theoretical and experimental studies above, we hypothesize that Cu/CN can act as an effective bactericidal agent when glucose present via a photocatalytic cascade reaction. To validate this concept, we conducted an in vitro antibacterial study using ESBL E. coli and MRSA as model bacteria. The results, presented in Figure 5a and 5c, d demonstrate that the bactericidal rates for both strains were significantly higher when treated with Cu/CN+vis compared to the control, GOx, and \(\mathrm{CN}_{700}\) groups, which were \(99.90\%\) and \(99.95\%\) , respectively. Notably, a significant difference (P \(< 0.5\) ) between GOx and \(\mathrm{CN}_{700}\) +vis groups (Figure 5b and 5d) was observed, which can be attributed to the enhanced bactericidal performance resulting from the \(\bullet \mathrm{OH}\) generated by the cascade catalysis. However, the cascade catalytic activity of \(\mathrm{CN}_{700}\) was relatively
+
+<--- Page Split --->
+
+low, resulting in a lower bactericidal rate than the \(\mathrm{Cu / CN + }\) - vis experimental group. This result is supported by the in situ EPR detection of the \(\bullet \mathrm{OH}\) activity produced by \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) photocatalysis (Figure S10). Furthermore, SEM images (Figure 5e) provide additional evidence, showing that MRSA and ESBL E. coli in the control group maintained their original plump structures with intact outer membranes, while treatment with GOx caused some structural distortion due to oxidative damage from \(\mathrm{H}_2\mathrm{O}_2\) .51 By contrast, bacteria treated with \(\mathrm{CN}_{700} + \mathrm{vis}\) and \(\mathrm{Cu / CN + }\) - vis exhibited significant morphological deformations, such as collapse, distortion, and breakage (highlighted by red arrows), indicating the highest bactericidal efficacy. Additionally, we also investigated the anti- bacterial biofilm capability of \(\mathrm{Cu / CN}\) under Xe light \((\lambda > 420 \mathrm{nm})\) radiation. As shown from Figure 5f, confocal laser scanning microscopy (CLSM) images of ESBL E.coli and MRSA biofilms showed a red fluorescence signal, confirming that \(\mathrm{Cu / CN}\) under \(\mathrm{Xe}(\lambda > 420 \mathrm{nm})\) light irradiation exhibits excellent antibiofilm performance. Therefore, the \(\mathrm{Cu / CN}\) photocatalytic therapeutic platform, which utilizes a photoswitch to control the generation of ROS to combat MDR bacteria, is highly effective. This platform operates through: i) photocatalytic cascade reaction that consume glucose to produce \(\bullet \mathrm{OH}\) , ii) photolysis of water to generate \(\bullet \mathrm{OH}\) , and iii) oxidation of \(\mathrm{O}_2\) to form \(\bullet \mathrm{O}_2\) (Figure 5g).
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 5. Evaluation of antibacterial capacity. Photographs of bacterial colonies formed by (a) ESBL E. coli and (c) MRSA after treatments with Cu/CN. The relative bacterial viability of (b) ESBL E. coli and (d) MRSA. Morphologies of (e) ESBL E. coli and MRSA treated or untreated with the Cu/CN group. (f) CLSM photos of biofilm treated with Cu/CN. (g) Schematic diagram of photocatalytic antibacterial control by photoswitch. </center>  
+
+### 2.6 Hemocompatibility and cytocompatibility investigation of the Cu/CN
+
+<--- Page Split --->
+
+Low cytotoxicity is essential for applying nanomaterials to skin wounds. \(^{52}\) The proliferation of 3T3 fibroblasts was assessed in vitro across different groups: control, GOx, CN, and Cu/CN treated ones (Figure 6a). After 24 hours of cultivation, the 3T3 fibroblast cells in all groups exhibited healthy growth and proliferation, indicating excellent cytocompatibility of Cu/CN. Additionally, after 24 hours, the viability of 3T3 fibroblast cells incubated with the CN and Cu/CN remained above \(95\%\) , as determined through Alamar Blue fluorescence testing (Figure 6b). These results suggest that Cu/CN is well- suited for treating chronic diabetic wounds due to its high biocompatibility. The hemolysis assay evaluated the effect of CN and Cu/CN on red blood cells (RBCs). As seen in Figure 6c, Triton X- 100 acted as the positive control, displaying red color due to complete hemolysis. In contrast, RBCs treated with GOx, CN, and Cu/CN appeared pale yellow and transparent, similar to Tris- HCl, the negative control. Hemolysis ratios of GOx, CN, and Cu/CN were \(1.5\%\) , \(2.8\%\) , and \(3.3\%\) , respectively, all below the permissible limit of \(5\%\) . This indicates that Cu/CN has good hemocompatibility and is suitable for treating diabetic wounds. The cell scratch test further examined the influence of CN and Cu/CN on cell migration. As shown in Figure 6d, the cell migration area after 48 hours of Cu/CN treatment was larger than in the other three groups, with the migration rates of \(68.0\%\) , \(45.8\%\) , \(75.4\%\) , and \(90.4\%\) respectively (Figure 6e). Cells treated with Cu/CN showed the highest cell migration rate, likely due to trace amounts of Cu promoting cell migration. \(^{53}\) Summarily, the excellent cell safety, anti- hemolytic and pro- migration effects of Cu/CN are highly favorable for diabetic wound healing.
+
+<--- Page Split --->
+![](images/Figure_6.jpg)
+
+<center>Figure 6. In vitro biocompatible evaluation of \(\mathrm{Cu / CN}\) . (a) Fluorescence images of 3T3 fibroblast cells cultured with hydrogel groups after 24 hours. (b) Cell viability of the \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) . (c) Hemolysis ratio of the \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) . (d) Cell migration of HUVEC cells in the control and the different groups at 0, 24, and 48 hours. (e) The cellular migration after various treatments for different time periods was identified by drawing the green shadow at the edge of cells in (d), and calculate the area of cell migration (\*p<0.5; \*\*p<0.01; \*\*\*p<0.001; \*\*\*\*p<0.0001). </center>  
+
+### 2.7 In vivo wound healing performance of \(\mathrm{Cu / CN}\)  
+
+A diabetic model in KM mice was established using STZ, selecting mice with blood glucose levels consistently above \(16.7 \mathrm{mM}\) for one week post- induction to create a wound model for subsequent treatment. \(^{14,54}\) Full- thickness dorsal skin punch wounds (6 mm in diameter) were created with a round punch, infected by MRSA to induce ulceration, and treated according to different protocols. The mice were randomly
+
+<--- Page Split --->
+
+allocated into ten distinct treatment groups, each with five mice: control, GOx, CN, and Cu/CN. Each group was treated with or without Xe light radiation ( \(\lambda > 420 \mathrm{nm}\) , 15 minutes). Photographs illustrating wound repair across treatment group are shown in Figure 7a, and quantitative analyses of wound area reduction in diabetic wounds are provided in Figure 7b. Remarkably, wounds treated with Cu/CN under Xe light irradiation demonstrated a significantly accelerated healing rate by day 7 compared to the other groups. The wound area ratios were \(62.8\%\) for the control, \(25.5\%\) for GOx, \(16.4\%\) for CN, and \(2.8\%\) for Cu/CN (Figure S13Figure 7e). In the absence of light, minimal difference observed in wound healing, indicating that Cu/CN requires NIR irradiation for effective activity (Figure S14). On day 14, wound tissue was collected for bacterial culture. As shown in Figure 7c, the control group exhibited a high concentration of MRSA, which impeded the wound healing process, whereas the Cu/CN group experienced a substantial reduction in bacterial counts. Notably, all MRSA bacteria on the skin treated with the Cu/CN group were eradicated, underscoring Cu/CN's efficacy in enhancing wound healing and providing photocatalytic antibacterial action in vivo. Cu/CN exhibits effective antibacterial activity upon exposure to Xe light ( \(\lambda > 420 \mathrm{nm}\) ) irradiation, while its nitrogen vacancy-rich CN component photocatalytically consumed glucose, generating \(\mathrm{H}_2\mathrm{O}_2\) through to normalize the glucose levels in diabetic wounds. Cu single- atom further catalyzed \(\mathrm{H}_2\mathrm{O}_2\) produce \(\bullet \mathrm{OH}\) for antibacterial purposes. The enhanced diabetic wound healing observed with Cu/CN treatment highlights its potential for combining hypoglycemic and antibacterial properties in MDR- infected diabetic wound healing applications.  
+
+The transition from the inflammatory to the proliferative stage is a key regulatory point in wound healing. Persistent inflammation is a prominent feature of chronic diabetic wounds. \(^{55}\) On day 14 of treatment, histological analyses were as performed. H&E and Masson staining of the wound tissue were utilized to examine the inflammatory reaction and evaluate collagen formation and deposition in wounds treated with control, GOx, CN, and Cu/CN, while immunofluorescence staining for cluster of differentiation 31 (CD31), tumor necrosis factor- alpha (TNF- \(\alpha\) ), and interleukin- 10 (IL- 10) was employed to assess angiogenesis and inflammation levels in the healed wounds of these treatment groups (Figure 7d). In the Cu/CN
+
+<--- Page Split --->
+
+treated group, complete wound healing was achieved by day 14, with numerous fibroblasts visible under the skin (Figure 7d). In contrast, wounds in the other groups retained a considerable number of inflammatory cells and showed incomplete healing. CD31 is commonly used to detect the presence of neovascularization in granulation tissue. As depicted in Figure 7d, the Cu/CN- treated group exhibited the highest number of vascular growths on days 14 compared to the other three groups, as indicated by CD31 immunohistochemical staining, with CD31 positivity reaching \(6.6\%\) (Figure 7e and Figure S15). TNF- \(\alpha\) , an inflammatory cytokine, can inhibit wound healing in diabetic conditions, while the anti- inflammatory M2 phenotype is usually activated by the Th2 cytokines of IL- 10. \(^{56}\) TNF- \(\alpha\) exacerbates inflammation by initiating inflammatory cascades, whereas IL- 10 acts as an anti- inflammatory cytokine by suppressing effector T cell activity. \(^{57}\) Immunohistochemical staining (Figure 7d) revealed that Cu/CN effectively decreased TNF- \(\alpha\) expression while increasing IL- 10 expression in inflammatory cells. On day 14, the positive proportion of TNF- \(\alpha\) decreased to \(1.0\%\) , while IL- 10 positivity increased to \(5.3\%\) (Figure 7e and Figure S15). This controlled treatment process is achieved by using a photocatalytic light switch, allowing ROS generation to be halted after irradiation, thus avoiding increased inflammation. Furthermore, after phototherapy completed, Cu/CN can scavenge excess ROS in the dark reaction, without the need for light. These findings suggest that Cu/CN promotes the transition of macrophages from the pro- inflammatory M1 to the anti- inflammatory M2 phenotype and modulates inflammatory cytokine expression (Figure 7f). This modulation led to anti- inflammatory effects, enhancing the capacity for skin repair in diabetic wounds.
+
+<--- Page Split --->
+![](images/Figure_7.jpg)
+
+<center>Figure 7. Evaluation of wound healing in diabetic mouse models. (a). Photographs of the wounds of wound closure treat with light \((\lambda >420 \mathrm{nm})\) . (b) Wound size ratios with different treatments at days 0, 3, 7, and 14. (c) Typical agar plate photos for the remaining MRSA colonies in the ulcer after different treatment on day 14. (d) Immunohistochemical studies of diabetic wound collected at day 14. Scale bar: \(1000 \mu \mathrm{m}\) for H&E and Masson, \(100 \mu \mathrm{m}\) for TNF- \(\alpha\) , IL-10 and CD31. (e) Overall performance of control, GOx, CN, and Cu/CN in the in vivo diabetic wound treatment. (f) Schematic diagram of photocatalytic therapy control by photoswitch. </center>
+
+<--- Page Split --->
+
+## 3. Conclusion  
+
+In conclusion, we have successfully developed a Cu/CN bionanomaterial that can effectively treat MDR bacteria infected diabetic wounds by simultaneously regulating blood glucose levels, bacterial infection, and persistent inflammation. The synergistic effects of nitrogen vacancies and single atoms in Cu/CN enabled the realization of a photoswitchable cascade reaction. The introduction of nitrogen vacancies enhances the activity of photocatalytic oxidation for glucose consumption, which helps to reduce blood glucose levels at the wound sites. Subsequently, the \(\mathrm{H}_2\mathrm{O}_2\) generated in the previous step is converted into •OH through a Cu single- atom photocatalytic cascade reaction. Concurrently, Cu/CN can directly produce •OH and • \(\mathrm{O}_2^{- }\) through photocatalytic water splitting. DFT calculations confirmed that glucose preferentially adsorbs at nitrogen vacancies, while \(\mathrm{H}_2\mathrm{O}_2\) adsorbs more readily at Cu single- atom. In vitro antibacterial results demonstrated that the Cu/CN photocatalyst had significant bactericidal effects on drug- resistant MRSA and ESBL E. coli and exhibited anti- biofilm activity. The in vivo study in a mice diabetic wound MRSA infection model shows that Cu/CN photocatalyst has good biological safety and excellent therapeutic efficacy with visible light irradiation. The excess ROS can be scavenged by Cu/CN without light, alleviating inflammation at the wound site and encouraging the transition of macrophages from the pro- inflammatory M1 phenotype to the anti- inflammatory M2 phenotype. This work presents a proof of concept for using a photoswitchable cascade reaction to treat MDR bacterial infected diabetic wound complications.  
+
+## Acknowledgements  
+
+This research was sponsored by the key research program of Ningbo (Grant No. 2023Z210), the Ningbo Natural Science Foundation (Grant No. 202003N4006), the National Foreign Expert Project (Grant No. H20240307), the National Natural Science Foundation of China (Grant No. 52473265), and the Shaanxi Provincial Science Fund for Distinguished Young Scholars (2023- JC- JQ- 32).
+
+<--- Page Split --->
+
+Received: ((will be filled in by the editorial staff))Revised: ((will be filled in by the editorial staff))Published online: ((will be filled in by the editorial staff))  
+
+## References  
+
+1. Armstrong DG, Tan T-W, Boulton AJM, Bus SA. Diabetic Foot Ulcers: A Review. JAMA-Journal of the American Medical Association 330, 62-75 (2023).  
+2. Falanga V, et al. Chronic wounds. Nature Reviews Disease Primers 8, 50-70 (2022).  
+3. Wang T, Li YR, Cornel EJ, Li C, Du JZ. Combined Antioxidant-Antibiotic Treatment for Effectively Healing Infected Diabetic Wounds Based on Polymer Vesicles. ACS Nano 15, 9027-9038 (2021).  
+4. Pranantyo D, et al. Hydrogel dressings with intrinsic antibiofilm and antioxidative dual functionalities accelerate infected diabetic wound healing. Nature Communications 15, 954-973 (2024).  
+5. Li JJ, et al. Lactose azocalixarene drug delivery system for the treatment of multidrug-resistant pseudomonas aeruginosa infected diabetic ulcer. Nature Communications 13, 6279-6291 (2022).  
+6. Zhang R, et al. Enhancement of antibacterial and wound healing capabilities of cotton fabric through the covalent bonding of antibiotic-loaded hydrophobized mesoporous silica nanoparticles. Fundamental Research, (2023).  
+7. Idrees M, et al. Bacterial spectrum from diabetic foot Ulcers: A study of antibiotic resistance patterns and phylogenetic diversity. Journal of King Saud University-Science 36, 103320-103330 (2024).  
+8. Wang SH, et al. Meticillin-resistant Staphylococcus aureus isolated from foot ulcers in diabetic patients in a Chinese care hospital: risk factors for infection and prevalence. Journal of Medical Microbiology 59, 1219-1224 (2010).  
+9. Chen LF, et al. A Glucose-Powered Activatable Nanozyme Breaking pH and H2O2 Limitations for Treating Diabetic Infections. Angewandte Chemie-International Edition 60, 23534-23539 (2021).  
+10. He SB, et al. Glucose Oxidase Energized Osmium with Dual-Active Centers and Triple Enzyme Activities for Infected Diabetic Wound Management. Advanced Healthcare Materials 13, 2303548-2303561 (2024).
+
+<--- Page Split --->
+
+11. Li YF, et al. A Guanosine-Quadruplex Hydrogel as Cascade Reaction Container Consuming Endogenous Glucose for Infected Wound Treatment-A Study in Diabetic Mice. Advanced Science 9, 2103485-2103498 (2022).  
+
+12. Liao YH, et al. Glucose oxidase: An emerging multidimensional treatment option for diabetic wound healing. Bioactive Materials 44, 131-151 (2025).  
+
+13. Shang L, et al. Ultrasound-Augmented Multienzyme-like Nanozyme Hydrogel Spray for Promoting Diabetic Wound Healing. ACS Nano 17, 15962-15977 (2023).  
+
+14. Sun XC, et al. Multifunctional Hydrogel Containing Oxygen Vacancy-Rich \(\mathrm{WO}_x\) for Synergistic Photocatalytic \(\mathrm{O}_2\) Production and Photothermal Therapy Promoting Bacteria-Infected Diabetic Wound Healing. Advanced Functional Materials, 2411117-2411132 (2024).  
+
+15. Deng MY, et al. Diabetes immunity-modulated multifunctional hydrogel with cascade enzyme catalytic activity for bacterial wound treatment. Biomaterials 289, 121790-121800 (2022).  
+
+16. Li HS, Li BY, Lv DL, Li WH, Lu YF, Luo GX. Biomaterials releasing drug responsively to promote wound healing via regulation of pathological microenvironment. Advanced Drug Delivery Reviews 196, 114778-114807 (2023).  
+
+17. Huang Y, Ren J, Qu X. Nanozymes: Classification, Catalytic Mechanisms, Activity Regulation, and Applications. Chemical Reviews 119, 4357-4412 (2019).  
+
+18. Min S, et al. Nanomaterials with Glucose Oxidase-Mimicking Activity for Biomedical Applications. Molecules 28, (2023).  
+
+19. Chong Y, Liu Q, Ge C. Advances in oxidase-mimicking nanozymes: Classification, activity regulation and biomedical applications. Nano Today 37, 101076-101092 (2021).  
+
+20. Wang M, et al. Collaboration in Contradiction: Self - Adaptive Synergistic ROS Generation and Scavenge Balancing Strategies Used for the Infected Wounds Treatment. Advanced Healthcare Materials, 2402579-2402596 (2024).  
+
+21. Zhang X, et al. Polarization of macrophages to an anti-cancer phenotype through in situ uncaging of a TLR 7/8 agonist using bioorthogonal nanozymes. Chemical Science 15, 2486-2494 (2024).  
+
+22. Chen S, et al. Photocatalytic glucose depletion and hydrogen generation for diabetic wound healing. Nature Communications 13, 5684-5695 (2022).
+
+<--- Page Split --->
+
+23. Liao GF, et al. Emerging graphitic carbon nitride-based materials for biomedical applications. Progress in Materials Science 112, 100666-100687 (2020).  
+
+24. Kong XY, Liu XM, Zheng YF, Chu PK, Zhang Y, Wu SL. Graphitic carbon nitride-based materials for photocatalytic antibacterial application. Materials Science & Engineering R-Reports 145, 100610-100639 (2021).  
+
+25. Chen Y, Tan CL, Zhang H, Wang LZ. Two-dimensional graphene analogues for biomedical applications. Chemical Society Reviews 44, 2681-2701 (2015).  
+
+26. Shi L, et al. Photoassisted Construction of Holey Defective g-C3N4 Photocatalysts for Efficient Visible-Light-Driven H2O2 Production. Small 14, 1703142-1703151 (2018).  
+
+27. Zhang P, et al. Modified carbon nitride nanozyme as bifunctional glucose oxidase-peroxidase for metal-free bioinspired cascade photocatalysis. Nature Communications 10, 940-954 (2019).  
+
+28. Niu P, Yin LC, Yang YQ, Liu G, Cheng HM. Increasing the visible light absorption of graphitic carbon nitride (melon) photocatalysts by homogeneous self-modification with nitrogen vacancies. Advanced Materials 26, 8046-8052 (2014).  
+
+29. Yu H, et al. Alkali-Assisted Synthesis of Nitrogen Deficient Graphitic Carbon Nitride with Tunable Band Structures for Efficient Visible-Light-Driven Hydrogen Evolution. Advanced Materials 29, 1605148-1605155 (2017).  
+
+30. Lin Z, Wang X. Nanostructure engineering and doping of conjugated carbon nitride semiconductors for hydrogen photosynthesis. Angewandte Chemie-International Edition 52, 1735-1738 (2013).  
+
+31. Kang Y, Yang Y, Yin LC, Kang X, Liu G, Cheng HM. An Amorphous Carbon Nitride Photocatalyst with Greatly Extended Visible-Light-Responsive Range for Photocatalytic Hydrogen Generation. Advanced Materials 27, 4572-4577 (2015).  
+
+32. Yang B, et al. Ultrathin porous carbon nitride nanosheets with well-tuned band structures via carbon vacancies and oxygen doping for significantly boosting H₂ production. Applied Catalysis B-Environment and Energy 314, 121521-121533 (2022).  
+
+33. Wang XY, et al. Controllable Approach to Carbon-Deficient and Oxygen-Doped Graphitic Carbon Nitride: Robust Photocatalyst Against Recalcitrant Organic Pollutants and the Mechanism Insight. Advanced Functional Materials 31, 2010763-2010779 (2021).
+
+<--- Page Split --->
+
+34. Zhang YZ, et al. Synergistic effect of nitrogen vacancy on ultrathin graphitic carbon nitride porous nanosheets for highly efficient photocatalytic \(\mathrm{H}_2\) evolution. Chemical Engineering Journal 431, 134101-134109 (2022).  
+
+35. Li JN, et al. Efficient photosynthesis of \(\mathrm{H}_2\mathrm{O}_2\) via two-electron oxygen reduction reaction by defective \(\mathrm{g - C_3N_4}\) with terminal cyano groups and nitrogen vacancies. Chemical Engineering Journal 463, 142512-142525 (2023).  
+
+36. Wang F, Xu J, Wang ZP, Lou Y, Pan CS, Zhu YF. Unprecedentedly efficient mineralization performance of photocatalysis-self-Fenton system towards organic pollutants over oxygen-doped porous \(\mathrm{g - C_3N_4}\) nanosheets. Appl Catal B-Environ 312, 121438-121447 (2022).  
+
+37. Zhang GQ, Xu YS, He CX, Zhang PX, Mi HW. Oxygen-doped crystalline carbon nitride with greatly extended visible-light-responsive range for photocatalytic \(\mathrm{H}_2\) generation. Appl Catal B-Environ 283, 119636-119645 (2021).  
+
+38. Chen X, Liu L, Yu PY, Mao SS. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 331, 746-750 (2011).  
+
+39. Wu XF, et al. Promoting effect of interfacial hole accumulation on photoelectrochemical water oxidation in \(\mathrm{BiVO_4}\) and Mo-doped \(\mathrm{BiVO_4}\) . Advanced Powder Materials 3, 100234-100241 (2024).  
+
+40. Mohamed MJS, Gondal MA. Nanostructured defects-rich black-ZnO/biowaste-derived carbon composites for efficient visible light-driven hydrogen generation: A study on the role of C-Zn@C-O-Zn interface. International Journal of Hydrogen Energy 92, 47-58 (2024).  
+
+41. Wang J, et al. A ruthenium single atom nanozyme-based antibiotic for the treatment of otitis media caused by Staphylococcus aureus. Frontiers in Chemistry 12, 1439039-1439053 (2024).  
+
+42. Fan YF, Gan XR, Zhao HM, Zeng ZX, You WJ, Quan X. Multiple application of SAzyme based on carbon nitride nanorod-supported Pt single-atom for \(\mathrm{H}_2\mathrm{O}_2\) detection, antibiotic detection and antibacterial therapy. Chemical Engineering Journal 427, 131572-131582 (2022).  
+
+43. Chen AT, et al. Regulating the charge density of Cu(I) single sites enriched on the surface of N3c Vacancies-engineered \(\mathrm{g - C_3N_4}\) for efficient Fenton-like reactions. Separation and Purification Technology 314, 123525-123538 (2023).  
+
+44. Niu P, Qiao M, Li Y, Huang L, Zhai T. Distinctive defects engineering in graphitic carbon nitride for greatly extended visible light photocatalytic hydrogen evolution. Nano Energy 44, 73-81 (2018).
+
+<--- Page Split --->
+
+45. Wei Y, et al. Oxygen-Doped Porous g-\(C_3N_4\) via Oxalic Acid-Assisted Thermal Polycondensation as a Visible Light-Driven Photocatalyst for Bisphenol A Degradation. ACS Applied Nano Materials 6, 16567-16579 (2023).  
+
+46. Muhammad P, et al. Defect Engineering in Nanocatalysts: From Design and Synthesis to Applications. Advanced Functional Materials 34, 2314686-2314742 (2024).  
+
+47. Kong Y, et al. Synergistic silver doping and N vacancy promoting photocatalytic performances of carbon nitride for pollutant oxidation and hydrogen production. Chemical Engineering Journal 479, 147676-147690 (2024).  
+
+48. Wang P, et al. CaCO(3)-encircled hollow CuS nanovehicles to suppress cervical cancer through enhanced calcium overload-triggered mitochondria damage. Asian Journal of Pharmaceutical Sciences 19, 100989-101005 (2024).  
+
+49. Chen Y, et al. Atomically Dispersed Cu Nanozyme with Intensive Ascorbate Peroxidase Mimic Activity Capable of Alleviating ROS - Mediated Oxidation Damage. Advanced Science 9, 2103977-2103986 (2021).  
+
+50. Yu B, et al. Using Cu - Based Metal-Organic Framework as a Comprehensive and Powerful Antioxidant Nanozyme for Efficient Osteoarthritis Treatment. Advanced Science 11, 2307798-2307810 (2024).  
+
+51. Huang Y, et al. Calcium Peroxide-Based Hydrogels Enable Biphasic Release of Hydrogen Peroxide for Infected Wound Healing. Advanced Materials, e2404813-e2404827 (2024).  
+
+52. Wang Q, et al. Mussel-inspired multifunctional hydrogel dressing with hemostasis, hypoglycemic, photothermal antibacterial properties on diabetic wounds. Biomaterials Science 10, 4796-4814 (2022).  
+
+53. Xiao J, et al. Copper Metal-Organic Framework Nanoparticles Stabilized with Folic Acid Improve Wound Healing in Diabetes. ACS Nano 12, 1023-1032 (2018).  
+
+54. Xu SJ, et al. Tanshinone IIA Ameliorates Streptozotocin-Induced Diabetic Nephropathy, Partly by Attenuating PERK Pathway-Induced Fibrosis. Drug Design Development and Therapy 14, 5773-5782 (2020).  
+
+55. Kim H, Wang SY, Kwak G, Yang Y, Kwon IC, Kim SH. Exosome - Guided Phenotypic Switch of M1 to M2 Macrophages for Cutaneous Wound Healing. Advanced Science 6, 1900513-1900526 (2019).
+
+<--- Page Split --->
+
+56. Li W, et al. Development of an Antiswelling Hydrogel System Incorporating M2-Exosomes and Photothermal Effect for Diabetic Wound Healing. ACS Nano 17, 22106-22120 (2023).  
+
+57. Wu Y, et al. A Versatile Glycopeptide Hydrogel Promotes Chronic Refractory Wound Healing Through Bacterial Elimination, Sustained Oxygenation, Immunoregulation, and Neovascularization. Advanced Functional Materials 33, 2305992-2306011 (2023).
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+SupportingInformation.pdf
+
+<--- Page Split --->

preprint/preprint__97ae97305ce10e362d0efa94157154334e4c670f53e025014566da79d85c660f/preprint__97ae97305ce10e362d0efa94157154334e4c670f53e025014566da79d85c660f_det.mmd

@@ -0,0 +1,425 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 910, 240]]<|/det|>
+# Cu Single-Atom Embedded g-C3N4 nanosheets Rehabilitate Multidrug-Resistant Bacteria Infected Diabetic Wounds via Photoswitchable Cascade Reaction  
+
+<|ref|>text<|/ref|><|det|>[[44, 263, 110, 282]]<|/det|>
+Peng Li  
+
+<|ref|>text<|/ref|><|det|>[[55, 291, 256, 308]]<|/det|>
+iampli@nwpu.edu.cn  
+
+<|ref|>text<|/ref|><|det|>[[44, 336, 748, 356]]<|/det|>
+Northwestern Polytechnical University https://orcid.org/0000- 0002- 5876- 2177  
+
+<|ref|>text<|/ref|><|det|>[[44, 361, 390, 401]]<|/det|>
+Xichen Sun Northwestern Polytechnical University  
+
+<|ref|>text<|/ref|><|det|>[[44, 407, 272, 447]]<|/det|>
+Pengqi Zhu Shanxi Bethune Hospital  
+
+<|ref|>text<|/ref|><|det|>[[44, 453, 390, 494]]<|/det|>
+Liuyan Tang Northwestern Polytechnical University  
+
+<|ref|>text<|/ref|><|det|>[[44, 500, 390, 540]]<|/det|>
+Pengfei Wang Northwestern Polytechnical University  
+
+<|ref|>text<|/ref|><|det|>[[44, 546, 390, 586]]<|/det|>
+Ningning Li Northwestern Polytechnical University  
+
+<|ref|>text<|/ref|><|det|>[[44, 592, 390, 632]]<|/det|>
+Qing Wang Northwestern Polytechnical University  
+
+<|ref|>text<|/ref|><|det|>[[44, 638, 600, 678]]<|/det|>
+Yan-Ru Lou University of Helsinki https://orcid.org/0000- 0001- 7717- 6010  
+
+<|ref|>text<|/ref|><|det|>[[44, 684, 750, 725]]<|/det|>
+Yuezhou Zhang Northwestern Polytechnical University https://orcid.org/0000- 0002- 8560- 5716  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 768, 103, 786]]<|/det|>
+## Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 806, 881, 847]]<|/det|>
+Keywords: nitrogen vacancy, graphitic carbon nitride, photocatalytic antibacterial, chronic wound infections, antibiofilm  
+
+<|ref|>text<|/ref|><|det|>[[44, 866, 336, 885]]<|/det|>
+Posted Date: February 19th, 2025  
+
+<|ref|>text<|/ref|><|det|>[[44, 904, 473, 923]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 5916168/v1
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 44, 916, 88]]<|/det|>
+License: © ① This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 105, 535, 126]]<|/det|>
+Additional Declarations: There is NO Competing Interest.  
+
+<|ref|>text<|/ref|><|det|>[[42, 161, 943, 205]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on October 16th, 2025. See the published version at https://doi.org/10.1038/s41467-025-64242- z.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 88, 812, 135]]<|/det|>
+## Cu Single-Atom Embedded g- \(\mathbf{C}_3\mathbf{N}_4\) nanosheets Rehabilitate Multidrug-Resistant Bacteria Infected Diabetic Wounds via Photoswitchable Cascade Reaction  
+
+<|ref|>text<|/ref|><|det|>[[115, 150, 881, 201]]<|/det|>
+Xichen Sun \(^{1,2\dagger}\) , Pengqi Zhu \(^{3,4\dagger}\) , Liuyan Tang \(^{1,2}\) , Pengfei Wang \(^{1,2}\) , Ningning Li \(^{1,2}\) , Qing Wang \(^{1,2}\) , Yan- Ru Lou \(^{5}\) , Yuezhou Zhang \(^{1,2*}\) , and Peng Li \(^{1,2,6*}\)  
+
+<|ref|>text<|/ref|><|det|>[[115, 216, 884, 370]]<|/det|>
+\(^{1}\) State Key Laboratory of Flexible Electronics (LOFE) & Institute of Flexible Electronics (IFE), Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 710072, China \(^{2}\) Ningbo Institute of Northwestern Polytechnical University, Frontiers Science Center for Flexible Electronics (FSCFE), Key laboratory of Flexible Electronics of Zhejiang Province, Ningbo Institute of Northwestern Polytechnical University, 218 Qingyi Road, Ningbo, 315103, China  
+
+<|ref|>text<|/ref|><|det|>[[115, 382, 884, 508]]<|/det|>
+\(^{3}\) Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, 99 Longcheng Street, Taiyuan, 030032, China \(^{4}\) Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430030, China  
+
+<|ref|>text<|/ref|><|det|>[[115, 494, 881, 600]]<|/det|>
+\(^{5}\) Faculty of Pharmacy, University of Helsinki, P.O. Box 56, FI- 00014 Helsinki, Finland \(^{6}\) School of Flexible Electronics (SoFE) and Henan Institute of Flexible Electronics (HIFE), Henan University, 379 Mingli Road, Zhengzhou, 450046, China \(^{\dagger}\) These authors contributed equally to this work.  
+
+<|ref|>text<|/ref|><|det|>[[115, 616, 588, 635]]<|/det|>
+E- mail: iamyzzhang@nwpu.edu.cn; iampli@nwpu.edu.cn  
+
+<|ref|>text<|/ref|><|det|>[[115, 653, 881, 700]]<|/det|>
+Keywords: nitrogen vacancy, graphitic carbon nitride, photocatalytic antibacterial, chronic wound infections, antibiofilm
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 87, 885, 441]]<|/det|>
+To tackle elevated blood glucose, multidrug- resistant (MDR) bacterial infections, and persistent inflammation in diabetic wounds, we present a therapeutic strategy that employs a photoswitch- controlled catalytic cascade reaction, utilizing a photocatalytic material engineered through the synergistic regulation of nitrogen vacancies and single- atom embedding. The nitrogen vacancies in \(\mathrm{g - C_3N_4}\) promise the photocatalytic glucose oxidation to \(\mathrm{H_2O_2}\) and facilitate its subsequent conversion into hydroxyl radicals (•OH) through a photocatalytic cascade reaction with Cu single- atom embedded \(\mathrm{g - C_3N_4}\) nanosheets (Cu/CN). Concurrently, the •OH and superoxide anions (• \(\mathrm{O_2}^-\) ) are obtained by photocatalytic water splitting over Cu/CN. Over \(99.9\%\) antibacterial activity and effective biofilm inhibition are achieved via photocatalytic cascade reaction. In the dark, excess ROS are scavenged by Cu/CN, reducing inflammation of wounds and promoting polarization of M2 macrophages. This photoswitchable cascade reaction effectively treated MDR bacterial- infected diabetic wounds, highlighting its potential for antibiotic- free diabetic wound therapy and its promising prospects for clinical applications.  
+
+<|ref|>sub_title<|/ref|><|det|>[[117, 488, 248, 504]]<|/det|>
+## 1. Introduction  
+
+<|ref|>text<|/ref|><|det|>[[114, 523, 885, 905]]<|/det|>
+Diabetes, along with its complications- particularly non- healing chronic wounds prone to bacterial infections- has emerged as a critical threat to public health, imposing a heavy burden on patients' life quality and social economy. \(^{1, 2}\) The diabetic wounds are characterized by excessive presence of glucose, persistent bacterial colonization, chronic inflammation, increased levels of oxidative stress, and abnormal angiogenesis, etc. \(^{3, 4}\) The rise of bacterial resistance has diminished the efficacy of conventional antibiotics, making bacterial infected diabetic wounds even harder to heal. \(^{5, 6}\) Staphylococcus aureus (Gram- positive) and Escherichia coli (Gram- negative) are the most frequent bacteria isolated from diabetic wounds, the fast emergence and spread of their multidrug- resistant (MDR) strains, such as methicillin- resistant S. aureus (MRSA) and extended- spectrum \(\beta\) - lactamases producing E. coli (ESBL E. coli), have intensified the demand for new antibacterial strategies. \(^{7}\) For example, a case study revealed that MRSA accounted for more than half of S. aureus isolated from diabetic foot ulcers. \(^{8}\) Novel strategies capable of effectively combating MDR bacteria, reducing glucose levels, controlling oxidative stress,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 87, 883, 134]]<|/det|>
+regulating inflammation, and promoting angiogenesis are highly desired for the treatment of diabetic wounds.  
+
+<|ref|>text<|/ref|><|det|>[[113, 144, 885, 612]]<|/det|>
+Glucose oxidase (GOx) has been employed to treat diabetic complications by catalyzing the oxidative depletion of glucose- an essential nutrient for bacterial growth- and generating \(\mathrm{H}_2\mathrm{O}_2\) , which adversely affects bacterial viability. \(^{9,10,11}\) Recently, a number of glucose oxidase- based platforms that effectively promoting diabetic wound healing have been developed. \(^{12,13,14,15,16}\) In addition, nanozymes with the similar activity and wider applicability have been developed to overcome some limitations of natural enzymes. For example, metal- based nanozymes (MnO \(_2\) , Au, TiO \(_2\) , Cu \(_2\) O, and CeO \(_2\) , etc.) have been constructed to mimic GOx activity. \(^{17,18,19}\) However, a challenge with these nanozymes is the lack of precise control over their reaction process, which typically results in non- stop catalysis and the production of excessive ROS at the wound site, thus exacerbating the inflammation and disturbing the wound healing process. \(^{20,21}\) A potential solution strategy involves the use of switchable processing methods that can regulate enzyme activity under specific stimuli. Among various stimulation methods, light stands out due to its spatiotemporal selectivity, enabling precise regulation of the activity of photocatalysis, thus playing a key role in the controllable treatments. For example, He and coworkers used visible light as a switch to activate photocatalytic titanium dioxide, which consumed glucose at the wound site to produce hydrogen and promote wound healing. \(^{22}\)  
+
+<|ref|>text<|/ref|><|det|>[[113, 623, 885, 923]]<|/det|>
+Graphitic carbon nitrides (g- C \(_3\) N \(_4\) ) have demonstrated exceptional and controllable photocatalytic performance, making them widely used in energy, environmental, and biomedical fields. \(^{23,24,25}\) g- C \(_3\) N \(_4\) exhibits the ability to photocatalytically oxidize glucose to generate \(\mathrm{H}_2\mathrm{O}_2\) , and has been employed as a GOx- mimicking photocatalyst for glucose detection. \(^{26,27}\) Defect engineering techniques \(^{28,29,30,31}\) such as carbon and nitrogen vacancies \(^{32,33,34,35}\) , as well as oxygen doping \(^{36,37}\) in g- C \(_3\) N \(_4\) , can effectively improve its photocatalytic performance by altering the electronic band structure \(^{38}\) , optimizing carrier transfer \(^{39}\) , and enhancing surface active sites \(^{40}\) . g- C \(_3\) N \(_4\) typically exhibits a low adsorption rate for \(\mathrm{H}_2\mathrm{O}_2\) and is unable to further consume the \(\mathrm{H}_2\mathrm{O}_2\) produced in the previous step through cascade reaction. By loading metal single- atoms such as Ru \(^{41}\) , Pt \(^{42}\) , and Cu \(^{43}\) onto g- C \(_3\) N \(_4\) , the resulting single- atom embedded g- C \(_3\) N \(_4\) demonstrates enhanced photocatalytic cascade reaction performance,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 87, 883, 190]]<|/det|>
+further consuming \(\mathrm{H}_2\mathrm{O}_2\) to produce ROS that are more effective at killing bacteria. However, few studies have been conducted that combine defect engineering with single- atom embedding to modulate the photocatalytic performance of \(\mathrm{g - C_3N_4}\) for the enhanced treatment of diabetic infected wounds.  
+
+<|ref|>text<|/ref|><|det|>[[114, 202, 886, 585]]<|/det|>
+Herein, we present a Cu single- atom embedded nitrogen vacancy- rich \(\mathrm{g - C_3N_4}\) photocatalyst (Cu/CN), which utilizes a cascade reaction controlled via photoswitchable treatment methods to effectively regulate glucose levels and antibacterial activity while preventing excessive ROS, thereby minimizes the risk of overtreatment and reduces inflammation at the wound site, ultimately enhancing chronic wound healing (Scheme 1). The nitrogen vacancies enhance photocatalytic glucose oxidation, effectively reducing blood glucose at the wound site. Meanwhile, the \(\bullet \mathrm{OH}\) produced from the photocatalytic decomposition of \(\mathrm{H}_2\mathrm{O}_2\) in the preceding step, combined with the \(\bullet \mathrm{OH}\) and \(\bullet \mathrm{O}_2^-\) produced through Cu single- atom- mediated photocatalytic water splitting, effectively eliminate MDR bacteria. In addition, Cu/CN scavenge excess ROS in dark reaction to alleviate inflammation at the wound site, thereby promoting the transition of macrophages from pro- inflammatory M1 phenotype to anti- inflammatory M2 phenotype. This photoswitchable cascade reaction of \(\mathrm{Cu / CN}\) thus represents a promising and innovative approach for treatment of chronic diabetic wounds.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[115, 90, 875, 633]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 650, 844, 693]]<|/det|>
+<center>Scheme 1. Schematic diagram of the preparation of Cu/CN and diabetic wound healing mechanism of Cu/CN. </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[117, 714, 332, 731]]<|/det|>
+## 2. Results and discussion  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 750, 650, 769]]<|/det|>
+### 2.1 Morphology, size, and composition characterization of CN  
+
+<|ref|>text<|/ref|><|det|>[[115, 787, 885, 918]]<|/det|>
+As shown in the scheme 1, CN with graphite- phase structure was prepared by thermal polymerization using melamine as the precursor, and the above materials were improved. To introduce nitrogen vacancies, the material was calcined at different temperatures, decomposing some CN under air atmosphere. \(^{44}\) The SEM images (Figure S1) and TEM images (Figure S2) reveal that the surface of CN becomes increasingly wrinkled and
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 87, 885, 498]]<|/det|>
+porous as the calcination temperature rise. Nitrogen adsorption experiments further confirm that higher calcination temperatures increase in surface area and pore size, potentially enhancing the material's capability to adsorb the reaction substrate (Figure S3 and Table S1). XRD results (Figure 1a) indicate that with increasing heat treatment temperature, the peak intensity of the material decreased at \(13^{\circ}\) (100) crystal plane (stacking of in-plane structural units) and \(27^{\circ}\) (002) crystal plane (stacking of interlayer structures), indicating a reduction in the crystallinity of carbon nitride. The XPS was employed to examine the chemical states of the samples, using the binding energy of the C1s line at \(284.8 \mathrm{eV}\) from alkyl or adventitious carbon as a reference. A comparison of the C 1s and N 1s spectra of \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) and \(\mathrm{CN}_{750}\) is shown in Figure 1bc. Both \(\mathrm{CN}_{550}\) and \(\mathrm{CN}_{700}\) exhibited characteristic binding configurations of carbon and nitrogen that are typical of graphitic carbon nitride. Specifically, in addition to the peak observed at \(284.8 \mathrm{eV}\) , the C1s spectrum for CN featured two additional peaks at \(286.1 \mathrm{eV}\) and \(288.05 \mathrm{eV}\) , which correspond to \(\mathrm{C - NH}_{2}\) and \(\mathrm{C(N)}_{3}\) moieties within the graphitic carbon nitride structure, respectively. \(^{45}\)  
+
+<|ref|>text<|/ref|><|det|>[[114, 513, 885, 923]]<|/det|>
+After heat treatment, the ultraviolet- visible absorption strength of the material was enhanced (Figure. 1d). The strong electron paramagnetic resonance (EPR) signal peak with a g value of 2.004 was clearly observed (Figure. 1e). As the heat treatment temperature increased, the N vacancies in the material first increase, peaking at \(\mathrm{CN}_{700}\) , and then decrease. FTIR provided further insights into these defects' states. The spectra obtained from the samples (Figure. 1f) displayed distinct peaks characteristic of graphitic carbon nitride. A peak around \(810 \mathrm{cm}^{- 1}\) was observed, indicating out- of- plane bending vibrations in heptazine ring systems. Additionally, peaks in the range of 900 \(\mathrm{cm}^{- 1}\) to \(1800 \mathrm{cm}^{- 1}\) were noted, which are related to \(\mathrm{C - N(- C) - C}\) vibrations within the heterocycles or bridging \(\mathrm{C - NH - C}\) units. Broad peaks spanning from \(3000 \mathrm{cm}^{- 1}\) to \(3500 \mathrm{cm}^{- 1}\) were attributed to the \(\mathrm{NH / NH}_{2}\) group. However, when the CN samples were subjected to rapid thermal treatment, marked changes were evident in the FTIR signals. A prominent peak appeared at \(2175 \mathrm{cm}^{- 1}\) , corresponding to \(- \mathrm{C} \equiv \mathrm{N}\) (cyano groups), and its intensity increased with higher treatment temperature increased. The emergence of these cyano groups disrupts the hydrogen bonds between the polymeric melon strands,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 87, 881, 134]]<|/det|>
+causing layers fluctuation and the collapse of the periodic stacking structure, as supported by the XRD findings.  
+
+<|ref|>image<|/ref|><|det|>[[115, 153, 884, 494]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 514, 881, 586]]<|/det|>
+<center>Figure 1. Composition characterization of CN. XRD patterns (a), C 1s XPS spectra (b), N 1s XPS spectra (c), UV-vis spectra (d), EPR spectra (e) and FT-IR spectra (f) of CN with different nitrogen vacancy concentrations. </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[117, 602, 436, 620]]<|/det|>
+### 2.2 VIS-photocatalytic performances  
+
+<|ref|>text<|/ref|><|det|>[[115, 637, 884, 910]]<|/det|>
+CN with varying concentrations of N vacancies was used to investigate the photocatalytic activity of glucose consumption, and it was found that \(\mathrm{CN}_{700}\) , with the highest concentration of N vacancies, exhibited the best photocatalytic activity (Figure 2a). Therefore, \(\mathrm{CN}_{700}\) was selected for loading Cu in subsequent experiments. Additionally, to assess the photocatalyst's capability, measurements of the transient photocurrent responses were conducted. As shown in Figure 2b, all samples exhibit reproducible and stable photocurrent signals, with photocurrent density directly proportional to the nitrogen vacancy density. \(\mathrm{CN}_{700}\) showed the highest photocurrent intensity, indicating a lower electron-hole pairs and improved migration efficiency. This is likely due to N vacancies acting as capture centers for photogenerated holes, which
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 78, 885, 696]]<|/det|>
+inhibit electron- hole recombination and promote separation, thus extending charge carrier lifetime. \(^{46, 47}\) Therefore, more photogenerated electrons and holes in \(\mathrm{CN}_{700}\) contribute to the photocatalytic consumption of glucose. Electron- transfer resistance was further investigated using electrochemical impedance spectroscopy (EIS). Typically, a smaller arc radius on the EIS Nyquist plot signifies reduced interfacial contact resistance between the electrode and the electrolyte. \(^{45}\) As shown in Figure 2c, the radius is inversely proportional to the N vacancy density, with \(\mathrm{CN}_{700}\) exhibiting the smallest radius of circularity, suggesting that its lower electron- transfer resistance facilitates carriers separation. According to \((\mathrm{Ahv})^2\) and the tangent intercept of the curve of light energy, the bandgap energy (Eg) was calculated. The Eg values of \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) were 2.8, 2.8, 2.77, 2.73, and 2.78 eV, respectively, indicating that the introduction of nitrogen vacancies slightly reduced the band gap of \(\mathrm{CN}_{700}\) and moderately increased its absorbance, consistent with UV- visible spectrum results. To further verify that the N vacancy modified samples enhance carrier separation, the charge carrier density \((\mathrm{N_d})\) was evaluated using Mott- Schottky (M- S) curves. Figure 2e reveals that the M- S curve has a positive slope, which is characteristic of n- type semiconductors like \(\mathrm{CN}_{700}\) , allowing \(\mathrm{N_d}\) values to be calculated by the Equation (1). The \(\mathrm{N_d}\) values for the \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) were estimated to be \(1.8522 \times 10^{21}\) , \(2.3153 \times 10^{21}\) , \(2.3453 \times 10^{21}\) , \(3.2008 \times 10^{21}\) , and \(2.0185 \times 10^{21}\) cm\(^{- 3}\) , respectively (Table S2). These \(\mathrm{N_d}\) values correlated with N vacancy density, with \(\mathrm{CN}_{700}\) having the highest \(\mathrm{N_d}\) , indicating that N vacancies in the catalyst can increase the carrier density.  
+
+<|ref|>text<|/ref|><|det|>[[115, 708, 885, 923]]<|/det|>
+Moreover, the catalyst's potential relative to the standard hydrogen electrode (SHE), was calculated from the Nernst equation (2) and the Mott- Schottky curve, allowing determination of the conduction band (CB) position. The intersection of the tangent with the Mott- Schottky curve's x- axis represents the catalyst's potential relative to the Ag/AgCl electrode; for n- type semiconductors, the CB position is 0.2 V lower than this potential. The CB positions of catalysts \(\mathrm{CN}_{550}\) , \(\mathrm{CN}_{600}\) , \(\mathrm{CN}_{650}\) , \(\mathrm{CN}_{700}\) , and \(\mathrm{CN}_{750}\) (relative to the SHE: \(- 0.17\) , \(- 0.19\) , \(- 0.26\) , \(- 0.19\) and \(- 0.37\) V, respectively) are shown in Table S3. Combining these \(\mathrm{E_g}\) values with the corresponding catalysts' Tauc curves (Figure
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 87, 884, 219]]<|/det|>
+2f), the valence band potentials are determined to be 2.63, 2.61, 2.50, 2.54, and 2.40 V, respectively. Thus, as N vacancies increase, the bandgap width narrows, consistent with the PBE calculation results (Figure S4). This narrowing allows \(\mathrm{CN}_{700}\) to utilize more low-energy photons, facilitating the photocatalytic consumption of glucose, which aligns with it observed the photocatalytic activity.  
+
+<|ref|>image<|/ref|><|det|>[[114, 232, 884, 574]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 593, 884, 750]]<|/det|>
+<center>Figure 2. Photoelectrochemical behaviours of modified CN. (a) Photocatalytic properties of different catalysts for glucose consumption. (b) Transient photocurrent responses. (c) Electrochemical impedance spectra (EIS). In the simulated electrical equivalent-circuit model (inset), RS, R1, and CPE represent as solution resistance, charge transfer resistance, and double layer capacitance, respectively. (d) The plots of \((\mathrm{Ahv})^2\) and \(hv\) of different catalysts. (e) Mott-Schottky curves of different catalysts. (f) Schematic illustration of band structure for the different catalysts. </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 765, 682, 784]]<|/det|>
+### 2.3 Morphology, size, and composition characterization of Cu/CN  
+
+<|ref|>text<|/ref|><|det|>[[115, 801, 884, 932]]<|/det|>
+CN rich in nitrogen vacancies was prepared by thermal etching, and single- atom Cu- dispersed Cu/CN was synthesized using impregnation method. From the kinetics of \(\mathrm{H}_2\mathrm{O}_2\) consumption by Cu/CN photocatalysis, it was found that \(7.4 \mu \mathrm{g} / \mathrm{L}\) of \(\mathrm{H}_2\mathrm{O}_2\) was consumed in 10 minutes (Figure S4). The visible light absorption intensity of Cu/CN in the UV- vis spectrum is stronger than \(\mathrm{CN}_{700}\) (Figure 3a), indicating that Cu/CN is more
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 78, 885, 917]]<|/det|>
+efficient in utilizing visible light for photocatalytic reaction, consistent with the PBE calculation results (Figure S5). The FT- IR spectra of Cu/CN show a reduced intensity in C- N vibration within the heterocycle, possibly due to the influence of Cu single- atom on the vibration modes (Figure 3b). The XRD pattern of Cu/CN shows no Cu crystal peaks (Figure 3c), indicating that Cu single- atom does not disrupt the material's crystal structure. The Cu 2p XPS spectra of Cu/CN confirm the presence of both +1 and +2 oxidation states (Figure 3d).<sup>48</sup> To further investigate the oxidation state of the Cu single atom in Cu/CN, we analyzed X- ray absorption energy near- edge structure (XANES) of Cu foil, Cu<sub>2</sub>O, CuO, and Cu/CN (Figure 3e). The findings suggest that the XANES spectrum of Cu/CN falls between the spectra of Cu<sub>2</sub>O and CuO, indicating that the oxidation state of the single copper atom is between +1 and +2. From the slope and valence plot, the valency of Cu was estimated to be +1.75 (Figure 3f). Extended X- ray absorption fine structure (EXAFS) was investigated to examine the coordination environment of the Cu sites. Quantitative EXAFS curve fitting analysis in R spaces was employed to characterize the coordination structure of the Cu atoms (Figure 3h and Figure S6). The first coordination shell, observed at 1.5 Å in Figure 3g, suggests the presence of Cu- N or/and Cu- O coordination. No significant Cu- Cu scattering was detected at 2.2 Å in Cu/CN samples, further confirming the single- atom dispersion of Cu. Wavelet transform of the EXAFS data distinguished backscattering atoms (Figure 3i), revealing maximum intensities at 7.2 and 10.6 Å<sup>-1</sup> for Cu foil and CuO, respectively, which correspond to the Cu- Cu configuration. In contrast, Cu/CN displayed a maximum intensity at 4.5 Å<sup>-1</sup>, attributed to Cu- N or/and Cu- O coordination. The N species produced from melamine during pyrolysis help fill vacancies, stabilizing the atomic Cu. EXAFS fitting results show the coordination number (CN) of Cu- N was approximately 4. Determining the specific CN for nitrogen and oxygen, however, is challenging due to their similar bond lengths to Cu. Based on previous studies, both N and O back- scattering paths were included for optimal fit; comparison with fixed CN values for N and O were also made (Table S4). By fitting the EXAFS spectra to different structural models that were optimized using DFT (Figure 4), we found that the optimal fit was achieved when the Cu was coordinated with 4 N atoms, as shown in the atomic Cu structure model for
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 87, 885, 384]]<|/det|>
+Cu/CN in the inset of Scheme 1. We also measured the EPR spectra to analyze the in- situ generation of \(\bullet \mathrm{OH}\) and \(\bullet \mathrm{O}_2^-\) by Cu/CN with and without light exposure (Figure S7). Under light, Cu/CN exhibited superior performance in photocatalytic decomposition of \(\mathrm{H}_2\mathrm{O}_2\) to generate \(\bullet \mathrm{OH}\) compared to \(\mathrm{CN}_{700}\) . Additionally, Cu/CN can directly photocatalyze water splitting to produce \(\bullet \mathrm{OH}\) and oxidize \(\mathrm{O}_2\) to produce \(\bullet \mathrm{O}_2^-\) , making it suitable for ROS generation for antibacterial purposes. However, excessive ROS at the wound site can cause inflammation, hindering the healing process. Utilizing DPPH (Figure S8), ABTS (Figure S9) assay, as well as RAW264.7 intracellular ROS scavenging experiments (Figure S10), it has been demonstrated that Cu/CN can effectively scavenge excessive ROS during dark reaction, independent of light exposure. \(^{49,50}\)  
+
+<|ref|>image<|/ref|><|det|>[[113, 398, 884, 770]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 784, 884, 883]]<|/det|>
+<center>Figure 3. Structural characterizations of Cu/CN. (a) UV-vis spectra of Cu/CN. (b) FT-IR spectra of Cu/CN. (c) XRD patterns of Cu/CN. (d) Cu 2p XPS spectra of Cu/CN. (e) XANES spectra at Cu K-edge of Cu/CN. (f) Standard curve of Cu valence state. (g) EXAFS spectra at Cu K-edge of Cu/CN. (h) EXAFS fitting result of Cu/CN at R space. (i) Wavelet transform of Cu foil, \(\mathrm{Cu}_2\mathrm{O}\) , \(\mathrm{CuO}\) and \(\mathrm{Cu/CN}\) . </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 899, 277, 916]]<|/det|>
+### 2.4 DFT of Cu/CN
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 87, 885, 526]]<|/det|>
+Density functional theory (DFT) calculations were conducted to investigate the effects of nitrogen- containing vacancies in CN and Cu single- atom in the adsorption energies of glucose and \(\mathrm{H}_2\mathrm{O}_2\) on Cu/CN. For glucose, there are two types of adsorption sites on the CN surface: i) glucose adsorbs at non- vacancy sites with an adsorption energy of \(- 0.15\) eV, and ii) glucose adsorbs at nitrogen vacancy sites with an adsorption energy of \(- 0.42\) eV (Figure 4a). For \(\mathrm{H}_2\mathrm{O}_2\) , there are three types of adsorption sites on the Cu/CN surface of: i) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs at nitrogen vacancies, ii) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs between nitrogen vacancies and Cu single- atom sites, and iii) \(\mathrm{H}_2\mathrm{O}_2\) adsorbs at Cu single- atom, with corresponding adsorption energies of \(- 0.26\) eV, \(- 0.74\) eV and \(- 1.39\) eV, respectively (Figure 4b). The more negative the adsorption energy, the more stable the adsorption, indicating that the presence of nitrogen vacancies enhance the catalyst's glucose adsorption capacity. Furthermore, charge transfer between CN and glucose was visualized using the Charge Density Difference (CDD) analysis. Figures S11 and S12 show electrons accumulation (yellow) at non- nitrogen vacancies site and electrons depletion (blue) at nitrogen vacancies, indicating that the presence of nitrogen vacancies are more effective in oxidizing glucose.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 88, 875, 505]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[114, 523, 884, 595]]<|/det|>
+<center>Figure 4. DFT of Cu/CN. (a) Adsorption energy of glucose at non vacancy and nitrogen vacancy sites on CN surface. (b) Adsorption energy of \(\mathrm{H}_2\mathrm{O}_2\) at nitrogen vacancy, non-vacancy, and Cu single-atom sites on Cu/CN surface. </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 611, 500, 629]]<|/det|>
+### 2.5 In vitro antibacterial property of Cu/CN  
+
+<|ref|>text<|/ref|><|det|>[[114, 647, 884, 918]]<|/det|>
+Motivated by the theoretical and experimental studies above, we hypothesize that Cu/CN can act as an effective bactericidal agent when glucose present via a photocatalytic cascade reaction. To validate this concept, we conducted an in vitro antibacterial study using ESBL E. coli and MRSA as model bacteria. The results, presented in Figure 5a and 5c, d demonstrate that the bactericidal rates for both strains were significantly higher when treated with Cu/CN+vis compared to the control, GOx, and \(\mathrm{CN}_{700}\) groups, which were \(99.90\%\) and \(99.95\%\) , respectively. Notably, a significant difference (P \(< 0.5\) ) between GOx and \(\mathrm{CN}_{700}\) +vis groups (Figure 5b and 5d) was observed, which can be attributed to the enhanced bactericidal performance resulting from the \(\bullet \mathrm{OH}\) generated by the cascade catalysis. However, the cascade catalytic activity of \(\mathrm{CN}_{700}\) was relatively
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 82, 885, 592]]<|/det|>
+low, resulting in a lower bactericidal rate than the \(\mathrm{Cu / CN + }\) - vis experimental group. This result is supported by the in situ EPR detection of the \(\bullet \mathrm{OH}\) activity produced by \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) photocatalysis (Figure S10). Furthermore, SEM images (Figure 5e) provide additional evidence, showing that MRSA and ESBL E. coli in the control group maintained their original plump structures with intact outer membranes, while treatment with GOx caused some structural distortion due to oxidative damage from \(\mathrm{H}_2\mathrm{O}_2\) .51 By contrast, bacteria treated with \(\mathrm{CN}_{700} + \mathrm{vis}\) and \(\mathrm{Cu / CN + }\) - vis exhibited significant morphological deformations, such as collapse, distortion, and breakage (highlighted by red arrows), indicating the highest bactericidal efficacy. Additionally, we also investigated the anti- bacterial biofilm capability of \(\mathrm{Cu / CN}\) under Xe light \((\lambda > 420 \mathrm{nm})\) radiation. As shown from Figure 5f, confocal laser scanning microscopy (CLSM) images of ESBL E.coli and MRSA biofilms showed a red fluorescence signal, confirming that \(\mathrm{Cu / CN}\) under \(\mathrm{Xe}(\lambda > 420 \mathrm{nm})\) light irradiation exhibits excellent antibiofilm performance. Therefore, the \(\mathrm{Cu / CN}\) photocatalytic therapeutic platform, which utilizes a photoswitch to control the generation of ROS to combat MDR bacteria, is highly effective. This platform operates through: i) photocatalytic cascade reaction that consume glucose to produce \(\bullet \mathrm{OH}\) , ii) photolysis of water to generate \(\bullet \mathrm{OH}\) , and iii) oxidation of \(\mathrm{O}_2\) to form \(\bullet \mathrm{O}_2\) (Figure 5g).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 88, 848, 710]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 737, 884, 840]]<|/det|>
+<center>Figure 5. Evaluation of antibacterial capacity. Photographs of bacterial colonies formed by (a) ESBL E. coli and (c) MRSA after treatments with Cu/CN. The relative bacterial viability of (b) ESBL E. coli and (d) MRSA. Morphologies of (e) ESBL E. coli and MRSA treated or untreated with the Cu/CN group. (f) CLSM photos of biofilm treated with Cu/CN. (g) Schematic diagram of photocatalytic antibacterial control by photoswitch. </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 858, 741, 877]]<|/det|>
+### 2.6 Hemocompatibility and cytocompatibility investigation of the Cu/CN
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 80, 885, 639]]<|/det|>
+Low cytotoxicity is essential for applying nanomaterials to skin wounds. \(^{52}\) The proliferation of 3T3 fibroblasts was assessed in vitro across different groups: control, GOx, CN, and Cu/CN treated ones (Figure 6a). After 24 hours of cultivation, the 3T3 fibroblast cells in all groups exhibited healthy growth and proliferation, indicating excellent cytocompatibility of Cu/CN. Additionally, after 24 hours, the viability of 3T3 fibroblast cells incubated with the CN and Cu/CN remained above \(95\%\) , as determined through Alamar Blue fluorescence testing (Figure 6b). These results suggest that Cu/CN is well- suited for treating chronic diabetic wounds due to its high biocompatibility. The hemolysis assay evaluated the effect of CN and Cu/CN on red blood cells (RBCs). As seen in Figure 6c, Triton X- 100 acted as the positive control, displaying red color due to complete hemolysis. In contrast, RBCs treated with GOx, CN, and Cu/CN appeared pale yellow and transparent, similar to Tris- HCl, the negative control. Hemolysis ratios of GOx, CN, and Cu/CN were \(1.5\%\) , \(2.8\%\) , and \(3.3\%\) , respectively, all below the permissible limit of \(5\%\) . This indicates that Cu/CN has good hemocompatibility and is suitable for treating diabetic wounds. The cell scratch test further examined the influence of CN and Cu/CN on cell migration. As shown in Figure 6d, the cell migration area after 48 hours of Cu/CN treatment was larger than in the other three groups, with the migration rates of \(68.0\%\) , \(45.8\%\) , \(75.4\%\) , and \(90.4\%\) respectively (Figure 6e). Cells treated with Cu/CN showed the highest cell migration rate, likely due to trace amounts of Cu promoting cell migration. \(^{53}\) Summarily, the excellent cell safety, anti- hemolytic and pro- migration effects of Cu/CN are highly favorable for diabetic wound healing.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[128, 81, 870, 562]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 570, 884, 727]]<|/det|>
+<center>Figure 6. In vitro biocompatible evaluation of \(\mathrm{Cu / CN}\) . (a) Fluorescence images of 3T3 fibroblast cells cultured with hydrogel groups after 24 hours. (b) Cell viability of the \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) . (c) Hemolysis ratio of the \(\mathrm{CN}_{700}\) and \(\mathrm{Cu / CN}\) . (d) Cell migration of HUVEC cells in the control and the different groups at 0, 24, and 48 hours. (e) The cellular migration after various treatments for different time periods was identified by drawing the green shadow at the edge of cells in (d), and calculate the area of cell migration (\*p<0.5; \*\*p<0.01; \*\*\*p<0.001; \*\*\*\*p<0.0001). </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 746, 544, 764]]<|/det|>
+### 2.7 In vivo wound healing performance of \(\mathrm{Cu / CN}\)  
+
+<|ref|>text<|/ref|><|det|>[[115, 782, 884, 912]]<|/det|>
+A diabetic model in KM mice was established using STZ, selecting mice with blood glucose levels consistently above \(16.7 \mathrm{mM}\) for one week post- induction to create a wound model for subsequent treatment. \(^{14,54}\) Full- thickness dorsal skin punch wounds (6 mm in diameter) were created with a round punch, infected by MRSA to induce ulceration, and treated according to different protocols. The mice were randomly
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 80, 885, 694]]<|/det|>
+allocated into ten distinct treatment groups, each with five mice: control, GOx, CN, and Cu/CN. Each group was treated with or without Xe light radiation ( \(\lambda > 420 \mathrm{nm}\) , 15 minutes). Photographs illustrating wound repair across treatment group are shown in Figure 7a, and quantitative analyses of wound area reduction in diabetic wounds are provided in Figure 7b. Remarkably, wounds treated with Cu/CN under Xe light irradiation demonstrated a significantly accelerated healing rate by day 7 compared to the other groups. The wound area ratios were \(62.8\%\) for the control, \(25.5\%\) for GOx, \(16.4\%\) for CN, and \(2.8\%\) for Cu/CN (Figure S13Figure 7e). In the absence of light, minimal difference observed in wound healing, indicating that Cu/CN requires NIR irradiation for effective activity (Figure S14). On day 14, wound tissue was collected for bacterial culture. As shown in Figure 7c, the control group exhibited a high concentration of MRSA, which impeded the wound healing process, whereas the Cu/CN group experienced a substantial reduction in bacterial counts. Notably, all MRSA bacteria on the skin treated with the Cu/CN group were eradicated, underscoring Cu/CN's efficacy in enhancing wound healing and providing photocatalytic antibacterial action in vivo. Cu/CN exhibits effective antibacterial activity upon exposure to Xe light ( \(\lambda > 420 \mathrm{nm}\) ) irradiation, while its nitrogen vacancy-rich CN component photocatalytically consumed glucose, generating \(\mathrm{H}_2\mathrm{O}_2\) through to normalize the glucose levels in diabetic wounds. Cu single- atom further catalyzed \(\mathrm{H}_2\mathrm{O}_2\) produce \(\bullet \mathrm{OH}\) for antibacterial purposes. The enhanced diabetic wound healing observed with Cu/CN treatment highlights its potential for combining hypoglycemic and antibacterial properties in MDR- infected diabetic wound healing applications.  
+
+<|ref|>text<|/ref|><|det|>[[115, 708, 885, 923]]<|/det|>
+The transition from the inflammatory to the proliferative stage is a key regulatory point in wound healing. Persistent inflammation is a prominent feature of chronic diabetic wounds. \(^{55}\) On day 14 of treatment, histological analyses were as performed. H&E and Masson staining of the wound tissue were utilized to examine the inflammatory reaction and evaluate collagen formation and deposition in wounds treated with control, GOx, CN, and Cu/CN, while immunofluorescence staining for cluster of differentiation 31 (CD31), tumor necrosis factor- alpha (TNF- \(\alpha\) ), and interleukin- 10 (IL- 10) was employed to assess angiogenesis and inflammation levels in the healed wounds of these treatment groups (Figure 7d). In the Cu/CN
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 80, 885, 666]]<|/det|>
+treated group, complete wound healing was achieved by day 14, with numerous fibroblasts visible under the skin (Figure 7d). In contrast, wounds in the other groups retained a considerable number of inflammatory cells and showed incomplete healing. CD31 is commonly used to detect the presence of neovascularization in granulation tissue. As depicted in Figure 7d, the Cu/CN- treated group exhibited the highest number of vascular growths on days 14 compared to the other three groups, as indicated by CD31 immunohistochemical staining, with CD31 positivity reaching \(6.6\%\) (Figure 7e and Figure S15). TNF- \(\alpha\) , an inflammatory cytokine, can inhibit wound healing in diabetic conditions, while the anti- inflammatory M2 phenotype is usually activated by the Th2 cytokines of IL- 10. \(^{56}\) TNF- \(\alpha\) exacerbates inflammation by initiating inflammatory cascades, whereas IL- 10 acts as an anti- inflammatory cytokine by suppressing effector T cell activity. \(^{57}\) Immunohistochemical staining (Figure 7d) revealed that Cu/CN effectively decreased TNF- \(\alpha\) expression while increasing IL- 10 expression in inflammatory cells. On day 14, the positive proportion of TNF- \(\alpha\) decreased to \(1.0\%\) , while IL- 10 positivity increased to \(5.3\%\) (Figure 7e and Figure S15). This controlled treatment process is achieved by using a photocatalytic light switch, allowing ROS generation to be halted after irradiation, thus avoiding increased inflammation. Furthermore, after phototherapy completed, Cu/CN can scavenge excess ROS in the dark reaction, without the need for light. These findings suggest that Cu/CN promotes the transition of macrophages from the pro- inflammatory M1 to the anti- inflammatory M2 phenotype and modulates inflammatory cytokine expression (Figure 7f). This modulation led to anti- inflammatory effects, enhancing the capacity for skin repair in diabetic wounds.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[135, 80, 860, 731]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 745, 884, 928]]<|/det|>
+<center>Figure 7. Evaluation of wound healing in diabetic mouse models. (a). Photographs of the wounds of wound closure treat with light \((\lambda >420 \mathrm{nm})\) . (b) Wound size ratios with different treatments at days 0, 3, 7, and 14. (c) Typical agar plate photos for the remaining MRSA colonies in the ulcer after different treatment on day 14. (d) Immunohistochemical studies of diabetic wound collected at day 14. Scale bar: \(1000 \mu \mathrm{m}\) for H&E and Masson, \(100 \mu \mathrm{m}\) for TNF- \(\alpha\) , IL-10 and CD31. (e) Overall performance of control, GOx, CN, and Cu/CN in the in vivo diabetic wound treatment. (f) Schematic diagram of photocatalytic therapy control by photoswitch. </center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 90, 235, 106]]<|/det|>
+## 3. Conclusion  
+
+<|ref|>text<|/ref|><|det|>[[114, 120, 885, 675]]<|/det|>
+In conclusion, we have successfully developed a Cu/CN bionanomaterial that can effectively treat MDR bacteria infected diabetic wounds by simultaneously regulating blood glucose levels, bacterial infection, and persistent inflammation. The synergistic effects of nitrogen vacancies and single atoms in Cu/CN enabled the realization of a photoswitchable cascade reaction. The introduction of nitrogen vacancies enhances the activity of photocatalytic oxidation for glucose consumption, which helps to reduce blood glucose levels at the wound sites. Subsequently, the \(\mathrm{H}_2\mathrm{O}_2\) generated in the previous step is converted into •OH through a Cu single- atom photocatalytic cascade reaction. Concurrently, Cu/CN can directly produce •OH and • \(\mathrm{O}_2^{- }\) through photocatalytic water splitting. DFT calculations confirmed that glucose preferentially adsorbs at nitrogen vacancies, while \(\mathrm{H}_2\mathrm{O}_2\) adsorbs more readily at Cu single- atom. In vitro antibacterial results demonstrated that the Cu/CN photocatalyst had significant bactericidal effects on drug- resistant MRSA and ESBL E. coli and exhibited anti- biofilm activity. The in vivo study in a mice diabetic wound MRSA infection model shows that Cu/CN photocatalyst has good biological safety and excellent therapeutic efficacy with visible light irradiation. The excess ROS can be scavenged by Cu/CN without light, alleviating inflammation at the wound site and encouraging the transition of macrophages from the pro- inflammatory M1 phenotype to the anti- inflammatory M2 phenotype. This work presents a proof of concept for using a photoswitchable cascade reaction to treat MDR bacterial infected diabetic wound complications.  
+
+<|ref|>sub_title<|/ref|><|det|>[[118, 728, 285, 744]]<|/det|>
+## Acknowledgements  
+
+<|ref|>text<|/ref|><|det|>[[115, 763, 884, 895]]<|/det|>
+This research was sponsored by the key research program of Ningbo (Grant No. 2023Z210), the Ningbo Natural Science Foundation (Grant No. 202003N4006), the National Foreign Expert Project (Grant No. H20240307), the National Natural Science Foundation of China (Grant No. 52473265), and the Shaanxi Provincial Science Fund for Distinguished Young Scholars (2023- JC- JQ- 32).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 88, 584, 162]]<|/det|>
+Received: ((will be filled in by the editorial staff))Revised: ((will be filled in by the editorial staff))Published online: ((will be filled in by the editorial staff))  
+
+<|ref|>sub_title<|/ref|><|det|>[[116, 209, 213, 225]]<|/det|>
+## References  
+
+<|ref|>text<|/ref|><|det|>[[110, 240, 880, 916]]<|/det|>
+1. Armstrong DG, Tan T-W, Boulton AJM, Bus SA. Diabetic Foot Ulcers: A Review. JAMA-Journal of the American Medical Association 330, 62-75 (2023).  
+2. Falanga V, et al. Chronic wounds. Nature Reviews Disease Primers 8, 50-70 (2022).  
+3. Wang T, Li YR, Cornel EJ, Li C, Du JZ. Combined Antioxidant-Antibiotic Treatment for Effectively Healing Infected Diabetic Wounds Based on Polymer Vesicles. ACS Nano 15, 9027-9038 (2021).  
+4. Pranantyo D, et al. Hydrogel dressings with intrinsic antibiofilm and antioxidative dual functionalities accelerate infected diabetic wound healing. Nature Communications 15, 954-973 (2024).  
+5. Li JJ, et al. Lactose azocalixarene drug delivery system for the treatment of multidrug-resistant pseudomonas aeruginosa infected diabetic ulcer. Nature Communications 13, 6279-6291 (2022).  
+6. Zhang R, et al. Enhancement of antibacterial and wound healing capabilities of cotton fabric through the covalent bonding of antibiotic-loaded hydrophobized mesoporous silica nanoparticles. Fundamental Research, (2023).  
+7. Idrees M, et al. Bacterial spectrum from diabetic foot Ulcers: A study of antibiotic resistance patterns and phylogenetic diversity. Journal of King Saud University-Science 36, 103320-103330 (2024).  
+8. Wang SH, et al. Meticillin-resistant Staphylococcus aureus isolated from foot ulcers in diabetic patients in a Chinese care hospital: risk factors for infection and prevalence. Journal of Medical Microbiology 59, 1219-1224 (2010).  
+9. Chen LF, et al. A Glucose-Powered Activatable Nanozyme Breaking pH and H2O2 Limitations for Treating Diabetic Infections. Angewandte Chemie-International Edition 60, 23534-23539 (2021).  
+10. He SB, et al. Glucose Oxidase Energized Osmium with Dual-Active Centers and Triple Enzyme Activities for Infected Diabetic Wound Management. Advanced Healthcare Materials 13, 2303548-2303561 (2024).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 83, 870, 140]]<|/det|>
+11. Li YF, et al. A Guanosine-Quadruplex Hydrogel as Cascade Reaction Container Consuming Endogenous Glucose for Infected Wound Treatment-A Study in Diabetic Mice. Advanced Science 9, 2103485-2103498 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 157, 870, 194]]<|/det|>
+12. Liao YH, et al. Glucose oxidase: An emerging multidimensional treatment option for diabetic wound healing. Bioactive Materials 44, 131-151 (2025).  
+
+<|ref|>text<|/ref|><|det|>[[115, 213, 870, 250]]<|/det|>
+13. Shang L, et al. Ultrasound-Augmented Multienzyme-like Nanozyme Hydrogel Spray for Promoting Diabetic Wound Healing. ACS Nano 17, 15962-15977 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 268, 880, 343]]<|/det|>
+14. Sun XC, et al. Multifunctional Hydrogel Containing Oxygen Vacancy-Rich \(\mathrm{WO}_x\) for Synergistic Photocatalytic \(\mathrm{O}_2\) Production and Photothermal Therapy Promoting Bacteria-Infected Diabetic Wound Healing. Advanced Functional Materials, 2411117-2411132 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 361, 878, 417]]<|/det|>
+15. Deng MY, et al. Diabetes immunity-modulated multifunctional hydrogel with cascade enzyme catalytic activity for bacterial wound treatment. Biomaterials 289, 121790-121800 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 435, 850, 491]]<|/det|>
+16. Li HS, Li BY, Lv DL, Li WH, Lu YF, Luo GX. Biomaterials releasing drug responsively to promote wound healing via regulation of pathological microenvironment. Advanced Drug Delivery Reviews 196, 114778-114807 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 509, 858, 547]]<|/det|>
+17. Huang Y, Ren J, Qu X. Nanozymes: Classification, Catalytic Mechanisms, Activity Regulation, and Applications. Chemical Reviews 119, 4357-4412 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[115, 564, 880, 602]]<|/det|>
+18. Min S, et al. Nanomaterials with Glucose Oxidase-Mimicking Activity for Biomedical Applications. Molecules 28, (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 620, 858, 676]]<|/det|>
+19. Chong Y, Liu Q, Ge C. Advances in oxidase-mimicking nanozymes: Classification, activity regulation and biomedical applications. Nano Today 37, 101076-101092 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[115, 694, 833, 750]]<|/det|>
+20. Wang M, et al. Collaboration in Contradiction: Self - Adaptive Synergistic ROS Generation and Scavenge Balancing Strategies Used for the Infected Wounds Treatment. Advanced Healthcare Materials, 2402579-2402596 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 769, 878, 824]]<|/det|>
+21. Zhang X, et al. Polarization of macrophages to an anti-cancer phenotype through in situ uncaging of a TLR 7/8 agonist using bioorthogonal nanozymes. Chemical Science 15, 2486-2494 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 843, 860, 880]]<|/det|>
+22. Chen S, et al. Photocatalytic glucose depletion and hydrogen generation for diabetic wound healing. Nature Communications 13, 5684-5695 (2022).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 83, 833, 120]]<|/det|>
+23. Liao GF, et al. Emerging graphitic carbon nitride-based materials for biomedical applications. Progress in Materials Science 112, 100666-100687 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[115, 139, 861, 194]]<|/det|>
+24. Kong XY, Liu XM, Zheng YF, Chu PK, Zhang Y, Wu SL. Graphitic carbon nitride-based materials for photocatalytic antibacterial application. Materials Science & Engineering R-Reports 145, 100610-100639 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[115, 214, 831, 250]]<|/det|>
+25. Chen Y, Tan CL, Zhang H, Wang LZ. Two-dimensional graphene analogues for biomedical applications. Chemical Society Reviews 44, 2681-2701 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[115, 269, 875, 305]]<|/det|>
+26. Shi L, et al. Photoassisted Construction of Holey Defective g-C3N4 Photocatalysts for Efficient Visible-Light-Driven H2O2 Production. Small 14, 1703142-1703151 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[115, 324, 877, 379]]<|/det|>
+27. Zhang P, et al. Modified carbon nitride nanozyme as bifunctional glucose oxidase-peroxidase for metal-free bioinspired cascade photocatalysis. Nature Communications 10, 940-954 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[115, 398, 881, 453]]<|/det|>
+28. Niu P, Yin LC, Yang YQ, Liu G, Cheng HM. Increasing the visible light absorption of graphitic carbon nitride (melon) photocatalysts by homogeneous self-modification with nitrogen vacancies. Advanced Materials 26, 8046-8052 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[115, 472, 876, 527]]<|/det|>
+29. Yu H, et al. Alkali-Assisted Synthesis of Nitrogen Deficient Graphitic Carbon Nitride with Tunable Band Structures for Efficient Visible-Light-Driven Hydrogen Evolution. Advanced Materials 29, 1605148-1605155 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[115, 546, 857, 601]]<|/det|>
+30. Lin Z, Wang X. Nanostructure engineering and doping of conjugated carbon nitride semiconductors for hydrogen photosynthesis. Angewandte Chemie-International Edition 52, 1735-1738 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[115, 620, 880, 675]]<|/det|>
+31. Kang Y, Yang Y, Yin LC, Kang X, Liu G, Cheng HM. An Amorphous Carbon Nitride Photocatalyst with Greatly Extended Visible-Light-Responsive Range for Photocatalytic Hydrogen Generation. Advanced Materials 27, 4572-4577 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[115, 694, 828, 767]]<|/det|>
+32. Yang B, et al. Ultrathin porous carbon nitride nanosheets with well-tuned band structures via carbon vacancies and oxygen doping for significantly boosting H₂ production. Applied Catalysis B-Environment and Energy 314, 121521-121533 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 788, 863, 861]]<|/det|>
+33. Wang XY, et al. Controllable Approach to Carbon-Deficient and Oxygen-Doped Graphitic Carbon Nitride: Robust Photocatalyst Against Recalcitrant Organic Pollutants and the Mechanism Insight. Advanced Functional Materials 31, 2010763-2010779 (2021).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 83, 861, 140]]<|/det|>
+34. Zhang YZ, et al. Synergistic effect of nitrogen vacancy on ultrathin graphitic carbon nitride porous nanosheets for highly efficient photocatalytic \(\mathrm{H}_2\) evolution. Chemical Engineering Journal 431, 134101-134109 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 157, 833, 213]]<|/det|>
+35. Li JN, et al. Efficient photosynthesis of \(\mathrm{H}_2\mathrm{O}_2\) via two-electron oxygen reduction reaction by defective \(\mathrm{g - C_3N_4}\) with terminal cyano groups and nitrogen vacancies. Chemical Engineering Journal 463, 142512-142525 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 231, 864, 306]]<|/det|>
+36. Wang F, Xu J, Wang ZP, Lou Y, Pan CS, Zhu YF. Unprecedentedly efficient mineralization performance of photocatalysis-self-Fenton system towards organic pollutants over oxygen-doped porous \(\mathrm{g - C_3N_4}\) nanosheets. Appl Catal B-Environ 312, 121438-121447 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 323, 840, 380]]<|/det|>
+37. Zhang GQ, Xu YS, He CX, Zhang PX, Mi HW. Oxygen-doped crystalline carbon nitride with greatly extended visible-light-responsive range for photocatalytic \(\mathrm{H}_2\) generation. Appl Catal B-Environ 283, 119636-119645 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[115, 397, 857, 435]]<|/det|>
+38. Chen X, Liu L, Yu PY, Mao SS. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 331, 746-750 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[115, 453, 840, 510]]<|/det|>
+39. Wu XF, et al. Promoting effect of interfacial hole accumulation on photoelectrochemical water oxidation in \(\mathrm{BiVO_4}\) and Mo-doped \(\mathrm{BiVO_4}\) . Advanced Powder Materials 3, 100234-100241 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 528, 853, 603]]<|/det|>
+40. Mohamed MJS, Gondal MA. Nanostructured defects-rich black-ZnO/biowaste-derived carbon composites for efficient visible light-driven hydrogen generation: A study on the role of C-Zn@C-O-Zn interface. International Journal of Hydrogen Energy 92, 47-58 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 620, 881, 676]]<|/det|>
+41. Wang J, et al. A ruthenium single atom nanozyme-based antibiotic for the treatment of otitis media caused by Staphylococcus aureus. Frontiers in Chemistry 12, 1439039-1439053 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 694, 840, 769]]<|/det|>
+42. Fan YF, Gan XR, Zhao HM, Zeng ZX, You WJ, Quan X. Multiple application of SAzyme based on carbon nitride nanorod-supported Pt single-atom for \(\mathrm{H}_2\mathrm{O}_2\) detection, antibiotic detection and antibacterial therapy. Chemical Engineering Journal 427, 131572-131582 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 788, 844, 844]]<|/det|>
+43. Chen AT, et al. Regulating the charge density of Cu(I) single sites enriched on the surface of N3c Vacancies-engineered \(\mathrm{g - C_3N_4}\) for efficient Fenton-like reactions. Separation and Purification Technology 314, 123525-123538 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 862, 850, 919]]<|/det|>
+44. Niu P, Qiao M, Li Y, Huang L, Zhai T. Distinctive defects engineering in graphitic carbon nitride for greatly extended visible light photocatalytic hydrogen evolution. Nano Energy 44, 73-81 (2018).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 100, 825, 157]]<|/det|>
+45. Wei Y, et al. Oxygen-Doped Porous g-\(C_3N_4\) via Oxalic Acid-Assisted Thermal Polycondensation as a Visible Light-Driven Photocatalyst for Bisphenol A Degradation. ACS Applied Nano Materials 6, 16567-16579 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 175, 877, 213]]<|/det|>
+46. Muhammad P, et al. Defect Engineering in Nanocatalysts: From Design and Synthesis to Applications. Advanced Functional Materials 34, 2314686-2314742 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 231, 835, 287]]<|/det|>
+47. Kong Y, et al. Synergistic silver doping and N vacancy promoting photocatalytic performances of carbon nitride for pollutant oxidation and hydrogen production. Chemical Engineering Journal 479, 147676-147690 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 305, 844, 361]]<|/det|>
+48. Wang P, et al. CaCO(3)-encircled hollow CuS nanovehicles to suppress cervical cancer through enhanced calcium overload-triggered mitochondria damage. Asian Journal of Pharmaceutical Sciences 19, 100989-101005 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 379, 814, 435]]<|/det|>
+49. Chen Y, et al. Atomically Dispersed Cu Nanozyme with Intensive Ascorbate Peroxidase Mimic Activity Capable of Alleviating ROS - Mediated Oxidation Damage. Advanced Science 9, 2103977-2103986 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[115, 453, 852, 510]]<|/det|>
+50. Yu B, et al. Using Cu - Based Metal-Organic Framework as a Comprehensive and Powerful Antioxidant Nanozyme for Efficient Osteoarthritis Treatment. Advanced Science 11, 2307798-2307810 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 527, 840, 583]]<|/det|>
+51. Huang Y, et al. Calcium Peroxide-Based Hydrogels Enable Biphasic Release of Hydrogen Peroxide for Infected Wound Healing. Advanced Materials, e2404813-e2404827 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[115, 601, 877, 657]]<|/det|>
+52. Wang Q, et al. Mussel-inspired multifunctional hydrogel dressing with hemostasis, hypoglycemic, photothermal antibacterial properties on diabetic wounds. Biomaterials Science 10, 4796-4814 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[115, 676, 862, 713]]<|/det|>
+53. Xiao J, et al. Copper Metal-Organic Framework Nanoparticles Stabilized with Folic Acid Improve Wound Healing in Diabetes. ACS Nano 12, 1023-1032 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[115, 731, 857, 787]]<|/det|>
+54. Xu SJ, et al. Tanshinone IIA Ameliorates Streptozotocin-Induced Diabetic Nephropathy, Partly by Attenuating PERK Pathway-Induced Fibrosis. Drug Design Development and Therapy 14, 5773-5782 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[115, 805, 810, 861]]<|/det|>
+55. Kim H, Wang SY, Kwak G, Yang Y, Kwon IC, Kim SH. Exosome - Guided Phenotypic Switch of M1 to M2 Macrophages for Cutaneous Wound Healing. Advanced Science 6, 1900513-1900526 (2019).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 83, 841, 139]]<|/det|>
+56. Li W, et al. Development of an Antiswelling Hydrogel System Incorporating M2-Exosomes and Photothermal Effect for Diabetic Wound Healing. ACS Nano 17, 22106-22120 (2023).  
+
+<|ref|>text<|/ref|><|det|>[[115, 157, 878, 231]]<|/det|>
+57. Wu Y, et al. A Versatile Glycopeptide Hydrogel Promotes Chronic Refractory Wound Healing Through Bacterial Elimination, Sustained Oxygenation, Immunoregulation, and Neovascularization. Advanced Functional Materials 33, 2305992-2306011 (2023).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[43, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[43, 92, 768, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 130, 318, 150]]<|/det|>
+SupportingInformation.pdf
+
+<--- Page Split --->

preprint/preprint__97c9a9f63f90540416a0527aa4a6d0fab88a8dcababe4ed5a1742c76e43c7542/images_list.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Fig. 1",
+    "footnote": [],
+    "bbox": [
+      [
+        218,
+        93,
+        794,
+        365
+      ]
+    ],
+    "page_idx": 33
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Fig. 2",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 33
+  }
+]

preprint/preprint__97f4ed0d700f6676eedcf080c87179c31138dbd7a7c41668e8ff0e450aee3754/images_list.json

@@ -0,0 +1,107 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1. Unusual binding and presentation of \\((\\mathbf{RA})_6\\mathbf{FAKKKYCL}\\) and \\((\\mathbf{RA})_6\\mathbf{FVKKKYCL}\\) A, Top panel, superimposition of bound \\((\\mathbf{RA})_6\\mathbf{FAKKKYCL}\\) (cyan) and \\((\\mathbf{RA})_6\\mathbf{FVKKKYCL}\\) (yellow) 20mer peptides. The backbone and side chain conformations of the peptides overlap between P3 and P8 but differ at P1, P2, and P-1 (P-2 was visible only in 20mer FV). Bottom panel, superimposition of bound 20mer peptides with 8mer FAKKKYCL (pink) and FVKKKYCL (green) control peptides. The four peptides overlap between P3 and P8 and are most divergent at P1. B, Interactions in the A pocket show that the main-chain nitrogen of P1 Phe FV 20mer (top panel) and FA 20mer (bottom panel) has rotated and forms a hydrogen bond with Asn63 (black dashed lines). The main-chain carbonyl oxygen in FV 20mer hydrogen bonds with Tyr159 and Tyr7, while the same atom in FA 20mer has undergone a very unusual rotation toward the \\(\\alpha 1\\) - helix and forms no interaction with MHC I residues. In both panels, extension residues protrude out of the groove. C, In the 20mer structures, the side chains of Arg62 have moved out of the canonical positions seen in the 8mer structures, which opens the A pocket and allows the extension residues \\((\\mathbf{RA})_6\\) to exit out.",
+    "footnote": [],
+    "bbox": [
+      [
+        120,
+        88,
+        875,
+        586
+      ]
+    ],
+    "page_idx": 23
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2. Top and side views of HLA-B8E76C groove in FA, FV, and AA 20mer structures. The figure shows peptide-induced structural distortions at the N-terminus of the groove. Residues Gln54 to Tyr59 (6 residues) and Ser42 to Tyr59 (18 residues) are not visible (shown as red dashed lines) in the FA and FV 20mer structures, respectively. This is in marked contrast to AA 20mer structure which has a correctly conformed groove<sup>29</sup>.",
+    "footnote": [],
+    "bbox": [
+      [
+        60,
+        93,
+        874,
+        423
+      ]
+    ],
+    "page_idx": 24
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3. Key roles of N-terminal MHC I residues Tyr59 and Ile52. A, Superimposition of FA and FV 20mer structures with AA 20mer (left panel) and FA 8mer (right panel) structures. The P1 Ala side chain of AA 20mer and P1 Phe side chain of FA 8mer occupy canonical positions and interact with Tyr59 (black dashed lines), while the bulky P1 Phe side chains of FA and FV 20mers clash with Tyr59 causing conformational disorders between Gln54 to Tyr59 (left and right panels). B, Same superimpositions as in A. The P1 Phe side chain of FA 20mer interacts with Ile52 of the \\(310\\) -helix (cyan dashed lines) which stabilizes Ser42 to Tyr59 (left and right panels), while a similar interaction involving Ile52 is not possible for FV 20mer resulting in conformational disorders between Ser42 to Tyr59 (left and right panels). The P1 Ala side chain of AA 20mer (left panel) and P1 Phe side chain of FA 8mer (right panel) have no interaction with Ile52.",
+    "footnote": [],
+    "bbox": [
+      [
+        135,
+        90,
+        861,
+        722
+      ]
+    ],
+    "page_idx": 25
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4. Molecular cross-talks between occupied pockets A and B. A, P1 phenyl side chain is oriented differently in FA 20mer (left panel) relative to FV 20mer (right panel). This is due to molecular cross-talks between peptide P1 and P2 residues and Ile152 and Tyr59 (missing) (see text and Fig. 3B). Overall, the network of hydrophobic (black dashed lines) and hydrogen bond (red dashed lines) interactions in the A and B pockets are different in these two structures. B, The network of interactions in FA 8mer (left panel) and FV 8mer (right panel) is quite similar overall in these conformed structures, in contrast to FA and FV 20mer structures shown in A.",
+    "footnote": [],
+    "bbox": [
+      [
+        125,
+        100,
+        870,
+        578
+      ]
+    ],
+    "page_idx": 26
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5. Properties of equilibrium ensembles of HLA-B8E76C-peptide complexes. A, RMSF values of individual residues along the heavy chain of HLA-B8E76C loaded with different peptides highlighting that the highest values are in the region of residues 41 to 62 (shown in a red box). B, A zoom-in of panel A reveals two distinct regions, residues 41 to 46 (peptide-independent) and residues 52 to 62 (peptide-dependent). C, Probability distributions of inter-residue distance between peptide P1 and Tyr59 and D, peptide P1 and Ile52 for HLA-B8E76C loaded with different peptides (see text).",
+    "footnote": [],
+    "bbox": [
+      [
+        130,
+        97,
+        850,
+        551
+      ]
+    ],
+    "page_idx": 27
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Figure 6. Comparisons with the structures of bat MHC I and human MHC II molecules. A, Superimposition of FA 8mer and bat MHC I molecule (Ptal-N\\*01:01; PDB code 6J2D) structures. The bat molecule has an additional turn (highlighted by a black box) in the extended region of Gln54 to Tyr59 (highlighted in red) that we identified as disordered (see Fig. 2). In this turn, Asp69 forms salt bridge interactions (black dashed lines) with Arg65 of the \\(\\alpha 1\\) -helix. B, Superimposition of FA 8mer and human MHC II molecule (HLA-DR1; PDB code 1DLH) structures showing that the HLA-DM susceptible region, i.e., \\(3_{10}\\) -helix and unstructured loop (highlighted in dark blue) overlaps with the \\(3_{10}\\) -helix and extended region (highlighted in red) that we identified as critical in shaping the A and B pockets (see Fig. 4).",
+    "footnote": [],
+    "bbox": [
+      [
+        112,
+        110,
+        872,
+        430
+      ]
+    ],
+    "page_idx": 28
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_7.jpg",
+    "caption": "Figure 7. A model of energy landscape. The model depicts FA and FV 20mer complexes as intermediate states that interconvert and are of higher energy relative to the more conformed states seen in the structures of FA and FV 8mers.",
+    "footnote": [],
+    "bbox": [
+      [
+        190,
+        90,
+        840,
+        520
+      ]
+    ],
+    "page_idx": 29
+  }
+]

preprint/preprint__97f4ed0d700f6676eedcf080c87179c31138dbd7a7c41668e8ff0e450aee3754/preprint__97f4ed0d700f6676eedcf080c87179c31138dbd7a7c41668e8ff0e450aee3754.mmd

@@ -0,0 +1,395 @@
+
+# Unusual crystal structures of MHC class I complexes reveal the elusive intermediate conformations explored during peptide editing in antigen presentation  
+
+Lenong Li University of Illinois at Chicago  
+
+Xubiao Peng Center for Quantum Technology Research  
+
+Mansoor Batliwala  
+
+Department of Microbiology and Immunology, University of Illinois at Chicago, 909 S. Wolcott Avenue, Chicago, IL 60612  
+
+Marlene Bouvier ( mbouvier@uic.edu )  
+
+Department of Microbiology and Immunology, University of Illinois at Chicago, 909 S. Wolcott Avenue, Chicago, IL 60612  
+
+## Article  
+
+Keywords:  
+
+Posted Date: January 26th, 2023  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 2500847/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: There is NO Competing Interest.  
+
+Version of Record: A version of this preprint was published at Nature Communications on August 18th, 2023. See the published version at https://doi.org/10.1038/s41467- 023- 40736- 6.
+
+<--- Page Split --->
+
+1 Unusual crystal structures of MHC class I complexes reveal the elusive intermediate 2 conformations explored during peptide editing in antigen presentation 3 4 5 6 7 8 9 Lenong Li', Xubiao Peng?, Mansoor Batliwala' & Marlene Bouvierl,\* 10 11 12 13 14 15 1 Department of Microbiology and Immunology, University of Illinois, Chicago, IL, 60612, USA 16 2 Center for Quantum Technology Research and Key Laboratory of Advanced Optoelectronic 17 Quantum Architecture and Measurements (MOE), School of Physics, Beijing Institute of 18 Technology, Beijing 100081, China 19 \* Corresponding author (mbouvier@uic.edu)
+
+<--- Page Split --->
+
+## Abstract  
+
+Studies have suggested that MHC class I (MHC I) molecules fluctuate rapidly between conformational states as they sample peptides for potential ligands. To date, MHC I intermediates are largely uncharacterized experimentally and remain elusive. We present x- ray crystal structures of HLA- B8 loaded with 20mer peptides that show significant conformational heterogeneity at the N- terminus of the groove. Long stretches of N- terminal residues were missing in the electron density maps creating an unstructured and widely open- ended groove. Our structures also revealed highly unusual features in MHC I and peptide conformations, and in MHC I- peptide interaction at the N- terminus of the groove. Molecular dynamics simulations showed that the complexes have varying degrees of flexibility in a manner consistent with the structures. We suggest that our structures represent transient substates explored by MHC I molecules during peptide editing. The visualization of peptide- dependent conformational flexibility in MHC I groove is a major step forward in our conceptual understanding of peptide repertoire development in antigen presentation. Our study also raises questions about the role of the N- terminus of the groove in peptide editing.
+
+<--- Page Split --->
+
+## Introduction  
+
+Antigen presentation by major histocompatibility class I (MHC I) molecules is central to adaptive immunity. MHC I molecules bind peptides and present them at the cell surface to specific receptors on \(\mathrm{CD8 + }\) T cells. This surveillance alerts the immune system to the presence of virally infected and transformed cells. MHC I molecules binds peptides in a groove that is lined with discrete pockets (A to F)'. The stability of MHC class I molecules is highly dependent on interaction with a bound peptide ligand2- 5. As such, within the endoplasmic reticulum (ER), there is an elaborate network of specialized proteins, referred to as the peptide- loading complex (PLC)6, that ensures MHC I molecules are loaded with high- affinity peptides prior to their transport to the cell surface. Studies of the mechanism by which high- affinity peptides become ligands of MHC I have highlighted that the F pocket at the C- terminal end of the groove is a critical region of conformational sensing7- 15. Indeed, the PLC proteins tapasin, ERp57, and calreticulin are spatially organized on MHC I molecules using sites of interaction at the C- terminal end of the groove16.  
+
+Molecular dynamic (MD) studies have suggested that the MHC I groove fluctuates rapidly between conformations, and it is the molecular features of such intermediate states that are recognized by specialized proteins, particularly tapasin7- 15. The binding of tapasin to immature MHC I molecules induces a widening of the groove thereby encouraging the dissociation of nonoptimally bound peptides17. Ultimately, under the action of tapasin, MHC I peptide repertoires are largely made up of highly stabilizing peptides, which ensures long- lived antigen presentation at the cell surface. An understanding of the molecular mechanism by which tapasin functions has been enriched from studies of TAPBPR18- 22, a tapasin homolog that works independently of the PLC. Although there is direct evidence that dynamics in MHC I play a fundamental role in mechanisms of peptide selection and exchange23, the molecular features of MHC I intermediate states at the core of these processes are elusive. Except for some information on the peptide- receptive form of MHC I24- 27, conformational intermediates of MHC I- peptide complexes are largely undefined due to the difficulty of probing such transient molecules. Furthermore, it is not well characterized if the region around the critical A and B pockets, at the N- terminal end of the groove, represents another site of conformational sensing and if it has a role in peptide editing.  
+
+In previous studies, we characterized crystallographically the presentation of long peptides based on the sequence \((\mathrm{RA})_n\mathrm{AAKKKYCL}\) by HLA- B8E76C28,29. We showed that these peptides adopt native bound conformations with their N- terminally elongated residues \((\mathrm{RA})_n\) protruding out of the groove. These structures provide a platform for the rationale design of peptides that have the potential to induce conformational perturbations at the N- terminal end of the groove and, therefore, that could provide direct insights into non- native MHC I conformations. Toward this goal, we substituted Ala at position 1 (P1) and P2 in \((\mathrm{RA})_6\mathrm{AAKKKYCL}\) 20mer peptide with the bulkier residues Phe and Val generating \((\mathrm{RA})_6\mathrm{FAKKKYCL}\) and \((\mathrm{RA})_6\mathrm{FVKKKYCL}\) . We present the x- ray crystal structures of HLA- B8 loaded with these long peptides that show highly unusual features in MHC I and peptide conformations and in MHC I- peptide interaction at the N- terminus of the groove. Our structures reveal for the first- time motions that MHC I molecules likely undergo transiently during the dynamic steps of peptide sampling. We combined these structural analyses with MD simulations which revealed other aspects of MHC I- peptide interaction in our complexes that are relevant for peptide editing, and consistent with the structures. Overall, our study provides a precise understanding of the link between molecular dynamics in MHC I molecules and their ability to adopt intermediate conformations for screening peptides efficiently, which is fundamentally important to ensure effective cytotoxic T cell responses to viral infections.
+
+<--- Page Split --->
+
+## Results  
+
+## Strategically designed peptides  
+
+We designed two HLA- B8- restricted 20mer peptides based on the sequence of HIV- 1 Gag epitope GGKKKYKL<sup>30</sup> in which P1 was substituted with Phe, P2 with either Ala or Val, and P7 with Cys. The resulting FAKKKYCL and FVKKKYCL peptides were N- terminally extended with twelve residues \((\mathrm{RA})_6\) generating \((\mathrm{RA})_6\mathrm{FAKKKYCL}\) and \((\mathrm{RA})_6\mathrm{FVKKKYCL}\) . The P7 Cys was introduced to form a disulfide bond with Cys76 in HLA- B8, after mutating Glu76, to prevent the dissociation of long peptides from the groove<sup>8,29</sup>. The 8mer FAKKKYCL and FVKKKYCL control peptides were also synthesized. The reconstitution of HLA- B8E76C- peptide complexes was carried out in vitro as we described previously<sup>28,29</sup>.  
+
+## Unconventional peptide binding and presentation  
+
+We determined the x- ray crystal structures of HLA- B8E76C loaded with the 20mer and 8mer peptides to high- resolution (Supplementary Table 1). The structures show that FA (cyan) and FV (yellow) 20mers adopt elongated conformations in which the core residues P1 to P8 are bound inside the groove and the extension residues P- 1 Ala (one position N- terminal to P1) and P- 2 Arg (visible only for FV 20mer) protrude out of the groove (Fig 1A, upper panel). The peptide backbones and side chains adopt nearly identical positions between P3 and P8, but clear differences are seen at P- 1, P1, and P2. The \(\mathrm{C}\alpha\) - atom positions have a 2.84- Å shift at P1 and 0.96- Å shift at P2. Comparisons of FA and FV 20mers with their corresponding FA (pink) and FV (green) 8mers (Fig 1A, lower panel) show that 8mer peptides adopt nearly identical bound conformations, and that P1 is the most divergent position in both 20/8mer pairs, with shifts in \(\mathrm{C}\alpha\) - atom of 3.13- Å for FAs and 1.42- Å for FVs. Finally, the electron density was clear over the entire length of the peptides, including the backbone and methyl side chain of P- 1 Ala and the backbone and part of the aliphatic side chain of P- 2 Arg (FV 20mer only) (Supplementary Fig. 1).  
+
+A close examination of how the FV 20mer peptide binds in the A pocket (Fig 1B, upper panel) shows that the main- chain nitrogen of P1 Phe is rotated in a position that is normally occupied by a canonical P1 side chain, as seen in FA and FV 8mers (Fig. 1A, lower panel). In this configuration, the P1 main- chain nitrogen forms a hydrogen bond with Asn63, and the extension residues P- 1 Ala and P- 2 Arg protrude out of the A pocket. Interestingly, the main- chain carbonyl oxygens of P- 1 and P- 2 residues form hydrogen bonds with the indole nitrogen of Trp167 (Fig 1B, upper panel), which likely stabilize the peptide backbone as it exits out of the groove. The structure also shows that the bulky P1 Phe side chain cannot occupy the terminal amino group canonical position, i.e., the cavity formed by residues Tyr7 and Tyr171, and instead the phenyl ring points toward the \(\alpha 1\) - helix (see also Fig. 4A, right panel). Finally, the main- chain carbonyl oxygen of P1 Phe forms hydrogen bonds with Tyr7 and Tyr159 (Fig 1B, upper panel). A similar rotation within the A pocket was also seen in the FA 20mer structure (Fig 1B, lower panel), with P1 main- chain nitrogen and P- 1 carbonyl oxygen forming hydrogen bonds to Asn63 and Trp167, respectively. In marked contrast to FV 20mer, however, the main- chain carbonyl group of P1 Phe is rotated toward the \(\alpha 1\) - helix, a highly unusual configuration, and surprisingly does not engage with any MHC I residues (see also Fig. 4A, left panel). Taken together, although P1 Phe residues of 20mer peptides have undergone a similar main- chain nitrogen rotation in the A pocket, there are clear differences in the binding mode of each peptide. Finally, a comparison of Arg62 in FA and FV 20mer structures relative to the 8mer structures (Fig. 1C) shows that the Arg side chains swing out of their canonical positions which creates an opening for \((\mathrm{RA})_6\) residues to exit out of the groove. A similar role of
+
+<--- Page Split --->
+
+residue 62 in opening the A pocket was observed in our structure of (RA)6AAKKKYCL 20mer peptide bound to HLA- B8E76C29.  
+
+## Significant peptide-induced conformational distortions at the N-terminus of the groove  
+
+The binding modes of FA and FV 20mers described in Fig. 1 are accompanied with significant structural changes in the MHC I groove. Figure 2 shows top and side views of the MHC I groove of FA and FV 20mer structures in which long stretches of N- terminal residues (shown by dashed red lines) in the \(\alpha 1\) - helix and loop connecting the \(\alpha 1\) - helix to \(\beta\) - strand of the floor were missing in the electron density maps. Specifically, the FA 20mer structure lacks 6 residues from Gln54 to Tyr59, and FV 20mer structure lacks as many as 18 residues from Ser42 to Tyr59 - electron densities at an acceptable \(1\sigma\) threshold were not visible for these residues in our structures. Consequently, the A pocket is unstructured and widely open- ended (Fig. 2 and Supplementary Fig. 2). In contrast, our previously determined AA 20mer structure showed that the N- terminus of the groove has a native structure (Fig. 2)29. Thus, the FA and FV 20mer structures provide direct evidence that the N- terminus of the groove has the potential to undergo significant peptide- induced structural distortions, highlighting its remarkable inherent plasticity. To the best of our knowledge, this is the first report of MHC I structures, with or without a bound peptide, showing such significant conformational distortions in the groove. Other than these differences, minor changes in the groove were detected between the FA and FV 20mer structures (r.m.s. deviation of 0.13- A). It is interesting that FA and FV 20mer peptides differ only by the nature of residues at P1 and P2 relative to our previous AA 20mer peptide29. Because all N- terminal MHC I residues were visible in the AA 20mer structure (Fig. 2), but not in the FA and FV 20mer structures, this strongly suggests that P1 and P2 residues play critical roles in the conformational maturation of the groove.  
+
+## Unusual MHC I-peptide interaction at the N-terminus of the groove  
+
+To understand the role of peptide P1 residue in modulating interactions with MHC I residues at the N- terminus of the groove, we analyzed the FA and FV 20mer structures in the context of both the AA 20mer29 and FA 8mer structures (Fig. 3). The structure of AA 20mer shows that the small P1 Ala methyl side chain forms a hydrophobic interaction with conserved residue Tyr59 (Fig. 3A, left panel). In contrast, the large P1 Phe side chains of FA and FV 20mers sterically clash with Tyr59 (Fig. 3A, left panel) and as such, residues Gln54 to Tyr59 become disorganized and are not visible in the FA and FV 20mer structures while these residues are conformed in AA 20mer structure. Interestingly, the native structure of FA 8mer (Fig. 3A, right panel) shows that the P1 side- chain phenyl ring occupies a canonical position and forms hydrophobic interactions with Tyr59, and that residues Gln54 to Tyr59 are visible. Similar observations were made in the FV 8mer structure (Supplementary Fig. 3). Taken together, our structures indicate that residues Gln54 to Tyr59, and especially Tyr59, form a region of remarkable conformational flexibility and structural plasticity that allows adaptations in response to size and configuration of peptide P1 side chains.  
+
+There is another unusual feature at the N- terminus of the groove in FA 20mer structure (Fig. 3B). In this structure (Fig. 3B, left panel), the P1 side- chain phenyl ring forms highly unusual hydrophobic interactions with the highly conserved Ile52 of the \(3_{10}\) - helix (a conserved element comprising residues 50 to 55). In the FV 20mer structure, however, the P1 phenyl ring is oriented differently (Fig. 3B, left panel) and, as such, it cannot engage with Ile52 thus causing residues Ser42 to Glu53 to become disordered. In the FA20mer structure, residues Ser42 to Glu53 were clearly visible due to P1 phenyl ring engagement with Ile52. Notably, interaction between a bound
+
+<--- Page Split --->
+
+peptide and Ile52, or any other residues of the \(3_{10}\) - helix, have not been reported before and are also not seen in FA and FV 8mer structures (Fig. 3B, right panel, and Supplementary Fig. 3). Taken together, there are clear molecular interplays involving peptide P1 residue and N- terminal residues Tyr59 and Ile52 that influence structural integrity at the N- terminal end of the groove (see also below).  
+
+## Molecular cross-talks between peptide P1 and P2 residues and N-terminal MHC I residues Ile52 and Tyr59  
+
+Results in Fig. 3 raise another important question: why is the P1 Phe side chain oriented differently in FA 20mer versus FV 20mer structures? This question is important given that this difference in orientation significantly affected the structural integrity of Ser42 to Glu53 in the FV 20mer structure. To address this, we examined peptide P2 residues, which is Ala in FA 20mer and Val in FV 20mer (Fig. 4A). In the FA 20mer structure (Fig. 4A, left panel), the small side- chain methyl group of P2 Ala allows the P1 main- chain carbonyl group to undergo an unusual rotation toward the \(\alpha 1\) - helix, which has the effect of orienting the P1 phenyl ring close to Ile52. In the FV 20mer structure, however, because P2 carries a larger Val side chain (Fig. 4A, right panel), the P1 main- chain carbonyl group cannot similarly rotate toward the \(\alpha 1\) - helix, and consequently the P1 phenyl ring is positioned further away from Ile52. Overall, different networks of interactions involving peptide P1 and P2 residues and N- terminal MHC I residues were established in FA and FV 20mer structures (Fig. 4A). Interestingly, using a thermal denaturation assay, we determined rather similar melting temperature (Tm), \(65.8^{\circ}\mathrm{C}\) for FA 20mer and \(67.5^{\circ}\mathrm{C}\) for FV 20mer. In contrast, the structures of FA and FV 8mer show very similar networks of interactions (Fig. 4B) and identical Tm values, \(71.1^{\circ}\mathrm{C}\) for FA 8mer and \(71.4^{\circ}\mathrm{C}\) for FV 8mer.  
+
+## Analysis of MHC I residues in pockets along the groove  
+
+Given the importance of the six binding pockets A to F in determining the peptide side chain specificities of HLA alleles, we compared the side chain orientations of MHC I residues in pockets A and B of FA and FV 20mer structures relative to the corresponding 8mer structures (Supplementary Fig. 4). Pocket A is made up of 9 residues and typically anchors the N- terminal amino group and P1 residue and closes the N- terminal end of the groove. The analysis shows that Tyr59 (conserved) and Asn63 (highly conserved) have the most divergent orientations in both FA and FV 20/8mer pairs, with some differences also seen in Tyr171 (conserved) and Trp167 (highly conserved). Pocket B is made up of 9 residues and binds the P2 peptide side chain that defines HLA binding motifs. For both the FA and FV 20/8mer pairs, all MHC I residues adopted very similar orientations, except for Asn63 at the boundary of the A and B pockets. Similar analyses in pockets C to F indicated that there were minimal changes in MHC I side chain orientations (data not shown). Taken together, the binding of 20mer peptides affects the configuration of MHC I residues in pocket A more significantly than those in pocket B, consistent with P1 being the most divergent position of these peptides (Fig. 1A). This analysis also highlighted a critical role for Asn63 in MHC I maturation; while Asn63 mediated several hydrophobic and hydrogen bond interactions with P1 and P2 residues in the FA and FV 20mer structures (Fig. 4A), Asn63 formed only one hydrogen bond with P2 main- chain nitrogen in the native FA and FV 8mer structures (Fig. 4B).
+
+<--- Page Split --->
+
+## MD simulations: conformational flexibility and geometrical parameters  
+
+To further characterize HLA- B8E76C- peptide complexes and interaction between peptide P1 residue and Tyr59 and Ile52, we analyzed thermal properties, i.e., conformational flexibility and geometrical parameters, from MD simulations at physiological temperature.  
+
+The flexibility of individual MHC I residues along the heavy chain is characterized by its root mean square fluctuation (RMSF) (Fig. 5A). Results show that RMSF values are clearly higher in the region comprising residues 41 to 62 (shown in a red box) for FA and FV 20mer complexes relative to other regions, indicating that residues 41 to 62 are conformationally more flexible and thermally more unstable. A zoom- in revealed that there are two distinct regions (Fig. 5B); residues 41 to 46 and residues 52 to 62. In the region of residues 52 to 62, the RMSF values for FA and FV 20mer complexes are much higher than those of the other complexes, indicating that the 20mer complexes are conformationally more flexible in this region. However, in the region of residues 41 to 46, all complexes have similarly high RMSF values, suggesting that conformational flexibility is independent on the bound peptide in this region.  
+
+We evaluated next the geometrical parameters of interaction between peptide P1 residue and Tyr59 and Ile52 to gain further insights into these complexes (see Methods for details). Figure 5C shows the probability distributions of inter- residue distance between P1 and Tyr59 for all five systems. Results show that the peak of the distribution is centered around 5Å for AA 20mer while the other distributions are centered around 6Å, suggesting a high probability of interaction between P1 and Tyr59 in all complexes when equilibrated at physiological temperature. However, because the distributions are broader for FA and FV 20mers than for the other systems, this suggests that interaction between P1 and Tyr59 is less stable in the 20mer complexes. We also calculated the probability distributions of relative orientation between the aromatic rings of P1 Phe and Tyr59 (Supplementary Fig. 5A). Results show that FA and FV 8mers have symmetric distributions centered around \(90^{\circ}\) , indicating that the two aromatic rings are essentially perpendicular to each other. In contrast, for FA and FV 20mers, the peaks of the distributions are positioned at \(110^{\circ}\) and \(150^{\circ}\) , respectively, showing that the aromatic rings of P1 Phe and Tyr59 are no longer perpendicular to each other. Accordingly, we conclude that the configuration of \(\pi - \pi\) interactions \(^{31}\) between P1 Phe and Tyr59 is more parallel- like in FV 20mer but changes toward T- shape- like in FA 20mer, and finally adopts the stable T- shaped structure in FA and FV 8mers. Finally, Fig. 5D clearly shows that the probability distributions of inter- residue distance between P1 and Ile52 are different among the five complexes. For FA 20mer, the distribution has a high peak centered around 5Å, indicating that there is a stable interaction between these two residues. For FV 20mer, the distribution still peaked around 5Å but it is much wider, indicating that interaction between P1 and Ile52 is less stable (Fig. 5D). In contrast, the distributions for FA and FV 8mers have high peaks centered around 10Å, indicating that there are no interactions between P1 and Ile52 in these complexes. For AA 20mer, since the distances are centered around 7Å, we infer that weak hydrophobic interaction between P1 and Ile52 may exist in this complex. To further characterize interaction between P1 and Ile52 in FA and FV 20mers, we determined the probability distributions of angle C- H- X (X being the center of Phe aromatic ring) (Supplementary Fig. 5B). Results show that the angles are mostly in the range \(120^{\circ}\) to \(150^{\circ}\) . As such, since the probability distributions of “distance” (Fig. 5D) and “defined angle” (Supplementary Fig. 5B) between P1 and Ile52 satisfy the geometric criterion of CH- \(\pi\) interaction \(^{32}\) , we conclude that CH- \(\pi\) interaction exists in both FA and FV 20mers.
+
+<--- Page Split --->
+
+## MD simulations: rotation of peptide terminal amino group in the A pocket  
+
+Given that the terminal amino groups of FA and FV 20mers adopt highly unusual configurations in the A pocket (Fig. 1B), we conducted MD simulations to assess whether these amino groups can undergo spontaneous rotations to canonical positions, i.e., pointing down in the A pocket. For these tests, we removed the extension \((\mathrm{RA})_6\) residues of FA and FV 20mer peptides in their structures, generating FA20.8mer and FV20.8mer (see Methods for details). The definition of the dihedral angle \(\omega\) for characterizing the rotation of the terminal amino group is shown in Fig. S6A, using as an example FV20.8mer. In the FA and FV 20mer structures, \(\omega\) values are \(- 110^{\circ}\) and \(- 73^{\circ}\) , respectively, indicating that both terminal amino groups point up, while in the native FA and FV 8mer structures, \(\omega\) values are \(96^{\circ}\) and \(98^{\circ}\) , respectively, indicating that the amino groups point down.  
+
+The evolution curves of the dihedral angle \(\omega\) in MD simulations for FA20.8mer and FV20.8mer are shown in Supplementary Figs. 6B and 6C, respectively. For FA20.8mer, the results show that \(\omega\) changes from about \(- 100^{\circ}\) to \(100^{\circ}\) in two out of the three replicate simulations within \(300~\mathrm{ns}\) . For FV20.8mer, \(\omega\) changes from about \(- 70^{\circ}\) to around \(- 260^{\circ}\) in two out of the three replicate simulations within \(300~\mathrm{ns}\) (note that \(\omega = - 260^{\circ}\) is equivalent to \(\omega = 100^{\circ}\) , when the periodicity of \(\omega\) is taken into consideration). Hence, we conclude that the terminal amino groups of FA20.8mer and FV20.8mer have high probability to rotate in the A pocket and point down as seen in native structures. In addition, it is interesting that different replicas gave different evolution curves of the dihedral angle \(\omega\) , indicating that the pathways of such rotations are unlikely to be unique.  
+
+## Analysis of our structures in the context of bat MHC I molecules and human MHC II molecules  
+
+The characterization of bat MHC I genes identified thus far, showed that many of these molecules contain a 3- or 5- amino acid insertion in their binding groove \(^{33}\) . Several structures of PtaI- N\*01:01 bat MHC I molecules with a 3- amino acid insertion have been determined and revealed that the insertion creates a turn at the N- terminus of the groove (Fig. 6A) \(^{34,35}\) , within the critical region of Gln54 to Tyr59 (shown in red). The bat structures also revealed that in this turn, Asp59 forms salt bride interactions with Arg65 of the \(\alpha 1\) - helix (Fig. 6A) \(^{34,35}\) . Such a pairing of charged residues at these positions is a highly conserved feature of bat MHC I molecules \(^{34,35}\) , and it is expected to add structural rigidity at the N- terminus of the groove relative to human MHC I molecules. As such, the groove of bat MHC I molecules may be more restricted in its ability to adopt alternate conformations, which could affect peptide editing and repertoire development with implication on bat adaptive immunity. Further investigation is required to understand these observations.  
+
+Finally, in MHC class II (MHC II) molecules, there is evidence of peptide- induced conformational flexibility at the N- terminus of the groove \(^{36 - 38}\) , i.e., the region recognized by the peptide- exchange catalyst HLA- DM \(^{39,40}\) . The binding of HLA- DM induces conformational changes in MHC II, particularly in the \(3_{10}\) - helix and unstructured loop \(^{41}\) (shown in dark blue in Fig. 6B). This critical region of MHC II overlaps with the region of MHC I ( \(3_{10}\) - helix and extended region) that we identified as important for shaping the A and B pockets (shown in red in Fig. 6B). This analysis lends support to the view that the N- terminal end of the groove likely has a role in peptide editing (see Discussion).
+
+<--- Page Split --->
+
+## Discussion  
+
+The crystal structures of native peptide- filled MHC I molecules have taught us a great deal about molecular recognition of bound peptides. These structures represent the endpoint of a complex intracellular maturation process whereby MHC I molecules acquire peptides of sufficiently high affinity to ensure efficient antigen presentation. Biophysical, NMR, and in silico studies have been consistent in demonstrating that MHC I molecules explore intermediate conformational states in solution during peptide binding. Given that the native crystal structures of peptide- filled MHC I molecules are always nearly identical, it has not been possible thus far to obtain detailed structural information of intermediate conformations explored by MHC I molecules. This makes the notion of functional dynamics and peptide- induced conformational motions elusive. We have determined the crystal structures of HLA- B8 loaded with 20mer peptides that reveal highly unusual conformational and structural features in both MHC I and peptides. We have also carried out MD simulations to gain further insights into peptide- dependent interaction in these complexes, which provided new information and helped refine our structural analyses.  
+
+The FA and FV 20mer structures showed that the N- terminus of the groove has undergone significant conformational distortions relative to native FA and FV 8mer structures (Fig. 2 and Supplementary Fig. 2). These differences were peptide specific. MD simulations identified that residues 52 to 62 are most conformationally flexible and thermally unstable in FA and FV 20mer complexes relative to the other complexes (Fig. 5B), consistent with the lack of electron density for residues 54 to 59 in FA and FV 20mer structures. MD simulations also identified a conformationally flexible region comprising residues 41 to 46 that seems more peptide independent. Consistent with this, it is interesting that several deposited structures of HLA- B8 loaded with 9mer peptides lack clear electron density between residues \(\sim 41\) to 49 (for example, 1M05, 3SKO, 4QRQ, and 5WMR), suggesting that some conformational fluctuations can persist at the N- terminus of the groove in native structures. Our structures also showed that the 20mer peptides, with only a single amino acid difference at P2, adopted different and unusual backbone and side chain orientations at P1 and P2 but overlapped almost identically between P3 to P8 (Fig. 1A, upper panel). Interestingly, both complexes had similar thermostabilities, \(65.8^{\circ}\mathrm{C}\) (FA 20mer) versus \(67.5^{\circ}\mathrm{C}\) (FV 20mer). We suggest that the FA and FV 20mer structures, although static snapshots, represent discrete states that are explored by these complexes as they navigate the conformational space to locate their low energy conformation ("native") (Fig. 7) (see below).  
+
+MHC I peptide ligands are usually 8 to 10 amino acids long and bind with their terminal amino group pointing down in the A pocket, as seen in the FA and FV 8mer structures (Fig. 4B). Because short linear peptides are generally unstructured in solution, it is reasonable to assume that they land as such within the immature MHC I groove. NMR and other biophysical studies showed that in the initial binding steps, incoming peptides are loosely accommodated in the groove until more specific conformational adaptions take place in both peptides and MHC I<sup>22,23,42,43</sup>. It is therefore plausible that when peptides of optimal lengths are first captured by MHC I, they adopt conformations that resemble those of our 20mer peptides, i.e., with an unusual rotation of the terminal amino group in the A pocket (Fig. 1B), until folding proceeds and peptides adopt a canonical conformation (or not). To test this, we simulated the relaxation process of FA20.8mer and FV20.8mer using plain MD simulations and the results showed that peptide terminal amino groups have a high probability to rotate to a canonical position from the unusual orientations observed in our structures. The simulations also suggested that the configurations of FA and FV 20mer peptides, as seen in the structures, represent "trapped" states (see below). Finally, it is worth
+
+<--- Page Split --->
+
+noting that naturally occurring HLA- B8- restricted 8 and 9mer peptides with a large residue at P1 or at P1/P2 have been reported: for example, Phe, Val, Leu, and His at P1<sup>44- 47</sup> or Tyr/Leu, Trp/Val, Phe/Leu, and Tyr/Ile at P1/P2<sup>44,48,49</sup>.  
+
+The FA and FV 20mer structures showed that strictly conserved Tyr59 and highly conserved Ile52 (3<sub>10</sub>- helix) act synergistically at the N- terminus of the groove. In native MHC I structures, Tyr59 stabilizes the peptide terminal amino group together with Tyr171, and Ile52 acts as a structural support to Tyr59 and Tyr171<sup>1</sup>. In our 20mer structures, Tyr59 could not adopt its native position because of the large size of P1 Phe side chain in the A pocket (Fig. 3A). This caused residues 54 to 59 to become disordered and created an open- ended A pocket. In the FA 20mer structure, this disordering was accompanied with Ile52 forming a highly unusual interaction with peptide P1 Phe side chain, which in turn was facilitated by the small peptide P2 Ala side chain in the adjacent B pocket (Fig. 4A). In the FV 20mer structure, however, similar molecular cross- talks between the A and B pockets were not possible because of the larger peptide P2 Val side chain in the B pocket that positioned P1 Phe side chain further away from Ile52 (Fig. 4A). Consequently, a significantly longer stretch of residues became disordered in the FV 20mer structure, generating a widely open- ended A pocket. In performing MD simulations, we were able to probe other aspects of the interaction between peptide P1 and MHC I residues Tyr59 and Ile52 in the thermal equilibrium ensembles of our complexes. The simulations revealed that 1. interaction between P1 and Tyr59 is characterized by the more stable T- shaped \(\pi\) - \(\pi\) configuration in FA and FV 8mer complexes relative to FA and FV 20mer complexes (Fig. 5C and Supplementary Fig. 5A), which is consistent with residues 54- 59 being ordered in the 8mer structures but disordered in the 20mer structures; and 2. CH- \(\pi\) interaction between P1 and Ile52 exists only in FA and FV 20mer complexes, and it is more stable in FA 20mer than FV 20mer (Fig. 5D and Supplementary Fig. 5B). This is also consistent with the FV 20mer structure showing a more highly disordered groove and widely open- ended A pocket. Taken together, we suggest that the 20mer structures represent "trapped" states, i.e., conformations in which Tyr59 cannot adopt a "closed" position (FA and FV 20mer structures) and 3<sub>10</sub>- helix cannot mature into its secondary fold (FV 20mer structure). In other words, formation of native A and B pockets requires the conformational transitions of Tyr59 into a "closed" position and 3<sub>10</sub>- helix (Ile52) into its native fold. A role for the 3<sub>10</sub>- helix in peptide- induced MHC I maturation was suggested previously<sup>50- 52</sup>. Because Tyr59 and 3<sub>10</sub>- helix are strictly conserved elements in human HLA alleles, these transitions are expected to be universal in MHC I maturation. Furthermore, that the 20mer structures can tolerate such high degrees of peptide- induced structural perturbations at the N- terminus of the groove is consistent with MD simulations of other groups showing that bound peptides can dissociate partially at the N- terminus of groove while remaining anchored within the F pocket<sup>53- 55</sup>, with potential implication in MHC I maturation (see below). Finally, the lower thermostabilities of FA and FV 20mer complexes relative to the 8mer complexes is consistent with the "trapped" states being higher energy conformations (Fig. 7), but of sufficiently low energy to be revealed crystallographically.  
+
+The conformational transitions discussed above are intricately coupled to occupancy of the A and B pockets. This is significant when considering that the crystal structure of a peptide- free MHC I molecule (HLA- A2) showed no substantial differences at the N- terminus (or C- terminus) relative to native HLA- A2 structures, except for minor differences in some side chain orientations within the A pocket<sup>27</sup>. It is also significant because, in native MHC I structures, the A and B pockets anchor the peptide terminal amino group and P2 residue, respectively, that contribute to protein stability. The FA and FV 20mer structures thus likely captured intermediate conformations adopted by MHC I when actively "evaluating" peptide P1 and P2 residues as part of editing, with
+
+<--- Page Split --->
+
+the exact molecular features of MHC I intermediates expected to be fluid as maturation proceeds based on our MD simulations.  
+
+Previous studies indicated that molecular dynamics around the F pocket is a major driving force for recognition of MHC I molecules by tapasin and TAPBRP7- 15. Conformational flexibility at the N- terminus of the groove of MHC II molecules is also a critical determinant of HLA- DM- mediated mechanism of peptide exchange41. As such, our structures of FA and FV 20mers thus raise the intriguing question of whether the region close to the A and B pockets represents a binding surface recognized by a protein, yet to be identified, with a role in stabilizing MHC I intermediates and/or peptide editing. It is interesting that the cryo- EM structure of the PLC showed that the long P- domain of CRT chaperone is positioned atop and across the MHC I groove with its tip interacting with ERp57'16. In this spatial organization, the P- domain could interact transiently with the partially folded region of the A and B pockets, as facilitated by the inherent structural plasticity of the P- domain and dynamic nature of the P- domain/ERp57 interaction56,57. More work is required to examine this idea. The possibility that ERAP1, ERAP2, and/or ERAP1/ERAP2 heterodimer play a functional role at the N- terminal end of the groove is also very reasonable. We showed previously in biochemical studies that ERAP1 and ERAP1/ERAP2 can actively trim the protruding N- terminal residues of long peptides, including the AA 20mer, while bound to HLA- B8E76C28,29. Others also showed that ERAP1 trims peptides bound to H2- Kb using cell- based assays53. Moreover, it was demonstrated that mouse ERAAP (equivalent to ERAP1) synergizes with tapasin to edit peptide repertoires58. In fact, a role for MHC I molecules in antigen processing has long been suggested59. Our current study supports this view and furthermore suggests that the ERAP enzymes are more likely engaging with intermediate forms of MHC I molecules, rather than correctly conformed molecules60.  
+
+In conclusion, our study provided a crystallographic and MD characterization of conformational substates in MHC I- peptide systems, and it brought into focus the N- terminal end of the groove in mechanisms of high- affinity peptide selection. This understanding is critical given the role of MHC I- restricted peptide repertoires for activation of adaptive immune responses to control viral infections. Finally, our work opens new avenues to examine chaperoning of the groove around the A pocket, and it also encourages further characterization of other MHC I molecules.
+
+<--- Page Split --->
+
+## Acknowledgements  
+
+We thank Dr. Bernard Santarsiero for expert assistance with remote x- ray data collection and discussion, and the staff at the Argonne National Laboratory (Argonne, IL) where all x- ray data were collected. This work was supported in whole, or in part, by NIAID, National Institutes of Health, Grants R01 AI114467, R01 AI108546, and R21 AI173863 (to MB).  
+
+## Author contributions  
+
+LL executed all experiments including x- ray crystallography and was assisted by Mansoor Batliwala in refolding MHC I- peptide complexes. LL and MB designed experiments and interpreted the structures. XP performed MD simulations and interpreted the results. MB wrote the manuscript with contributions from all authors.  
+
+Competing financial interests: The authors declare no competing financial interests.
+
+<--- Page Split --->
+
+## Methods  
+
+Synthetic peptides. Peptides were synthesized by the solid- phase methodology (GenScript Biotech Co.) and purified by reverse- phase chromatography on a C18 HPLC column. Stock solutions of peptides in DMSO were stored at \(- 80^{\circ}\mathrm{C}\) .  
+
+Refolding of HLA- B\\*0801E76C complexes. Using the crystal structure of HLA- B\\*0801/GGRKKYKL (PDB code 1AGB), we identified residue Glu76 to be geometrically well- positioned to form a disulfide bond with the side chain of P7 peptide residue, after mutation with a cysteine, as we described previously<sup>28</sup>. The HLA- B\\*0801E76C heavy chain mutant was generated as described previously<sup>28</sup>. HLA- B\\*0801E76C complexes were reconstituted from ureasolubilized inclusion bodies of HLAB\\*0801E76C heavy chain (1 \(\mu \mathrm{M}\) ) and \(\beta_{2}\) - microglobulin (2 \(\mu \mathrm{M}\) ) with a synthetic Cys- P7 peptide (10 \(\mu \mathrm{M}\) ) in an oxidative refolding buffer at \(4^{\circ}\mathrm{C}^{61}\) . After 48 hours, the crude refolding mixture of HLA- B\\*0801E76C complexes was purified on a Superdex- 200 size exclusion chromatography column by FPLC. Stock solutions of purified complexes (10- 30 mg/ml) in 20 mM Tris- HCl, pH 7.5, 150 mM NaCl, were kept at \(- 80^{\circ}\mathrm{C}\) .  
+
+Crystallization. The initial crystallization condition of HLA- B\\*0801E76C/(RA)<sub>6</sub> FAKKKYCL (10 mg/ml) was identified using the Crystal Screen<sup>TM</sup> (Hampton Research, Riverside, CA) as solution #9 (0.2 M ammonium acetate, 0.1 M sodium citrate tribasic dihydrate, pH 5.6, 30% (w/v) PEG 4000) via the hanging- drop vapor diffusion method at room temperature. The initial crystals were optimized using different pH values (4.5- 7.0) and PEGs (6000- 10000; 10- 30%). These optimized crystals were used to generate a seeding solution in solution #9. Crystals used for data collection were grown by mixing 2 \(\mu \mathrm{l}\) of 10 mg/ml protein solution with 2 \(\mu \mathrm{l}\) of 0.2 M ammonium acetate, 18% PEG 4000, 0.1 M sodium citrate, pH 5.7, and 0.5 \(\mu \mathrm{l}\) of seeding solution. Similar crystallization conditions were used to collect data for HLA- B\\*0801E76C loaded with FA and FV 8mers and FV 20mer.  
+
+Data collection, structural determination, and refinement. X- ray diffraction data sets were collected with a MAR- 225 CCD detector at the LS- CAT beamline 21- ID- G (or 21- ID- F) of the Advanced Photon Source (Argonne National Laboratory, Argonne, IL). Data were integrated and scaled with the HKL2000 program package<sup>62</sup> or XDS<sup>63</sup>. Details of data processing are indicated in Supplementary Table 1. The structures of all complexes were solved by molecular replacement using Phaser<sup>64</sup> (the initial search model was HLA- B\\*0801E76C/R(N- Me)AAAKKKYCL (PDB code 6P2S). Structure refinement of all models was carried out in Phenix (or Refmac in CCP4)<sup>65- 67</sup> and manual building with COOT<sup>68</sup>. Final refinement statistics are summarized in Supplementary Table 1. The atomic coordinates of all structures have been deposited in the Protein Data Bank with the following accession codes: FA 8mer (8E13), FA 20mer (8E2Z), FV 8mer (8E81), and FV 20mer (8EC5).  
+
+Thermal denaturation assay. A thermal denaturation assay was performed using reaction mixtures consisting of 7 \(\mu \mathrm{l}\) of a complex (final concentration of 2 \(\mu \mathrm{M}\) ), 7 \(\mu \mathrm{l}\) of 10x SYPRO orange dye (5000x, Thermo Fisher Scientific, Waltham, MA) and 7 \(\mu \mathrm{l}\) of 50 mM HEPES, pH 7.2, 150 mM NaCl. Each mixture (total volume 21 \(\mu \mathrm{l}\) ) was analyzed in quadruplicate using an ABI ViiA7 RT- PCR instrument (Life Technologies, Inc., Carlsbad, CA). A temperature gradient from 25 to \(95^{\circ}\mathrm{C}\) with continuous increment of \(0.06^{\circ}\mathrm{C / sec}\) was used to generate the denaturation curves. The
+
+<--- Page Split --->
+
+averaged denaturation curves were plotted as “fluorescence intensity” versus “temperature”, and the minimum point of the first derivative of each curve provided the melting temperature.  
+
+MD initial structures preparation. The x- ray crystal structures of HLA- B8E76C loaded with five different peptides were analyzed by MD simulations: FA and FV 8mers (this study), AA 20mer (6P2C) \(^{29}\) , and FA and FV 20mers (this study). The missing regions of HLA- B8E76C in the structures of FA and FV 20mers were complemented using software UCSF Chimera \(^{69}\) with the structure of AA 20mer as the template. The missing regions of the bound 20mer peptides were complemented using Modeller \(^{70}\) integrated in software UCSF Chimera \(^{69}\) . In the simulations, the peptide termini were neutralized to exclude artificial charge effects.  
+
+MD simulations set- up. Depending on the purpose, two different MD simulations were performed: replica exchange MD (REMD) and plain MD. The REMD simulations were used for the thermal properties analysis due to its efficiency in thermal equilibrated conformational ensemble sampling, while the plain MD simulations were used for simulating the rotation of the terminal amino group of bound FA20.8mer and FV20.8mer from their unusual up orientations. Both types of MD simulations included common settings, as follows. Simulations were performed with explicit solvent using the software package GROMACS 5.1. \(^{71,72}\) . The force field CHARMM36m \(^{73}\) together with its own modified TIP3P water model \(^{74}\) was also used. The LINCS algorithm was applied to constrain the covalent bonds with H- atoms, and the time step in simulation was 2 fs. The protein was simulated in 0.1 M aqueous NaCl solution. After a short energy minimization, an NVT simulation of 100 ps with the V- rescale temperature coupling at 310 K was performed, followed by an NPT simulation of 300 ps with the Parrinello- Rahman coupling method at a reference pressure of 1 bar. The relaxation times for the temperature coupling and pressure coupling are 0.1 ps and 2 ps, respectively. During NVT and NPT simulations, the protein backbone is constrained to its initial structure. At the end, we removed the constraints and performed production simulations at the same temperature and pressure. The time interval for conformational sampling in simulations was 20 ps.  
+
+REMD simulations. After a short thermal equilibration process with NVT and NPT, as described above, we set up 30 replicas with temperature distributed from 300 K to 340 K following the webserver (https://virtualchemistry.org/remd- temperature- generator/) \(^{75}\) with an attempt swap duration of 1 ps between two neighboring temperatures. The average acceptance probability for the replica exchanges was about 30%. Each replica ran for 40 ns, and thus we have a simulation with total time up to 1.2 μs. The thermal equilibrated ensembles were collected from the replica at temperature 310 K from the REMD simulations, from which the flexibility of HLA- B8E76C heavy chain and interactions between peptide residue P1 and MHC I residues Tyr59 and Ile52 were analyzed.  
+
+Plain MD simulations. Using the FA and FV 20mer structures, as generated in the “Initial structures preparation” stage, we generated the initial configurations of FA20.8mer and FV20.8mer by deleting the extension (RA) \(_{6}\) residues. After a short thermal equilibration process with NVT and NPT as described above, we performed the production run for 300 ns to simulate the relaxation process of FA20.8mer and FV20.8mer. During the simulations, we observed rotations of the terminal amino groups in FA20.8mer and FV20.8mer peptides. We repeated the simulation three times for each system to ensure producibility.
+
+<--- Page Split --->
+
+Interaction Analysis. We identified possible interactions between peptide residue P1 and MHC I residues Tyr59 and Ile52 according to the geometrical properties, i.e., distances and relative orientations between the two residues. For interaction between P1 Phe and Tyr59, the distance was defined between the geometrical centers of the two aromatic rings, while the relative orientation was defined by the angle between the normal directions of the aromatic rings. For interaction between P1 Ala of AA 20mer and Tyr59, the distance was defined between the Cβ atom of P1 Ala and the geometrical center of the aromatic ring of Tyr59. For interaction between P1 Phe and Ile52, the distance was defined between the geometrical center of the aromatic ring in P1 Phe and the Cδ atom of Ile 52, while the angle C- H- X was defined by the atoms Cδ of Ile52, hydrogen atom covalently bonded to Cδ and the center of the Phe aromatic ring. Finally, for interaction between P1 Ala of AA 20mer and Ile52, the distance was defined between the Cβ atom in P1 Ala and the Cδ atom of Ile52.  
+
+Analysis of rotations of peptide terminal amino groups. We probed rotations of terminal amino groups in pocket A by monitoring the evolution of the dihedral angle \(\omega\) formed by atoms Cys76:CA- P7:CA- P1:CA- P1:N (Supplementary Fig. 6A) in the plain MD simulations. When drawing evolution curves, we reasonably required that the rotation between two neighboring frames be less than \(180^{\circ}\) , i.e., \(|\omega (t_{i + 1}) - \omega (t_{i})| < 180^{\circ}\) . Otherwise, we added/subtracted the dihedral angle \(\omega (t_{i + 1})\) by \(360^{\circ}\) . In the end, to eliminate thermal fluctuations and visualize the rotation pathways better, we smoothed the evolution curves by an average sliding window of 10 ns.
+
+<--- Page Split --->
+
+References  
+
+1. Saper, M. A., Bjorkman, P. J. & Wiley, D. C. Refined structure of the human histocompatibility antigen HLA-A2 at 2.6 Å resolution. J. Mol. Biol. 219, 277-319 (1991).  
+
+2. Matsumura, M., Fremont, D. H., Peterson, P. A. & Wilson, I. A. Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257, 927-934 (1992).  
+
+3. Fahnestock, M. L., Johnson, J. L., Feldman, R. M., Tsomides, T. J., Mayer, J., Nahri, L. O. & Bjorkman, P. J. Effects of peptide length and composition on binding to an empty class I MHC heterodimer. Biochem. 33, 8149-8158 (1994).  
+
+4. Madden, D. R. The three-dimensional structure of peptide-MHC-complexes. Annu. Rev. Immunol. 13, 587-622 (1995).  
+
+5. Bouvier, M. & Wiley D. C. Importance of peptide amino and carboxyl termini to the stability of MHC class I molecules. Science 265, 398-402 (1994).  
+
+6. Cresswell, P., Bangia, N., Dick, T & Diedrich, G. The nature of the MHC class I peptide loading complex. Immunol. Rev. 171, 21-28 (1999).  
+
+7. Van Hateren, A., Bailey, A., Werner, J. M. & Elliott T. Plasticity of empty major histocompatibility complex molecules determines peptide-selector function. Molec. Immunol. 68, 98-101 (2015).  
+
+8. Ayres, C. M., Corcelli, S. A. & Baker B. M. peptide and peptide-dependent motions in MHC proteins: immunological and biophysical underpinnings. Front. Immunol. 8, 935 (2017).  
+
+9. Yanaka, S. & Sugase, K. Exploration of the conformational dynamics of major histocompatibility complex molecules. Frontiers Immunol. 8, 632 (2017).  
+
+10. Wieczorek, M. Abualrous, E. T., Stich, J., Alvaro-Benito, M., Stolzenberg, S., Noe, F. & Freund, C. Major histocompatibility complex (MHC) class I and MHC class II proteins: conformational plasticity in antigen presentation. Front. Immunol. 8, 292 (2017).  
+
+11. Fisette, O., Wingbermuhle, Tampe R. & Schafer, L.V. Molecular mechanism of peptide editing in the tapasin-MHC I complex. Sc. Reports 19085 (2016).  
+
+12. Truong, H. V. & Sgourakis, N. G. Dynamics of MHC I-molecules in the antigen processing and presentation pathway. Curr. Opin. Immunol. 70, 122-128 (2021).  
+
+13. Jantz-Naeem, N. & Springer, S. Venus flytrap or pas de trois? The dynamics of MHC class I molecules. Curr. Opin. Immunol. 70, 82-89 (2021).  
+
+14. Thomas, C. & Tampe, R. MHC I assembly and peptide editing - chaperones, clients, and molecular plasticity in immunity. Curr. Opin. Immunol. 70, 48-56 (2021).
+
+<--- Page Split --->
+
+15. Margulies, D. H., Taylor, D. K., Jiang, J., Boyd, L. F., Ahmad, J., Mage, M. G. & Natarajan, K. Chaperones and catalysts: How antigen presentation pathways cope with biological necessity. Front. Immunol. 13, 859782 (2022).  
+
+16. Blees, A., Januliene, D., Hofmann, T., Koller, N., Schmidt, C., Trowitzsch, S., Moeller, A. & Tampe, R. Structure of the human MHC I peptide-loading complex. Nature 551, 525–528 (2017).  
+
+17. Chen, M. & Bouvier, M. Analysis of interaction in a tapasin/class I complex provides a mechanism for peptide selection. EMBO J. 26, 1681–1690 (2007).  
+
+18. Herman, C., Strittmatter, L. M., Deanne, J. E. & Boyle, L. H. The binding of TAPBPR and tapasin to MHC class I is mutually exclusive. J. Immunol. 191, 5743–5750 (2013).  
+
+19. Herman, C., van Hateren, A., Trautwein, N., Neerincx, A., Duriez, P. J., Stevanovic, S., Trowsdale, J., Deane, J. E., Elliott, T. & Boyle L. H. TAPBPR alters MHC class I peptide presentation by functioning as a peptide catalyst. eLife 4, e09617 (2015).  
+
+20. Thomas, C. & Tampe, R. Structure of the TAPBPR-MHC I complex defines the mechanism of peptide loading and editing. Science 358, 1060–1064 (2017).  
+
+21. Jiang, J., Natarajan, K., Boyd, L. F., Morozov, G. I., Mage, M. G. & Margulies, D. H. Crystal structure of the TAPBPR-MHC I complex reveals the mechanism of peptide editing in antigen presentation. Science 358, 1064–1068 (2017).  
+
+22. McShan, A., Devlin, C. A., Overall, S. A., Park, S. A., Toor, J. S., Moschidi, D., Flores-Solis, D., Choi, H., Tripathi, S., Procko, S. & Sgourakis, N. G. Molecular determinants of chaperone interactions on MHC-I folding and antigen repertoire selection. Proc. Natl. Acad. Sci. USA 116, 25602–25613 (2019).  
+
+23. Van Hateren, A., Anderson, M., Bailey, A., Werner, J. M. Skipp, P. & Elliott, T. Direct evidence for conformational dynamics in major histocompatibility complex class I molecules. J. Biol. Chem. 292, 20255–20269 (2017).  
+
+24. Bouvier, M. & Wiley D. C. Structural characterization of a soluble and partial folded class I major histocompatibility heavy chain/β₂m heterodimer. Nat. Struct. Biol. 5, 377–384 (1998).  
+
+25. Fahnestock, M. L., Tamir, I., Narhi, L. & Bjorkman, P. J. Thermal stability comparisons of purified empty and peptide-filled forms of a class I MHC molecule. Science 258, 1658–1662 (1992).  
+
+26. Kurimoto, E., Kuroki, K., Yamaguchi, Y., Yagi-Utsumi, M., Igaki, T., Iguchi, T., Maenaka, K. & Kato, K. Structural and functional mosaic nature of MHC class I molecules in their peptide-free form. Molec. Immunol. 55, 393–399 (2013).  
+
+27. Anjanappa, R., Garcia-Alai, M., Kopicki, J.-D., Lockhauserbaumer, J., Aboelmagd, M., Hinrichs, J., Nemtanu, I. M., Uetrecht, C., Zacharias, M., Springer, S. & Meijers, R. Structures of
+
+<--- Page Split --->
+
+peptide- free and partially loaded MHC class I molecules reveal mechanisms of peptide selection. Nat. Comm. 11, 1314 (2020).  
+
+28. Chen, H., Li, L., Weimershaus, M., Evnouchidou, I., van Endert P. & Bouvier, M. ERAP1-ERAP2 dimers trim MHC I-bound precursor peptides; implications for understanding peptide editing. Sc. Rep. 6, 28902 (2016).  
+
+29. Li, L., Batliwala, M. & Bouvier, M. ERAP1 enzyme-mediated trimming and structural analyses of MHC I-bound precursor peptides yield novel insights into antigen processing and presentation. J. Biol. Chem. 294, 18534-18544 (2019).  
+
+30. Phillips, R. E., Rowland-Jones, S., Nixon, D. F., Gotch, F. M., Edwards, J. P., Ongulesi, A. O., Elvin, J. G., Rothbard, J. A., Bangham, C. R., Rizza, C. R. & McMichael, A. J. Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition. Nature 354, 453-459 (1991).  
+
+31. Tehrani, A. Z. & Kim, K. S. Functional molecules and materials by \(\pi\) -interaction based quantum theoretical design. Int. J. Quantum Chem. 116, 622-633 (2016).  
+
+32. Wang J. & Yao, L. Dissecting C-H···π and N-H···π interactions in two proteins using a combined experimental and computational approach. Sci. Rep. 9, 20149 (2019).  
+
+33. Ng, J. H. J., Tachedjian, M., Deakin, J., Wynne, J. W., Cui, J., Haring, V., Broz, I., Chen, H., Belov, K., Wang, L.-F. & Baker, M. L. Evolution and comparative analysis of the bat MHC -I region. Sc. Rep. 6, 21256 (2016).  
+
+34. Lu, D., Liu, K., Zhang, D., Yue, C., Lu, Q., Cheng, H., Wang, L., Chai, Y., Qi, J., Wang, L.-F., Gao, G. F. & Liu, W. J. Peptide presentation by bat MHC class I provides new insights into the antiviral immunity of bats. PLoS Biol. 9, e30000436 (2019).  
+
+35. Qu, Z., Li, Z., Ma, L., Wei, X., Zhang, L., Liang, R., Meng, G., Zhang, N. & Xia, C. Structure and peptidome of the bat MHC class I molecule reveal a novel mechanism leading to high-affinity peptide binding. J. Immunol. 202, 3493-3506 (2019).  
+
+36. Natarajan, S. K., Stern, L. J. & Sadegh-Nasseri, S. Sodium dodecyl sulfate stability of HLA-DR1 complexes correlates with burial of hydrophobic residues in pocket 1. J. Immunol. 162, 3463-3470 (1999).  
+
+37. Zarutskie, J. A., Sato, A. K., Rusche, M. M., Chan, C., Lomakin, A., Benedek, G. B. & Stern, L. J. A conformational change in the human major histocompatibility complex protein HLA-DR1 induced by peptide binding. Biochem. 38, 5878-5887 (1999).  
+
+38. Sato, A. K., Zarutskie, J. A., Rusche, M. M., Lomakin, A., Natarajan, S. K., Sadegh-Nasseri, S., Benedek, G. B. & Stern, L. J. Determinants of the peptide-induced conformational change in the human class II major histocompatibility complex protein HLA-DR1. J. Biol. Chem. 275, 2165-2173 (2000).
+
+<--- Page Split --->
+
+39. Chou, C. L. & Sadegh-Nasseri, S. HLA-DM recognized the flexible conformation of major histocompatibility complex class II. J. Expm. Med. 192, 1697-1706 (2000).  
+
+40. Pos, W., Sethi, D. K., Call, M. J., Schulze, M.-S. E. D., Anders, A.-K., Pyrdol, J. & Wucherpfennig, K. W. Crystal structure of the HLA-DM-HLA-DR1 complex defines mechanism for rapid peptide selection. Cell 151, 1557-1568 (2012).  
+
+41. Painter, C. A., Negroni, M. P., Kellesberger, K. A., Zavala-Ruiz, Z., Evans, J. E. & Stern, L. J. Conformational lability in the class II MHC \(3_{10}\) helix and adjacent extended strand dictate HLA-DM susceptibility and peptide exchange. Proc. Natl. Acad. Sci. USA 108, 19329-19334 (2011).  
+
+42. Yanaka, S., Ueno, T., Shi, Y., Qi, J., Gao, G. F., Tsumoto, K. & Sugase, K. Peptide-dependent conformational fluctuation determines the stability of the human leukocyte antigen class I complex. J. Biol. Chem. 289, 24680-24690 (2014).  
+
+43. Hawse, W., Gloor, B. E., Ayres, A. M., Kho, K., Nuter, E. & Baker, B. M. Peptide modulation of class I major histocompatibility complex protein molecular flexibility and the implications for immune recognition. J. Biol. Chem. 288, 24372-24381 (2013).  
+
+44. Flad, T., Spengler, B., Kalbacher, H., Brossart, P., Baier, D., Kaufmann, R., Bold, P., Metzger, S., Bluggel, M., Meyer, H. E., Kurz, B. & Muller, C. A. Direct identification of major histocompatibility complex class I-bound tumor-associated peptide antigens of a renal carcinoma cell line by a novel mass spectrometric method. Cancer Res. 58, 5803-5811 (1998).  
+
+45. Stickel, J. S., Weinzierl, A. O., Hillen, O., Drews, O., Schuler, M. M., Hennenlotter, J., Wernet, D., Muller, C. A., Stenzl, A., Rammensee, H.-G. & Stevanovic, S. HLA-ligand profiles of primary renal cell carcinoma maintained in metastases. Cancer Immunol. Immunother. 58, 1407-1417 (2009).  
+
+46. Adair, S. J., Carr, T. M., Fink, M. J., Slingluff Jr, C. L. & Hogan, K. T. The TAG family of cancer/testis antigens is widely expressed in a variety of malignancies and gives rise to HLA-A2-restricted epitopes. J. Immunol. 31, 7-17 (2008).  
+
+47. Koziel, M. J., Dudley, D., Afdhal, N., Grakuoi, A., Rice, C. M., Choo, Q. L., Houghton, M. & Walker, B. D. HLA class I-restricted cytotoxic T lymphocytes specific for hepatitis C virus. Identification of multiple epitopes and characterization of patterns of cytokine release. J. Clin. Invest. 96, 2311-2321 (1995).  
+
+48. Burrows, S. R., Sculley, T. B., Misko, I. S., Schmidt, C. & Moss, D. J. An Epstein-Barr virus-specific cytotoxic T cell epitope in EBV nuclear antigen 3 (EBNA 3). J. Exp. Med. 171, 345-349 (1990).  
+
+49. Steven, N. M., Leese, A. M., Annels, N. E., Lee, S. P. & Rickinson, A. B. epitope focusing in the primary cytotoxic T cell response to Epstein-Barr virus and its relationship to T cell memory. J. Expm. Med. 184, 1801-1813 (1996).
+
+<--- Page Split --->
+
+50. Hansen, T. H., Lybarger, L., Yu, L., Mitaksov, V. & Fremont, D. H. Recognition of open conformers of classical MHC by chaperones and monoclonal antibodies. Immunol. Rev. 207, 100–111 (2005).  
+
+51. Mage, M. G., Dolan, M. A., Wang, R., Boyd, L. F. Revilleza, M. J., Robinson, Natarajan, K., Myers, N. B., Hansen, T. H. & Margulies, D. H. The peptide-receptive transition state of MHC class I molecules: insights from structures and molecular dynamics. J. Immunol. 189, 1391–1399 (2012).  
+
+52. Mage, M. G., Dolan, M. A., Wang, R., Boyd, L. F. Revilleza, M. J., Robinson, Natarajan, K., Myers, N. B., Hansen, T. H. & Margulies, D. H. A structural and molecular dynamics approach to understanding the peptide-receptive transition state od MHC-I molecules. Molec. Immunol. 55, 123–125 (2013).  
+
+53. Papakyriakou, A., Reeves, E., Beton, H., Mikolajek, H., Douglas, L., Cooper, G., Elliott, T., Werner, J. M. & James, E. The partial dissociation of MHC class I-bound peptides exposes their N terminus to trimming by endoplasmic reticulum aminopeptidase 1. J. Biol. Chem. 293, 7538–7648 (2018).  
+
+54. Fisette, O., Wingbermuhle, & Schafer, L.V. Partial dissociation of truncated peptide influences the structural dynamics of the MHC I binding groove. Front. Immunol. 8, 408 (2017).  
+
+55. Wingbermuhle, S. & Schafer, L. V. Partial peptide dissociation and binding groove plasticity in two major histocompatibility complex class I alleles – differences between alleles versus force field and sampling effects. RSC Adv. 12, 29908–29914 (2022).  
+
+56. Ellgaard, L., McNaul, N., Chatsisvili, A., Braakman, I. Co- and post-translational protein folding in the ER. Traffic 17, 615–638 (2016).  
+
+57. Fisette, O. Schroder, G. F. & Schafer L. V. Atomistic structure and dynamics of the human MHC-I peptide-loading complex. Proc. Natl. Acad. Sci. USA 117, 20597–20606 (2020).  
+
+58. Kanaseki, T., Blanchard, N., Hammer, G. E., Gonzalez, F. & Shastri N. ERAPP synergizes with MHC class I molecules to make the final cut in the antigenic peptide precursors in the endoplasmic reticulum. Immunity 25, 795–806 (2006).  
+
+59. Falk, K., Rotzschke, O. & Rammensee, H. G. Cellular peptide composition governed by major histocompatibility complex class I molecules. Nature 348, 248–251 (1990).  
+
+60. Mavridis, G., Arya, R., Domnick, A., Zoidakis, J., Makridakis, M., Vlahou, A., Mpakali, A., Lelis, A., Georgiadis, D., Tampe, R., Papakyriakou, A., Stern, L. J. & Stratikos, E. A systematic re-examination of processing of MHC I-bound antigenic peptide precursors by endoplasmic reticulum aminopeptidase 1. J. Biol. Chem. 295, 7193-7210 (2020).
+
+<--- Page Split --->
+
+61. Garboczi, D. N., Hung, D. T. & Wiley, D. C. HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc. Natl. Acad. Sci. U.S.A. 89, 3429-3433 (1992).  
+
+62. Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307-326 (1997).  
+
+63. Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125-132 (2010).  
+
+64. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658-674 (2007).  
+
+65. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213-221 (2010).  
+
+66. Winn, M. D., Murshudov, G. N. & Papiz, M. Z. Macromolecular TLS refinement in REFMAC at moderate resolutions. Methods Enzymol. 374, 300-321 (2003).  
+
+67. Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. Overview of the CCP4 suite and current developments. Acta Crystallogr. D Biol. Crystallogr. 67, 235-242 (2011).  
+
+68. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126-2132 (2004).  
+
+69. Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C. & Ferrin, T. E. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605-1612 (2004).  
+
+70. Sali, A., & Blundell, T. L. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779-815 (1993).  
+
+71. Pronk, S., Pall, S., Schulz, R., Larsson, P., Bjelkmar, P., Apostolov, R., Shirts, M. R., Smith, J. C., Kasson, P. M., van der Spoel, D., Hess, B. & Lindahl, E. GROMACS 4.5: A high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 29, 845-85 (2013).  
+
+72. Abraham, M. J., Murtola, T., Schulz, R., Pall, S., Smith, J. C., Hess, B. & Lindahl, E. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1-2, 19-25 (2015).  
+
+73. Huang, J., Sarah, R., Grzegorz, N., Ting, R., Michael, F., de Groot, B. L., Helmut, G. & MacKerell Jr., A. D. CHARMM36m: an improved force field for folded and intrinsically disordered proteins, Nat. Methods 14, 71-76 (2017).
+
+<--- Page Split --->
+
+834 74. Mark, P. & Nilsson, L. Structure and dynamics of liquid water with different long-range 835 interaction truncation and temperature control methods in molecular dynamics simulations. J. 836 Comput. Chem. 23, 1211–1219 (2002).  
+
+837 75. Patriksson, A. & van der Spoel, D. A temperature predictor for parallel tempering simulations 839 Phys. Chem. Chem. Phys. 10, 2073–2077 (2008).
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1. Unusual binding and presentation of \((\mathbf{RA})_6\mathbf{FAKKKYCL}\) and \((\mathbf{RA})_6\mathbf{FVKKKYCL}\) A, Top panel, superimposition of bound \((\mathbf{RA})_6\mathbf{FAKKKYCL}\) (cyan) and \((\mathbf{RA})_6\mathbf{FVKKKYCL}\) (yellow) 20mer peptides. The backbone and side chain conformations of the peptides overlap between P3 and P8 but differ at P1, P2, and P-1 (P-2 was visible only in 20mer FV). Bottom panel, superimposition of bound 20mer peptides with 8mer FAKKKYCL (pink) and FVKKKYCL (green) control peptides. The four peptides overlap between P3 and P8 and are most divergent at P1. B, Interactions in the A pocket show that the main-chain nitrogen of P1 Phe FV 20mer (top panel) and FA 20mer (bottom panel) has rotated and forms a hydrogen bond with Asn63 (black dashed lines). The main-chain carbonyl oxygen in FV 20mer hydrogen bonds with Tyr159 and Tyr7, while the same atom in FA 20mer has undergone a very unusual rotation toward the \(\alpha 1\) - helix and forms no interaction with MHC I residues. In both panels, extension residues protrude out of the groove. C, In the 20mer structures, the side chains of Arg62 have moved out of the canonical positions seen in the 8mer structures, which opens the A pocket and allows the extension residues \((\mathbf{RA})_6\) to exit out. </center>
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2. Top and side views of HLA-B8E76C groove in FA, FV, and AA 20mer structures. The figure shows peptide-induced structural distortions at the N-terminus of the groove. Residues Gln54 to Tyr59 (6 residues) and Ser42 to Tyr59 (18 residues) are not visible (shown as red dashed lines) in the FA and FV 20mer structures, respectively. This is in marked contrast to AA 20mer structure which has a correctly conformed groove<sup>29</sup>. </center>
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3. Key roles of N-terminal MHC I residues Tyr59 and Ile52. A, Superimposition of FA and FV 20mer structures with AA 20mer (left panel) and FA 8mer (right panel) structures. The P1 Ala side chain of AA 20mer and P1 Phe side chain of FA 8mer occupy canonical positions and interact with Tyr59 (black dashed lines), while the bulky P1 Phe side chains of FA and FV 20mers clash with Tyr59 causing conformational disorders between Gln54 to Tyr59 (left and right panels). B, Same superimpositions as in A. The P1 Phe side chain of FA 20mer interacts with Ile52 of the \(310\) -helix (cyan dashed lines) which stabilizes Ser42 to Tyr59 (left and right panels), while a similar interaction involving Ile52 is not possible for FV 20mer resulting in conformational disorders between Ser42 to Tyr59 (left and right panels). The P1 Ala side chain of AA 20mer (left panel) and P1 Phe side chain of FA 8mer (right panel) have no interaction with Ile52. </center>
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4. Molecular cross-talks between occupied pockets A and B. A, P1 phenyl side chain is oriented differently in FA 20mer (left panel) relative to FV 20mer (right panel). This is due to molecular cross-talks between peptide P1 and P2 residues and Ile152 and Tyr59 (missing) (see text and Fig. 3B). Overall, the network of hydrophobic (black dashed lines) and hydrogen bond (red dashed lines) interactions in the A and B pockets are different in these two structures. B, The network of interactions in FA 8mer (left panel) and FV 8mer (right panel) is quite similar overall in these conformed structures, in contrast to FA and FV 20mer structures shown in A. </center>
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 5. Properties of equilibrium ensembles of HLA-B8E76C-peptide complexes. A, RMSF values of individual residues along the heavy chain of HLA-B8E76C loaded with different peptides highlighting that the highest values are in the region of residues 41 to 62 (shown in a red box). B, A zoom-in of panel A reveals two distinct regions, residues 41 to 46 (peptide-independent) and residues 52 to 62 (peptide-dependent). C, Probability distributions of inter-residue distance between peptide P1 and Tyr59 and D, peptide P1 and Ile52 for HLA-B8E76C loaded with different peptides (see text). </center>
+
+<--- Page Split --->
+![](images/Figure_6.jpg)
+
+<center>Figure 6. Comparisons with the structures of bat MHC I and human MHC II molecules. A, Superimposition of FA 8mer and bat MHC I molecule (Ptal-N\*01:01; PDB code 6J2D) structures. The bat molecule has an additional turn (highlighted by a black box) in the extended region of Gln54 to Tyr59 (highlighted in red) that we identified as disordered (see Fig. 2). In this turn, Asp69 forms salt bridge interactions (black dashed lines) with Arg65 of the \(\alpha 1\) -helix. B, Superimposition of FA 8mer and human MHC II molecule (HLA-DR1; PDB code 1DLH) structures showing that the HLA-DM susceptible region, i.e., \(3_{10}\) -helix and unstructured loop (highlighted in dark blue) overlaps with the \(3_{10}\) -helix and extended region (highlighted in red) that we identified as critical in shaping the A and B pockets (see Fig. 4). </center>
+
+<--- Page Split --->
+![](images/Figure_7.jpg)
+
+<center>Figure 7. A model of energy landscape. The model depicts FA and FV 20mer complexes as intermediate states that interconvert and are of higher energy relative to the more conformed states seen in the structures of FA and FV 8mers. </center>
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+SuppfilesJan18. pdf
+
+<--- Page Split --->

preprint/preprint__97f4ed0d700f6676eedcf080c87179c31138dbd7a7c41668e8ff0e450aee3754/preprint__97f4ed0d700f6676eedcf080c87179c31138dbd7a7c41668e8ff0e450aee3754_det.mmd

@@ -0,0 +1,527 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 896, 244]]<|/det|>
+# Unusual crystal structures of MHC class I complexes reveal the elusive intermediate conformations explored during peptide editing in antigen presentation  
+
+<|ref|>text<|/ref|><|det|>[[44, 263, 420, 305]]<|/det|>
+Lenong Li University of Illinois at Chicago  
+
+<|ref|>text<|/ref|><|det|>[[44, 310, 420, 351]]<|/det|>
+Xubiao Peng Center for Quantum Technology Research  
+
+<|ref|>text<|/ref|><|det|>[[44, 356, 210, 375]]<|/det|>
+Mansoor Batliwala  
+
+<|ref|>text<|/ref|><|det|>[[44, 378, 936, 420]]<|/det|>
+Department of Microbiology and Immunology, University of Illinois at Chicago, 909 S. Wolcott Avenue, Chicago, IL 60612  
+
+<|ref|>text<|/ref|><|det|>[[44, 425, 405, 445]]<|/det|>
+Marlene Bouvier ( mbouvier@uic.edu )  
+
+<|ref|>text<|/ref|><|det|>[[44, 448, 936, 490]]<|/det|>
+Department of Microbiology and Immunology, University of Illinois at Chicago, 909 S. Wolcott Avenue, Chicago, IL 60612  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 531, 102, 549]]<|/det|>
+## Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 569, 137, 588]]<|/det|>
+Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 606, 330, 626]]<|/det|>
+Posted Date: January 26th, 2023  
+
+<|ref|>text<|/ref|><|det|>[[44, 644, 473, 664]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 2500847/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 681, 910, 725]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[44, 742, 530, 762]]<|/det|>
+Additional Declarations: There is NO Competing Interest.  
+
+<|ref|>text<|/ref|><|det|>[[42, 798, 930, 842]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on August 18th, 2023. See the published version at https://doi.org/10.1038/s41467- 023- 40736- 6.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[60, 88, 857, 420]]<|/det|>
+1 Unusual crystal structures of MHC class I complexes reveal the elusive intermediate 2 conformations explored during peptide editing in antigen presentation 3 4 5 6 7 8 9 Lenong Li', Xubiao Peng?, Mansoor Batliwala' & Marlene Bouvierl,\* 10 11 12 13 14 15 1 Department of Microbiology and Immunology, University of Illinois, Chicago, IL, 60612, USA 16 2 Center for Quantum Technology Research and Key Laboratory of Advanced Optoelectronic 17 Quantum Architecture and Measurements (MOE), School of Physics, Beijing Institute of 18 Technology, Beijing 100081, China 19 \* Corresponding author (mbouvier@uic.edu)
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 191, 106]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[112, 107, 882, 352]]<|/det|>
+Studies have suggested that MHC class I (MHC I) molecules fluctuate rapidly between conformational states as they sample peptides for potential ligands. To date, MHC I intermediates are largely uncharacterized experimentally and remain elusive. We present x- ray crystal structures of HLA- B8 loaded with 20mer peptides that show significant conformational heterogeneity at the N- terminus of the groove. Long stretches of N- terminal residues were missing in the electron density maps creating an unstructured and widely open- ended groove. Our structures also revealed highly unusual features in MHC I and peptide conformations, and in MHC I- peptide interaction at the N- terminus of the groove. Molecular dynamics simulations showed that the complexes have varying degrees of flexibility in a manner consistent with the structures. We suggest that our structures represent transient substates explored by MHC I molecules during peptide editing. The visualization of peptide- dependent conformational flexibility in MHC I groove is a major step forward in our conceptual understanding of peptide repertoire development in antigen presentation. Our study also raises questions about the role of the N- terminus of the groove in peptide editing.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 223, 106]]<|/det|>
+## Introduction  
+
+<|ref|>text<|/ref|><|det|>[[113, 107, 882, 316]]<|/det|>
+Antigen presentation by major histocompatibility class I (MHC I) molecules is central to adaptive immunity. MHC I molecules bind peptides and present them at the cell surface to specific receptors on \(\mathrm{CD8 + }\) T cells. This surveillance alerts the immune system to the presence of virally infected and transformed cells. MHC I molecules binds peptides in a groove that is lined with discrete pockets (A to F)'. The stability of MHC class I molecules is highly dependent on interaction with a bound peptide ligand2- 5. As such, within the endoplasmic reticulum (ER), there is an elaborate network of specialized proteins, referred to as the peptide- loading complex (PLC)6, that ensures MHC I molecules are loaded with high- affinity peptides prior to their transport to the cell surface. Studies of the mechanism by which high- affinity peptides become ligands of MHC I have highlighted that the F pocket at the C- terminal end of the groove is a critical region of conformational sensing7- 15. Indeed, the PLC proteins tapasin, ERp57, and calreticulin are spatially organized on MHC I molecules using sites of interaction at the C- terminal end of the groove16.  
+
+<|ref|>text<|/ref|><|det|>[[113, 315, 882, 577]]<|/det|>
+Molecular dynamic (MD) studies have suggested that the MHC I groove fluctuates rapidly between conformations, and it is the molecular features of such intermediate states that are recognized by specialized proteins, particularly tapasin7- 15. The binding of tapasin to immature MHC I molecules induces a widening of the groove thereby encouraging the dissociation of nonoptimally bound peptides17. Ultimately, under the action of tapasin, MHC I peptide repertoires are largely made up of highly stabilizing peptides, which ensures long- lived antigen presentation at the cell surface. An understanding of the molecular mechanism by which tapasin functions has been enriched from studies of TAPBPR18- 22, a tapasin homolog that works independently of the PLC. Although there is direct evidence that dynamics in MHC I play a fundamental role in mechanisms of peptide selection and exchange23, the molecular features of MHC I intermediate states at the core of these processes are elusive. Except for some information on the peptide- receptive form of MHC I24- 27, conformational intermediates of MHC I- peptide complexes are largely undefined due to the difficulty of probing such transient molecules. Furthermore, it is not well characterized if the region around the critical A and B pockets, at the N- terminal end of the groove, represents another site of conformational sensing and if it has a role in peptide editing.  
+
+<|ref|>text<|/ref|><|det|>[[113, 577, 882, 874]]<|/det|>
+In previous studies, we characterized crystallographically the presentation of long peptides based on the sequence \((\mathrm{RA})_n\mathrm{AAKKKYCL}\) by HLA- B8E76C28,29. We showed that these peptides adopt native bound conformations with their N- terminally elongated residues \((\mathrm{RA})_n\) protruding out of the groove. These structures provide a platform for the rationale design of peptides that have the potential to induce conformational perturbations at the N- terminal end of the groove and, therefore, that could provide direct insights into non- native MHC I conformations. Toward this goal, we substituted Ala at position 1 (P1) and P2 in \((\mathrm{RA})_6\mathrm{AAKKKYCL}\) 20mer peptide with the bulkier residues Phe and Val generating \((\mathrm{RA})_6\mathrm{FAKKKYCL}\) and \((\mathrm{RA})_6\mathrm{FVKKKYCL}\) . We present the x- ray crystal structures of HLA- B8 loaded with these long peptides that show highly unusual features in MHC I and peptide conformations and in MHC I- peptide interaction at the N- terminus of the groove. Our structures reveal for the first- time motions that MHC I molecules likely undergo transiently during the dynamic steps of peptide sampling. We combined these structural analyses with MD simulations which revealed other aspects of MHC I- peptide interaction in our complexes that are relevant for peptide editing, and consistent with the structures. Overall, our study provides a precise understanding of the link between molecular dynamics in MHC I molecules and their ability to adopt intermediate conformations for screening peptides efficiently, which is fundamentally important to ensure effective cytotoxic T cell responses to viral infections.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 228, 106]]<|/det|>
+## Results  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 108, 376, 125]]<|/det|>
+## Strategically designed peptides  
+
+<|ref|>text<|/ref|><|det|>[[115, 125, 882, 265]]<|/det|>
+We designed two HLA- B8- restricted 20mer peptides based on the sequence of HIV- 1 Gag epitope GGKKKYKL<sup>30</sup> in which P1 was substituted with Phe, P2 with either Ala or Val, and P7 with Cys. The resulting FAKKKYCL and FVKKKYCL peptides were N- terminally extended with twelve residues \((\mathrm{RA})_6\) generating \((\mathrm{RA})_6\mathrm{FAKKKYCL}\) and \((\mathrm{RA})_6\mathrm{FVKKKYCL}\) . The P7 Cys was introduced to form a disulfide bond with Cys76 in HLA- B8, after mutating Glu76, to prevent the dissociation of long peptides from the groove<sup>8,29</sup>. The 8mer FAKKKYCL and FVKKKYCL control peptides were also synthesized. The reconstitution of HLA- B8E76C- peptide complexes was carried out in vitro as we described previously<sup>28,29</sup>.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 281, 531, 299]]<|/det|>
+## Unconventional peptide binding and presentation  
+
+<|ref|>text<|/ref|><|det|>[[115, 299, 882, 525]]<|/det|>
+We determined the x- ray crystal structures of HLA- B8E76C loaded with the 20mer and 8mer peptides to high- resolution (Supplementary Table 1). The structures show that FA (cyan) and FV (yellow) 20mers adopt elongated conformations in which the core residues P1 to P8 are bound inside the groove and the extension residues P- 1 Ala (one position N- terminal to P1) and P- 2 Arg (visible only for FV 20mer) protrude out of the groove (Fig 1A, upper panel). The peptide backbones and side chains adopt nearly identical positions between P3 and P8, but clear differences are seen at P- 1, P1, and P2. The \(\mathrm{C}\alpha\) - atom positions have a 2.84- Å shift at P1 and 0.96- Å shift at P2. Comparisons of FA and FV 20mers with their corresponding FA (pink) and FV (green) 8mers (Fig 1A, lower panel) show that 8mer peptides adopt nearly identical bound conformations, and that P1 is the most divergent position in both 20/8mer pairs, with shifts in \(\mathrm{C}\alpha\) - atom of 3.13- Å for FAs and 1.42- Å for FVs. Finally, the electron density was clear over the entire length of the peptides, including the backbone and methyl side chain of P- 1 Ala and the backbone and part of the aliphatic side chain of P- 2 Arg (FV 20mer only) (Supplementary Fig. 1).  
+
+<|ref|>text<|/ref|><|det|>[[113, 525, 882, 875]]<|/det|>
+A close examination of how the FV 20mer peptide binds in the A pocket (Fig 1B, upper panel) shows that the main- chain nitrogen of P1 Phe is rotated in a position that is normally occupied by a canonical P1 side chain, as seen in FA and FV 8mers (Fig. 1A, lower panel). In this configuration, the P1 main- chain nitrogen forms a hydrogen bond with Asn63, and the extension residues P- 1 Ala and P- 2 Arg protrude out of the A pocket. Interestingly, the main- chain carbonyl oxygens of P- 1 and P- 2 residues form hydrogen bonds with the indole nitrogen of Trp167 (Fig 1B, upper panel), which likely stabilize the peptide backbone as it exits out of the groove. The structure also shows that the bulky P1 Phe side chain cannot occupy the terminal amino group canonical position, i.e., the cavity formed by residues Tyr7 and Tyr171, and instead the phenyl ring points toward the \(\alpha 1\) - helix (see also Fig. 4A, right panel). Finally, the main- chain carbonyl oxygen of P1 Phe forms hydrogen bonds with Tyr7 and Tyr159 (Fig 1B, upper panel). A similar rotation within the A pocket was also seen in the FA 20mer structure (Fig 1B, lower panel), with P1 main- chain nitrogen and P- 1 carbonyl oxygen forming hydrogen bonds to Asn63 and Trp167, respectively. In marked contrast to FV 20mer, however, the main- chain carbonyl group of P1 Phe is rotated toward the \(\alpha 1\) - helix, a highly unusual configuration, and surprisingly does not engage with any MHC I residues (see also Fig. 4A, left panel). Taken together, although P1 Phe residues of 20mer peptides have undergone a similar main- chain nitrogen rotation in the A pocket, there are clear differences in the binding mode of each peptide. Finally, a comparison of Arg62 in FA and FV 20mer structures relative to the 8mer structures (Fig. 1C) shows that the Arg side chains swing out of their canonical positions which creates an opening for \((\mathrm{RA})_6\) residues to exit out of the groove. A similar role of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 881, 125]]<|/det|>
+residue 62 in opening the A pocket was observed in our structure of (RA)6AAKKKYCL 20mer peptide bound to HLA- B8E76C29.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 142, 843, 160]]<|/det|>
+## Significant peptide-induced conformational distortions at the N-terminus of the groove  
+
+<|ref|>text<|/ref|><|det|>[[113, 160, 882, 490]]<|/det|>
+The binding modes of FA and FV 20mers described in Fig. 1 are accompanied with significant structural changes in the MHC I groove. Figure 2 shows top and side views of the MHC I groove of FA and FV 20mer structures in which long stretches of N- terminal residues (shown by dashed red lines) in the \(\alpha 1\) - helix and loop connecting the \(\alpha 1\) - helix to \(\beta\) - strand of the floor were missing in the electron density maps. Specifically, the FA 20mer structure lacks 6 residues from Gln54 to Tyr59, and FV 20mer structure lacks as many as 18 residues from Ser42 to Tyr59 - electron densities at an acceptable \(1\sigma\) threshold were not visible for these residues in our structures. Consequently, the A pocket is unstructured and widely open- ended (Fig. 2 and Supplementary Fig. 2). In contrast, our previously determined AA 20mer structure showed that the N- terminus of the groove has a native structure (Fig. 2)29. Thus, the FA and FV 20mer structures provide direct evidence that the N- terminus of the groove has the potential to undergo significant peptide- induced structural distortions, highlighting its remarkable inherent plasticity. To the best of our knowledge, this is the first report of MHC I structures, with or without a bound peptide, showing such significant conformational distortions in the groove. Other than these differences, minor changes in the groove were detected between the FA and FV 20mer structures (r.m.s. deviation of 0.13- A). It is interesting that FA and FV 20mer peptides differ only by the nature of residues at P1 and P2 relative to our previous AA 20mer peptide29. Because all N- terminal MHC I residues were visible in the AA 20mer structure (Fig. 2), but not in the FA and FV 20mer structures, this strongly suggests that P1 and P2 residues play critical roles in the conformational maturation of the groove.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 508, 685, 525]]<|/det|>
+## Unusual MHC I-peptide interaction at the N-terminus of the groove  
+
+<|ref|>text<|/ref|><|det|>[[113, 526, 882, 770]]<|/det|>
+To understand the role of peptide P1 residue in modulating interactions with MHC I residues at the N- terminus of the groove, we analyzed the FA and FV 20mer structures in the context of both the AA 20mer29 and FA 8mer structures (Fig. 3). The structure of AA 20mer shows that the small P1 Ala methyl side chain forms a hydrophobic interaction with conserved residue Tyr59 (Fig. 3A, left panel). In contrast, the large P1 Phe side chains of FA and FV 20mers sterically clash with Tyr59 (Fig. 3A, left panel) and as such, residues Gln54 to Tyr59 become disorganized and are not visible in the FA and FV 20mer structures while these residues are conformed in AA 20mer structure. Interestingly, the native structure of FA 8mer (Fig. 3A, right panel) shows that the P1 side- chain phenyl ring occupies a canonical position and forms hydrophobic interactions with Tyr59, and that residues Gln54 to Tyr59 are visible. Similar observations were made in the FV 8mer structure (Supplementary Fig. 3). Taken together, our structures indicate that residues Gln54 to Tyr59, and especially Tyr59, form a region of remarkable conformational flexibility and structural plasticity that allows adaptations in response to size and configuration of peptide P1 side chains.  
+
+<|ref|>text<|/ref|><|det|>[[114, 770, 882, 892]]<|/det|>
+There is another unusual feature at the N- terminus of the groove in FA 20mer structure (Fig. 3B). In this structure (Fig. 3B, left panel), the P1 side- chain phenyl ring forms highly unusual hydrophobic interactions with the highly conserved Ile52 of the \(3_{10}\) - helix (a conserved element comprising residues 50 to 55). In the FV 20mer structure, however, the P1 phenyl ring is oriented differently (Fig. 3B, left panel) and, as such, it cannot engage with Ile52 thus causing residues Ser42 to Glu53 to become disordered. In the FA20mer structure, residues Ser42 to Glu53 were clearly visible due to P1 phenyl ring engagement with Ile52. Notably, interaction between a bound
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 882, 177]]<|/det|>
+peptide and Ile52, or any other residues of the \(3_{10}\) - helix, have not been reported before and are also not seen in FA and FV 8mer structures (Fig. 3B, right panel, and Supplementary Fig. 3). Taken together, there are clear molecular interplays involving peptide P1 residue and N- terminal residues Tyr59 and Ile52 that influence structural integrity at the N- terminal end of the groove (see also below).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 194, 881, 228]]<|/det|>
+## Molecular cross-talks between peptide P1 and P2 residues and N-terminal MHC I residues Ile52 and Tyr59  
+
+<|ref|>text<|/ref|><|det|>[[113, 230, 882, 490]]<|/det|>
+Results in Fig. 3 raise another important question: why is the P1 Phe side chain oriented differently in FA 20mer versus FV 20mer structures? This question is important given that this difference in orientation significantly affected the structural integrity of Ser42 to Glu53 in the FV 20mer structure. To address this, we examined peptide P2 residues, which is Ala in FA 20mer and Val in FV 20mer (Fig. 4A). In the FA 20mer structure (Fig. 4A, left panel), the small side- chain methyl group of P2 Ala allows the P1 main- chain carbonyl group to undergo an unusual rotation toward the \(\alpha 1\) - helix, which has the effect of orienting the P1 phenyl ring close to Ile52. In the FV 20mer structure, however, because P2 carries a larger Val side chain (Fig. 4A, right panel), the P1 main- chain carbonyl group cannot similarly rotate toward the \(\alpha 1\) - helix, and consequently the P1 phenyl ring is positioned further away from Ile52. Overall, different networks of interactions involving peptide P1 and P2 residues and N- terminal MHC I residues were established in FA and FV 20mer structures (Fig. 4A). Interestingly, using a thermal denaturation assay, we determined rather similar melting temperature (Tm), \(65.8^{\circ}\mathrm{C}\) for FA 20mer and \(67.5^{\circ}\mathrm{C}\) for FV 20mer. In contrast, the structures of FA and FV 8mer show very similar networks of interactions (Fig. 4B) and identical Tm values, \(71.1^{\circ}\mathrm{C}\) for FA 8mer and \(71.4^{\circ}\mathrm{C}\) for FV 8mer.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 508, 581, 525]]<|/det|>
+## Analysis of MHC I residues in pockets along the groove  
+
+<|ref|>text<|/ref|><|det|>[[113, 526, 882, 840]]<|/det|>
+Given the importance of the six binding pockets A to F in determining the peptide side chain specificities of HLA alleles, we compared the side chain orientations of MHC I residues in pockets A and B of FA and FV 20mer structures relative to the corresponding 8mer structures (Supplementary Fig. 4). Pocket A is made up of 9 residues and typically anchors the N- terminal amino group and P1 residue and closes the N- terminal end of the groove. The analysis shows that Tyr59 (conserved) and Asn63 (highly conserved) have the most divergent orientations in both FA and FV 20/8mer pairs, with some differences also seen in Tyr171 (conserved) and Trp167 (highly conserved). Pocket B is made up of 9 residues and binds the P2 peptide side chain that defines HLA binding motifs. For both the FA and FV 20/8mer pairs, all MHC I residues adopted very similar orientations, except for Asn63 at the boundary of the A and B pockets. Similar analyses in pockets C to F indicated that there were minimal changes in MHC I side chain orientations (data not shown). Taken together, the binding of 20mer peptides affects the configuration of MHC I residues in pocket A more significantly than those in pocket B, consistent with P1 being the most divergent position of these peptides (Fig. 1A). This analysis also highlighted a critical role for Asn63 in MHC I maturation; while Asn63 mediated several hydrophobic and hydrogen bond interactions with P1 and P2 residues in the FA and FV 20mer structures (Fig. 4A), Asn63 formed only one hydrogen bond with P2 main- chain nitrogen in the native FA and FV 8mer structures (Fig. 4B).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 715, 109]]<|/det|>
+## MD simulations: conformational flexibility and geometrical parameters  
+
+<|ref|>text<|/ref|><|det|>[[115, 108, 882, 161]]<|/det|>
+To further characterize HLA- B8E76C- peptide complexes and interaction between peptide P1 residue and Tyr59 and Ile52, we analyzed thermal properties, i.e., conformational flexibility and geometrical parameters, from MD simulations at physiological temperature.  
+
+<|ref|>text<|/ref|><|det|>[[115, 161, 882, 334]]<|/det|>
+The flexibility of individual MHC I residues along the heavy chain is characterized by its root mean square fluctuation (RMSF) (Fig. 5A). Results show that RMSF values are clearly higher in the region comprising residues 41 to 62 (shown in a red box) for FA and FV 20mer complexes relative to other regions, indicating that residues 41 to 62 are conformationally more flexible and thermally more unstable. A zoom- in revealed that there are two distinct regions (Fig. 5B); residues 41 to 46 and residues 52 to 62. In the region of residues 52 to 62, the RMSF values for FA and FV 20mer complexes are much higher than those of the other complexes, indicating that the 20mer complexes are conformationally more flexible in this region. However, in the region of residues 41 to 46, all complexes have similarly high RMSF values, suggesting that conformational flexibility is independent on the bound peptide in this region.  
+
+<|ref|>text<|/ref|><|det|>[[113, 334, 882, 857]]<|/det|>
+We evaluated next the geometrical parameters of interaction between peptide P1 residue and Tyr59 and Ile52 to gain further insights into these complexes (see Methods for details). Figure 5C shows the probability distributions of inter- residue distance between P1 and Tyr59 for all five systems. Results show that the peak of the distribution is centered around 5Å for AA 20mer while the other distributions are centered around 6Å, suggesting a high probability of interaction between P1 and Tyr59 in all complexes when equilibrated at physiological temperature. However, because the distributions are broader for FA and FV 20mers than for the other systems, this suggests that interaction between P1 and Tyr59 is less stable in the 20mer complexes. We also calculated the probability distributions of relative orientation between the aromatic rings of P1 Phe and Tyr59 (Supplementary Fig. 5A). Results show that FA and FV 8mers have symmetric distributions centered around \(90^{\circ}\) , indicating that the two aromatic rings are essentially perpendicular to each other. In contrast, for FA and FV 20mers, the peaks of the distributions are positioned at \(110^{\circ}\) and \(150^{\circ}\) , respectively, showing that the aromatic rings of P1 Phe and Tyr59 are no longer perpendicular to each other. Accordingly, we conclude that the configuration of \(\pi - \pi\) interactions \(^{31}\) between P1 Phe and Tyr59 is more parallel- like in FV 20mer but changes toward T- shape- like in FA 20mer, and finally adopts the stable T- shaped structure in FA and FV 8mers. Finally, Fig. 5D clearly shows that the probability distributions of inter- residue distance between P1 and Ile52 are different among the five complexes. For FA 20mer, the distribution has a high peak centered around 5Å, indicating that there is a stable interaction between these two residues. For FV 20mer, the distribution still peaked around 5Å but it is much wider, indicating that interaction between P1 and Ile52 is less stable (Fig. 5D). In contrast, the distributions for FA and FV 8mers have high peaks centered around 10Å, indicating that there are no interactions between P1 and Ile52 in these complexes. For AA 20mer, since the distances are centered around 7Å, we infer that weak hydrophobic interaction between P1 and Ile52 may exist in this complex. To further characterize interaction between P1 and Ile52 in FA and FV 20mers, we determined the probability distributions of angle C- H- X (X being the center of Phe aromatic ring) (Supplementary Fig. 5B). Results show that the angles are mostly in the range \(120^{\circ}\) to \(150^{\circ}\) . As such, since the probability distributions of “distance” (Fig. 5D) and “defined angle” (Supplementary Fig. 5B) between P1 and Ile52 satisfy the geometric criterion of CH- \(\pi\) interaction \(^{32}\) , we conclude that CH- \(\pi\) interaction exists in both FA and FV 20mers.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 90, 740, 108]]<|/det|>
+## MD simulations: rotation of peptide terminal amino group in the A pocket  
+
+<|ref|>text<|/ref|><|det|>[[115, 108, 882, 281]]<|/det|>
+Given that the terminal amino groups of FA and FV 20mers adopt highly unusual configurations in the A pocket (Fig. 1B), we conducted MD simulations to assess whether these amino groups can undergo spontaneous rotations to canonical positions, i.e., pointing down in the A pocket. For these tests, we removed the extension \((\mathrm{RA})_6\) residues of FA and FV 20mer peptides in their structures, generating FA20.8mer and FV20.8mer (see Methods for details). The definition of the dihedral angle \(\omega\) for characterizing the rotation of the terminal amino group is shown in Fig. S6A, using as an example FV20.8mer. In the FA and FV 20mer structures, \(\omega\) values are \(- 110^{\circ}\) and \(- 73^{\circ}\) , respectively, indicating that both terminal amino groups point up, while in the native FA and FV 8mer structures, \(\omega\) values are \(96^{\circ}\) and \(98^{\circ}\) , respectively, indicating that the amino groups point down.  
+
+<|ref|>text<|/ref|><|det|>[[115, 281, 882, 457]]<|/det|>
+The evolution curves of the dihedral angle \(\omega\) in MD simulations for FA20.8mer and FV20.8mer are shown in Supplementary Figs. 6B and 6C, respectively. For FA20.8mer, the results show that \(\omega\) changes from about \(- 100^{\circ}\) to \(100^{\circ}\) in two out of the three replicate simulations within \(300~\mathrm{ns}\) . For FV20.8mer, \(\omega\) changes from about \(- 70^{\circ}\) to around \(- 260^{\circ}\) in two out of the three replicate simulations within \(300~\mathrm{ns}\) (note that \(\omega = - 260^{\circ}\) is equivalent to \(\omega = 100^{\circ}\) , when the periodicity of \(\omega\) is taken into consideration). Hence, we conclude that the terminal amino groups of FA20.8mer and FV20.8mer have high probability to rotate in the A pocket and point down as seen in native structures. In addition, it is interesting that different replicas gave different evolution curves of the dihedral angle \(\omega\) , indicating that the pathways of such rotations are unlikely to be unique.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 472, 881, 506]]<|/det|>
+## Analysis of our structures in the context of bat MHC I molecules and human MHC II molecules  
+
+<|ref|>text<|/ref|><|det|>[[115, 507, 882, 716]]<|/det|>
+The characterization of bat MHC I genes identified thus far, showed that many of these molecules contain a 3- or 5- amino acid insertion in their binding groove \(^{33}\) . Several structures of PtaI- N\*01:01 bat MHC I molecules with a 3- amino acid insertion have been determined and revealed that the insertion creates a turn at the N- terminus of the groove (Fig. 6A) \(^{34,35}\) , within the critical region of Gln54 to Tyr59 (shown in red). The bat structures also revealed that in this turn, Asp59 forms salt bride interactions with Arg65 of the \(\alpha 1\) - helix (Fig. 6A) \(^{34,35}\) . Such a pairing of charged residues at these positions is a highly conserved feature of bat MHC I molecules \(^{34,35}\) , and it is expected to add structural rigidity at the N- terminus of the groove relative to human MHC I molecules. As such, the groove of bat MHC I molecules may be more restricted in its ability to adopt alternate conformations, which could affect peptide editing and repertoire development with implication on bat adaptive immunity. Further investigation is required to understand these observations.  
+
+<|ref|>text<|/ref|><|det|>[[115, 717, 882, 856]]<|/det|>
+Finally, in MHC class II (MHC II) molecules, there is evidence of peptide- induced conformational flexibility at the N- terminus of the groove \(^{36 - 38}\) , i.e., the region recognized by the peptide- exchange catalyst HLA- DM \(^{39,40}\) . The binding of HLA- DM induces conformational changes in MHC II, particularly in the \(3_{10}\) - helix and unstructured loop \(^{41}\) (shown in dark blue in Fig. 6B). This critical region of MHC II overlaps with the region of MHC I ( \(3_{10}\) - helix and extended region) that we identified as important for shaping the A and B pockets (shown in red in Fig. 6B). This analysis lends support to the view that the N- terminal end of the groove likely has a role in peptide editing (see Discussion).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 205, 106]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[115, 108, 882, 352]]<|/det|>
+The crystal structures of native peptide- filled MHC I molecules have taught us a great deal about molecular recognition of bound peptides. These structures represent the endpoint of a complex intracellular maturation process whereby MHC I molecules acquire peptides of sufficiently high affinity to ensure efficient antigen presentation. Biophysical, NMR, and in silico studies have been consistent in demonstrating that MHC I molecules explore intermediate conformational states in solution during peptide binding. Given that the native crystal structures of peptide- filled MHC I molecules are always nearly identical, it has not been possible thus far to obtain detailed structural information of intermediate conformations explored by MHC I molecules. This makes the notion of functional dynamics and peptide- induced conformational motions elusive. We have determined the crystal structures of HLA- B8 loaded with 20mer peptides that reveal highly unusual conformational and structural features in both MHC I and peptides. We have also carried out MD simulations to gain further insights into peptide- dependent interaction in these complexes, which provided new information and helped refine our structural analyses.  
+
+<|ref|>text<|/ref|><|det|>[[113, 352, 882, 650]]<|/det|>
+The FA and FV 20mer structures showed that the N- terminus of the groove has undergone significant conformational distortions relative to native FA and FV 8mer structures (Fig. 2 and Supplementary Fig. 2). These differences were peptide specific. MD simulations identified that residues 52 to 62 are most conformationally flexible and thermally unstable in FA and FV 20mer complexes relative to the other complexes (Fig. 5B), consistent with the lack of electron density for residues 54 to 59 in FA and FV 20mer structures. MD simulations also identified a conformationally flexible region comprising residues 41 to 46 that seems more peptide independent. Consistent with this, it is interesting that several deposited structures of HLA- B8 loaded with 9mer peptides lack clear electron density between residues \(\sim 41\) to 49 (for example, 1M05, 3SKO, 4QRQ, and 5WMR), suggesting that some conformational fluctuations can persist at the N- terminus of the groove in native structures. Our structures also showed that the 20mer peptides, with only a single amino acid difference at P2, adopted different and unusual backbone and side chain orientations at P1 and P2 but overlapped almost identically between P3 to P8 (Fig. 1A, upper panel). Interestingly, both complexes had similar thermostabilities, \(65.8^{\circ}\mathrm{C}\) (FA 20mer) versus \(67.5^{\circ}\mathrm{C}\) (FV 20mer). We suggest that the FA and FV 20mer structures, although static snapshots, represent discrete states that are explored by these complexes as they navigate the conformational space to locate their low energy conformation ("native") (Fig. 7) (see below).  
+
+<|ref|>text<|/ref|><|det|>[[113, 650, 882, 892]]<|/det|>
+MHC I peptide ligands are usually 8 to 10 amino acids long and bind with their terminal amino group pointing down in the A pocket, as seen in the FA and FV 8mer structures (Fig. 4B). Because short linear peptides are generally unstructured in solution, it is reasonable to assume that they land as such within the immature MHC I groove. NMR and other biophysical studies showed that in the initial binding steps, incoming peptides are loosely accommodated in the groove until more specific conformational adaptions take place in both peptides and MHC I<sup>22,23,42,43</sup>. It is therefore plausible that when peptides of optimal lengths are first captured by MHC I, they adopt conformations that resemble those of our 20mer peptides, i.e., with an unusual rotation of the terminal amino group in the A pocket (Fig. 1B), until folding proceeds and peptides adopt a canonical conformation (or not). To test this, we simulated the relaxation process of FA20.8mer and FV20.8mer using plain MD simulations and the results showed that peptide terminal amino groups have a high probability to rotate to a canonical position from the unusual orientations observed in our structures. The simulations also suggested that the configurations of FA and FV 20mer peptides, as seen in the structures, represent "trapped" states (see below). Finally, it is worth
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 881, 142]]<|/det|>
+noting that naturally occurring HLA- B8- restricted 8 and 9mer peptides with a large residue at P1 or at P1/P2 have been reported: for example, Phe, Val, Leu, and His at P1<sup>44- 47</sup> or Tyr/Leu, Trp/Val, Phe/Leu, and Tyr/Ile at P1/P2<sup>44,48,49</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[112, 143, 882, 753]]<|/det|>
+The FA and FV 20mer structures showed that strictly conserved Tyr59 and highly conserved Ile52 (3<sub>10</sub>- helix) act synergistically at the N- terminus of the groove. In native MHC I structures, Tyr59 stabilizes the peptide terminal amino group together with Tyr171, and Ile52 acts as a structural support to Tyr59 and Tyr171<sup>1</sup>. In our 20mer structures, Tyr59 could not adopt its native position because of the large size of P1 Phe side chain in the A pocket (Fig. 3A). This caused residues 54 to 59 to become disordered and created an open- ended A pocket. In the FA 20mer structure, this disordering was accompanied with Ile52 forming a highly unusual interaction with peptide P1 Phe side chain, which in turn was facilitated by the small peptide P2 Ala side chain in the adjacent B pocket (Fig. 4A). In the FV 20mer structure, however, similar molecular cross- talks between the A and B pockets were not possible because of the larger peptide P2 Val side chain in the B pocket that positioned P1 Phe side chain further away from Ile52 (Fig. 4A). Consequently, a significantly longer stretch of residues became disordered in the FV 20mer structure, generating a widely open- ended A pocket. In performing MD simulations, we were able to probe other aspects of the interaction between peptide P1 and MHC I residues Tyr59 and Ile52 in the thermal equilibrium ensembles of our complexes. The simulations revealed that 1. interaction between P1 and Tyr59 is characterized by the more stable T- shaped \(\pi\) - \(\pi\) configuration in FA and FV 8mer complexes relative to FA and FV 20mer complexes (Fig. 5C and Supplementary Fig. 5A), which is consistent with residues 54- 59 being ordered in the 8mer structures but disordered in the 20mer structures; and 2. CH- \(\pi\) interaction between P1 and Ile52 exists only in FA and FV 20mer complexes, and it is more stable in FA 20mer than FV 20mer (Fig. 5D and Supplementary Fig. 5B). This is also consistent with the FV 20mer structure showing a more highly disordered groove and widely open- ended A pocket. Taken together, we suggest that the 20mer structures represent "trapped" states, i.e., conformations in which Tyr59 cannot adopt a "closed" position (FA and FV 20mer structures) and 3<sub>10</sub>- helix cannot mature into its secondary fold (FV 20mer structure). In other words, formation of native A and B pockets requires the conformational transitions of Tyr59 into a "closed" position and 3<sub>10</sub>- helix (Ile52) into its native fold. A role for the 3<sub>10</sub>- helix in peptide- induced MHC I maturation was suggested previously<sup>50- 52</sup>. Because Tyr59 and 3<sub>10</sub>- helix are strictly conserved elements in human HLA alleles, these transitions are expected to be universal in MHC I maturation. Furthermore, that the 20mer structures can tolerate such high degrees of peptide- induced structural perturbations at the N- terminus of the groove is consistent with MD simulations of other groups showing that bound peptides can dissociate partially at the N- terminus of groove while remaining anchored within the F pocket<sup>53- 55</sup>, with potential implication in MHC I maturation (see below). Finally, the lower thermostabilities of FA and FV 20mer complexes relative to the 8mer complexes is consistent with the "trapped" states being higher energy conformations (Fig. 7), but of sufficiently low energy to be revealed crystallographically.  
+
+<|ref|>text<|/ref|><|det|>[[115, 752, 882, 891]]<|/det|>
+The conformational transitions discussed above are intricately coupled to occupancy of the A and B pockets. This is significant when considering that the crystal structure of a peptide- free MHC I molecule (HLA- A2) showed no substantial differences at the N- terminus (or C- terminus) relative to native HLA- A2 structures, except for minor differences in some side chain orientations within the A pocket<sup>27</sup>. It is also significant because, in native MHC I structures, the A and B pockets anchor the peptide terminal amino group and P2 residue, respectively, that contribute to protein stability. The FA and FV 20mer structures thus likely captured intermediate conformations adopted by MHC I when actively "evaluating" peptide P1 and P2 residues as part of editing, with
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 882, 125]]<|/det|>
+the exact molecular features of MHC I intermediates expected to be fluid as maturation proceeds based on our MD simulations.  
+
+<|ref|>text<|/ref|><|det|>[[112, 125, 883, 490]]<|/det|>
+Previous studies indicated that molecular dynamics around the F pocket is a major driving force for recognition of MHC I molecules by tapasin and TAPBRP7- 15. Conformational flexibility at the N- terminus of the groove of MHC II molecules is also a critical determinant of HLA- DM- mediated mechanism of peptide exchange41. As such, our structures of FA and FV 20mers thus raise the intriguing question of whether the region close to the A and B pockets represents a binding surface recognized by a protein, yet to be identified, with a role in stabilizing MHC I intermediates and/or peptide editing. It is interesting that the cryo- EM structure of the PLC showed that the long P- domain of CRT chaperone is positioned atop and across the MHC I groove with its tip interacting with ERp57'16. In this spatial organization, the P- domain could interact transiently with the partially folded region of the A and B pockets, as facilitated by the inherent structural plasticity of the P- domain and dynamic nature of the P- domain/ERp57 interaction56,57. More work is required to examine this idea. The possibility that ERAP1, ERAP2, and/or ERAP1/ERAP2 heterodimer play a functional role at the N- terminal end of the groove is also very reasonable. We showed previously in biochemical studies that ERAP1 and ERAP1/ERAP2 can actively trim the protruding N- terminal residues of long peptides, including the AA 20mer, while bound to HLA- B8E76C28,29. Others also showed that ERAP1 trims peptides bound to H2- Kb using cell- based assays53. Moreover, it was demonstrated that mouse ERAAP (equivalent to ERAP1) synergizes with tapasin to edit peptide repertoires58. In fact, a role for MHC I molecules in antigen processing has long been suggested59. Our current study supports this view and furthermore suggests that the ERAP enzymes are more likely engaging with intermediate forms of MHC I molecules, rather than correctly conformed molecules60.  
+
+<|ref|>text<|/ref|><|det|>[[114, 490, 883, 612]]<|/det|>
+In conclusion, our study provided a crystallographic and MD characterization of conformational substates in MHC I- peptide systems, and it brought into focus the N- terminal end of the groove in mechanisms of high- affinity peptide selection. This understanding is critical given the role of MHC I- restricted peptide repertoires for activation of adaptive immune responses to control viral infections. Finally, our work opens new avenues to examine chaperoning of the groove around the A pocket, and it also encourages further characterization of other MHC I molecules.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 280, 106]]<|/det|>
+## Acknowledgements  
+
+<|ref|>text<|/ref|><|det|>[[115, 108, 883, 177]]<|/det|>
+We thank Dr. Bernard Santarsiero for expert assistance with remote x- ray data collection and discussion, and the staff at the Argonne National Laboratory (Argonne, IL) where all x- ray data were collected. This work was supported in whole, or in part, by NIAID, National Institutes of Health, Grants R01 AI114467, R01 AI108546, and R21 AI173863 (to MB).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 196, 296, 211]]<|/det|>
+## Author contributions  
+
+<|ref|>text<|/ref|><|det|>[[115, 212, 882, 281]]<|/det|>
+LL executed all experiments including x- ray crystallography and was assisted by Mansoor Batliwala in refolding MHC I- peptide complexes. LL and MB designed experiments and interpreted the structures. XP performed MD simulations and interpreted the results. MB wrote the manuscript with contributions from all authors.  
+
+<|ref|>text<|/ref|><|det|>[[112, 299, 789, 317]]<|/det|>
+Competing financial interests: The authors declare no competing financial interests.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 191, 106]]<|/det|>
+## Methods  
+
+<|ref|>text<|/ref|><|det|>[[115, 108, 882, 160]]<|/det|>
+Synthetic peptides. Peptides were synthesized by the solid- phase methodology (GenScript Biotech Co.) and purified by reverse- phase chromatography on a C18 HPLC column. Stock solutions of peptides in DMSO were stored at \(- 80^{\circ}\mathrm{C}\) .  
+
+<|ref|>text<|/ref|><|det|>[[115, 177, 882, 350]]<|/det|>
+Refolding of HLA- B\\*0801E76C complexes. Using the crystal structure of HLA- B\\*0801/GGRKKYKL (PDB code 1AGB), we identified residue Glu76 to be geometrically well- positioned to form a disulfide bond with the side chain of P7 peptide residue, after mutation with a cysteine, as we described previously<sup>28</sup>. The HLA- B\\*0801E76C heavy chain mutant was generated as described previously<sup>28</sup>. HLA- B\\*0801E76C complexes were reconstituted from ureasolubilized inclusion bodies of HLAB\\*0801E76C heavy chain (1 \(\mu \mathrm{M}\) ) and \(\beta_{2}\) - microglobulin (2 \(\mu \mathrm{M}\) ) with a synthetic Cys- P7 peptide (10 \(\mu \mathrm{M}\) ) in an oxidative refolding buffer at \(4^{\circ}\mathrm{C}^{61}\) . After 48 hours, the crude refolding mixture of HLA- B\\*0801E76C complexes was purified on a Superdex- 200 size exclusion chromatography column by FPLC. Stock solutions of purified complexes (10- 30 mg/ml) in 20 mM Tris- HCl, pH 7.5, 150 mM NaCl, were kept at \(- 80^{\circ}\mathrm{C}\) .  
+
+<|ref|>text<|/ref|><|det|>[[115, 367, 882, 543]]<|/det|>
+Crystallization. The initial crystallization condition of HLA- B\\*0801E76C/(RA)<sub>6</sub> FAKKKYCL (10 mg/ml) was identified using the Crystal Screen<sup>TM</sup> (Hampton Research, Riverside, CA) as solution #9 (0.2 M ammonium acetate, 0.1 M sodium citrate tribasic dihydrate, pH 5.6, 30% (w/v) PEG 4000) via the hanging- drop vapor diffusion method at room temperature. The initial crystals were optimized using different pH values (4.5- 7.0) and PEGs (6000- 10000; 10- 30%). These optimized crystals were used to generate a seeding solution in solution #9. Crystals used for data collection were grown by mixing 2 \(\mu \mathrm{l}\) of 10 mg/ml protein solution with 2 \(\mu \mathrm{l}\) of 0.2 M ammonium acetate, 18% PEG 4000, 0.1 M sodium citrate, pH 5.7, and 0.5 \(\mu \mathrm{l}\) of seeding solution. Similar crystallization conditions were used to collect data for HLA- B\\*0801E76C loaded with FA and FV 8mers and FV 20mer.  
+
+<|ref|>text<|/ref|><|det|>[[115, 560, 882, 750]]<|/det|>
+Data collection, structural determination, and refinement. X- ray diffraction data sets were collected with a MAR- 225 CCD detector at the LS- CAT beamline 21- ID- G (or 21- ID- F) of the Advanced Photon Source (Argonne National Laboratory, Argonne, IL). Data were integrated and scaled with the HKL2000 program package<sup>62</sup> or XDS<sup>63</sup>. Details of data processing are indicated in Supplementary Table 1. The structures of all complexes were solved by molecular replacement using Phaser<sup>64</sup> (the initial search model was HLA- B\\*0801E76C/R(N- Me)AAAKKKYCL (PDB code 6P2S). Structure refinement of all models was carried out in Phenix (or Refmac in CCP4)<sup>65- 67</sup> and manual building with COOT<sup>68</sup>. Final refinement statistics are summarized in Supplementary Table 1. The atomic coordinates of all structures have been deposited in the Protein Data Bank with the following accession codes: FA 8mer (8E13), FA 20mer (8E2Z), FV 8mer (8E81), and FV 20mer (8EC5).  
+
+<|ref|>text<|/ref|><|det|>[[115, 768, 882, 873]]<|/det|>
+Thermal denaturation assay. A thermal denaturation assay was performed using reaction mixtures consisting of 7 \(\mu \mathrm{l}\) of a complex (final concentration of 2 \(\mu \mathrm{M}\) ), 7 \(\mu \mathrm{l}\) of 10x SYPRO orange dye (5000x, Thermo Fisher Scientific, Waltham, MA) and 7 \(\mu \mathrm{l}\) of 50 mM HEPES, pH 7.2, 150 mM NaCl. Each mixture (total volume 21 \(\mu \mathrm{l}\) ) was analyzed in quadruplicate using an ABI ViiA7 RT- PCR instrument (Life Technologies, Inc., Carlsbad, CA). A temperature gradient from 25 to \(95^{\circ}\mathrm{C}\) with continuous increment of \(0.06^{\circ}\mathrm{C / sec}\) was used to generate the denaturation curves. The
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 882, 125]]<|/det|>
+averaged denaturation curves were plotted as “fluorescence intensity” versus “temperature”, and the minimum point of the first derivative of each curve provided the melting temperature.  
+
+<|ref|>text<|/ref|><|det|>[[113, 142, 882, 264]]<|/det|>
+MD initial structures preparation. The x- ray crystal structures of HLA- B8E76C loaded with five different peptides were analyzed by MD simulations: FA and FV 8mers (this study), AA 20mer (6P2C) \(^{29}\) , and FA and FV 20mers (this study). The missing regions of HLA- B8E76C in the structures of FA and FV 20mers were complemented using software UCSF Chimera \(^{69}\) with the structure of AA 20mer as the template. The missing regions of the bound 20mer peptides were complemented using Modeller \(^{70}\) integrated in software UCSF Chimera \(^{69}\) . In the simulations, the peptide termini were neutralized to exclude artificial charge effects.  
+
+<|ref|>text<|/ref|><|det|>[[112, 281, 882, 577]]<|/det|>
+MD simulations set- up. Depending on the purpose, two different MD simulations were performed: replica exchange MD (REMD) and plain MD. The REMD simulations were used for the thermal properties analysis due to its efficiency in thermal equilibrated conformational ensemble sampling, while the plain MD simulations were used for simulating the rotation of the terminal amino group of bound FA20.8mer and FV20.8mer from their unusual up orientations. Both types of MD simulations included common settings, as follows. Simulations were performed with explicit solvent using the software package GROMACS 5.1. \(^{71,72}\) . The force field CHARMM36m \(^{73}\) together with its own modified TIP3P water model \(^{74}\) was also used. The LINCS algorithm was applied to constrain the covalent bonds with H- atoms, and the time step in simulation was 2 fs. The protein was simulated in 0.1 M aqueous NaCl solution. After a short energy minimization, an NVT simulation of 100 ps with the V- rescale temperature coupling at 310 K was performed, followed by an NPT simulation of 300 ps with the Parrinello- Rahman coupling method at a reference pressure of 1 bar. The relaxation times for the temperature coupling and pressure coupling are 0.1 ps and 2 ps, respectively. During NVT and NPT simulations, the protein backbone is constrained to its initial structure. At the end, we removed the constraints and performed production simulations at the same temperature and pressure. The time interval for conformational sampling in simulations was 20 ps.  
+
+<|ref|>text<|/ref|><|det|>[[113, 594, 882, 751]]<|/det|>
+REMD simulations. After a short thermal equilibration process with NVT and NPT, as described above, we set up 30 replicas with temperature distributed from 300 K to 340 K following the webserver (https://virtualchemistry.org/remd- temperature- generator/) \(^{75}\) with an attempt swap duration of 1 ps between two neighboring temperatures. The average acceptance probability for the replica exchanges was about 30%. Each replica ran for 40 ns, and thus we have a simulation with total time up to 1.2 ��s. The thermal equilibrated ensembles were collected from the replica at temperature 310 K from the REMD simulations, from which the flexibility of HLA- B8E76C heavy chain and interactions between peptide residue P1 and MHC I residues Tyr59 and Ile52 were analyzed.  
+
+<|ref|>text<|/ref|><|det|>[[113, 768, 882, 891]]<|/det|>
+Plain MD simulations. Using the FA and FV 20mer structures, as generated in the “Initial structures preparation” stage, we generated the initial configurations of FA20.8mer and FV20.8mer by deleting the extension (RA) \(_{6}\) residues. After a short thermal equilibration process with NVT and NPT as described above, we performed the production run for 300 ns to simulate the relaxation process of FA20.8mer and FV20.8mer. During the simulations, we observed rotations of the terminal amino groups in FA20.8mer and FV20.8mer peptides. We repeated the simulation three times for each system to ensure producibility.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 106, 883, 316]]<|/det|>
+Interaction Analysis. We identified possible interactions between peptide residue P1 and MHC I residues Tyr59 and Ile52 according to the geometrical properties, i.e., distances and relative orientations between the two residues. For interaction between P1 Phe and Tyr59, the distance was defined between the geometrical centers of the two aromatic rings, while the relative orientation was defined by the angle between the normal directions of the aromatic rings. For interaction between P1 Ala of AA 20mer and Tyr59, the distance was defined between the Cβ atom of P1 Ala and the geometrical center of the aromatic ring of Tyr59. For interaction between P1 Phe and Ile52, the distance was defined between the geometrical center of the aromatic ring in P1 Phe and the Cδ atom of Ile 52, while the angle C- H- X was defined by the atoms Cδ of Ile52, hydrogen atom covalently bonded to Cδ and the center of the Phe aromatic ring. Finally, for interaction between P1 Ala of AA 20mer and Ile52, the distance was defined between the Cβ atom in P1 Ala and the Cδ atom of Ile52.  
+
+<|ref|>text<|/ref|><|det|>[[113, 332, 883, 475]]<|/det|>
+Analysis of rotations of peptide terminal amino groups. We probed rotations of terminal amino groups in pocket A by monitoring the evolution of the dihedral angle \(\omega\) formed by atoms Cys76:CA- P7:CA- P1:CA- P1:N (Supplementary Fig. 6A) in the plain MD simulations. When drawing evolution curves, we reasonably required that the rotation between two neighboring frames be less than \(180^{\circ}\) , i.e., \(|\omega (t_{i + 1}) - \omega (t_{i})| < 180^{\circ}\) . Otherwise, we added/subtracted the dihedral angle \(\omega (t_{i + 1})\) by \(360^{\circ}\) . In the end, to eliminate thermal fluctuations and visualize the rotation pathways better, we smoothed the evolution curves by an average sliding window of 10 ns.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 90, 210, 106]]<|/det|>
+References  
+
+<|ref|>text<|/ref|><|det|>[[111, 107, 883, 144]]<|/det|>
+1. Saper, M. A., Bjorkman, P. J. & Wiley, D. C. Refined structure of the human histocompatibility antigen HLA-A2 at 2.6 Å resolution. J. Mol. Biol. 219, 277-319 (1991).  
+
+<|ref|>text<|/ref|><|det|>[[111, 158, 883, 196]]<|/det|>
+2. Matsumura, M., Fremont, D. H., Peterson, P. A. & Wilson, I. A. Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257, 927-934 (1992).  
+
+<|ref|>text<|/ref|><|det|>[[111, 210, 883, 264]]<|/det|>
+3. Fahnestock, M. L., Johnson, J. L., Feldman, R. M., Tsomides, T. J., Mayer, J., Nahri, L. O. & Bjorkman, P. J. Effects of peptide length and composition on binding to an empty class I MHC heterodimer. Biochem. 33, 8149-8158 (1994).  
+
+<|ref|>text<|/ref|><|det|>[[111, 280, 883, 316]]<|/det|>
+4. Madden, D. R. The three-dimensional structure of peptide-MHC-complexes. Annu. Rev. Immunol. 13, 587-622 (1995).  
+
+<|ref|>text<|/ref|><|det|>[[111, 331, 883, 368]]<|/det|>
+5. Bouvier, M. & Wiley D. C. Importance of peptide amino and carboxyl termini to the stability of MHC class I molecules. Science 265, 398-402 (1994).  
+
+<|ref|>text<|/ref|><|det|>[[111, 383, 883, 420]]<|/det|>
+6. Cresswell, P., Bangia, N., Dick, T & Diedrich, G. The nature of the MHC class I peptide loading complex. Immunol. Rev. 171, 21-28 (1999).  
+
+<|ref|>text<|/ref|><|det|>[[111, 435, 883, 489]]<|/det|>
+7. Van Hateren, A., Bailey, A., Werner, J. M. & Elliott T. Plasticity of empty major histocompatibility complex molecules determines peptide-selector function. Molec. Immunol. 68, 98-101 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[111, 504, 883, 541]]<|/det|>
+8. Ayres, C. M., Corcelli, S. A. & Baker B. M. peptide and peptide-dependent motions in MHC proteins: immunological and biophysical underpinnings. Front. Immunol. 8, 935 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 556, 883, 593]]<|/det|>
+9. Yanaka, S. & Sugase, K. Exploration of the conformational dynamics of major histocompatibility complex molecules. Frontiers Immunol. 8, 632 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 608, 883, 662]]<|/det|>
+10. Wieczorek, M. Abualrous, E. T., Stich, J., Alvaro-Benito, M., Stolzenberg, S., Noe, F. & Freund, C. Major histocompatibility complex (MHC) class I and MHC class II proteins: conformational plasticity in antigen presentation. Front. Immunol. 8, 292 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 677, 883, 714]]<|/det|>
+11. Fisette, O., Wingbermuhle, Tampe R. & Schafer, L.V. Molecular mechanism of peptide editing in the tapasin-MHC I complex. Sc. Reports 19085 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[111, 729, 883, 766]]<|/det|>
+12. Truong, H. V. & Sgourakis, N. G. Dynamics of MHC I-molecules in the antigen processing and presentation pathway. Curr. Opin. Immunol. 70, 122-128 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[111, 782, 883, 818]]<|/det|>
+13. Jantz-Naeem, N. & Springer, S. Venus flytrap or pas de trois? The dynamics of MHC class I molecules. Curr. Opin. Immunol. 70, 82-89 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[111, 834, 883, 870]]<|/det|>
+14. Thomas, C. & Tampe, R. MHC I assembly and peptide editing - chaperones, clients, and molecular plasticity in immunity. Curr. Opin. Immunol. 70, 48-56 (2021).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 90, 884, 145]]<|/det|>
+15. Margulies, D. H., Taylor, D. K., Jiang, J., Boyd, L. F., Ahmad, J., Mage, M. G. & Natarajan, K. Chaperones and catalysts: How antigen presentation pathways cope with biological necessity. Front. Immunol. 13, 859782 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[111, 158, 884, 196]]<|/det|>
+16. Blees, A., Januliene, D., Hofmann, T., Koller, N., Schmidt, C., Trowitzsch, S., Moeller, A. & Tampe, R. Structure of the human MHC I peptide-loading complex. Nature 551, 525–528 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 210, 884, 247]]<|/det|>
+17. Chen, M. & Bouvier, M. Analysis of interaction in a tapasin/class I complex provides a mechanism for peptide selection. EMBO J. 26, 1681–1690 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[111, 262, 884, 299]]<|/det|>
+18. Herman, C., Strittmatter, L. M., Deanne, J. E. & Boyle, L. H. The binding of TAPBPR and tapasin to MHC class I is mutually exclusive. J. Immunol. 191, 5743–5750 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[111, 313, 884, 369]]<|/det|>
+19. Herman, C., van Hateren, A., Trautwein, N., Neerincx, A., Duriez, P. J., Stevanovic, S., Trowsdale, J., Deane, J. E., Elliott, T. & Boyle L. H. TAPBPR alters MHC class I peptide presentation by functioning as a peptide catalyst. eLife 4, e09617 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[111, 384, 884, 421]]<|/det|>
+20. Thomas, C. & Tampe, R. Structure of the TAPBPR-MHC I complex defines the mechanism of peptide loading and editing. Science 358, 1060–1064 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 435, 884, 490]]<|/det|>
+21. Jiang, J., Natarajan, K., Boyd, L. F., Morozov, G. I., Mage, M. G. & Margulies, D. H. Crystal structure of the TAPBPR-MHC I complex reveals the mechanism of peptide editing in antigen presentation. Science 358, 1064–1068 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 505, 884, 577]]<|/det|>
+22. McShan, A., Devlin, C. A., Overall, S. A., Park, S. A., Toor, J. S., Moschidi, D., Flores-Solis, D., Choi, H., Tripathi, S., Procko, S. & Sgourakis, N. G. Molecular determinants of chaperone interactions on MHC-I folding and antigen repertoire selection. Proc. Natl. Acad. Sci. USA 116, 25602–25613 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[111, 592, 884, 646]]<|/det|>
+23. Van Hateren, A., Anderson, M., Bailey, A., Werner, J. M. Skipp, P. & Elliott, T. Direct evidence for conformational dynamics in major histocompatibility complex class I molecules. J. Biol. Chem. 292, 20255–20269 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 661, 884, 698]]<|/det|>
+24. Bouvier, M. & Wiley D. C. Structural characterization of a soluble and partial folded class I major histocompatibility heavy chain/β₂m heterodimer. Nat. Struct. Biol. 5, 377–384 (1998).  
+
+<|ref|>text<|/ref|><|det|>[[111, 712, 884, 767]]<|/det|>
+25. Fahnestock, M. L., Tamir, I., Narhi, L. & Bjorkman, P. J. Thermal stability comparisons of purified empty and peptide-filled forms of a class I MHC molecule. Science 258, 1658–1662 (1992).  
+
+<|ref|>text<|/ref|><|det|>[[111, 783, 884, 837]]<|/det|>
+26. Kurimoto, E., Kuroki, K., Yamaguchi, Y., Yagi-Utsumi, M., Igaki, T., Iguchi, T., Maenaka, K. & Kato, K. Structural and functional mosaic nature of MHC class I molecules in their peptide-free form. Molec. Immunol. 55, 393–399 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[111, 853, 884, 889]]<|/det|>
+27. Anjanappa, R., Garcia-Alai, M., Kopicki, J.-D., Lockhauserbaumer, J., Aboelmagd, M., Hinrichs, J., Nemtanu, I. M., Uetrecht, C., Zacharias, M., Springer, S. & Meijers, R. Structures of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 90, 881, 126]]<|/det|>
+peptide- free and partially loaded MHC class I molecules reveal mechanisms of peptide selection. Nat. Comm. 11, 1314 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[111, 141, 881, 195]]<|/det|>
+28. Chen, H., Li, L., Weimershaus, M., Evnouchidou, I., van Endert P. & Bouvier, M. ERAP1-ERAP2 dimers trim MHC I-bound precursor peptides; implications for understanding peptide editing. Sc. Rep. 6, 28902 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[111, 211, 881, 264]]<|/det|>
+29. Li, L., Batliwala, M. & Bouvier, M. ERAP1 enzyme-mediated trimming and structural analyses of MHC I-bound precursor peptides yield novel insights into antigen processing and presentation. J. Biol. Chem. 294, 18534-18544 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[111, 279, 881, 350]]<|/det|>
+30. Phillips, R. E., Rowland-Jones, S., Nixon, D. F., Gotch, F. M., Edwards, J. P., Ongulesi, A. O., Elvin, J. G., Rothbard, J. A., Bangham, C. R., Rizza, C. R. & McMichael, A. J. Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition. Nature 354, 453-459 (1991).  
+
+<|ref|>text<|/ref|><|det|>[[111, 366, 881, 402]]<|/det|>
+31. Tehrani, A. Z. & Kim, K. S. Functional molecules and materials by \(\pi\) -interaction based quantum theoretical design. Int. J. Quantum Chem. 116, 622-633 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[111, 417, 881, 454]]<|/det|>
+32. Wang J. & Yao, L. Dissecting C-H···π and N-H···π interactions in two proteins using a combined experimental and computational approach. Sci. Rep. 9, 20149 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[111, 470, 881, 523]]<|/det|>
+33. Ng, J. H. J., Tachedjian, M., Deakin, J., Wynne, J. W., Cui, J., Haring, V., Broz, I., Chen, H., Belov, K., Wang, L.-F. & Baker, M. L. Evolution and comparative analysis of the bat MHC -I region. Sc. Rep. 6, 21256 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[111, 539, 881, 592]]<|/det|>
+34. Lu, D., Liu, K., Zhang, D., Yue, C., Lu, Q., Cheng, H., Wang, L., Chai, Y., Qi, J., Wang, L.-F., Gao, G. F. & Liu, W. J. Peptide presentation by bat MHC class I provides new insights into the antiviral immunity of bats. PLoS Biol. 9, e30000436 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[111, 608, 881, 661]]<|/det|>
+35. Qu, Z., Li, Z., Ma, L., Wei, X., Zhang, L., Liang, R., Meng, G., Zhang, N. & Xia, C. Structure and peptidome of the bat MHC class I molecule reveal a novel mechanism leading to high-affinity peptide binding. J. Immunol. 202, 3493-3506 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[111, 678, 881, 731]]<|/det|>
+36. Natarajan, S. K., Stern, L. J. & Sadegh-Nasseri, S. Sodium dodecyl sulfate stability of HLA-DR1 complexes correlates with burial of hydrophobic residues in pocket 1. J. Immunol. 162, 3463-3470 (1999).  
+
+<|ref|>text<|/ref|><|det|>[[111, 747, 881, 800]]<|/det|>
+37. Zarutskie, J. A., Sato, A. K., Rusche, M. M., Chan, C., Lomakin, A., Benedek, G. B. & Stern, L. J. A conformational change in the human major histocompatibility complex protein HLA-DR1 induced by peptide binding. Biochem. 38, 5878-5887 (1999).  
+
+<|ref|>text<|/ref|><|det|>[[111, 816, 881, 888]]<|/det|>
+38. Sato, A. K., Zarutskie, J. A., Rusche, M. M., Lomakin, A., Natarajan, S. K., Sadegh-Nasseri, S., Benedek, G. B. & Stern, L. J. Determinants of the peptide-induced conformational change in the human class II major histocompatibility complex protein HLA-DR1. J. Biol. Chem. 275, 2165-2173 (2000).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 106, 884, 144]]<|/det|>
+39. Chou, C. L. & Sadegh-Nasseri, S. HLA-DM recognized the flexible conformation of major histocompatibility complex class II. J. Expm. Med. 192, 1697-1706 (2000).  
+
+<|ref|>text<|/ref|><|det|>[[112, 157, 882, 211]]<|/det|>
+40. Pos, W., Sethi, D. K., Call, M. J., Schulze, M.-S. E. D., Anders, A.-K., Pyrdol, J. & Wucherpfennig, K. W. Crystal structure of the HLA-DM-HLA-DR1 complex defines mechanism for rapid peptide selection. Cell 151, 1557-1568 (2012).  
+
+<|ref|>text<|/ref|><|det|>[[112, 226, 882, 281]]<|/det|>
+41. Painter, C. A., Negroni, M. P., Kellesberger, K. A., Zavala-Ruiz, Z., Evans, J. E. & Stern, L. J. Conformational lability in the class II MHC \(3_{10}\) helix and adjacent extended strand dictate HLA-DM susceptibility and peptide exchange. Proc. Natl. Acad. Sci. USA 108, 19329-19334 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[112, 295, 882, 349]]<|/det|>
+42. Yanaka, S., Ueno, T., Shi, Y., Qi, J., Gao, G. F., Tsumoto, K. & Sugase, K. Peptide-dependent conformational fluctuation determines the stability of the human leukocyte antigen class I complex. J. Biol. Chem. 289, 24680-24690 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[112, 364, 882, 418]]<|/det|>
+43. Hawse, W., Gloor, B. E., Ayres, A. M., Kho, K., Nuter, E. & Baker, B. M. Peptide modulation of class I major histocompatibility complex protein molecular flexibility and the implications for immune recognition. J. Biol. Chem. 288, 24372-24381 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[112, 433, 882, 505]]<|/det|>
+44. Flad, T., Spengler, B., Kalbacher, H., Brossart, P., Baier, D., Kaufmann, R., Bold, P., Metzger, S., Bluggel, M., Meyer, H. E., Kurz, B. & Muller, C. A. Direct identification of major histocompatibility complex class I-bound tumor-associated peptide antigens of a renal carcinoma cell line by a novel mass spectrometric method. Cancer Res. 58, 5803-5811 (1998).  
+
+<|ref|>text<|/ref|><|det|>[[112, 520, 882, 592]]<|/det|>
+45. Stickel, J. S., Weinzierl, A. O., Hillen, O., Drews, O., Schuler, M. M., Hennenlotter, J., Wernet, D., Muller, C. A., Stenzl, A., Rammensee, H.-G. & Stevanovic, S. HLA-ligand profiles of primary renal cell carcinoma maintained in metastases. Cancer Immunol. Immunother. 58, 1407-1417 (2009).  
+
+<|ref|>text<|/ref|><|det|>[[112, 608, 882, 662]]<|/det|>
+46. Adair, S. J., Carr, T. M., Fink, M. J., Slingluff Jr, C. L. & Hogan, K. T. The TAG family of cancer/testis antigens is widely expressed in a variety of malignancies and gives rise to HLA-A2-restricted epitopes. J. Immunol. 31, 7-17 (2008).  
+
+<|ref|>text<|/ref|><|det|>[[112, 677, 882, 748]]<|/det|>
+47. Koziel, M. J., Dudley, D., Afdhal, N., Grakuoi, A., Rice, C. M., Choo, Q. L., Houghton, M. & Walker, B. D. HLA class I-restricted cytotoxic T lymphocytes specific for hepatitis C virus. Identification of multiple epitopes and characterization of patterns of cytokine release. J. Clin. Invest. 96, 2311-2321 (1995).  
+
+<|ref|>text<|/ref|><|det|>[[112, 765, 882, 818]]<|/det|>
+48. Burrows, S. R., Sculley, T. B., Misko, I. S., Schmidt, C. & Moss, D. J. An Epstein-Barr virus-specific cytotoxic T cell epitope in EBV nuclear antigen 3 (EBNA 3). J. Exp. Med. 171, 345-349 (1990).  
+
+<|ref|>text<|/ref|><|det|>[[112, 834, 882, 887]]<|/det|>
+49. Steven, N. M., Leese, A. M., Annels, N. E., Lee, S. P. & Rickinson, A. B. epitope focusing in the primary cytotoxic T cell response to Epstein-Barr virus and its relationship to T cell memory. J. Expm. Med. 184, 1801-1813 (1996).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 105, 884, 161]]<|/det|>
+50. Hansen, T. H., Lybarger, L., Yu, L., Mitaksov, V. & Fremont, D. H. Recognition of open conformers of classical MHC by chaperones and monoclonal antibodies. Immunol. Rev. 207, 100–111 (2005).  
+
+<|ref|>text<|/ref|><|det|>[[112, 175, 883, 248]]<|/det|>
+51. Mage, M. G., Dolan, M. A., Wang, R., Boyd, L. F. Revilleza, M. J., Robinson, Natarajan, K., Myers, N. B., Hansen, T. H. & Margulies, D. H. The peptide-receptive transition state of MHC class I molecules: insights from structures and molecular dynamics. J. Immunol. 189, 1391–1399 (2012).  
+
+<|ref|>text<|/ref|><|det|>[[112, 262, 883, 333]]<|/det|>
+52. Mage, M. G., Dolan, M. A., Wang, R., Boyd, L. F. Revilleza, M. J., Robinson, Natarajan, K., Myers, N. B., Hansen, T. H. & Margulies, D. H. A structural and molecular dynamics approach to understanding the peptide-receptive transition state od MHC-I molecules. Molec. Immunol. 55, 123–125 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[112, 348, 883, 420]]<|/det|>
+53. Papakyriakou, A., Reeves, E., Beton, H., Mikolajek, H., Douglas, L., Cooper, G., Elliott, T., Werner, J. M. & James, E. The partial dissociation of MHC class I-bound peptides exposes their N terminus to trimming by endoplasmic reticulum aminopeptidase 1. J. Biol. Chem. 293, 7538–7648 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[112, 435, 883, 472]]<|/det|>
+54. Fisette, O., Wingbermuhle, & Schafer, L.V. Partial dissociation of truncated peptide influences the structural dynamics of the MHC I binding groove. Front. Immunol. 8, 408 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[112, 487, 882, 542]]<|/det|>
+55. Wingbermuhle, S. & Schafer, L. V. Partial peptide dissociation and binding groove plasticity in two major histocompatibility complex class I alleles – differences between alleles versus force field and sampling effects. RSC Adv. 12, 29908–29914 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[112, 557, 882, 594]]<|/det|>
+56. Ellgaard, L., McNaul, N., Chatsisvili, A., Braakman, I. Co- and post-translational protein folding in the ER. Traffic 17, 615–638 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[112, 608, 882, 646]]<|/det|>
+57. Fisette, O. Schroder, G. F. & Schafer L. V. Atomistic structure and dynamics of the human MHC-I peptide-loading complex. Proc. Natl. Acad. Sci. USA 117, 20597–20606 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[112, 660, 882, 715]]<|/det|>
+58. Kanaseki, T., Blanchard, N., Hammer, G. E., Gonzalez, F. & Shastri N. ERAPP synergizes with MHC class I molecules to make the final cut in the antigenic peptide precursors in the endoplasmic reticulum. Immunity 25, 795–806 (2006).  
+
+<|ref|>text<|/ref|><|det|>[[112, 730, 882, 767]]<|/det|>
+59. Falk, K., Rotzschke, O. & Rammensee, H. G. Cellular peptide composition governed by major histocompatibility complex class I molecules. Nature 348, 248–251 (1990).  
+
+<|ref|>text<|/ref|><|det|>[[112, 782, 882, 854]]<|/det|>
+60. Mavridis, G., Arya, R., Domnick, A., Zoidakis, J., Makridakis, M., Vlahou, A., Mpakali, A., Lelis, A., Georgiadis, D., Tampe, R., Papakyriakou, A., Stern, L. J. & Stratikos, E. A systematic re-examination of processing of MHC I-bound antigenic peptide precursors by endoplasmic reticulum aminopeptidase 1. J. Biol. Chem. 295, 7193-7210 (2020).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 90, 884, 145]]<|/det|>
+61. Garboczi, D. N., Hung, D. T. & Wiley, D. C. HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc. Natl. Acad. Sci. U.S.A. 89, 3429-3433 (1992).  
+
+<|ref|>text<|/ref|><|det|>[[111, 159, 884, 196]]<|/det|>
+62. Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307-326 (1997).  
+
+<|ref|>text<|/ref|><|det|>[[111, 211, 754, 230]]<|/det|>
+63. Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125-132 (2010).  
+
+<|ref|>text<|/ref|><|det|>[[111, 245, 883, 282]]<|/det|>
+64. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658-674 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[111, 297, 883, 370]]<|/det|>
+65. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213-221 (2010).  
+
+<|ref|>text<|/ref|><|det|>[[111, 385, 883, 421]]<|/det|>
+66. Winn, M. D., Murshudov, G. N. & Papiz, M. Z. Macromolecular TLS refinement in REFMAC at moderate resolutions. Methods Enzymol. 374, 300-321 (2003).  
+
+<|ref|>text<|/ref|><|det|>[[111, 437, 883, 508]]<|/det|>
+67. Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. Overview of the CCP4 suite and current developments. Acta Crystallogr. D Biol. Crystallogr. 67, 235-242 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[111, 524, 881, 560]]<|/det|>
+68. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126-2132 (2004).  
+
+<|ref|>text<|/ref|><|det|>[[111, 575, 881, 629]]<|/det|>
+69. Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C. & Ferrin, T. E. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605-1612 (2004).  
+
+<|ref|>text<|/ref|><|det|>[[111, 644, 881, 680]]<|/det|>
+70. Sali, A., & Blundell, T. L. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779-815 (1993).  
+
+<|ref|>text<|/ref|><|det|>[[111, 696, 881, 750]]<|/det|>
+71. Pronk, S., Pall, S., Schulz, R., Larsson, P., Bjelkmar, P., Apostolov, R., Shirts, M. R., Smith, J. C., Kasson, P. M., van der Spoel, D., Hess, B. & Lindahl, E. GROMACS 4.5: A high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 29, 845-85 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[111, 766, 881, 820]]<|/det|>
+72. Abraham, M. J., Murtola, T., Schulz, R., Pall, S., Smith, J. C., Hess, B. & Lindahl, E. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1-2, 19-25 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[111, 836, 881, 890]]<|/det|>
+73. Huang, J., Sarah, R., Grzegorz, N., Ting, R., Michael, F., de Groot, B. L., Helmut, G. & MacKerell Jr., A. D. CHARMM36m: an improved force field for folded and intrinsically disordered proteins, Nat. Methods 14, 71-76 (2017).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 105, 884, 160]]<|/det|>
+834 74. Mark, P. & Nilsson, L. Structure and dynamics of liquid water with different long-range 835 interaction truncation and temperature control methods in molecular dynamics simulations. J. 836 Comput. Chem. 23, 1211–1219 (2002).  
+
+<|ref|>text<|/ref|><|det|>[[111, 178, 884, 213]]<|/det|>
+837 75. Patriksson, A. & van der Spoel, D. A temperature predictor for parallel tempering simulations 839 Phys. Chem. Chem. Phys. 10, 2073–2077 (2008).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 88, 875, 586]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 590, 883, 840]]<|/det|>
+<center>Figure 1. Unusual binding and presentation of \((\mathbf{RA})_6\mathbf{FAKKKYCL}\) and \((\mathbf{RA})_6\mathbf{FVKKKYCL}\) A, Top panel, superimposition of bound \((\mathbf{RA})_6\mathbf{FAKKKYCL}\) (cyan) and \((\mathbf{RA})_6\mathbf{FVKKKYCL}\) (yellow) 20mer peptides. The backbone and side chain conformations of the peptides overlap between P3 and P8 but differ at P1, P2, and P-1 (P-2 was visible only in 20mer FV). Bottom panel, superimposition of bound 20mer peptides with 8mer FAKKKYCL (pink) and FVKKKYCL (green) control peptides. The four peptides overlap between P3 and P8 and are most divergent at P1. B, Interactions in the A pocket show that the main-chain nitrogen of P1 Phe FV 20mer (top panel) and FA 20mer (bottom panel) has rotated and forms a hydrogen bond with Asn63 (black dashed lines). The main-chain carbonyl oxygen in FV 20mer hydrogen bonds with Tyr159 and Tyr7, while the same atom in FA 20mer has undergone a very unusual rotation toward the \(\alpha 1\) - helix and forms no interaction with MHC I residues. In both panels, extension residues protrude out of the groove. C, In the 20mer structures, the side chains of Arg62 have moved out of the canonical positions seen in the 8mer structures, which opens the A pocket and allows the extension residues \((\mathbf{RA})_6\) to exit out. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[60, 93, 874, 423]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 436, 881, 526]]<|/det|>
+<center>Figure 2. Top and side views of HLA-B8E76C groove in FA, FV, and AA 20mer structures. The figure shows peptide-induced structural distortions at the N-terminus of the groove. Residues Gln54 to Tyr59 (6 residues) and Ser42 to Tyr59 (18 residues) are not visible (shown as red dashed lines) in the FA and FV 20mer structures, respectively. This is in marked contrast to AA 20mer structure which has a correctly conformed groove<sup>29</sup>. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[135, 90, 861, 722]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 730, 884, 908]]<|/det|>
+<center>Figure 3. Key roles of N-terminal MHC I residues Tyr59 and Ile52. A, Superimposition of FA and FV 20mer structures with AA 20mer (left panel) and FA 8mer (right panel) structures. The P1 Ala side chain of AA 20mer and P1 Phe side chain of FA 8mer occupy canonical positions and interact with Tyr59 (black dashed lines), while the bulky P1 Phe side chains of FA and FV 20mers clash with Tyr59 causing conformational disorders between Gln54 to Tyr59 (left and right panels). B, Same superimpositions as in A. The P1 Phe side chain of FA 20mer interacts with Ile52 of the \(310\) -helix (cyan dashed lines) which stabilizes Ser42 to Tyr59 (left and right panels), while a similar interaction involving Ile52 is not possible for FV 20mer resulting in conformational disorders between Ser42 to Tyr59 (left and right panels). The P1 Ala side chain of AA 20mer (left panel) and P1 Phe side chain of FA 8mer (right panel) have no interaction with Ile52. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[125, 100, 870, 578]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 586, 883, 708]]<|/det|>
+<center>Figure 4. Molecular cross-talks between occupied pockets A and B. A, P1 phenyl side chain is oriented differently in FA 20mer (left panel) relative to FV 20mer (right panel). This is due to molecular cross-talks between peptide P1 and P2 residues and Ile152 and Tyr59 (missing) (see text and Fig. 3B). Overall, the network of hydrophobic (black dashed lines) and hydrogen bond (red dashed lines) interactions in the A and B pockets are different in these two structures. B, The network of interactions in FA 8mer (left panel) and FV 8mer (right panel) is quite similar overall in these conformed structures, in contrast to FA and FV 20mer structures shown in A. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[130, 97, 850, 551]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 562, 883, 686]]<|/det|>
+<center>Figure 5. Properties of equilibrium ensembles of HLA-B8E76C-peptide complexes. A, RMSF values of individual residues along the heavy chain of HLA-B8E76C loaded with different peptides highlighting that the highest values are in the region of residues 41 to 62 (shown in a red box). B, A zoom-in of panel A reveals two distinct regions, residues 41 to 46 (peptide-independent) and residues 52 to 62 (peptide-dependent). C, Probability distributions of inter-residue distance between peptide P1 and Tyr59 and D, peptide P1 and Ile52 for HLA-B8E76C loaded with different peptides (see text). </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[112, 110, 872, 430]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 446, 884, 604]]<|/det|>
+<center>Figure 6. Comparisons with the structures of bat MHC I and human MHC II molecules. A, Superimposition of FA 8mer and bat MHC I molecule (Ptal-N\*01:01; PDB code 6J2D) structures. The bat molecule has an additional turn (highlighted by a black box) in the extended region of Gln54 to Tyr59 (highlighted in red) that we identified as disordered (see Fig. 2). In this turn, Asp69 forms salt bridge interactions (black dashed lines) with Arg65 of the \(\alpha 1\) -helix. B, Superimposition of FA 8mer and human MHC II molecule (HLA-DR1; PDB code 1DLH) structures showing that the HLA-DM susceptible region, i.e., \(3_{10}\) -helix and unstructured loop (highlighted in dark blue) overlaps with the \(3_{10}\) -helix and extended region (highlighted in red) that we identified as critical in shaping the A and B pockets (see Fig. 4). </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[190, 90, 840, 520]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 528, 884, 584]]<|/det|>
+<center>Figure 7. A model of energy landscape. The model depicts FA and FV 20mer complexes as intermediate states that interconvert and are of higher energy relative to the more conformed states seen in the structures of FA and FV 8mers. </center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 93, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[61, 130, 261, 150]]<|/det|>
+SuppfilesJan18. pdf
+
+<--- Page Split --->

preprint/preprint__9821a16de1c5dff3e3be578498bc5241bb7e4f77f30df9398e8b1aafba548010/images_list.json

@@ -0,0 +1,107 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1: Paternal MTX injection affects offspring's cranial cartilages lengths. (A) scheme of experimental design. (B-J) Violin plots represent the measurement of different cranial cartilages lengths, angles and areas on control (Ctrl), 10mg/kg MTX and 50 mg/kg MTX. Statistical analyses were performed using ANOVA one-way followed by multiple comparison Tukey's test. \\(*P = 0.0164\\) , \\(***P = 0.0008\\) , \\(****P < 0.0001\\) , ns: \\(P > 0.05\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        100,
+        375,
+        900,
+        820
+      ]
+    ],
+    "page_idx": 7
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2: Paternal MTX injection produced offspring's cranial cartilages malformations. (A-B) Dorsal view of normal and affected anterior orbital (ao) showing a meandering shape mostly observed on the offspring from MTX treated males. (C) Quantification of the percentage of embryos presenting affected or normal anterior orbital cartilages. Numbers in the graph represent the analyzed embryos. (D) Lateral view of larvae presenting normal (trowel shape), mild (bended shape) and strong (hook shape) deformities of the basihyal cartilage. (E) Quantification of the percentage of embryos presenting normal (non-affected), mild or strong basihyal cartilage abnormalities observed on the offspring from MTX treated males. Numbers in the graph represent the analyzed embryos. Statistical analyses were performed using a contingency table followed by Fisher's exact test. \\(*P = 0.0329\\) , \\(**P = 0.0059\\) , \\(***P = 0.0006\\) , ns: \\(P > 0.05\\) . Means \\(\\pm\\) SEM.",
+    "footnote": [],
+    "bbox": [
+      [
+        123,
+        225,
+        852,
+        560
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3: Paternal MTX injection alters sperm sncRNAs. (A) Histogram representing the comparison from sperm sncRNA on control (Ctrl1-3) and MTX (MTX-1-3) treated males. (B) Histogram displaying biotypes of tRNAs from sperm of MTX treated males. See also supplementary table 1 for A and B. (C) Volcano plot of depicting the fold changes in sperm sncRNAs identified as being differentially expressed within control versus MTX-treated males. (D) MA plot displaying normalized counts (base mean) for different sncRNAs. Dotted lines depict thresholds values for significantly up and down-regulated (± ≥1 log₂ fold change and -log₁₀Pvalue ≥1.3). See also supplementary table 2 for C and D.",
+    "footnote": [],
+    "bbox": [
+      [
+        231,
+        82,
+        805,
+        396
+      ]
+    ],
+    "page_idx": 10
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4. Paternal MTX injection induced the expression and cleavage of specific tRNAs. (A) Volcano plot of depicting the fold changes in sperm 5' and 3' tRNA halves as being differentially expressed within control versus MTX treated males. (B) MA plot displaying normalized counts (base mean) for different 5' and 3' tRNA halves. Dotted lines depict thresholds values for significantly up and down-regulated \\((\\pm \\geq 1\\) log2 fold change and -log10Pvalue \\(\\geq 1.3\\) ). See also supplementary table 3 for A and B. (C) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from different tRNAs affected by MTX treatment. Asterisk indicated significant differences analyzed by multiple unpaired t-student' test ( \\(^{*}P< 0.05\\) ). (D) Histogram showing the length variation of mapped tRNA reads on control and MTX treated males. Histogram showing the read coverage (E) and size (F) distribution for the most abundant and having a significant increase in the 5'tsRNA (tRNA-GluCUC, tRNA-AspGUC, and tRNA-GlyGCC) between control and MTX. See also supplementary table 4 for C, D and E.",
+    "footnote": [],
+    "bbox": [
+      [
+        93,
+        214,
+        904,
+        720
+      ]
+    ],
+    "page_idx": 11
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5: Paternal MTX injection alters RNA modifications in sperm tRNA fraction and the testicular expression of RNA-methyltransferases. Histogram comparing sperm RNA methylations on control and MTX analyzed by UHPLC-MS-MS in 50-90nt (A) and 20-50nt (B) fractions. (C) Schematic representation of modified nucleotides in the tRNA at secondary and tertiary structure. (D) RT-qPCR for methyltransferases of m1A (TRMT6) and m5C (DNMT2 and NSUM2) on testis from control and MTX treated males, gene expression was normalized using Rpl7 and Ef1 as housekeeping genes. Statistical analysis was performed by using the unpaired \\(t\\) -student' test. \\(*P<\\) 0.05, \\(^{**}P< 0.01\\) , ns: \\(P > 0.05\\) . Means \\(\\pm\\) SEM. (E) Proposed model summarizing the results.",
+    "footnote": [],
+    "bbox": [
+      [
+        135,
+        216,
+        884,
+        609
+      ]
+    ],
+    "page_idx": 13
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "Figure S1: (A) Histogram showing the percentage of fertilization utilizing the sperm from control and MTX treated males. (B) Kaplan Meier plot displaying the survival of fertilized eggs until the beginning of hatch. Histogram showing the percentage of hatching embryos (C) and the day of hatching (D) on control and MTX group. Statistics for A and C were generated by contingency table followed by Chi-square test. Statistics for B and D were generated by using the Long-rank (Matel-Cox) test, Gehan-Breslow-Wilcoxon test. ns: \\(P > 0.05\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        186,
+        111,
+        773,
+        528
+      ]
+    ],
+    "page_idx": 30
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_1.jpg",
+    "caption": "Figure S2: (A) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from tRNAs not affected (tRNA-Pro<sup>UGG</sup>, -Arg<sup>UCU</sup>, -Val<sup>AAC</sup>) or having a reduction (tRNA-Ser<sup>GCU</sup>) on MTX treatment. (B) Histogram showing the read coverage for tRNA-Glu<sup>CUC</sup> and -Ser<sup>GCU</sup>. (C) Histogram showing the lack of length variation of mapped rRNAs on control and MTX treated males. See also supplementary table 5.",
+    "footnote": [],
+    "bbox": [
+      [
+        90,
+        85,
+        880,
+        640
+      ]
+    ],
+    "page_idx": 31
+  }
+]

preprint/preprint__9821a16de1c5dff3e3be578498bc5241bb7e4f77f30df9398e8b1aafba548010/preprint__9821a16de1c5dff3e3be578498bc5241bb7e4f77f30df9398e8b1aafba548010.mmd

@@ -0,0 +1,372 @@
+
+# Paternal methotrexate exposure affects sperm small RNA content and causes craniofacial defects in the offspring  
+
+Nagif Alata Jimenez Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM) Mauricio Castellano  
+
+Functional Genomics Unit. Instituto Pasteur de Montevideo. School of Science. Universidad de la Republica.  
+
+Emilio Santillan Johns Hopkins School of Medicine  
+
+Konstantinos Boulias Boston Children's Hospital  
+
+Agustin Boan Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+Luisa Arias Padilla Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM) https://orcid.org/0000- 0003- 4689- 2561  
+
+Juan Femandino Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+Eric Greer Harvard/Boston Children's Hospital https://orcid.org/0000- 0002- 7501- 7371  
+
+Juan Tosar Universidad de la Republica https://orcid.org/0000- 0002- 2021- 2479  
+
+Luisa Cochella Johns Hopkins School of Medicine https://orcid.org/0000- 0003- 4018- 7722  
+
+Pablo strobl- mazzulla ( strobl@intech.gov.ar) Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+## Article  
+
+Keywords: methotrexate, paternal exposure, offspring craniofacial defects, tsRNAs, RNA methylation, medaka  
+
+Posted Date: July 18th, 2022
+
+<--- Page Split --->
+
+DOI: https://doi.org/10.21203/rs.3.rs-1841878/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+
+# Paternal methotrexate exposure affects sperm small RNA content and causes craniofacial defects in the offspring 
+
+Nagif Alata Jimenez1,2; Mauricio Castellano3,4; Emilio M. Santillan5; Konstantinos Boulias6,7; 
+
+Agustín Boan1,2; Luisa A. Padilla1,2; Juan I. Fernando1,2; Eric Lieberman Greer6,7; Juan P. 
+
+Tosar3,4; Luisa Cochella5; Pablo H. Strobl-Mazzulla1,2* 
+
+*Author for correspondence: strobl@intech.gov.ar. 
+
+1 Laboratory of Developmental Biology. Instituto de Investigaciones Biotecnológicas- Instituto 
+
+Tecnológico de Chascomús (CONICET-UNSAM). Chascomús, ARGENTINA. 
+
+2Escuela de Bio y Nanotecnologías (UNSAM). Chascomús, ARGENTINA. 
+
+3Functional Genomics Unit. Instituto Pasteur de Montevideo. Montevideo. URUGUAY. 
+
+4School of Science. Universidad de la República. Montevideo. URUGUAY. 
+
+5Department of Molecular Biology and Genetics, Johns Hopkins University School of 
+
+Medicine, Baltimore, Maryland, USA 
+
+6Department of Pediatrics, HMS Initiative for RNA Medicine, Harvard Medical School, Boston 
+
+MA, USA. 
+
+7Division of Newborn Medicine, Boston Children’s Hospital, Boston MA, USA. 
+
+Keywords: methotrexate, paternal exposure, offspring craniofacial defects, tsRNAs, RNA 
+
+methylation, medaka.
+
+<--- Page Split --->
+
+## Abstract  
+
+Folate is an essential vitamin for vertebrate embryo development. Methotrexate (MTX) is a folate antagonist that is widely prescribed for autoimmune diseases, blood and solid organ malignancies, and dermatologic diseases. Although it is highly contraindicated for pregnant women, because it is associated with an increased risk of multiple birth defects, the effect of paternal MTX exposure on their offspring has been largely unexplored. Here, we found MTX treatment of adult medaka male fish (Oryzias latipes) causes cranial cartilage defects in their offspring. Small non- coding RNA (sncRNAs) sequencing in the sperm of MTX treated males identify differential expression of a subset of tRNAs, with higher abundance for specific 5' tRNA halves. Sperm RNA methylation analysis on MTX treated males shows that m5C is the most abundant and differential modification found in RNAs ranging in size from 50 to 90 nucleotides, predominantly tRNAs, and that it correlates with greater testicular Dnmt2 methyltransferase expression. Overall, our data suggest that paternal MTX exposure alters sperm sncRNAs expression and modifications that may contribute to developmental defects in their offspring.
+
+<--- Page Split --->
+
+## Introduction  
+
+Folate is a water- soluble vitamin obtained from the diet that is essential for vertebrates. It is incorporated as an essential cofactor for the synthesis of nucleotides and the generation of S- adenosylmethionine (SAM) which serves as a universal donor of methyl groups for DNA, RNA and proteins implicated in gene regulation during early development 1- 4. Maternal folate deficiency leads to severe neural tube defects and craniofacial anomalies of descendants 5- 7. Importantly, the prevalence of these defects is highly reduced by folic acid supplementation prior and during pregnancy 8,9. Despite global efforts to supplement the maternal diets with folate, there is still a worldwide prevalence of these congenital defects 10- 12. Methotrexate (MTX) is a recognized teratogenic folic acid antagonist that has been linked to an elevated incidence of congenital anomalies in children born from exposed women. Intrauterine MTX exposure has been linked to craniofacial and limb defects, as well as developmental delays 13,14. In addition to oral clefts, folic acid antagonists may raise the risk of cardiovascular, neural tube, and urinary tract abnormalities 15. As a result, current recommendations urge that mothers stop using MTX at least three months before conception 16. Prior research has also identified a variety of issues concerning MTX use and a probable genotoxic effect on sperm, which might result in chronic disease or congenital anomalies 17. However, medical care recommendations for males taking MTX while trying to conceive are less clear.  
+
+For decades, the sperm genome has been considered transcriptionally quiescent and solely contributing to the restoration of the ploidy of the zygote. However, more recently, a set of functional RNAs have been characterized in mature spermatozoa that are delivered to the oocyte upon fertilization, contributing to early embryo development and thus, influencing the phenotypic outcome of the offspring 18- 24. Intriguingly, paternal folate concentrations can affect the sperm epigenome 25,26. Whereas the direct impact of these changes is expected to be minimal given the protamine exchange and resetting of DNA methylation during
+
+<--- Page Split --->
+
+spermatogenesis and early development \(^{27 - 29}\) , we wondered whether paternal folate levels may also affect the RNA composition of mature sperm.  
+
+Small non- coding RNAs (sncRNAs) are a particularly attractive potential carrier of nongenetic information in the spermatozoa. In particular, tRNA- derived small RNAs (tsRNAs) and microRNAs (miRNAs) are the most abundant in mature spermatozoa \(^{30,31}\) ; and have been identified as molecular carriers of paternal experiences, including high fat diet \(^{22,24,32}\) , low protein diet \(^{33}\) , stress \(^{21,34}\) , and odoriferous sensitivity to chemicals \(^{23}\) . Small RNA biogenesis, stability and functionality are highly dependent on their post- transcriptional modification status, primarily methylation \(^{35 - 37}\) . Furthermore, transmission of paternally acquired metabolic disorders is dependent on the presence of post- transcriptional modifications in sperm sncRNAs \(^{19,24}\) .  
+
+Here, we explored the intriguing possibility that paternal folate deficiency impacts the offspring's development, and that it may do so through changes in sncRNA abundance and methylation levels. We injected medaka male fish with the folate inhibitor methotrexate (MTX) and characterized their offspring's developmental defects. Next, we analyzed and compared the abundance and modifications of sncRNAs present in the sperm of MTX- treated males to test the idea that they work as mediators of congenital developmental defects.  
+
+## Results  
+
+Paternal folate deficiency induced cranial cartilage malformations in their offspringTo investigate the impact of paternal folate deficiency on the development of their progeny, we administered medaka male fish with an intraperitoneal injection of methotrexate (MTX), a well- known folate inhibitor \(^{38 - 40}\) , at 10mg of MTX per Kg of body weight (10MTX) and 50mg/Kg MTX (50MTX)(Fig. 1A). After 7 days, we fertilized wild type oocytes with sperm extracted
+
+<--- Page Split --->
+
+from treated and untreated males. None of our treatments had a significant impact on sperm fecundity, hatching time, or overall embryo hatching (Fig. S1). Several studies have shown that folate is an important vitamin for neural and neural crest development in several vertebrate species including humans \(^{5,41,42}\) . Moreover, maternal folate deficiency during pregnancy leads to abnormal development of neural crest derivatives such as cranial cartilages \(^{43 - 46}\) . Taking this into account, we first evaluated the effect of paternal folate deficiency on the development of the offspring's cranial cartilages by performing alcian blue staining at 3 days post hatching-stage (3dph). We measured the length of three dorsal cartilages (anterior orbital, epiphyseal bar and posterior orbital), four ventral cartilages (Meckel, ceratohyal, basibranchial and palatoguadrate), and the Meckel's area and ceratohyal angle (Fig. 1B-J). From the dorsal cartilages, we found a significant reduction in the length of the anterior orbital (also known as taenia marginalis anterior) in the 50MTX group (115.02 \(\mu \mathrm{m} \pm 9.03\) , one- way ANOVA followed by multiple comparison Tukey's test \(p = 0.0164\) ) when compared to the 10MTX (130.02 \(\mu \mathrm{m} \pm 10.22\) ) and control (125.98 \(\mu \mathrm{m} \pm 15.77\) ). On the ventral side, the basibranchial and Meckel's cartilages were not affected. However, the ceratohyal was reduced to almost half the length, at both 10MTX (192.99 \(\mu \mathrm{m} \pm 7.55\) , \(p < 0.0001\) ) and 50MTX (185.42 \(\mu \mathrm{m} \pm 8.71\) , \(p < 0.0001\) ) compared with control (363.64 \(\mu \mathrm{m} \pm\)
+
+<--- Page Split --->
+
+18.96). Interestingly, when we looked at the morphology of those cranial cartilages, we found that two of them, the anterior orbital and basihyal, presented an abnormal shape (Fig. 2). In particular, the anterior orbital has an abnormal serpentine shape, compared with the normal curved shape (Fig. 2A-B). This phenotype was significantly prevalent ( \(p = 0.0059\) ) at the 50MTX group (Fig. 2C). However, one of the most drastically affected cartilages was the basihyal, whose phenotypes presented a curved trowel shape (mild) or hook shape (strong) (Fig. 2D). Quantitation of those phenotypes evidences a significant increase in the severity of them at both 10MTX ( \(p = 0.0329\) ) and 50MTX ( \(p = 0.0006\) ) compared with Control group (Fig. 2E). Overall, these findings support the notion that paternal MTX exposure affects the  
+
+![](images/Figure_1.jpg)
+
+<center>Figure 1: Paternal MTX injection affects offspring's cranial cartilages lengths. (A) scheme of experimental design. (B-J) Violin plots represent the measurement of different cranial cartilages lengths, angles and areas on control (Ctrl), 10mg/kg MTX and 50 mg/kg MTX. Statistical analyses were performed using ANOVA one-way followed by multiple comparison Tukey's test. \(*P = 0.0164\) , \(***P = 0.0008\) , \(****P < 0.0001\) , ns: \(P > 0.05\) . </center>
+
+<--- Page Split --->
+
+development of the offspring's cranial cartilage, indicating that sperm may convey some information involved in the observed phenotypic inheritance.  
+
+## SncRNAs abundance is altered in the sperm of MTX-treated males  
+
+![](images/Figure_2.jpg)
+
+<center>Figure 2: Paternal MTX injection produced offspring's cranial cartilages malformations. (A-B) Dorsal view of normal and affected anterior orbital (ao) showing a meandering shape mostly observed on the offspring from MTX treated males. (C) Quantification of the percentage of embryos presenting affected or normal anterior orbital cartilages. Numbers in the graph represent the analyzed embryos. (D) Lateral view of larvae presenting normal (trowel shape), mild (bended shape) and strong (hook shape) deformities of the basihyal cartilage. (E) Quantification of the percentage of embryos presenting normal (non-affected), mild or strong basihyal cartilage abnormalities observed on the offspring from MTX treated males. Numbers in the graph represent the analyzed embryos. Statistical analyses were performed using a contingency table followed by Fisher's exact test. \(*P = 0.0329\) , \(**P = 0.0059\) , \(***P = 0.0006\) , ns: \(P > 0.05\) . Means \(\pm\) SEM. </center>  
+
+Epigenetic information, including histone modifications and DNA methylation, particularly from the paternal lineage, is largely reprogrammed during germline and early embryo development. However, increasing evidence indicates that sncRNAs are a carrier of epigenetic information across generations and may act as mediators of paternally inherited traits \(^{18 - 23,47}\) . To assess if paternal folate deficiency affects the small RNA content, we sequenced size selected (\~18- 30 nt long) RNAs from sperm of 10MTX and control males.
+
+<--- Page Split --->
+
+Based on the analysis of three biological replicated for each group, we were able to identify different populations of sncRNAs including: transfer RNA fragments, miRNAs, small nucleolar RNAs (snoRNAs), small nuclear RNAs (snRNAs) and other miscellaneous RNAs (miscRNA) (Fig. 3A). The comparative analysis showed that tRNA fragments were the most abundant (58.47% ± 6.6) and became further enriched in response to 10MTX treatment (87.26% ± 2.24). Interestingly, some of the most abundant tRNA fragments, aspartic acid (having the anticodon Asp<sup>GUC</sup>), glutamic acid (Glu<sup>CUC</sup> and Glu<sup>UUC</sup>), lysine (Lys<sup>CUU</sup>) and glycine (Gly<sup>GCC</sup>) Fig. 3B), became further significantly enriched upon MTX treatment (Fig. 3C- D). Together, these results demonstrate that paternal MTX exposure affects the relative abundance of specific sncRNAs in the sperm, with tRNA fragments being the most affected population.  
+
+## 5' halves of particular tRNAs are preferentially affected by methotrexate treatment  
+
+tRNAs can be cleaved into 5' and 3' halves, known as tsRNAs, in response to stress or other external factors<sup>19,24,35,48</sup>. Of particular interest in the sperm RNA content is the large abundance of those tsRNA fragments and their potential regulatory roles in early embryo
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3: Paternal MTX injection alters sperm sncRNAs. (A) Histogram representing the comparison from sperm sncRNA on control (Ctrl1-3) and MTX (MTX-1-3) treated males. (B) Histogram displaying biotypes of tRNAs from sperm of MTX treated males. See also supplementary table 1 for A and B. (C) Volcano plot of depicting the fold changes in sperm sncRNAs identified as being differentially expressed within control versus MTX-treated males. (D) MA plot displaying normalized counts (base mean) for different sncRNAs. Dotted lines depict thresholds values for significantly up and down-regulated (± ≥1 log₂ fold change and -log₁₀Pvalue ≥1.3). See also supplementary table 2 for C and D. </center>  
+
+development \(^{19,20,24,30,31}\) . Interestingly, we observed that 5' tsRNA fragments, product of the cleavage of three of the most abundant tRNAs (5' tsRNA of Asp<sup>GUC</sup>, Lys<sup>CUU</sup> and Glu<sup>CUC</sup>), were significantly increased in MTX sperm, without a concomitant increase in their respective 3' tsRNA fragments (Fig. 4A-B, E). Quantification of the proportion of 5' halves relative to their corresponding 3' halves showed a significant increase in the percentage of 5' halves for tRNA Asp<sup>GUC</sup> and Gly<sup>GCC</sup> in the MTX treated samples (Fig. 4C, E). It is interesting to note that for some tRNAs (i.e., tRNA Glu<sup>CUC</sup>, Glu<sup>UUC</sup> and Gly<sup>UCC</sup>) we mostly retrieve reads for their 5' halves, but their corresponding 3' halves are almost undetected for both control and MTX. On the other hand, the 5' halves of many (most) other tRNAs did not show differences compared with their 3' halves (tRNA Pro<sup>UGG</sup>, Arg<sup>UCU</sup>), or a major proportion of their 3' halves (tRNA Ser<sup>GCU</sup>) (Fig. S2). These results suggest changes in processing or stability of specific tRNA fragments as a consequence of the MTX treatment.
+
+<--- Page Split --->
+
+154 Production of 5' tsRNAs fragments in the 15- 22 nt range occurs in multiple tissues and cell 155 lines \(^{49}\) , whereas longer 5' tsRNAs (31- 40 nt long) are preferentially generated in response 156 to different stresses \(^{50,51}\) . We thus compared the length distribution of tRNAs- derived 157 fragments in both conditions and observed a shift towards longer fragments in 10MTX relative  
+
+![](images/Figure_4.jpg)
+
+<center>Figure 4. Paternal MTX injection induced the expression and cleavage of specific tRNAs. (A) Volcano plot of depicting the fold changes in sperm 5' and 3' tRNA halves as being differentially expressed within control versus MTX treated males. (B) MA plot displaying normalized counts (base mean) for different 5' and 3' tRNA halves. Dotted lines depict thresholds values for significantly up and down-regulated \((\pm \geq 1\) log2 fold change and -log10Pvalue \(\geq 1.3\) ). See also supplementary table 3 for A and B. (C) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from different tRNAs affected by MTX treatment. Asterisk indicated significant differences analyzed by multiple unpaired t-student' test ( \(^{*}P< 0.05\) ). (D) Histogram showing the length variation of mapped tRNA reads on control and MTX treated males. Histogram showing the read coverage (E) and size (F) distribution for the most abundant and having a significant increase in the 5'tsRNA (tRNA-GluCUC, tRNA-AspGUC, and tRNA-GlyGCC) between control and MTX. See also supplementary table 4 for C, D and E. </center>
+
+<--- Page Split --->
+
+to control samples (26- 35 vs. 18- 29 nt) (Fig. 4D). A similar situation is found for the tRNAs Glu<sup>CUC</sup>, Asp<sup>GUC</sup> and Lys<sup>CUU</sup> where their 5'tsRNAs are significantly increased and their coverage lengths are greater in MTX- treated samples than in controls (Fig. 4F). There is a chance that bias size selection occurred when separating the small RNAs from the gel, resulting in these differences. However, when we looked at the length of mapped reads for ribosomal RNAs (rRNAs), we found no consistent changes (Fig. S2C). These findings suggest that paternal MTX exposure alters the abundance and cleavage site of specific 5'tsRNAs in the sperm.  
+
+## m5C modifications are increased by methotrexate  
+
+Post- transcriptional modifications of tRNAs, including methylation, are important for their specific cleavage, stability and functionality, as well as for the transmission of paternal experiences to the offspring \(^{35 - 37,52}\) . We thus evaluated the methylation status of two populations of RNAs isolated from polyacrylamide: a 20- 50 nt population (mostly enriched for miRNAs and tsRNAs), and a 50- 90 nt population (mostly enriched for mature tRNAs). Within the 20- 50 nt RNA population we did not observe significant differences in the abundance of any of the analyzed methylation events between MTX and control groups (Fig. 5A). Conversely, within the 50- 90 nt population, we found that MTX treatment led to a significant increase in the relative abundance of several modifications (Fig. 5B). From the two most abundant modifications analyzed (m1A and m5C) only m5C was significantly increased (p = 0.0155) in MTX treated samples. From the other less abundant modifications only m2G, m7G and m2'2G presented a significant increase in MTX samples (p value = 0.0027; 0.0063; and 0.0090, respectively). Interestingly, the most abundant modification observed to be differentially detected in MTX samples has been described to be located at the 3' ends of tRNAs \(^{53,54}\) (Fig. 5C).
+
+<--- Page Split --->
+
+Given the observed differences for certain RNA modifications, we examined the expression of specific RNA methyltransferases on the testis of control and 10MTX treated males (Fig. 5D). In agreement with our results, there was no significant change in the expression of Trmt6 which catalyzes m1A methylation. Conversely, the expression of the enzymes that catalyzed  
+
+![](images/Figure_5.jpg)
+
+<center>Figure 5: Paternal MTX injection alters RNA modifications in sperm tRNA fraction and the testicular expression of RNA-methyltransferases. Histogram comparing sperm RNA methylations on control and MTX analyzed by UHPLC-MS-MS in 50-90nt (A) and 20-50nt (B) fractions. (C) Schematic representation of modified nucleotides in the tRNA at secondary and tertiary structure. (D) RT-qPCR for methyltransferases of m1A (TRMT6) and m5C (DNMT2 and NSUM2) on testis from control and MTX treated males, gene expression was normalized using Rpl7 and Ef1 as housekeeping genes. Statistical analysis was performed by using the unpaired \(t\) -student' test. \(*P<\) 0.05, \(^{**}P< 0.01\) , ns: \(P > 0.05\) . Means \(\pm\) SEM. (E) Proposed model summarizing the results. </center>  
+
+m5C was only significantly upregulated for Dnmt2 (p = 0.01), but not for Nsun2 (p = 0.46).  
+
+These results suggest that an increase of RNA methyltransferase expression leads to changes in the methylation status of sperm tRNAs upon MTX treatment.  
+
+## Discussion
+
+<--- Page Split --->
+
+MTX binds to and inhibits dihydrofolate reductase activity, preventing folic acid from performing its biological tasks. For more than 30 years, this drug has been used to treat immunological illnesses (including rheumatoid arthritis), blood and solid organ cancers, dermatologic diseases, and pregnancy termination \(^{55,56}\) . Despite the drug's contraindication for pregnant women due to the risk of miscarriage and birth abnormalities, the paternal influence of MTX on their offspring was largely unknown. In addition to this, the vast majority of studies in fish models such as medaka and zebrafish has been performed during embryological stages \(^{57 - 60}\) , while few have evaluated the effect on adults \(^{61}\) and the consequences on their offspring.  
+
+In agreement with other studies in mice \(^{62,63}\) , we found that paternal MTX treatments had no effect on the fertility or survival of their progeny during the early embryonic stages. Lifetime exposure to folic acid- deficient diets, on the other hand, lead to lower sperm counts, negative consequences in progeny, and epigenetic changes \(^{62,63}\) . However, this may be due to folate deficiency during embryonic and post- embryonic development, which could compromise early germ cell formation and adult spermatogenesis. Moreover, major epigenetic reprogramming occurs at these periods, and multiple imprinting areas may be altered as a result of the prolonged folate shortage.  
+
+For many years, paternal contribution to offspring's health was thought to be restricted to the haploid genome of spermatozoa, whereas mother health and nutrition were linked to offspring's wellness. However, multiple recent studies have revealed that spermatozoa carry a variety of RNAs \(^{18,64 - 66}\) capable to transmit paternal experiences \(^{19,22 - 24,32}\) . In this regard, our work illustrated the critical significance of MTX therapy and its impact on sperm small non- coding RNA content as a possible mechanism underlying the observed craniofacial abnormalities or possibly other unexplored effects of this treatment. We discovered that tsRNAs and miRNAs are the most common small non- coding RNA in medaka sperm, which
+
+<--- Page Split --->
+
+is consistent with past findings in mammals \(^{19,30,31,66,67}\) . Furthermore, we revealed that tsRNAs halves changed significantly owing MTX treatment, which is in agreement with previous studies showing that tsRNAs are a dynamic population that responds to a variety of environmental stressors \(^{19,24,68}\) . Particularly, we observed a higher abundance of certain 5'tsRNAs, where 5'tsRNA- AspGUC was the most abundant. This result is in concordance with several studies where external factors also modulated the abundance of 5'tsRNA- AspGUC \(^{20,24,31}\) , thus highlighting the idea that certain tRNAs may be preferentially cleaved and their 5' halves have a longer half- life compared to their respective 3' halves.  
+
+TsRNAs can be generated through multistep cleavages, through the formation of various intermediates. Moreover, there is growing evidence that regulatory factors, such as RNA modifications and specific RNases, have a role in their specific cleavage and stability \(^{66}\) . Interestingly, we found that 5'tsRNAs from AspGUC and GlyGCC are consistently longer ( \(\sim 35\) nt) on the sperm of MTX- treated males. This is in agreement with the discovery that small 15- 22 nucleotide long fragments are normally formed in multiple tissues and cell lines \(^{49}\) , whereas longer 31- 40 nucleotide tRNA halves are preferentially cleaved in response to different stresses \(^{50,51}\) .  
+
+Up to now, the tsRNA functions are very speculative, but have been associated to translation, ribosome biogenesis, retrotransposition, cell- cell communication, and epigenetic inheritance, as well as how tsRNA dysregulation are related to a variety of human disorders (summarized in recent reviews \(^{69,70}\) ). Importantly, both tsRNAs and their precursor tRNAs are heavily modified, which contributes to multiple aspects of their function, biogenesis, stability, amino acid charging, and translational accuracy \(^{71,72}\) . Our initial hypothesis was that MTX treatment may reduce the tRNA methylation thus inducing their cleavage. This is based on previous studies where the addition of m5C, which is controlled by DNMT2 and NSUN2, increase tRNA stability in flies and mice, but its deletion makes them more likely to be cleaved into tsRNAs
+
+<--- Page Split --->
+
+under stress conditions \(^{19,36,37}\) . However, and contrarily to our predictions, we observed that tRNAs- enriched samples (\~50- 70 nt) derived from MTX- treated sperm showed significantly greater levels of methylation in m5C, m2G, m7G and m2'2G. Increased levels of m5C and m2G have been observed in the 30- 40 nucleotide fraction of sperm RNAs (predominantly tRNAs) in mice fed with high- fat diets compared with those from males fed with normal diets \(^{24}\) . However, it is important to mention that in our MTX- treatment we fail to observe differences in the methylation levels from the tsRNAs/miRNAs fraction (\~20- 50 nt). We speculate that because the most abundant tRNA modifications found in our study (5mC) are stated to be positioned at the 3' end of tRNAs (positions 38C, 48C, 49C, 50C) \(^{73}\) , then the cleaved 3'tsRNAs halves, which accumulate the bulk of these methylations, may be preferentially degraded.  
+
+The high levels of 5mC in our 50- 90nt fraction from MTX- treated males correlated with the higher expression of Dnmt2 (also known as Trdmt1), but not Nsun2. Dnmt2 is structurally close to other DNA methyltransferases but rather methylates only one tRNA, specifically at the cytosine 38 in the anticodon loop of aspartic acid (tRNA- Asp) \(^{74}\) . Interestingly, here we have found that 5'tsRNA- Asp<sup>GUC</sup> was the second most abundant tRNA in MTX- treated males and presenting a significant increase respect to their 3'tsRNA- Asp<sup>GUC</sup> half. In contradiction to our finding, Schaefe et al. \(^{37}\) demonstrated that m5C modification mediated by DNMT2 improves tRNA stability, where tRNA- Asp is protected from angiogenin cleavage during the heat shock response in Drosophila. In mammals, it is well known that angiogenin activity, RNase that cleaves tRNAs, is also inhibited by the presence of 5mC \(^{19,36}\) . However, it is important to mention that since the endonuclease targeting the anti- codon loop of Drosophila tRNAs has not been identified yet, the authors analyzed the cleavage of tRNA- Asp induced by the addition of recombinant human angiogenin into Drosophila S2 cells \(^{37}\) , which may not reflect the truly physiological condition. Moreover, it has been shown that the presence of
+
+<--- Page Split --->
+
+angiogenin is not mandatory for the generation of tsRNAs, and other RNases (Dicer, RNase T2, L) can also cleave tRNAs \(^{69,75,76}\) . In that sense, fish does not have angiogenin, but instead orthologues genes with the capacity to cleave tRNAs have been found \(^{77 - 79}\) suggesting that the generation of tRNA fragments is an evolutive response against environmental stressors. In addition, it is important to mention that the activity \(^{77}\) , structure \(^{78 - 80}\) , and targeted dinucleotides for cleavage are different in between fish and mammals RNases \(^{80}\) . These facts suggest that the overall generation of tRNA fragments is an ancient response where RNases have maintained their main role and have evolved as the organisms did it. On the other hand, the presence of 5mC, and/or other modifications, might affect their activity in a different way as it was speculated by Barraud and Tisné \(^{81}\) . These authors stated that tRNA modifications are critical features of the cellular stress responses, and described the existence of a streaky crosstalk among them regulating the tRNAs stability \(^{81}\) . As a result, modifications may act as a "barcode" to regulate the specific tRNA cleavage and stability resulting in the accumulation of specific tsRNAs in the sperm, which could affect the phenotype of their offspring.  
+
+In summary, our data suggests that paternal MTX- exposure influenced sperm tRNA methylation, as a result of alterations in the expression of certain RNA methyltransferases. These epitranscriptomic changes may cause the selective tRNA cleavage and the maintenance of certain 5' tRNA halves. These changes in the sperm RNA content and modifications might affect transcriptional cascades in the fertilized oocyte, with possible implications in cranial cartilage formation (see hypothetical model in figure 5E). The understanding of how tRNA modifications and their derived fragments impact on the transcriptional cascades occurring during early embryo will provide valuable insights into several diseases and it is expected that this will be a main focus of research in this field in the near future.
+
+<--- Page Split --->
+
+## Methods  
+
+## Medaka Husbandry  
+
+All experiments were performed with medaka fish (Oryzias latipes) (strain hi- medaka, ID:MT835) supplied by the National BioResource Project (NBRP) Medaka (http://www.shingen.nig.ac.jp/medaka/). Fish were maintained and fed following standard protocols for medaka 82. Fish were handled on the Care and Management of Laboratory Animals (http://www.ufaw.org.uk) and internal regulations. Adult fishes were divided and acclimatized in 4L fish tank during 3 weeks under a constant photoperiod (14L:10D) and controlled temperature (26 ± 0.5°C), prior to experimental procedures.  
+
+## Experimental design  
+
+Adults medaka fish were divided into 3 groups composed by 3 independent replicates having 9 males and 2 females per tank (all of them having a body weight of \(\sim 200 \text{mg}\) ). After the acclimatizing period, each male was intraperitoneal injected with control solution (PBS/1%DMSO), 10mg of MTX per kg of body weight (10mg/kg MTX) or 50mg/kg MTX (A6770- SigmaAldrich, diluted in PBS/1%DMSO). Briefly, males were anesthetized with 1% benzoacaine solution (Parafarm), gently dried with a paper towel and placed in a dampened sponge ventral side up, with their anal fin and cloaca exposed. Immediately, using a 10ul syringe (Hamilton), fish were injected using a binocular stereoscope (Nikon SMZ745) and then returned to their tanks for 7 days until sperm collection for in vitro fertilization and small RNA extraction.  
+
+## In vitro fertilization
+
+<--- Page Split --->
+
+Sperm collection was carried by anesthetizing the fish and placed in a dampened sponge ventral side up following published protocols for medaka 82. A micro- forceps was used to gently strip the fish and the released semen was collected by using a micropipette ( \(\sim 0.5\mu \mathrm{l} / \mathrm{fish}\) ) and pooled for the posterior in vitro fertilization and small RNA extraction. For the in vitro fertilization, \(0.2\mu \mathrm{l}\) from obtained sperm were used to fertilize a pool of 24- 28 eggs collected from mature untreated females. Fertilized eggs were immediately transferred and incubated in \(60~\mathrm{mm}\) petri dishes with embryo media (17mM NaCl, \(0.4\mathrm{mM}\) KCl, \(0.27\mathrm{mM}\) \(\mathrm{CaCl_2.2H_2O}\) , and \(0.66\mathrm{mM}\) MgSO4; pH:7) until 3 days' post hatching (dph). Incubation was monitored and the percentage of fertilization and survival until hatching was evaluated.  
+
+## Alcian blue staining  
+
+Cartilages from embryos were analyzed at 3 dph by using alcian blue staining. Larvae were fixed in \(4\%\) paraformaldehyde overnight at \(4^{\circ}\mathrm{C}\) and washed three times with PBSw (PBS- \(0.1\%\) tween20). After that, embryos were incubated in a bleaching solution (0.5X SSC, \(5\%\) formamide, \(10\%\) hydrogen peroxide) and exposed to light during 2 hours. Larvae were washed several times with PBSw and immediately incubated in alcian Blue solution ( \(0.1\% \mathrm{p} / \mathrm{v}\) alcian blue, \(0.37\% \mathrm{v} / \mathrm{v}\) HCl, \(70\% \mathrm{v} / \mathrm{v}\) EtOH) for 1 hour on a nutator. Then, larvae were washed five times with \(01\% \mathrm{v} / \mathrm{v}\) HCl- \(70\% \mathrm{v} / \mathrm{v}\) EtOH for 30 minutes on a nutator; the last wash was left overnight at room temperature. Next, larvae were washed six times with \(50\% \mathrm{v} / \mathrm{v}\) glycerol- \(0.5\% \mathrm{v} / \mathrm{v}\) KOH for 30 min on a nutator and the last wash was left overnight. Finally, larvae were washed four times with the same solution and left in \(90\%\) glycerol- \(10\%\) EtOH for imaging processing and phenotype analysis. Larvae were photographed at ventral, dorsal and lateral view by using a trinocular stereoscope (SteREO Discovery v20. Zeiss) and analyzed using the ImageJ software 83.
+
+<--- Page Split --->
+
+## Small RNA extraction and library preparation  
+
+Small RNAs were isolated from sperm following manufacturer's instructions (illustra RNAspin Mini RNA isolation kit- GE Healthcare). The 3' adapters were ligated using SRBC barcode adapters for each sample, additionally 18- mer and 30- mer markers were ligated and used as control for the ligation process and markers for the product correct size. The 3'- ligated small RNAs were size selected using \(15\%\) denaturing urea polyacrylamide gels at a constant power of 40- 50W for \(\sim 30\) minutes and stained by using SYBR Gold \(0.05\% V / V\) in TBE \(0.5X\) and the 3' ligated RNAs ranging from 18- 30 mer were cut out. RNAs were purified using Zymo PAGE elution kit (ZRTM small RNA PAGE recovery kit) according to manufacturer's instructions, the elute 3'- ligated small RNAs were elute in 5' linker mix containing 5' adaptor. The 3'- ligated RNAs + 5' adapter were denaturated for 5 min at \(70^{\circ}C\) , cooled on ice immediately, ligated with T4 RNA ligase (NEB) and incubated at 16 degrees overnight. Ligated small RNAs were purified by using MBS beads, briefly: MBS buffer, MBS bead slurry (beads + buffer), mixed by vortexing, added isopropanol and incubated at room temperature. Beads were separated on magnet and the supernatant was removed, after several washes with ethanol the RNA was eluted with ultrapure water and transferred into PCR strip. For reverse transcription, small- RNAseq RT primer to each sample were used and a negative control without reverse transcriptase was included, Superscript II reverse transcriptase was used to obtain the synthesis of the first strand. To amplify cDNA libraries, KAPA HiFi Real Time Library Amplification Kit (Roche) was used; PCR were performed using TruSeq Universal Adapter primer (Solexa_PCR_fwd) and TruSeq Index reverse primers (Solexa_IDX_rev), this latter includes barcodes assigned to each different sample. Briefly: master mix was added and TruSeq Index reverse primer were added to PCR strips containing cDNAs; then KAPA HiFi HS RM and Truseq Universal Adapter primer were added to the mix. The cycling program was: Denaturation at \(98^{\circ}C\) for 45 sec; 20 cycles of \(98^{\circ}C\) for 15 sec, \(65^{\circ}C\) for 30 sec, \(72^{\circ}C\) for
+
+<--- Page Split --->
+
+30 sec, \(72^{\circ}C\) for 10 sec; and a final extension at \(72^{\circ}C\) for 1 min. The amplified cDNA was purified by using \(3\%\) Low- Range Ultra Agarose gel (Bio- Rad) according to the manufacturer's instructions at constant 80- 100 V using GeneRuler 50bp DNA Ladder (ThermoFischer Scientific) as molecular marker. Gel was visualized on a long wave UV transilluminator and DNA band between 150- 200 bp were excised using a clean scalpel blade and put into a clean 15ml Falcon tube; the DNA was purified using the Zymoclean Gel DNA recovery kit (Zymo Research) according to manufacturer's instructions.  
+
+## Bioinformatics Analysis  
+
+Adapters from reads were removed using CUTADAPT, the output were reads \(\geq 15\) bp, reads whose adapters were not identified were discarded. The output of 15bp were used to analyze differential expression of sncRNAs (miRNAs, tRNAs, snRNAs, snoRNAs and rRNAs) and differential expression of tRNAs, 5' tRNA halves and 3' tRNA halves by different strategies. First, differential expression of sncRNAs was analyzed on reads where the random nucleotides on 5' (4bp) and 3' (6pb) were cut using FASTQ Trimmer. The obtained reads having \(< 19\) bp were discarded using Filter Fastq and the remaining reads were aligned against the medaka genome (Assembly ASM00223467v1) with RNA STAR (allowing multimapping reads, 1 mismatch, and not allowing introns). Expression of sncRNAs was analyzed using FeatureCounts (allowing multimapping reads to be counted, and assigning 1/n fractions to multimapping reads) with Ensembl annotation (Release v102). Differential expression of sncRNAs was calculated using DESEQ2. Second, to analyze differential expression of 5' and 3' tRNA halves, an additional 3 base pairs were removed with FASTQ Trimmer from the 3' end of all reads. Reads having less than 15bp were discarded using Filter Fastq. The output was aligned and analyzed as mentioned before using custom GTF files with genomic coordinates for either 5' or 3' tRNA halves. To determine the sequence length of mapped
+
+<--- Page Split --->
+
+tRNA reads, BAM files were filtered (using GTFs files containing genomic coordinates for full length tRNAs), reads were extracted, converted to fasta and their length computed with in-house scripts. tRNA read coverage was calculated with BamCoverage (bin size 1, no smoothing).  
+
+## UHPLC-MS-MS  
+
+The analysis of modified ribonucleotides from spermatic RNAs were performed by UHPLC- MS- MS. For that purpose, \(\sim 1.5 \mu \mathrm{g}\) of total RNAs were isolated from two independent pools of stripped sperm from \(\sim 9\) control and 10MTX- treated males and run in denaturant polyacrylamide gel (15%, 7M Urea). The gel was then stained with ethidium bromide and RNAs that have a ranged size from 20- 50nt and 50- 90nt were cut and recovered using the ZR small- RNA™ PAGE Recovery Kit (Zymo Research) by following the manufacturer's instruction. Approximately 100ng of RNA was obtained on each fraction and utilized for UHPLC- MS- MS analysis. Then, 100 ng purified RNA samples were digested to nucleosides for 2 hr at 37°C using the Nucleoside Digestion mix (NEB, M069S). Quantifications were performed as in 84, briefly: digested RNA samples were diluted to 100 μl with ddH2O and filtered through 0.22 μm Millex Syringe Filters. 5 μl of the filtered solution was injected for LC- MS/MS analysis using the Agilent 1290 UHPLC- MS/MS system with a Hypersil Gold C18 reversed- phase column (2.1 x 150 mm, 3 μm). Mobile phase A consisted of water with 0.1% (v/v) formic acid and mobile phase B consisted of acetonitrile with 0.1% (v/v) formic acid. Mass spectrometry detection was performed using an Agilent 6470 triple quadrupole mass spectrometer in positive electrospray ionization mode and data were quantified in dynamic multiple reaction monitoring (dMRM) mode, by monitoring the mass transitions 268 \(\square\) II 136 for Adenosine (A), 282 \(\square\) II 150 for N6- methyladenosine (m1A), 244 \(\square\) II 112 for Cytidine (C), 258 \(\square\) II 126 for C5- methylcytidine (m5C), 284 \(\square\) II 152 for Guanosine (G), 298 \(\square\) II 166 for
+
+<--- Page Split --->
+
+N7'- methyladenosine (m7G) and N2- methylguanosine (m2G), 312 \(\square \square 180\) for N2,N2- dimethylguanosine (m2'2G), 282 \(\square \square 136\) for 2'- O- methyladenosine (Am), 258 \(\square \square 112\) for 2'- O- methylcytidine (Cm) and 298 \(\square \square 152\) for 2'- O- methylguanosine (Gm). To quantify the concentrations of the various methylation modifications we used pure nucleosides of A, C, G, m1A, m5C, m7G, m2G m2'2G, Am, Cm and Gm to generate calibration standard curves through serial dilution.  
+
+## RNA quantification by RT-PCR  
+
+Dissected testis from adult males (Control and 10MTX groups) were used for gene expression analysis. For this purpose, a complete testis from 8 males were individually grinded in \(300\mu l\) of TRlzol Reagent (Life Technologies) and total RNAs ( \(\sim 800ng\) /testis) were isolated and retrotranscribed \(^{85}\) . Expression of target genes were measured by qPCR using Fast Start Universal SYBR green Supermix (Roche Diagnostics, USA) on Thermal Cycler StepOne Plus (Applied Biosystem,USA), using ribosomal protein L7 (rpl7) and elongation factor 1 alpha (ef1α) as reference genes with the geometric mean calculation as described by Padilla et al. \(^{85}\) . Real- time PCR primers are listed in Supplementary Table 6. Each sample was run in duplicate and a PCR reaction without the addition of template, was used as negative control. The amplification protocol consisted of an initial cycle of 1min at \(95^{\circ}C\) followed by 40 cycles: 15 s at \(95^{\circ}C\) and 30 s at \(60^{\circ}C\) . After the amplification, a melt curve was performed by 1 cycle: 15 s at \(95^{\circ}C\) , 60 s at \(60^{\circ}C\) and 15 s at \(95^{\circ}C\) enabling confirmation of amplification of single products. Gene expression levels were calculated by the \(2^{-\Delta \Delta Ct}\) comparative threshold cycle (Ct) method (where \(\Delta \Delta Ct = \Delta Ct\) sample - \(\Delta Ct\) reference). The efficiency of amplification ranged 95- 105% for all genes studied. The expression level in each group was normalized to the control and was presented as a fold of change \(^{86}\) .
+
+<--- Page Split --->
+
+441 1. Sánchez-Vásquez E, Alata Jimenez N, Vázquez NA, Strobl-Mazzulla PH. Emerging role of 442 dynamic RNA modifications during animal development. Mechanisms of Development. 443 2018;154. doi:10.1016/j.mod.2018.04.002 444 2. Greenberg MVC, Bourc'his D. The diverse roles of DNA methylation in mammalian 445 development and disease. Nature Reviews Molecular Cell Biology. 2019;20(10). 446 doi:10.1038/s41580-019-0159-6 447 3. Bernstein C, Bernstein H. Demethylation in Early Embryonic Development and Memory. In: 448 DNA Methylation Mechanism. ; 2020. doi:10.5772/intechopen.90306 449 4. O'Neill C. The epigenetics of embryo development. Animal Frontiers. 2015;5(1). 450 doi:10.2527/af.2015-0007 451 5. Li J, Shi Y, Sun J, Zhang Y, Mao B. Xenopus reduced folate carrier regulates neural crest 452 development epigenetically. PLoS ONE. 2011;6(11). doi:10.1371/journal.pone.0027198 453 6. Shaw GM, Schaffer D, Velie EM, Morland K, Harris JA. Periconceptional vitamin use, dietary 454 folate, and the occurrence of neural tube defects. Epidemiology. 1995;6(3). 455 doi:10.1097/00001648- 199505000- 00005 456 7. Tolarova M. PERICONCEPTIONAL SUPPLEMENTATION WITH VITAMINS AND FOLIC ACID TO 457 PREVENT RECURRENCE OF CLEFT LIP. The Lancet. 1982;320(8291). doi:10.1016/S0140- 458 6736(82)91063- 7 459 8. McDonald SD, Ferguson S, Tam L, Lougheed J, Walker MC. The prevention of congenital 460 anomalies with periconceptional folic acid supplementation. Journal of obstetrics and 461 gynaecology Canada : JOGC = Journal d'obstetrique et gynécologie du Canada : JOGC. 462 2003;25(2). doi:10.1016/S1701- 2163(16)30207- 9 463 9. Vergel RG, Sanchez LR, Heredero BL, Rodriguez PL, Martinez AJ. Primary prevention of neural 464 tube defects with folic acid supplementation: Cuban experience. Prenatal Diagnosis. 465 1990;10(3). doi:10.1002/pd.1970100303 466 10. Blencowe H, Kanchera V, Moorthie S, Darlison MW, Modell B. Estimates of global and 467 regional prevalence of neural tube defects for 2015: a systematic analysis. Ann N Y Acad Sci. 468 2018;1414(1). doi:10.1111/nyas.13548 469 11. Salari N, Darvishi N, Heydari M, Bokaees S, Darvishi F, Mohammadi M. Global prevalence of 470 cleft palate, cleft lip and cleft palate and lip: A comprehensive systematic review and meta- 471 analysis. Journal of Stomatology, Oral and Maxillofacial Surgery. 2022;123(2). 472 doi:10.1016/j.jormas.2021.05.008 473 12. Impellizzeri A, Giannantoni I, Polimeni A, Barbato E, Galluccio G. Epidemiological 474 characteristic of Orofacial clefts and its associated congenital anomalies: Retrospective study. 475 BMC Oral Health. 2019;19(1). doi:10.1186/s12903- 019-0980-5 476 13. Mahadevan U, McConnell RA, Chambers CD. Drug Safety and Risk of Adverse Outcomes for 477 Pregnant Patients With Inflammatory Bowel Disease. Gastroenterology. 2017;152(2). 478 doi:10.1053/j.gastro.2016.10.013 479 14. Hyoun SC, Obican SG, Scialli AR. Teratogen update: Methotrexate. Birth Defects Research 480 Part A - Clinical and Molecular Teratology. 2012;94(4). doi:10.1002/bdra.23003 481 15. Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Folic Acid Antagonists during 482 Pregnancy and the Risk of Birth Defects. New England Journal of Medicine. 2000;343(22). 483 doi:10.1056/nejm200011303432204
+
+<--- Page Split --->
+
+484 16. Nguyen GC, Seow CH, Maxwell C, et al. The Toronto Consensus Statements for the 485 Management of Inflammatory Bowel Disease in Pregnancy. Gastroenterology. 2016;150(3). 486 doi:10.1053/j.gastro.2015.12.003 487 17. Beghin D, Cournot MP, Vauzelle C, Elefant E. Paternal exposure to methotrexate and 488 pregnancy outcomes. Journal of Rheumatology. 2011;38(4). doi:10.3899/jrheum.100600 489 18. Sharma U, Sun F, Conine CC, et al. Small RNAs Are Trafficked from the Epididymis to 490 Developing Mammalian Sperm. Developmental Cell. 2018;46(4). 491 doi:10.1016/j.devcel.2018.06.023 492 19. Zhang Y, Zhang X, Shi J, et al. Dnmt2 mediates intergenerational transmission of paternally 493 acquired metabolic disorders through sperm small non-coding RNAs. Nature Cell Biology. 494 2018;20(5). doi:10.1038/s41556-018-0087-2 495 20. Chen X, Zheng Y, Lei A, et al. Early cleavage of preimplantation embryos is regulated by 496 tRNAGln-TTG-derived small RNAs present in mature spermatozoa. Journal of Biological 497 Chemistry. 2020;295(32). doi:10.1074/jbc.ra120.013003 498 21. Rodgers AB, Morgan CP, Leu NA, Bale TL. Transgenerational epigenetic programming via 499 sperm microRNA recapitulates effects of paternal stress. Proc Natl Acad Sci U S A. Published 500 online 2015. doi:10.1073/pnas.1508347112 501 22. Fullston T, Teague EMCO, Palmer NO, et al. Paternal obesity initiates metabolic disturbances 502 in two generations of mice with incomplete penetrance to the F2 generation and alters the 503 transcriptional profile of testis and sperm microRNA content. FASEB Journal. Published online 504 2013. doi:10.1096/fj.12-224048 505 23. Gapp K, Jawaid A, Sarkies P, et al. Implication of sperm RNAs in transgenerational inheritance 506 of the effects of early trauma in mice. Nature Neuroscience. Published online 2014. 507 doi:10.1038/nn.3695 508 24. Chen Q, Yan M, Cao Z, et al. Sperm tsRNAs contribute to intergenerational inheritance of an 509 acquired metabolic disorder. Science (1979). Published online 2016. 510 doi:10.1126/science.aad7977 511 25. Aarabi M, San Gabrie MC, Chan D, et al. High-dose folic acid supplementation alters the 512 human sperm methylome and is influenced by the MTHFR C677T polymorphism. Human 513 Molecular Genetics. 2015;24(22). doi:10.1093/hmg/ddv338 514 26. Boxmeer JC, Smit M, Utomo E, et al. Low folate in seminal plasma is associated with 515 increased sperm DNA damage. Fertility and Sterility. 2009;92(2). 516 doi:10.1016/j.fertnstert.2008.06.010 517 27. Wang Y, Wang H, Zhang Y, et al. Reprogramming of Meiotic Chromatin Architecture during 518 Spermatogenesis. Molecular Cell. 2019;73(3). doi:10.1016/j.molcel.2018.11.019 519 28. Santos F, Hendrich B, Reik W, Dean W. Dynamic reprogramming of DNA methylation in the 520 early mouse embryo. Developmental Biology. 2002;241(1). doi:10.1006/dbio.2001.0501 521 29. Feng S, Jacobsen SE, Reik W. Epigenetic reprogramming in plant and animal development. 522 Science (1979). Published online 2010. doi:10.1126/science.1190614 523 30. Chen Q, Yan W, Duan E. Epigenetic inheritance of acquired traits through sperm RNAs and 524 sperm RNA modifications. Nature Reviews Genetics. 2016;17(12). doi:10.1038/nrg.2016.106 525 31. Sharma U, Conine CC, Shea JM, et al. Biogenesis and function of tRNA fragments during 526 sperm maturation and fertilization in mammals. Science (1979). 2016;351(6271). 527 doi:10.1126/science.aad6780 528 32. Grandjean V, Fourré S, de Abreu DAF, Derieppe MA, Remy JJ, Rassoulzadegan M. RNA- 529 mediated paternal heredity of diet-induced obesity and metabolic disorders. Scientific 530 Reports. Published online 2015. doi:10.1038/srep18193
+
+<--- Page Split --->
+
+531 33. Carone BR, Fauquier L, Habib N, et al. Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals. Cell. Published online 2010. doi:10.1016/j.cell.2010.12.008
+534 34. Wu L, Lu Y, Jiao Y, et al. Paternal Psychological Stress Reprograms Hepatic Gluconeogenesis in Offspring. Cell Metabolism. Published online 2016. doi:10.1016/j.cmet.2016.01.014
+535 35. Blanco S, Dietmann S, Flores J v, et al. Aberrant methylation of t RNA s links cellular stress to neuro-developmental disorders. The EMBO Journal. Published online 2014. doi:10.15252/embj.201489282
+536 36. Tuorto F, Liebers R, Musch T, et al. RNA cytosine methylation by Dnmt2 and NSun2 promotes tRNA stability and protein synthesis. Nature Structural and Molecular Biology. 2012;19(9). doi:10.1038/nsmb.2357
+537 37. Schaefer M, Pollex T, Hanna K, et al. RNA methylation by Dnmt2 protects transfer RNAs against stress-induced cleavage. Genes and Development. 2010;24(15). doi:10.1101/gad.586710
+538 38. Kao TT, Lee GH, Fu CC, Chen BH, Chen LT, Fu TF. Methotrexate-Induced Decrease in Embryonic 5-Methyl-Tetrahydrofolate Is Irreversible with Leucovorin Supplementation. Zebrafish. 2013;10(3). doi:10.1089/zeB.2013.0876
+539 39. Cronstein BN, Aune TM. Methotrexate and its mechanisms of action in inflammatory arthritis. Nature Reviews Rheumatology. 2020;16(3). doi:10.1038/s41584-020-0373-9
+540 40. Dervieux T, Greenstein N, Kremer J. Pharmacogenomic and metabolic biomarkers in the folate pathway and their association with methotrexate effects during dosage escalation in rheumatoid arthritis. Arthritis and Rheumatism. 2006;54(10). doi:10.1002/art.22129
+541 41. Alata Jimenez N, Torres Perez SA, Sanchez-Vasquez E, Fernandino JI, Strobl-Mazzulla PH. Folate deficiency prevents neural crest fate by disturbing the epigenetic Sox2 repression on the dorsal neural tube. Developmental Biology. Published online 2018. doi:10.1016/j.ydbio.2018.08.001
+542 42. Antony AC, Hansen DK. Hypothesis: Folate-responsive neural tube defects and neurocristopathies. Teratology. Published online 2000. doi:10.1002/1096-9926(200007)62:1<42::AID-TERA9>3.0.CO;2-U
+543 43. van Rooij IALM, Ocké MC, Straatman H, Zielhuis GA, Merkus HMWM, Steegers-Theunissen RPM. Periconceptional folate intake by supplement and food reduces the risk of nonsyndromic cleft lip with or without cleft palate. Preventive Medicine. 2004;39(4). doi:10.1016/j.ypmed.2004.02.036
+544 44. Wilcox AJ, Lie RT, Solvoll K, et al. Folic acid supplements and risk of facial clefts: National population based case-control study. British Medical Journal. 2007;334(7591). doi:10.1136/bmj.39079.618287.0B
+545 45. Burgoon JM, Selhub J, Nadeau M, Sadler TW. Investigation of the effects of folate deficiency on embryonic development through the establishment of a folate deficient mouse model. Teratology. 2002;65(5). doi:10.1002/tera.10040
+546 46. Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. The Lancet. 1991;338(8760). doi:10.1016/0140-6736(91)90133-A
+547 47. Ord J, Heath PR, Fazeli A, Watt PJ. Paternal effects in a wild-type zebrafish implicate a role of sperm-derived small RNAs. Molecular Ecology. 2020;29(14). doi:10.1111/mec.15505
+548 48. Rashad S, Han X, Sato K, et al. The stress specific impact of ALKBH1 on tRNA cleavage and tRNA generation. RNA Biology. 2020;17(8). doi:10.1080/15476286.2020.1779492
+
+<--- Page Split --->
+
+49. Kumar P, Anaya J, Mudunuri SB, Dutta A. Meta-analysis of tRNA derived RNA fragments reveals that they are evolutionarily conserved and associate with AGO proteins to recognize specific RNA targets. BMC Medicine. 2014;12(1). doi:10.1186/s12915-014-0078-0
+50. Thompson DM, Lu C, Green PJ, Parker R. tRNA cleavage is a conserved response to oxidative stress in eukaryotes. RNA. 2008;14(10). doi:10.1261/rna.1232808
+51. Yamasaki S, Ivanov P, Hu GF, Anderson P. Angiogenin cleaves tRNA and promotes stress-induced translational repression. Journal of Cell Biology. 2009;185(1). doi:10.1083/jcb.200811106
+52. Mishima E, Inoue C, Saigusa D, et al. Conformational change in transfer RNA is an early indicator of acute cellular damage. Journal of the American Society of Nephrology. Published online 2014. doi:10.1681/ASN.2013091001
+53. Huber SM, Leonardi A, Dedon PC, Begley TJ. The versatile roles of the tRNA epitranscriptome during cellular responses to toxic exposures and environmental stress. Toxics. 2019;7(1). doi:10.3390/toxics7010017
+54. Hori H. Methylated nucleosides in tRNA and tRNA methyltransferases. Frontiers in Genetics. 2014;5(MAY). doi:10.3389/fgene.2014.00144
+55. Smith SW, Nelson LS. Case files of the New York City Poison Control Center: antidotal strategies for the management of methotrexate toxicity. J Med Toxicol. 2008;4(2). doi:10.1007/BF03160968
+56. LoVecchio F, Katz K, Watts D, Wood I. Four-year experience with methotrexate exposures. J Med Toxicol. 2008;4(3). doi:10.1007/BF03161192
+57. Jacabacci J, Guzzardi M, D'Aversa TG. Exogenous folic acid is not able to inhibit the effects of methotrexate on the folate pathway in zebrafish embryos. BIOS. 2015;86(3). doi:10.1893/bios-d-14-00027.1
+58. Liu S, Narumi R, Ikeda N, Morita O, Tasaki J. Chemical-induced craniofacial anomalies caused by disruption of neural crest cell development in a zebrafish model. Developmental Dynamics. 2020;249(7). doi:10.1002/dvdy.179
+59. Sun S, Gui Y, Wang Y, et al. Effects of methotrexate on the developments of heart and vessel in zebrafish. Acta Biochimica et Biophysica Sinica. 2009;41(1). doi:10.1093/abs/gmn010
+60. Lee MS, Bonner JR, Bernard DJ, et al. Disruption of the folate pathway in zebrafish causes developmental defects. BMC Developmental Biology. Published online 2012. doi:10.1186/1471-213X-12-12
+61. Hirako A, Takeoka Y, Furukawa S, Sugiyama A. Effects of methotrexate exposure on medaka (Oryzias latipes) testes. Journal of Toxicologic Pathology. 2017;30(4). doi:10.1293/tox.2017-0029
+62. Lambrot R, Xu C, Saint-Phar S, et al. Low paternal dietary folate alters the mouse sperm epigenome and is associated with negative pregnancy outcomes. Nature Communications. 2013;4. doi:10.1038/ncomms3889
+63. Ly L, Chan D, Aarabi M, et al. Intergenerational impact of paternal lifetime exposures to both folic acid deficiency and supplementation on reproductive outcomes and imprinted gene methylation. Molecular Human Reproduction. 2017;23(7). doi:10.1093/molehr/gax029
+64. Zhang Y, Shi J, Rassoulzadegan M, Tuorto F, Chen Q. Sperm RNA code programmes the metabolic health of offspring. Nature Reviews Endocrinology. 2019;15(8). doi:10.1038/s41574-019-0226-2
+65. Johnson GD, Mackie P, Jodar M, Moskovtsev S, Krawetz SA. Chromatin and extracellular vesicle associated sperm RNAs. Nucleic Acids Research. 2015;43(14). doi:10.1093/nar/gkv591
+
+<--- Page Split --->
+
+66. Peng H, Shi J, Zhang Y, et al. A novel class of tRNA-derived small RNAs extremely enriched in mature mouse sperm. Cell Research. 2012;22(11). doi:10.1038/cr.2012.141
+67. Kawano M, Kawaji H, Grandjean V, Kiani J, Rassoulzadegan M. Novel Small Noncoding RNAs in Mouse Spermatozoa, Zygotes and Early Embryos. PLoS ONE. 2012;7(9). doi:10.1371/journal.pone.0044542
+68. Fu H, Feng J, Liu Q, et al. Stress induces tRNA cleavage by angiogenin in mammalian cells. FEBS Letters. 2009;583(2). doi:10.1016/j.febslet.2008.12.043
+69. Su Z, Wilson B, Kumar P, Dutta A. Noncanonical Roles of tRNAs: TRNA Fragments and beyond. Annual Review of Genetics. 2020;54. doi:10.1146/annurev-genet-022620-101840
+70. Schimmel P. RNA Processing and Modifications: The emerging complexity of the tRNA world: Mammalian tRNAs beyond protein synthesis. Nature Reviews Molecular Cell Biology. 2018;19(1). doi:10.1038/nrm.2017.77
+71. Pan T. Modifications and functional genomics of human transfer RNA. Cell Research. 2018;28(4). doi:10.1038/s41422-018-0013-y
+72. Suzuki T. The expanding world of tRNA modifications and their disease relevance. Nature Reviews Molecular Cell Biology. 2021;22(6). doi:10.1038/s41580-021-00342-0
+73. Pereira M, Francisco S, Varanda AS, Santos M, Santos MAS, Soares AR. Impact of tRNA modifications and tRNA-modifying enzymes on proteostasis and human disease. International Journal of Molecular Sciences. 2018;19(12). doi:10.3390/ijms19123738
+74. Goll MG, Kirpekar F, Maggert KA, et al. Methylation of tRNAAsp by the DNA methyltransferase homolog Dnmt2. Science (1979). 2006;311(5759). doi:10.1126/science.1120976
+75. Chen Q, Zhang X, Shi J, Yan M, Zhou T. Origins and evolving functionalities of tRNA-derived small RNAs. Trends in Biochemical Sciences. 2021;46(10). doi:10.1016/j.tibs.2021.05.001
+76. Akiyama Y, Lyons S, Fay MM, Abe T, Anderson P, Ivanov P. Multiple ribonuclease A family members cleave transfer RNAs in response to stress. bioRxiv. 2019;(1).
+77. Cho S, Zhang J. Zebrafish ribonucleasees are bactericidal: Implications for the origin of the vertebrate RNase a superfamily. Molecular Biology and Evolution. 2007;24(5). doi:10.1093/molbev/msm047
+78. Pizzo E, Buonanno P, di Maro A, et al. Ribonucleasees and angiogenins from fish. Journal of Biological Chemistry. 2006;281(37). doi:10.1074/jbc.M605505200
+79. Kazakou K, Holloway DE, Prior SH, Subramanian V, Acharya KR. Ribonuclease A Homologues of the Zebrafish: Polymorphism, Crystal Structures of Two Representatives and their Evolutionary Implications. Journal of Molecular Biology. 2008;380(1). doi:10.1016/j.jmb.2008.04.070
+80. Prats-Ejarque G, Lu L, Salazar VA, Moussaoui M, Boix E. Evolutionary trends in RNA base selectivity within the RNase A superfamily. Frontiers in Pharmacology. 2019;10. doi:10.3389/fphar.2019.01170
+81. Barraud P, Tisné C. To be or not to be modified: Miscellaneous aspects influencing nucleotide modifications in tRNAs. IUBMB Life. 2019;71(8). doi:10.1002/iub.2041
+82. Kinoshita M, Murata K, Naruse K, Tanaka M. Medaka: Biology, Management, and Experimental Protocols.; 2012. doi:10.1002/9780813818849
+83. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nature Methods. 2012;9(7). doi:10.1038/nmeth.2089
+84. Liberman N, Gerashchenko M v., Boulias K, et al. Intergenerational hormesis is regulated by heritable 18S rRNA methylation. bioRxiv. Published online 2021.
+
+<--- Page Split --->
+
+85. Padilla LFA, Castañeda-Cortés DC, Rosa IF, et al. Cystic proliferation of germline stem cells is necessary to reproductive success and normal mating behavior in Medaka. Elife. 2021;10. doi:10.7554/elife.62757  
+
+70. doi:10.7554/elife.62757  
+
+86. Schmittgen TD, Livak KJ. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4).  
+
+Acknowledgements: We thank all the authors and members in LBD for their contribution and helpful discussions during the course of our study.  
+
+Authors contribution: N.A.J. and P.H.S-M. designed, performed the experiments and wrote the manuscript; L.C. and E.S. contributed making the sRNAs libraries and sequencing; J.P.T and M.C. contributed on the bioinformatics analysis of smallRNAs; E.L.G. and K.B. contributed on the RNA methylation analysis; J.I.F. and L.A.P. contributed on the fish handling, sperm acquisition and eggs fertilization. All the authors contributed on the final manuscript edition.  
+
+Funding: This work was supported by the Agencia Nacional de Promoción Científica y Tecnológica (PICT 2018- 1879 to P.H.S-M.). J.P.T. would like to acknowledge support from Universidad de la República, Uruguay (CSIC I+D_2020_433). Work in the Greer lab was supported by an NIH grant (DP2AG055947) to E.L.G.  
+
+Competing interests: The authors declare no competing interests.  
+
+Materials & correspondence: Correspondence and material requests should be send to strobl@intech.gov.ar.  
+
+Data availability: Trimmed sequencing data that was generated in this study for the initial bioinformatics analysis have been submitted to the NCBI Sequence Read Archive under BioProject ID PRJNA857097 (http://www.ncbi.nlm.nih.gov/bioproject/857097).
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+<center>Figure S1: (A) Histogram showing the percentage of fertilization utilizing the sperm from control and MTX treated males. (B) Kaplan Meier plot displaying the survival of fertilized eggs until the beginning of hatch. Histogram showing the percentage of hatching embryos (C) and the day of hatching (D) on control and MTX group. Statistics for A and C were generated by contingency table followed by Chi-square test. Statistics for B and D were generated by using the Long-rank (Matel-Cox) test, Gehan-Breslow-Wilcoxon test. ns: \(P > 0.05\) . </center>
+
+<--- Page Split --->
+![](images/Figure_unknown_1.jpg)
+
+<center>Figure S2: (A) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from tRNAs not affected (tRNA-Pro<sup>UGG</sup>, -Arg<sup>UCU</sup>, -Val<sup>AAC</sup>) or having a reduction (tRNA-Ser<sup>GCU</sup>) on MTX treatment. (B) Histogram showing the read coverage for tRNA-Glu<sup>CUC</sup> and -Ser<sup>GCU</sup>. (C) Histogram showing the lack of length variation of mapped rRNAs on control and MTX treated males. See also supplementary table 5.</center>
+
+<--- Page Split --->
+
+## Supplementary tables:  
+
+Supplementary table 1: Relative abundance of sncRNA (tRNA, miRNA, snoRNA, snRNA, miscRNA) for control and MTX groups. Values are indicated for each replicate. Relative abundance of tRNAs, mean values are indicated.  
+
+Supplementary table 2: List of differentially expressed sncRNAs comparing MTX and control groups.  
+
+Supplementary table 3: List of differentially expressed 3' and 5' tsRNAs comparing MTX and control group.  
+
+Supplementary table 4: Proportion of 5' halves for the most abundant tRNAs, overall coverage length for tRNAs, and coverage- length for tRNA- AspGUC, - GlyGCC, - LysCUU.  
+
+Supplementary table 5: Proportion of 5' halves for tRNA- SerGcU, - ProUGG, - ArgUCU, - ValAAC. Size distribution of rRNAs reads, and coverage- length for tRNA- GluCUU, - SerGcU.  
+
+Supplementary table 6: List of primers and oligo adapters for RT- qPCR and smallRNA- seq.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+Supplementarytable1. xlsx Supplementarytable2. xlsx Supplementarytable3. xlsx Supplementarytable4. xlsx SupplementaryTable5. xlsx Supplementarytable6. docx
+
+<--- Page Split --->

preprint/preprint__9821a16de1c5dff3e3be578498bc5241bb7e4f77f30df9398e8b1aafba548010/preprint__9821a16de1c5dff3e3be578498bc5241bb7e4f77f30df9398e8b1aafba548010_det.mmd

@@ -0,0 +1,456 @@
+<|ref|>title<|/ref|><|det|>[[44, 108, 938, 208]]<|/det|>
+# Paternal methotrexate exposure affects sperm small RNA content and causes craniofacial defects in the offspring  
+
+<|ref|>text<|/ref|><|det|>[[44, 230, 930, 300]]<|/det|>
+Nagif Alata Jimenez Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM) Mauricio Castellano  
+
+<|ref|>text<|/ref|><|det|>[[44, 297, 895, 339]]<|/det|>
+Functional Genomics Unit. Instituto Pasteur de Montevideo. School of Science. Universidad de la Republica.  
+
+<|ref|>text<|/ref|><|det|>[[44, 345, 362, 385]]<|/det|>
+Emilio Santillan Johns Hopkins School of Medicine  
+
+<|ref|>text<|/ref|><|det|>[[44, 391, 285, 432]]<|/det|>
+Konstantinos Boulias Boston Children's Hospital  
+
+<|ref|>text<|/ref|><|det|>[[44, 438, 931, 480]]<|/det|>
+Agustin Boan Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+<|ref|>text<|/ref|><|det|>[[44, 485, 930, 549]]<|/det|>
+Luisa Arias Padilla Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM) https://orcid.org/0000- 0003- 4689- 2561  
+
+<|ref|>text<|/ref|><|det|>[[44, 554, 930, 597]]<|/det|>
+Juan Femandino Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+<|ref|>text<|/ref|><|det|>[[44, 601, 720, 642]]<|/det|>
+Eric Greer Harvard/Boston Children's Hospital https://orcid.org/0000- 0002- 7501- 7371  
+
+<|ref|>text<|/ref|><|det|>[[44, 648, 656, 689]]<|/det|>
+Juan Tosar Universidad de la Republica https://orcid.org/0000- 0002- 2021- 2479  
+
+<|ref|>text<|/ref|><|det|>[[44, 694, 720, 735]]<|/det|>
+Luisa Cochella Johns Hopkins School of Medicine https://orcid.org/0000- 0003- 4018- 7722  
+
+<|ref|>text<|/ref|><|det|>[[44, 739, 930, 781]]<|/det|>
+Pablo strobl- mazzulla ( strobl@intech.gov.ar) Instituto Tecnologico de Chascomus (CONICET- UNSAM). Escuela de Bio y Nanotecnologias (UNSAM)  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 822, 101, 839]]<|/det|>
+## Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 860, 916, 901]]<|/det|>
+Keywords: methotrexate, paternal exposure, offspring craniofacial defects, tsRNAs, RNA methylation, medaka  
+
+<|ref|>text<|/ref|><|det|>[[44, 920, 296, 939]]<|/det|>
+Posted Date: July 18th, 2022
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[44, 46, 475, 64]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs-1841878/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 82, 911, 125]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[135, 81, 873, 128]]<|/det|>
+# Paternal methotrexate exposure affects sperm small RNA content and causes craniofacial defects in the offspring 
+
+<|ref|>text<|/ref|><|det|>[[92, 142, 912, 161]]<|/det|>
+Nagif Alata Jimenez1,2; Mauricio Castellano3,4; Emilio M. Santillan5; Konstantinos Boulias6,7; 
+
+<|ref|>text<|/ref|><|det|>[[92, 177, 910, 196]]<|/det|>
+Agustín Boan1,2; Luisa A. Padilla1,2; Juan I. Fernando1,2; Eric Lieberman Greer6,7; Juan P. 
+
+<|ref|>text<|/ref|><|det|>[[260, 211, 745, 229]]<|/det|>
+Tosar3,4; Luisa Cochella5; Pablo H. Strobl-Mazzulla1,2* 
+
+<|ref|>text<|/ref|><|det|>[[92, 277, 544, 295]]<|/det|>
+*Author for correspondence: strobl@intech.gov.ar. 
+
+<|ref|>text<|/ref|><|det|>[[92, 310, 914, 328]]<|/det|>
+1 Laboratory of Developmental Biology. Instituto de Investigaciones Biotecnológicas- Instituto 
+
+<|ref|>text<|/ref|><|det|>[[92, 343, 785, 361]]<|/det|>
+Tecnológico de Chascomús (CONICET-UNSAM). Chascomús, ARGENTINA. 
+
+<|ref|>text<|/ref|><|det|>[[92, 376, 754, 394]]<|/det|>
+2Escuela de Bio y Nanotecnologías (UNSAM). Chascomús, ARGENTINA. 
+
+<|ref|>text<|/ref|><|det|>[[92, 409, 861, 427]]<|/det|>
+3Functional Genomics Unit. Instituto Pasteur de Montevideo. Montevideo. URUGUAY. 
+
+<|ref|>text<|/ref|><|det|>[[92, 442, 759, 460]]<|/det|>
+4School of Science. Universidad de la República. Montevideo. URUGUAY. 
+
+<|ref|>text<|/ref|><|det|>[[92, 475, 914, 493]]<|/det|>
+5Department of Molecular Biology and Genetics, Johns Hopkins University School of 
+
+<|ref|>text<|/ref|><|det|>[[92, 508, 420, 526]]<|/det|>
+Medicine, Baltimore, Maryland, USA 
+
+<|ref|>text<|/ref|><|det|>[[92, 541, 914, 559]]<|/det|>
+6Department of Pediatrics, HMS Initiative for RNA Medicine, Harvard Medical School, Boston 
+
+<|ref|>text<|/ref|><|det|>[[92, 574, 184, 591]]<|/det|>
+MA, USA. 
+
+<|ref|>text<|/ref|><|det|>[[92, 606, 783, 625]]<|/det|>
+7Division of Newborn Medicine, Boston Children’s Hospital, Boston MA, USA. 
+
+<|ref|>text<|/ref|><|det|>[[92, 770, 914, 786]]<|/det|>
+Keywords: methotrexate, paternal exposure, offspring craniofacial defects, tsRNAs, RNA 
+
+<|ref|>text<|/ref|><|det|>[[92, 802, 286, 819]]<|/det|>
+methylation, medaka.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 82, 178, 100]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[88, 110, 920, 563]]<|/det|>
+Folate is an essential vitamin for vertebrate embryo development. Methotrexate (MTX) is a folate antagonist that is widely prescribed for autoimmune diseases, blood and solid organ malignancies, and dermatologic diseases. Although it is highly contraindicated for pregnant women, because it is associated with an increased risk of multiple birth defects, the effect of paternal MTX exposure on their offspring has been largely unexplored. Here, we found MTX treatment of adult medaka male fish (Oryzias latipes) causes cranial cartilage defects in their offspring. Small non- coding RNA (sncRNAs) sequencing in the sperm of MTX treated males identify differential expression of a subset of tRNAs, with higher abundance for specific 5' tRNA halves. Sperm RNA methylation analysis on MTX treated males shows that m5C is the most abundant and differential modification found in RNAs ranging in size from 50 to 90 nucleotides, predominantly tRNAs, and that it correlates with greater testicular Dnmt2 methyltransferase expression. Overall, our data suggest that paternal MTX exposure alters sperm sncRNAs expression and modifications that may contribute to developmental defects in their offspring.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 83, 233, 103]]<|/det|>
+## Introduction  
+
+<|ref|>text<|/ref|><|det|>[[88, 113, 919, 680]]<|/det|>
+Folate is a water- soluble vitamin obtained from the diet that is essential for vertebrates. It is incorporated as an essential cofactor for the synthesis of nucleotides and the generation of S- adenosylmethionine (SAM) which serves as a universal donor of methyl groups for DNA, RNA and proteins implicated in gene regulation during early development 1- 4. Maternal folate deficiency leads to severe neural tube defects and craniofacial anomalies of descendants 5- 7. Importantly, the prevalence of these defects is highly reduced by folic acid supplementation prior and during pregnancy 8,9. Despite global efforts to supplement the maternal diets with folate, there is still a worldwide prevalence of these congenital defects 10- 12. Methotrexate (MTX) is a recognized teratogenic folic acid antagonist that has been linked to an elevated incidence of congenital anomalies in children born from exposed women. Intrauterine MTX exposure has been linked to craniofacial and limb defects, as well as developmental delays 13,14. In addition to oral clefts, folic acid antagonists may raise the risk of cardiovascular, neural tube, and urinary tract abnormalities 15. As a result, current recommendations urge that mothers stop using MTX at least three months before conception 16. Prior research has also identified a variety of issues concerning MTX use and a probable genotoxic effect on sperm, which might result in chronic disease or congenital anomalies 17. However, medical care recommendations for males taking MTX while trying to conceive are less clear.  
+
+<|ref|>text<|/ref|><|det|>[[88, 685, 919, 901]]<|/det|>
+For decades, the sperm genome has been considered transcriptionally quiescent and solely contributing to the restoration of the ploidy of the zygote. However, more recently, a set of functional RNAs have been characterized in mature spermatozoa that are delivered to the oocyte upon fertilization, contributing to early embryo development and thus, influencing the phenotypic outcome of the offspring 18- 24. Intriguingly, paternal folate concentrations can affect the sperm epigenome 25,26. Whereas the direct impact of these changes is expected to be minimal given the protamine exchange and resetting of DNA methylation during
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 134]]<|/det|>
+spermatogenesis and early development \(^{27 - 29}\) , we wondered whether paternal folate levels may also affect the RNA composition of mature sperm.  
+
+<|ref|>text<|/ref|><|det|>[[88, 145, 919, 437]]<|/det|>
+Small non- coding RNAs (sncRNAs) are a particularly attractive potential carrier of nongenetic information in the spermatozoa. In particular, tRNA- derived small RNAs (tsRNAs) and microRNAs (miRNAs) are the most abundant in mature spermatozoa \(^{30,31}\) ; and have been identified as molecular carriers of paternal experiences, including high fat diet \(^{22,24,32}\) , low protein diet \(^{33}\) , stress \(^{21,34}\) , and odoriferous sensitivity to chemicals \(^{23}\) . Small RNA biogenesis, stability and functionality are highly dependent on their post- transcriptional modification status, primarily methylation \(^{35 - 37}\) . Furthermore, transmission of paternally acquired metabolic disorders is dependent on the presence of post- transcriptional modifications in sperm sncRNAs \(^{19,24}\) .  
+
+<|ref|>text<|/ref|><|det|>[[88, 450, 919, 636]]<|/det|>
+Here, we explored the intriguing possibility that paternal folate deficiency impacts the offspring's development, and that it may do so through changes in sncRNA abundance and methylation levels. We injected medaka male fish with the folate inhibitor methotrexate (MTX) and characterized their offspring's developmental defects. Next, we analyzed and compared the abundance and modifications of sncRNAs present in the sperm of MTX- treated males to test the idea that they work as mediators of congenital developmental defects.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 682, 170, 699]]<|/det|>
+## Results  
+
+<|ref|>text<|/ref|><|det|>[[88, 712, 919, 866]]<|/det|>
+Paternal folate deficiency induced cranial cartilage malformations in their offspringTo investigate the impact of paternal folate deficiency on the development of their progeny, we administered medaka male fish with an intraperitoneal injection of methotrexate (MTX), a well- known folate inhibitor \(^{38 - 40}\) , at 10mg of MTX per Kg of body weight (10MTX) and 50mg/Kg MTX (50MTX)(Fig. 1A). After 7 days, we fertilized wild type oocytes with sperm extracted
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 78, 920, 636]]<|/det|>
+from treated and untreated males. None of our treatments had a significant impact on sperm fecundity, hatching time, or overall embryo hatching (Fig. S1). Several studies have shown that folate is an important vitamin for neural and neural crest development in several vertebrate species including humans \(^{5,41,42}\) . Moreover, maternal folate deficiency during pregnancy leads to abnormal development of neural crest derivatives such as cranial cartilages \(^{43 - 46}\) . Taking this into account, we first evaluated the effect of paternal folate deficiency on the development of the offspring's cranial cartilages by performing alcian blue staining at 3 days post hatching-stage (3dph). We measured the length of three dorsal cartilages (anterior orbital, epiphyseal bar and posterior orbital), four ventral cartilages (Meckel, ceratohyal, basibranchial and palatoguadrate), and the Meckel's area and ceratohyal angle (Fig. 1B-J). From the dorsal cartilages, we found a significant reduction in the length of the anterior orbital (also known as taenia marginalis anterior) in the 50MTX group (115.02 \(\mu \mathrm{m} \pm 9.03\) , one- way ANOVA followed by multiple comparison Tukey's test \(p = 0.0164\) ) when compared to the 10MTX (130.02 \(\mu \mathrm{m} \pm 10.22\) ) and control (125.98 \(\mu \mathrm{m} \pm 15.77\) ). On the ventral side, the basibranchial and Meckel's cartilages were not affected. However, the ceratohyal was reduced to almost half the length, at both 10MTX (192.99 \(\mu \mathrm{m} \pm 7.55\) , \(p < 0.0001\) ) and 50MTX (185.42 \(\mu \mathrm{m} \pm 8.71\) , \(p < 0.0001\) ) compared with control (363.64 \(\mu \mathrm{m} \pm\)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 365]]<|/det|>
+18.96). Interestingly, when we looked at the morphology of those cranial cartilages, we found that two of them, the anterior orbital and basihyal, presented an abnormal shape (Fig. 2). In particular, the anterior orbital has an abnormal serpentine shape, compared with the normal curved shape (Fig. 2A-B). This phenotype was significantly prevalent ( \(p = 0.0059\) ) at the 50MTX group (Fig. 2C). However, one of the most drastically affected cartilages was the basihyal, whose phenotypes presented a curved trowel shape (mild) or hook shape (strong) (Fig. 2D). Quantitation of those phenotypes evidences a significant increase in the severity of them at both 10MTX ( \(p = 0.0329\) ) and 50MTX ( \(p = 0.0006\) ) compared with Control group (Fig. 2E). Overall, these findings support the notion that paternal MTX exposure affects the  
+
+<|ref|>image<|/ref|><|det|>[[100, 375, 900, 820]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[103, 832, 911, 911]]<|/det|>
+<center>Figure 1: Paternal MTX injection affects offspring's cranial cartilages lengths. (A) scheme of experimental design. (B-J) Violin plots represent the measurement of different cranial cartilages lengths, angles and areas on control (Ctrl), 10mg/kg MTX and 50 mg/kg MTX. Statistical analyses were performed using ANOVA one-way followed by multiple comparison Tukey's test. \(*P = 0.0164\) , \(***P = 0.0008\) , \(****P < 0.0001\) , ns: \(P > 0.05\) . </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 81, 916, 135]]<|/det|>
+development of the offspring's cranial cartilage, indicating that sperm may convey some information involved in the observed phenotypic inheritance.  
+
+<|ref|>sub_title<|/ref|><|det|>[[90, 180, 730, 200]]<|/det|>
+## SncRNAs abundance is altered in the sperm of MTX-treated males  
+
+<|ref|>image<|/ref|><|det|>[[123, 225, 852, 560]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[77, 575, 896, 728]]<|/det|>
+<center>Figure 2: Paternal MTX injection produced offspring's cranial cartilages malformations. (A-B) Dorsal view of normal and affected anterior orbital (ao) showing a meandering shape mostly observed on the offspring from MTX treated males. (C) Quantification of the percentage of embryos presenting affected or normal anterior orbital cartilages. Numbers in the graph represent the analyzed embryos. (D) Lateral view of larvae presenting normal (trowel shape), mild (bended shape) and strong (hook shape) deformities of the basihyal cartilage. (E) Quantification of the percentage of embryos presenting normal (non-affected), mild or strong basihyal cartilage abnormalities observed on the offspring from MTX treated males. Numbers in the graph represent the analyzed embryos. Statistical analyses were performed using a contingency table followed by Fisher's exact test. \(*P = 0.0329\) , \(**P = 0.0059\) , \(***P = 0.0006\) , ns: \(P > 0.05\) . Means \(\pm\) SEM. </center>  
+
+<|ref|>text<|/ref|><|det|>[[80, 731, 916, 912]]<|/det|>
+Epigenetic information, including histone modifications and DNA methylation, particularly from the paternal lineage, is largely reprogrammed during germline and early embryo development. However, increasing evidence indicates that sncRNAs are a carrier of epigenetic information across generations and may act as mediators of paternally inherited traits \(^{18 - 23,47}\) . To assess if paternal folate deficiency affects the small RNA content, we sequenced size selected (\~18- 30 nt long) RNAs from sperm of 10MTX and control males.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 920, 400]]<|/det|>
+Based on the analysis of three biological replicated for each group, we were able to identify different populations of sncRNAs including: transfer RNA fragments, miRNAs, small nucleolar RNAs (snoRNAs), small nuclear RNAs (snRNAs) and other miscellaneous RNAs (miscRNA) (Fig. 3A). The comparative analysis showed that tRNA fragments were the most abundant (58.47% ± 6.6) and became further enriched in response to 10MTX treatment (87.26% ± 2.24). Interestingly, some of the most abundant tRNA fragments, aspartic acid (having the anticodon Asp<sup>GUC</sup>), glutamic acid (Glu<sup>CUC</sup> and Glu<sup>UUC</sup>), lysine (Lys<sup>CUU</sup>) and glycine (Gly<sup>GCC</sup>) Fig. 3B), became further significantly enriched upon MTX treatment (Fig. 3C- D). Together, these results demonstrate that paternal MTX exposure affects the relative abundance of specific sncRNAs in the sperm, with tRNA fragments being the most affected population.  
+
+<|ref|>sub_title<|/ref|><|det|>[[90, 442, 881, 462]]<|/det|>
+## 5' halves of particular tRNAs are preferentially affected by methotrexate treatment  
+
+<|ref|>text<|/ref|><|det|>[[90, 473, 919, 562]]<|/det|>
+tRNAs can be cleaved into 5' and 3' halves, known as tsRNAs, in response to stress or other external factors<sup>19,24,35,48</sup>. Of particular interest in the sperm RNA content is the large abundance of those tsRNA fragments and their potential regulatory roles in early embryo
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[231, 82, 805, 396]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[92, 405, 912, 529]]<|/det|>
+<center>Figure 3: Paternal MTX injection alters sperm sncRNAs. (A) Histogram representing the comparison from sperm sncRNA on control (Ctrl1-3) and MTX (MTX-1-3) treated males. (B) Histogram displaying biotypes of tRNAs from sperm of MTX treated males. See also supplementary table 1 for A and B. (C) Volcano plot of depicting the fold changes in sperm sncRNAs identified as being differentially expressed within control versus MTX-treated males. (D) MA plot displaying normalized counts (base mean) for different sncRNAs. Dotted lines depict thresholds values for significantly up and down-regulated (± ≥1 log₂ fold change and -log₁₀Pvalue ≥1.3). See also supplementary table 2 for C and D. </center>  
+
+<|ref|>text<|/ref|><|det|>[[92, 530, 916, 910]]<|/det|>
+development \(^{19,20,24,30,31}\) . Interestingly, we observed that 5' tsRNA fragments, product of the cleavage of three of the most abundant tRNAs (5' tsRNA of Asp<sup>GUC</sup>, Lys<sup>CUU</sup> and Glu<sup>CUC</sup>), were significantly increased in MTX sperm, without a concomitant increase in their respective 3' tsRNA fragments (Fig. 4A-B, E). Quantification of the proportion of 5' halves relative to their corresponding 3' halves showed a significant increase in the percentage of 5' halves for tRNA Asp<sup>GUC</sup> and Gly<sup>GCC</sup> in the MTX treated samples (Fig. 4C, E). It is interesting to note that for some tRNAs (i.e., tRNA Glu<sup>CUC</sup>, Glu<sup>UUC</sup> and Gly<sup>UCC</sup>) we mostly retrieve reads for their 5' halves, but their corresponding 3' halves are almost undetected for both control and MTX. On the other hand, the 5' halves of many (most) other tRNAs did not show differences compared with their 3' halves (tRNA Pro<sup>UGG</sup>, Arg<sup>UCU</sup>), or a major proportion of their 3' halves (tRNA Ser<sup>GCU</sup>) (Fig. S2). These results suggest changes in processing or stability of specific tRNA fragments as a consequence of the MTX treatment.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[30, 81, 919, 201]]<|/det|>
+154 Production of 5' tsRNAs fragments in the 15- 22 nt range occurs in multiple tissues and cell 155 lines \(^{49}\) , whereas longer 5' tsRNAs (31- 40 nt long) are preferentially generated in response 156 to different stresses \(^{50,51}\) . We thus compared the length distribution of tRNAs- derived 157 fragments in both conditions and observed a shift towards longer fragments in 10MTX relative  
+
+<|ref|>image<|/ref|><|det|>[[93, 214, 904, 720]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[84, 728, 911, 910]]<|/det|>
+<center>Figure 4. Paternal MTX injection induced the expression and cleavage of specific tRNAs. (A) Volcano plot of depicting the fold changes in sperm 5' and 3' tRNA halves as being differentially expressed within control versus MTX treated males. (B) MA plot displaying normalized counts (base mean) for different 5' and 3' tRNA halves. Dotted lines depict thresholds values for significantly up and down-regulated \((\pm \geq 1\) log2 fold change and -log10Pvalue \(\geq 1.3\) ). See also supplementary table 3 for A and B. (C) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from different tRNAs affected by MTX treatment. Asterisk indicated significant differences analyzed by multiple unpaired t-student' test ( \(^{*}P< 0.05\) ). (D) Histogram showing the length variation of mapped tRNA reads on control and MTX treated males. Histogram showing the read coverage (E) and size (F) distribution for the most abundant and having a significant increase in the 5'tsRNA (tRNA-GluCUC, tRNA-AspGUC, and tRNA-GlyGCC) between control and MTX. See also supplementary table 4 for C, D and E. </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 330]]<|/det|>
+to control samples (26- 35 vs. 18- 29 nt) (Fig. 4D). A similar situation is found for the tRNAs Glu<sup>CUC</sup>, Asp<sup>GUC</sup> and Lys<sup>CUU</sup> where their 5'tsRNAs are significantly increased and their coverage lengths are greater in MTX- treated samples than in controls (Fig. 4F). There is a chance that bias size selection occurred when separating the small RNAs from the gel, resulting in these differences. However, when we looked at the length of mapped reads for ribosomal RNAs (rRNAs), we found no consistent changes (Fig. S2C). These findings suggest that paternal MTX exposure alters the abundance and cleavage site of specific 5'tsRNAs in the sperm.  
+
+<|ref|>sub_title<|/ref|><|det|>[[90, 376, 574, 396]]<|/det|>
+## m5C modifications are increased by methotrexate  
+
+<|ref|>text<|/ref|><|det|>[[88, 405, 919, 890]]<|/det|>
+Post- transcriptional modifications of tRNAs, including methylation, are important for their specific cleavage, stability and functionality, as well as for the transmission of paternal experiences to the offspring \(^{35 - 37,52}\) . We thus evaluated the methylation status of two populations of RNAs isolated from polyacrylamide: a 20- 50 nt population (mostly enriched for miRNAs and tsRNAs), and a 50- 90 nt population (mostly enriched for mature tRNAs). Within the 20- 50 nt RNA population we did not observe significant differences in the abundance of any of the analyzed methylation events between MTX and control groups (Fig. 5A). Conversely, within the 50- 90 nt population, we found that MTX treatment led to a significant increase in the relative abundance of several modifications (Fig. 5B). From the two most abundant modifications analyzed (m1A and m5C) only m5C was significantly increased (p = 0.0155) in MTX treated samples. From the other less abundant modifications only m2G, m7G and m2'2G presented a significant increase in MTX samples (p value = 0.0027; 0.0063; and 0.0090, respectively). Interestingly, the most abundant modification observed to be differentially detected in MTX samples has been described to be located at the 3' ends of tRNAs \(^{53,54}\) (Fig. 5C).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[30, 80, 919, 201]]<|/det|>
+Given the observed differences for certain RNA modifications, we examined the expression of specific RNA methyltransferases on the testis of control and 10MTX treated males (Fig. 5D). In agreement with our results, there was no significant change in the expression of Trmt6 which catalyzes m1A methylation. Conversely, the expression of the enzymes that catalyzed  
+
+<|ref|>image<|/ref|><|det|>[[135, 216, 884, 609]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[90, 613, 911, 730]]<|/det|>
+<center>Figure 5: Paternal MTX injection alters RNA modifications in sperm tRNA fraction and the testicular expression of RNA-methyltransferases. Histogram comparing sperm RNA methylations on control and MTX analyzed by UHPLC-MS-MS in 50-90nt (A) and 20-50nt (B) fractions. (C) Schematic representation of modified nucleotides in the tRNA at secondary and tertiary structure. (D) RT-qPCR for methyltransferases of m1A (TRMT6) and m5C (DNMT2 and NSUM2) on testis from control and MTX treated males, gene expression was normalized using Rpl7 and Ef1 as housekeeping genes. Statistical analysis was performed by using the unpaired \(t\) -student' test. \(*P<\) 0.05, \(^{**}P< 0.01\) , ns: \(P > 0.05\) . Means \(\pm\) SEM. (E) Proposed model summarizing the results. </center>  
+
+<|ref|>text<|/ref|><|det|>[[90, 733, 916, 754]]<|/det|>
+m5C was only significantly upregulated for Dnmt2 (p = 0.01), but not for Nsun2 (p = 0.46).  
+
+<|ref|>text<|/ref|><|det|>[[90, 770, 916, 822]]<|/det|>
+These results suggest that an increase of RNA methyltransferase expression leads to changes in the methylation status of sperm tRNAs upon MTX treatment.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 875, 220, 895]]<|/det|>
+## Discussion
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 366]]<|/det|>
+MTX binds to and inhibits dihydrofolate reductase activity, preventing folic acid from performing its biological tasks. For more than 30 years, this drug has been used to treat immunological illnesses (including rheumatoid arthritis), blood and solid organ cancers, dermatologic diseases, and pregnancy termination \(^{55,56}\) . Despite the drug's contraindication for pregnant women due to the risk of miscarriage and birth abnormalities, the paternal influence of MTX on their offspring was largely unknown. In addition to this, the vast majority of studies in fish models such as medaka and zebrafish has been performed during embryological stages \(^{57 - 60}\) , while few have evaluated the effect on adults \(^{61}\) and the consequences on their offspring.  
+
+<|ref|>text<|/ref|><|det|>[[88, 375, 919, 628]]<|/det|>
+In agreement with other studies in mice \(^{62,63}\) , we found that paternal MTX treatments had no effect on the fertility or survival of their progeny during the early embryonic stages. Lifetime exposure to folic acid- deficient diets, on the other hand, lead to lower sperm counts, negative consequences in progeny, and epigenetic changes \(^{62,63}\) . However, this may be due to folate deficiency during embryonic and post- embryonic development, which could compromise early germ cell formation and adult spermatogenesis. Moreover, major epigenetic reprogramming occurs at these periods, and multiple imprinting areas may be altered as a result of the prolonged folate shortage.  
+
+<|ref|>text<|/ref|><|det|>[[88, 638, 919, 893]]<|/det|>
+For many years, paternal contribution to offspring's health was thought to be restricted to the haploid genome of spermatozoa, whereas mother health and nutrition were linked to offspring's wellness. However, multiple recent studies have revealed that spermatozoa carry a variety of RNAs \(^{18,64 - 66}\) capable to transmit paternal experiences \(^{19,22 - 24,32}\) . In this regard, our work illustrated the critical significance of MTX therapy and its impact on sperm small non- coding RNA content as a possible mechanism underlying the observed craniofacial abnormalities or possibly other unexplored effects of this treatment. We discovered that tsRNAs and miRNAs are the most common small non- coding RNA in medaka sperm, which
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 339]]<|/det|>
+is consistent with past findings in mammals \(^{19,30,31,66,67}\) . Furthermore, we revealed that tsRNAs halves changed significantly owing MTX treatment, which is in agreement with previous studies showing that tsRNAs are a dynamic population that responds to a variety of environmental stressors \(^{19,24,68}\) . Particularly, we observed a higher abundance of certain 5'tsRNAs, where 5'tsRNA- AspGUC was the most abundant. This result is in concordance with several studies where external factors also modulated the abundance of 5'tsRNA- AspGUC \(^{20,24,31}\) , thus highlighting the idea that certain tRNAs may be preferentially cleaved and their 5' halves have a longer half- life compared to their respective 3' halves.  
+
+<|ref|>text<|/ref|><|det|>[[88, 348, 919, 599]]<|/det|>
+TsRNAs can be generated through multistep cleavages, through the formation of various intermediates. Moreover, there is growing evidence that regulatory factors, such as RNA modifications and specific RNases, have a role in their specific cleavage and stability \(^{66}\) . Interestingly, we found that 5'tsRNAs from AspGUC and GlyGCC are consistently longer ( \(\sim 35\) nt) on the sperm of MTX- treated males. This is in agreement with the discovery that small 15- 22 nucleotide long fragments are normally formed in multiple tissues and cell lines \(^{49}\) , whereas longer 31- 40 nucleotide tRNA halves are preferentially cleaved in response to different stresses \(^{50,51}\) .  
+
+<|ref|>text<|/ref|><|det|>[[88, 608, 919, 895]]<|/det|>
+Up to now, the tsRNA functions are very speculative, but have been associated to translation, ribosome biogenesis, retrotransposition, cell- cell communication, and epigenetic inheritance, as well as how tsRNA dysregulation are related to a variety of human disorders (summarized in recent reviews \(^{69,70}\) ). Importantly, both tsRNAs and their precursor tRNAs are heavily modified, which contributes to multiple aspects of their function, biogenesis, stability, amino acid charging, and translational accuracy \(^{71,72}\) . Our initial hypothesis was that MTX treatment may reduce the tRNA methylation thus inducing their cleavage. This is based on previous studies where the addition of m5C, which is controlled by DNMT2 and NSUN2, increase tRNA stability in flies and mice, but its deletion makes them more likely to be cleaved into tsRNAs
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 80, 919, 430]]<|/det|>
+under stress conditions \(^{19,36,37}\) . However, and contrarily to our predictions, we observed that tRNAs- enriched samples (\~50- 70 nt) derived from MTX- treated sperm showed significantly greater levels of methylation in m5C, m2G, m7G and m2'2G. Increased levels of m5C and m2G have been observed in the 30- 40 nucleotide fraction of sperm RNAs (predominantly tRNAs) in mice fed with high- fat diets compared with those from males fed with normal diets \(^{24}\) . However, it is important to mention that in our MTX- treatment we fail to observe differences in the methylation levels from the tsRNAs/miRNAs fraction (\~20- 50 nt). We speculate that because the most abundant tRNA modifications found in our study (5mC) are stated to be positioned at the 3' end of tRNAs (positions 38C, 48C, 49C, 50C) \(^{73}\) , then the cleaved 3'tsRNAs halves, which accumulate the bulk of these methylations, may be preferentially degraded.  
+
+<|ref|>text<|/ref|><|det|>[[88, 440, 919, 891]]<|/det|>
+The high levels of 5mC in our 50- 90nt fraction from MTX- treated males correlated with the higher expression of Dnmt2 (also known as Trdmt1), but not Nsun2. Dnmt2 is structurally close to other DNA methyltransferases but rather methylates only one tRNA, specifically at the cytosine 38 in the anticodon loop of aspartic acid (tRNA- Asp) \(^{74}\) . Interestingly, here we have found that 5'tsRNA- Asp<sup>GUC</sup> was the second most abundant tRNA in MTX- treated males and presenting a significant increase respect to their 3'tsRNA- Asp<sup>GUC</sup> half. In contradiction to our finding, Schaefe et al. \(^{37}\) demonstrated that m5C modification mediated by DNMT2 improves tRNA stability, where tRNA- Asp is protected from angiogenin cleavage during the heat shock response in Drosophila. In mammals, it is well known that angiogenin activity, RNase that cleaves tRNAs, is also inhibited by the presence of 5mC \(^{19,36}\) . However, it is important to mention that since the endonuclease targeting the anti- codon loop of Drosophila tRNAs has not been identified yet, the authors analyzed the cleavage of tRNA- Asp induced by the addition of recombinant human angiogenin into Drosophila S2 cells \(^{37}\) , which may not reflect the truly physiological condition. Moreover, it has been shown that the presence of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 78, 920, 530]]<|/det|>
+angiogenin is not mandatory for the generation of tsRNAs, and other RNases (Dicer, RNase T2, L) can also cleave tRNAs \(^{69,75,76}\) . In that sense, fish does not have angiogenin, but instead orthologues genes with the capacity to cleave tRNAs have been found \(^{77 - 79}\) suggesting that the generation of tRNA fragments is an evolutive response against environmental stressors. In addition, it is important to mention that the activity \(^{77}\) , structure \(^{78 - 80}\) , and targeted dinucleotides for cleavage are different in between fish and mammals RNases \(^{80}\) . These facts suggest that the overall generation of tRNA fragments is an ancient response where RNases have maintained their main role and have evolved as the organisms did it. On the other hand, the presence of 5mC, and/or other modifications, might affect their activity in a different way as it was speculated by Barraud and Tisné \(^{81}\) . These authors stated that tRNA modifications are critical features of the cellular stress responses, and described the existence of a streaky crosstalk among them regulating the tRNAs stability \(^{81}\) . As a result, modifications may act as a "barcode" to regulate the specific tRNA cleavage and stability resulting in the accumulation of specific tsRNAs in the sperm, which could affect the phenotype of their offspring.  
+
+<|ref|>text<|/ref|><|det|>[[88, 538, 920, 855]]<|/det|>
+In summary, our data suggests that paternal MTX- exposure influenced sperm tRNA methylation, as a result of alterations in the expression of certain RNA methyltransferases. These epitranscriptomic changes may cause the selective tRNA cleavage and the maintenance of certain 5' tRNA halves. These changes in the sperm RNA content and modifications might affect transcriptional cascades in the fertilized oocyte, with possible implications in cranial cartilage formation (see hypothetical model in figure 5E). The understanding of how tRNA modifications and their derived fragments impact on the transcriptional cascades occurring during early embryo will provide valuable insights into several diseases and it is expected that this will be a main focus of research in this field in the near future.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 83, 193, 103]]<|/det|>
+## Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 120, 281, 139]]<|/det|>
+## Medaka Husbandry  
+
+<|ref|>text<|/ref|><|det|>[[91, 150, 919, 374]]<|/det|>
+All experiments were performed with medaka fish (Oryzias latipes) (strain hi- medaka, ID:MT835) supplied by the National BioResource Project (NBRP) Medaka (http://www.shingen.nig.ac.jp/medaka/). Fish were maintained and fed following standard protocols for medaka 82. Fish were handled on the Care and Management of Laboratory Animals (http://www.ufaw.org.uk) and internal regulations. Adult fishes were divided and acclimatized in 4L fish tank during 3 weeks under a constant photoperiod (14L:10D) and controlled temperature (26 ± 0.5°C), prior to experimental procedures.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 419, 293, 438]]<|/det|>
+## Experimental design  
+
+<|ref|>text<|/ref|><|det|>[[91, 449, 919, 764]]<|/det|>
+Adults medaka fish were divided into 3 groups composed by 3 independent replicates having 9 males and 2 females per tank (all of them having a body weight of \(\sim 200 \text{mg}\) ). After the acclimatizing period, each male was intraperitoneal injected with control solution (PBS/1%DMSO), 10mg of MTX per kg of body weight (10mg/kg MTX) or 50mg/kg MTX (A6770- SigmaAldrich, diluted in PBS/1%DMSO). Briefly, males were anesthetized with 1% benzoacaine solution (Parafarm), gently dried with a paper towel and placed in a dampened sponge ventral side up, with their anal fin and cloaca exposed. Immediately, using a 10ul syringe (Hamilton), fish were injected using a binocular stereoscope (Nikon SMZ745) and then returned to their tanks for 7 days until sperm collection for in vitro fertilization and small RNA extraction.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 813, 276, 831]]<|/det|>
+## In vitro fertilization
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 81, 919, 366]]<|/det|>
+Sperm collection was carried by anesthetizing the fish and placed in a dampened sponge ventral side up following published protocols for medaka 82. A micro- forceps was used to gently strip the fish and the released semen was collected by using a micropipette ( \(\sim 0.5\mu \mathrm{l} / \mathrm{fish}\) ) and pooled for the posterior in vitro fertilization and small RNA extraction. For the in vitro fertilization, \(0.2\mu \mathrm{l}\) from obtained sperm were used to fertilize a pool of 24- 28 eggs collected from mature untreated females. Fertilized eggs were immediately transferred and incubated in \(60~\mathrm{mm}\) petri dishes with embryo media (17mM NaCl, \(0.4\mathrm{mM}\) KCl, \(0.27\mathrm{mM}\) \(\mathrm{CaCl_2.2H_2O}\) , and \(0.66\mathrm{mM}\) MgSO4; pH:7) until 3 days' post hatching (dph). Incubation was monitored and the percentage of fertilization and survival until hatching was evaluated.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 411, 286, 430]]<|/det|>
+## Alcian blue staining  
+
+<|ref|>text<|/ref|><|det|>[[88, 441, 919, 858]]<|/det|>
+Cartilages from embryos were analyzed at 3 dph by using alcian blue staining. Larvae were fixed in \(4\%\) paraformaldehyde overnight at \(4^{\circ}\mathrm{C}\) and washed three times with PBSw (PBS- \(0.1\%\) tween20). After that, embryos were incubated in a bleaching solution (0.5X SSC, \(5\%\) formamide, \(10\%\) hydrogen peroxide) and exposed to light during 2 hours. Larvae were washed several times with PBSw and immediately incubated in alcian Blue solution ( \(0.1\% \mathrm{p} / \mathrm{v}\) alcian blue, \(0.37\% \mathrm{v} / \mathrm{v}\) HCl, \(70\% \mathrm{v} / \mathrm{v}\) EtOH) for 1 hour on a nutator. Then, larvae were washed five times with \(01\% \mathrm{v} / \mathrm{v}\) HCl- \(70\% \mathrm{v} / \mathrm{v}\) EtOH for 30 minutes on a nutator; the last wash was left overnight at room temperature. Next, larvae were washed six times with \(50\% \mathrm{v} / \mathrm{v}\) glycerol- \(0.5\% \mathrm{v} / \mathrm{v}\) KOH for 30 min on a nutator and the last wash was left overnight. Finally, larvae were washed four times with the same solution and left in \(90\%\) glycerol- \(10\%\) EtOH for imaging processing and phenotype analysis. Larvae were photographed at ventral, dorsal and lateral view by using a trinocular stereoscope (SteREO Discovery v20. Zeiss) and analyzed using the ImageJ software 83.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[91, 82, 527, 101]]<|/det|>
+## Small RNA extraction and library preparation  
+
+<|ref|>text<|/ref|><|det|>[[88, 110, 919, 895]]<|/det|>
+Small RNAs were isolated from sperm following manufacturer's instructions (illustra RNAspin Mini RNA isolation kit- GE Healthcare). The 3' adapters were ligated using SRBC barcode adapters for each sample, additionally 18- mer and 30- mer markers were ligated and used as control for the ligation process and markers for the product correct size. The 3'- ligated small RNAs were size selected using \(15\%\) denaturing urea polyacrylamide gels at a constant power of 40- 50W for \(\sim 30\) minutes and stained by using SYBR Gold \(0.05\% V / V\) in TBE \(0.5X\) and the 3' ligated RNAs ranging from 18- 30 mer were cut out. RNAs were purified using Zymo PAGE elution kit (ZRTM small RNA PAGE recovery kit) according to manufacturer's instructions, the elute 3'- ligated small RNAs were elute in 5' linker mix containing 5' adaptor. The 3'- ligated RNAs + 5' adapter were denaturated for 5 min at \(70^{\circ}C\) , cooled on ice immediately, ligated with T4 RNA ligase (NEB) and incubated at 16 degrees overnight. Ligated small RNAs were purified by using MBS beads, briefly: MBS buffer, MBS bead slurry (beads + buffer), mixed by vortexing, added isopropanol and incubated at room temperature. Beads were separated on magnet and the supernatant was removed, after several washes with ethanol the RNA was eluted with ultrapure water and transferred into PCR strip. For reverse transcription, small- RNAseq RT primer to each sample were used and a negative control without reverse transcriptase was included, Superscript II reverse transcriptase was used to obtain the synthesis of the first strand. To amplify cDNA libraries, KAPA HiFi Real Time Library Amplification Kit (Roche) was used; PCR were performed using TruSeq Universal Adapter primer (Solexa_PCR_fwd) and TruSeq Index reverse primers (Solexa_IDX_rev), this latter includes barcodes assigned to each different sample. Briefly: master mix was added and TruSeq Index reverse primer were added to PCR strips containing cDNAs; then KAPA HiFi HS RM and Truseq Universal Adapter primer were added to the mix. The cycling program was: Denaturation at \(98^{\circ}C\) for 45 sec; 20 cycles of \(98^{\circ}C\) for 15 sec, \(65^{\circ}C\) for 30 sec, \(72^{\circ}C\) for
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 80, 919, 299]]<|/det|>
+30 sec, \(72^{\circ}C\) for 10 sec; and a final extension at \(72^{\circ}C\) for 1 min. The amplified cDNA was purified by using \(3\%\) Low- Range Ultra Agarose gel (Bio- Rad) according to the manufacturer's instructions at constant 80- 100 V using GeneRuler 50bp DNA Ladder (ThermoFischer Scientific) as molecular marker. Gel was visualized on a long wave UV transilluminator and DNA band between 150- 200 bp were excised using a clean scalpel blade and put into a clean 15ml Falcon tube; the DNA was purified using the Zymoclean Gel DNA recovery kit (Zymo Research) according to manufacturer's instructions.  
+
+<|ref|>sub_title<|/ref|><|det|>[[93, 345, 327, 364]]<|/det|>
+## Bioinformatics Analysis  
+
+<|ref|>text<|/ref|><|det|>[[88, 375, 919, 896]]<|/det|>
+Adapters from reads were removed using CUTADAPT, the output were reads \(\geq 15\) bp, reads whose adapters were not identified were discarded. The output of 15bp were used to analyze differential expression of sncRNAs (miRNAs, tRNAs, snRNAs, snoRNAs and rRNAs) and differential expression of tRNAs, 5' tRNA halves and 3' tRNA halves by different strategies. First, differential expression of sncRNAs was analyzed on reads where the random nucleotides on 5' (4bp) and 3' (6pb) were cut using FASTQ Trimmer. The obtained reads having \(< 19\) bp were discarded using Filter Fastq and the remaining reads were aligned against the medaka genome (Assembly ASM00223467v1) with RNA STAR (allowing multimapping reads, 1 mismatch, and not allowing introns). Expression of sncRNAs was analyzed using FeatureCounts (allowing multimapping reads to be counted, and assigning 1/n fractions to multimapping reads) with Ensembl annotation (Release v102). Differential expression of sncRNAs was calculated using DESEQ2. Second, to analyze differential expression of 5' and 3' tRNA halves, an additional 3 base pairs were removed with FASTQ Trimmer from the 3' end of all reads. Reads having less than 15bp were discarded using Filter Fastq. The output was aligned and analyzed as mentioned before using custom GTF files with genomic coordinates for either 5' or 3' tRNA halves. To determine the sequence length of mapped
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 81, 917, 201]]<|/det|>
+tRNA reads, BAM files were filtered (using GTFs files containing genomic coordinates for full length tRNAs), reads were extracted, converted to fasta and their length computed with in-house scripts. tRNA read coverage was calculated with BamCoverage (bin size 1, no smoothing).  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 247, 240, 265]]<|/det|>
+## UHPLC-MS-MS  
+
+<|ref|>text<|/ref|><|det|>[[88, 275, 918, 895]]<|/det|>
+The analysis of modified ribonucleotides from spermatic RNAs were performed by UHPLC- MS- MS. For that purpose, \(\sim 1.5 \mu \mathrm{g}\) of total RNAs were isolated from two independent pools of stripped sperm from \(\sim 9\) control and 10MTX- treated males and run in denaturant polyacrylamide gel (15%, 7M Urea). The gel was then stained with ethidium bromide and RNAs that have a ranged size from 20- 50nt and 50- 90nt were cut and recovered using the ZR small- RNA™ PAGE Recovery Kit (Zymo Research) by following the manufacturer's instruction. Approximately 100ng of RNA was obtained on each fraction and utilized for UHPLC- MS- MS analysis. Then, 100 ng purified RNA samples were digested to nucleosides for 2 hr at 37°C using the Nucleoside Digestion mix (NEB, M069S). Quantifications were performed as in 84, briefly: digested RNA samples were diluted to 100 μl with ddH2O and filtered through 0.22 μm Millex Syringe Filters. 5 μl of the filtered solution was injected for LC- MS/MS analysis using the Agilent 1290 UHPLC- MS/MS system with a Hypersil Gold C18 reversed- phase column (2.1 x 150 mm, 3 μm). Mobile phase A consisted of water with 0.1% (v/v) formic acid and mobile phase B consisted of acetonitrile with 0.1% (v/v) formic acid. Mass spectrometry detection was performed using an Agilent 6470 triple quadrupole mass spectrometer in positive electrospray ionization mode and data were quantified in dynamic multiple reaction monitoring (dMRM) mode, by monitoring the mass transitions 268 \(\square\) II 136 for Adenosine (A), 282 \(\square\) II 150 for N6- methyladenosine (m1A), 244 \(\square\) II 112 for Cytidine (C), 258 \(\square\) II 126 for C5- methylcytidine (m5C), 284 \(\square\) II 152 for Guanosine (G), 298 \(\square\) II 166 for
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 81, 917, 266]]<|/det|>
+N7'- methyladenosine (m7G) and N2- methylguanosine (m2G), 312 \(\square \square 180\) for N2,N2- dimethylguanosine (m2'2G), 282 \(\square \square 136\) for 2'- O- methyladenosine (Am), 258 \(\square \square 112\) for 2'- O- methylcytidine (Cm) and 298 \(\square \square 152\) for 2'- O- methylguanosine (Gm). To quantify the concentrations of the various methylation modifications we used pure nucleosides of A, C, G, m1A, m5C, m7G, m2G m2'2G, Am, Cm and Gm to generate calibration standard curves through serial dilution.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 310, 389, 330]]<|/det|>
+## RNA quantification by RT-PCR  
+
+<|ref|>text<|/ref|><|det|>[[88, 340, 917, 860]]<|/det|>
+Dissected testis from adult males (Control and 10MTX groups) were used for gene expression analysis. For this purpose, a complete testis from 8 males were individually grinded in \(300\mu l\) of TRlzol Reagent (Life Technologies) and total RNAs ( \(\sim 800ng\) /testis) were isolated and retrotranscribed \(^{85}\) . Expression of target genes were measured by qPCR using Fast Start Universal SYBR green Supermix (Roche Diagnostics, USA) on Thermal Cycler StepOne Plus (Applied Biosystem,USA), using ribosomal protein L7 (rpl7) and elongation factor 1 alpha (ef1α) as reference genes with the geometric mean calculation as described by Padilla et al. \(^{85}\) . Real- time PCR primers are listed in Supplementary Table 6. Each sample was run in duplicate and a PCR reaction without the addition of template, was used as negative control. The amplification protocol consisted of an initial cycle of 1min at \(95^{\circ}C\) followed by 40 cycles: 15 s at \(95^{\circ}C\) and 30 s at \(60^{\circ}C\) . After the amplification, a melt curve was performed by 1 cycle: 15 s at \(95^{\circ}C\) , 60 s at \(60^{\circ}C\) and 15 s at \(95^{\circ}C\) enabling confirmation of amplification of single products. Gene expression levels were calculated by the \(2^{-\Delta \Delta Ct}\) comparative threshold cycle (Ct) method (where \(\Delta \Delta Ct = \Delta Ct\) sample - \(\Delta Ct\) reference). The efficiency of amplification ranged 95- 105% for all genes studied. The expression level in each group was normalized to the control and was presented as a fold of change \(^{86}\) .
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[30, 120, 911, 870]]<|/det|>
+441 1. Sánchez-Vásquez E, Alata Jimenez N, Vázquez NA, Strobl-Mazzulla PH. Emerging role of 442 dynamic RNA modifications during animal development. Mechanisms of Development. 443 2018;154. doi:10.1016/j.mod.2018.04.002 444 2. Greenberg MVC, Bourc'his D. The diverse roles of DNA methylation in mammalian 445 development and disease. Nature Reviews Molecular Cell Biology. 2019;20(10). 446 doi:10.1038/s41580-019-0159-6 447 3. Bernstein C, Bernstein H. Demethylation in Early Embryonic Development and Memory. In: 448 DNA Methylation Mechanism. ; 2020. doi:10.5772/intechopen.90306 449 4. O'Neill C. The epigenetics of embryo development. Animal Frontiers. 2015;5(1). 450 doi:10.2527/af.2015-0007 451 5. Li J, Shi Y, Sun J, Zhang Y, Mao B. Xenopus reduced folate carrier regulates neural crest 452 development epigenetically. PLoS ONE. 2011;6(11). doi:10.1371/journal.pone.0027198 453 6. Shaw GM, Schaffer D, Velie EM, Morland K, Harris JA. Periconceptional vitamin use, dietary 454 folate, and the occurrence of neural tube defects. Epidemiology. 1995;6(3). 455 doi:10.1097/00001648- 199505000- 00005 456 7. Tolarova M. PERICONCEPTIONAL SUPPLEMENTATION WITH VITAMINS AND FOLIC ACID TO 457 PREVENT RECURRENCE OF CLEFT LIP. The Lancet. 1982;320(8291). doi:10.1016/S0140- 458 6736(82)91063- 7 459 8. McDonald SD, Ferguson S, Tam L, Lougheed J, Walker MC. The prevention of congenital 460 anomalies with periconceptional folic acid supplementation. Journal of obstetrics and 461 gynaecology Canada : JOGC = Journal d'obstetrique et gynécologie du Canada : JOGC. 462 2003;25(2). doi:10.1016/S1701- 2163(16)30207- 9 463 9. Vergel RG, Sanchez LR, Heredero BL, Rodriguez PL, Martinez AJ. Primary prevention of neural 464 tube defects with folic acid supplementation: Cuban experience. Prenatal Diagnosis. 465 1990;10(3). doi:10.1002/pd.1970100303 466 10. Blencowe H, Kanchera V, Moorthie S, Darlison MW, Modell B. Estimates of global and 467 regional prevalence of neural tube defects for 2015: a systematic analysis. Ann N Y Acad Sci. 468 2018;1414(1). doi:10.1111/nyas.13548 469 11. Salari N, Darvishi N, Heydari M, Bokaees S, Darvishi F, Mohammadi M. Global prevalence of 470 cleft palate, cleft lip and cleft palate and lip: A comprehensive systematic review and meta- 471 analysis. Journal of Stomatology, Oral and Maxillofacial Surgery. 2022;123(2). 472 doi:10.1016/j.jormas.2021.05.008 473 12. Impellizzeri A, Giannantoni I, Polimeni A, Barbato E, Galluccio G. Epidemiological 474 characteristic of Orofacial clefts and its associated congenital anomalies: Retrospective study. 475 BMC Oral Health. 2019;19(1). doi:10.1186/s12903- 019-0980-5 476 13. Mahadevan U, McConnell RA, Chambers CD. Drug Safety and Risk of Adverse Outcomes for 477 Pregnant Patients With Inflammatory Bowel Disease. Gastroenterology. 2017;152(2). 478 doi:10.1053/j.gastro.2016.10.013 479 14. Hyoun SC, Obican SG, Scialli AR. Teratogen update: Methotrexate. Birth Defects Research 480 Part A - Clinical and Molecular Teratology. 2012;94(4). doi:10.1002/bdra.23003 481 15. Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Folic Acid Antagonists during 482 Pregnancy and the Risk of Birth Defects. New England Journal of Medicine. 2000;343(22). 483 doi:10.1056/nejm200011303432204
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[30, 78, 914, 904]]<|/det|>
+484 16. Nguyen GC, Seow CH, Maxwell C, et al. The Toronto Consensus Statements for the 485 Management of Inflammatory Bowel Disease in Pregnancy. Gastroenterology. 2016;150(3). 486 doi:10.1053/j.gastro.2015.12.003 487 17. Beghin D, Cournot MP, Vauzelle C, Elefant E. Paternal exposure to methotrexate and 488 pregnancy outcomes. Journal of Rheumatology. 2011;38(4). doi:10.3899/jrheum.100600 489 18. Sharma U, Sun F, Conine CC, et al. Small RNAs Are Trafficked from the Epididymis to 490 Developing Mammalian Sperm. Developmental Cell. 2018;46(4). 491 doi:10.1016/j.devcel.2018.06.023 492 19. Zhang Y, Zhang X, Shi J, et al. Dnmt2 mediates intergenerational transmission of paternally 493 acquired metabolic disorders through sperm small non-coding RNAs. Nature Cell Biology. 494 2018;20(5). doi:10.1038/s41556-018-0087-2 495 20. Chen X, Zheng Y, Lei A, et al. Early cleavage of preimplantation embryos is regulated by 496 tRNAGln-TTG-derived small RNAs present in mature spermatozoa. Journal of Biological 497 Chemistry. 2020;295(32). doi:10.1074/jbc.ra120.013003 498 21. Rodgers AB, Morgan CP, Leu NA, Bale TL. Transgenerational epigenetic programming via 499 sperm microRNA recapitulates effects of paternal stress. Proc Natl Acad Sci U S A. Published 500 online 2015. doi:10.1073/pnas.1508347112 501 22. Fullston T, Teague EMCO, Palmer NO, et al. Paternal obesity initiates metabolic disturbances 502 in two generations of mice with incomplete penetrance to the F2 generation and alters the 503 transcriptional profile of testis and sperm microRNA content. FASEB Journal. Published online 504 2013. doi:10.1096/fj.12-224048 505 23. Gapp K, Jawaid A, Sarkies P, et al. Implication of sperm RNAs in transgenerational inheritance 506 of the effects of early trauma in mice. Nature Neuroscience. Published online 2014. 507 doi:10.1038/nn.3695 508 24. Chen Q, Yan M, Cao Z, et al. Sperm tsRNAs contribute to intergenerational inheritance of an 509 acquired metabolic disorder. Science (1979). Published online 2016. 510 doi:10.1126/science.aad7977 511 25. Aarabi M, San Gabrie MC, Chan D, et al. High-dose folic acid supplementation alters the 512 human sperm methylome and is influenced by the MTHFR C677T polymorphism. Human 513 Molecular Genetics. 2015;24(22). doi:10.1093/hmg/ddv338 514 26. Boxmeer JC, Smit M, Utomo E, et al. Low folate in seminal plasma is associated with 515 increased sperm DNA damage. Fertility and Sterility. 2009;92(2). 516 doi:10.1016/j.fertnstert.2008.06.010 517 27. Wang Y, Wang H, Zhang Y, et al. Reprogramming of Meiotic Chromatin Architecture during 518 Spermatogenesis. Molecular Cell. 2019;73(3). doi:10.1016/j.molcel.2018.11.019 519 28. Santos F, Hendrich B, Reik W, Dean W. Dynamic reprogramming of DNA methylation in the 520 early mouse embryo. Developmental Biology. 2002;241(1). doi:10.1006/dbio.2001.0501 521 29. Feng S, Jacobsen SE, Reik W. Epigenetic reprogramming in plant and animal development. 522 Science (1979). Published online 2010. doi:10.1126/science.1190614 523 30. Chen Q, Yan W, Duan E. Epigenetic inheritance of acquired traits through sperm RNAs and 524 sperm RNA modifications. Nature Reviews Genetics. 2016;17(12). doi:10.1038/nrg.2016.106 525 31. Sharma U, Conine CC, Shea JM, et al. Biogenesis and function of tRNA fragments during 526 sperm maturation and fertilization in mammals. Science (1979). 2016;351(6271). 527 doi:10.1126/science.aad6780 528 32. Grandjean V, Fourré S, de Abreu DAF, Derieppe MA, Remy JJ, Rassoulzadegan M. RNA- 529 mediated paternal heredity of diet-induced obesity and metabolic disorders. Scientific 530 Reports. Published online 2015. doi:10.1038/srep18193
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[29, 80, 910, 870]]<|/det|>
+531 33. Carone BR, Fauquier L, Habib N, et al. Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals. Cell. Published online 2010. doi:10.1016/j.cell.2010.12.008
+534 34. Wu L, Lu Y, Jiao Y, et al. Paternal Psychological Stress Reprograms Hepatic Gluconeogenesis in Offspring. Cell Metabolism. Published online 2016. doi:10.1016/j.cmet.2016.01.014
+535 35. Blanco S, Dietmann S, Flores J v, et al. Aberrant methylation of t RNA s links cellular stress to neuro-developmental disorders. The EMBO Journal. Published online 2014. doi:10.15252/embj.201489282
+536 36. Tuorto F, Liebers R, Musch T, et al. RNA cytosine methylation by Dnmt2 and NSun2 promotes tRNA stability and protein synthesis. Nature Structural and Molecular Biology. 2012;19(9). doi:10.1038/nsmb.2357
+537 37. Schaefer M, Pollex T, Hanna K, et al. RNA methylation by Dnmt2 protects transfer RNAs against stress-induced cleavage. Genes and Development. 2010;24(15). doi:10.1101/gad.586710
+538 38. Kao TT, Lee GH, Fu CC, Chen BH, Chen LT, Fu TF. Methotrexate-Induced Decrease in Embryonic 5-Methyl-Tetrahydrofolate Is Irreversible with Leucovorin Supplementation. Zebrafish. 2013;10(3). doi:10.1089/zeB.2013.0876
+539 39. Cronstein BN, Aune TM. Methotrexate and its mechanisms of action in inflammatory arthritis. Nature Reviews Rheumatology. 2020;16(3). doi:10.1038/s41584-020-0373-9
+540 40. Dervieux T, Greenstein N, Kremer J. Pharmacogenomic and metabolic biomarkers in the folate pathway and their association with methotrexate effects during dosage escalation in rheumatoid arthritis. Arthritis and Rheumatism. 2006;54(10). doi:10.1002/art.22129
+541 41. Alata Jimenez N, Torres Perez SA, Sanchez-Vasquez E, Fernandino JI, Strobl-Mazzulla PH. Folate deficiency prevents neural crest fate by disturbing the epigenetic Sox2 repression on the dorsal neural tube. Developmental Biology. Published online 2018. doi:10.1016/j.ydbio.2018.08.001
+542 42. Antony AC, Hansen DK. Hypothesis: Folate-responsive neural tube defects and neurocristopathies. Teratology. Published online 2000. doi:10.1002/1096-9926(200007)62:1<42::AID-TERA9>3.0.CO;2-U
+543 43. van Rooij IALM, Ocké MC, Straatman H, Zielhuis GA, Merkus HMWM, Steegers-Theunissen RPM. Periconceptional folate intake by supplement and food reduces the risk of nonsyndromic cleft lip with or without cleft palate. Preventive Medicine. 2004;39(4). doi:10.1016/j.ypmed.2004.02.036
+544 44. Wilcox AJ, Lie RT, Solvoll K, et al. Folic acid supplements and risk of facial clefts: National population based case-control study. British Medical Journal. 2007;334(7591). doi:10.1136/bmj.39079.618287.0B
+545 45. Burgoon JM, Selhub J, Nadeau M, Sadler TW. Investigation of the effects of folate deficiency on embryonic development through the establishment of a folate deficient mouse model. Teratology. 2002;65(5). doi:10.1002/tera.10040
+546 46. Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. The Lancet. 1991;338(8760). doi:10.1016/0140-6736(91)90133-A
+547 47. Ord J, Heath PR, Fazeli A, Watt PJ. Paternal effects in a wild-type zebrafish implicate a role of sperm-derived small RNAs. Molecular Ecology. 2020;29(14). doi:10.1111/mec.15505
+548 48. Rashad S, Han X, Sato K, et al. The stress specific impact of ALKBH1 on tRNA cleavage and tRNA generation. RNA Biology. 2020;17(8). doi:10.1080/15476286.2020.1779492
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[26, 80, 910, 884]]<|/det|>
+49. Kumar P, Anaya J, Mudunuri SB, Dutta A. Meta-analysis of tRNA derived RNA fragments reveals that they are evolutionarily conserved and associate with AGO proteins to recognize specific RNA targets. BMC Medicine. 2014;12(1). doi:10.1186/s12915-014-0078-0
+50. Thompson DM, Lu C, Green PJ, Parker R. tRNA cleavage is a conserved response to oxidative stress in eukaryotes. RNA. 2008;14(10). doi:10.1261/rna.1232808
+51. Yamasaki S, Ivanov P, Hu GF, Anderson P. Angiogenin cleaves tRNA and promotes stress-induced translational repression. Journal of Cell Biology. 2009;185(1). doi:10.1083/jcb.200811106
+52. Mishima E, Inoue C, Saigusa D, et al. Conformational change in transfer RNA is an early indicator of acute cellular damage. Journal of the American Society of Nephrology. Published online 2014. doi:10.1681/ASN.2013091001
+53. Huber SM, Leonardi A, Dedon PC, Begley TJ. The versatile roles of the tRNA epitranscriptome during cellular responses to toxic exposures and environmental stress. Toxics. 2019;7(1). doi:10.3390/toxics7010017
+54. Hori H. Methylated nucleosides in tRNA and tRNA methyltransferases. Frontiers in Genetics. 2014;5(MAY). doi:10.3389/fgene.2014.00144
+55. Smith SW, Nelson LS. Case files of the New York City Poison Control Center: antidotal strategies for the management of methotrexate toxicity. J Med Toxicol. 2008;4(2). doi:10.1007/BF03160968
+56. LoVecchio F, Katz K, Watts D, Wood I. Four-year experience with methotrexate exposures. J Med Toxicol. 2008;4(3). doi:10.1007/BF03161192
+57. Jacabacci J, Guzzardi M, D'Aversa TG. Exogenous folic acid is not able to inhibit the effects of methotrexate on the folate pathway in zebrafish embryos. BIOS. 2015;86(3). doi:10.1893/bios-d-14-00027.1
+58. Liu S, Narumi R, Ikeda N, Morita O, Tasaki J. Chemical-induced craniofacial anomalies caused by disruption of neural crest cell development in a zebrafish model. Developmental Dynamics. 2020;249(7). doi:10.1002/dvdy.179
+59. Sun S, Gui Y, Wang Y, et al. Effects of methotrexate on the developments of heart and vessel in zebrafish. Acta Biochimica et Biophysica Sinica. 2009;41(1). doi:10.1093/abs/gmn010
+60. Lee MS, Bonner JR, Bernard DJ, et al. Disruption of the folate pathway in zebrafish causes developmental defects. BMC Developmental Biology. Published online 2012. doi:10.1186/1471-213X-12-12
+61. Hirako A, Takeoka Y, Furukawa S, Sugiyama A. Effects of methotrexate exposure on medaka (Oryzias latipes) testes. Journal of Toxicologic Pathology. 2017;30(4). doi:10.1293/tox.2017-0029
+62. Lambrot R, Xu C, Saint-Phar S, et al. Low paternal dietary folate alters the mouse sperm epigenome and is associated with negative pregnancy outcomes. Nature Communications. 2013;4. doi:10.1038/ncomms3889
+63. Ly L, Chan D, Aarabi M, et al. Intergenerational impact of paternal lifetime exposures to both folic acid deficiency and supplementation on reproductive outcomes and imprinted gene methylation. Molecular Human Reproduction. 2017;23(7). doi:10.1093/molehr/gax029
+64. Zhang Y, Shi J, Rassoulzadegan M, Tuorto F, Chen Q. Sperm RNA code programmes the metabolic health of offspring. Nature Reviews Endocrinology. 2019;15(8). doi:10.1038/s41574-019-0226-2
+65. Johnson GD, Mackie P, Jodar M, Moskovtsev S, Krawetz SA. Chromatin and extracellular vesicle associated sperm RNAs. Nucleic Acids Research. 2015;43(14). doi:10.1093/nar/gkv591
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[30, 78, 920, 888]]<|/det|>
+66. Peng H, Shi J, Zhang Y, et al. A novel class of tRNA-derived small RNAs extremely enriched in mature mouse sperm. Cell Research. 2012;22(11). doi:10.1038/cr.2012.141
+67. Kawano M, Kawaji H, Grandjean V, Kiani J, Rassoulzadegan M. Novel Small Noncoding RNAs in Mouse Spermatozoa, Zygotes and Early Embryos. PLoS ONE. 2012;7(9). doi:10.1371/journal.pone.0044542
+68. Fu H, Feng J, Liu Q, et al. Stress induces tRNA cleavage by angiogenin in mammalian cells. FEBS Letters. 2009;583(2). doi:10.1016/j.febslet.2008.12.043
+69. Su Z, Wilson B, Kumar P, Dutta A. Noncanonical Roles of tRNAs: TRNA Fragments and beyond. Annual Review of Genetics. 2020;54. doi:10.1146/annurev-genet-022620-101840
+70. Schimmel P. RNA Processing and Modifications: The emerging complexity of the tRNA world: Mammalian tRNAs beyond protein synthesis. Nature Reviews Molecular Cell Biology. 2018;19(1). doi:10.1038/nrm.2017.77
+71. Pan T. Modifications and functional genomics of human transfer RNA. Cell Research. 2018;28(4). doi:10.1038/s41422-018-0013-y
+72. Suzuki T. The expanding world of tRNA modifications and their disease relevance. Nature Reviews Molecular Cell Biology. 2021;22(6). doi:10.1038/s41580-021-00342-0
+73. Pereira M, Francisco S, Varanda AS, Santos M, Santos MAS, Soares AR. Impact of tRNA modifications and tRNA-modifying enzymes on proteostasis and human disease. International Journal of Molecular Sciences. 2018;19(12). doi:10.3390/ijms19123738
+74. Goll MG, Kirpekar F, Maggert KA, et al. Methylation of tRNAAsp by the DNA methyltransferase homolog Dnmt2. Science (1979). 2006;311(5759). doi:10.1126/science.1120976
+75. Chen Q, Zhang X, Shi J, Yan M, Zhou T. Origins and evolving functionalities of tRNA-derived small RNAs. Trends in Biochemical Sciences. 2021;46(10). doi:10.1016/j.tibs.2021.05.001
+76. Akiyama Y, Lyons S, Fay MM, Abe T, Anderson P, Ivanov P. Multiple ribonuclease A family members cleave transfer RNAs in response to stress. bioRxiv. 2019;(1).
+77. Cho S, Zhang J. Zebrafish ribonucleasees are bactericidal: Implications for the origin of the vertebrate RNase a superfamily. Molecular Biology and Evolution. 2007;24(5). doi:10.1093/molbev/msm047
+78. Pizzo E, Buonanno P, di Maro A, et al. Ribonucleasees and angiogenins from fish. Journal of Biological Chemistry. 2006;281(37). doi:10.1074/jbc.M605505200
+79. Kazakou K, Holloway DE, Prior SH, Subramanian V, Acharya KR. Ribonuclease A Homologues of the Zebrafish: Polymorphism, Crystal Structures of Two Representatives and their Evolutionary Implications. Journal of Molecular Biology. 2008;380(1). doi:10.1016/j.jmb.2008.04.070
+80. Prats-Ejarque G, Lu L, Salazar VA, Moussaoui M, Boix E. Evolutionary trends in RNA base selectivity within the RNase A superfamily. Frontiers in Pharmacology. 2019;10. doi:10.3389/fphar.2019.01170
+81. Barraud P, Tisné C. To be or not to be modified: Miscellaneous aspects influencing nucleotide modifications in tRNAs. IUBMB Life. 2019;71(8). doi:10.1002/iub.2041
+82. Kinoshita M, Murata K, Naruse K, Tanaka M. Medaka: Biology, Management, and Experimental Protocols.; 2012. doi:10.1002/9780813818849
+83. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nature Methods. 2012;9(7). doi:10.1038/nmeth.2089
+84. Liberman N, Gerashchenko M v., Boulias K, et al. Intergenerational hormesis is regulated by heritable 18S rRNA methylation. bioRxiv. Published online 2021.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 82, 910, 118]]<|/det|>
+85. Padilla LFA, Castañeda-Cortés DC, Rosa IF, et al. Cystic proliferation of germline stem cells is necessary to reproductive success and normal mating behavior in Medaka. Elife. 2021;10. doi:10.7554/elife.62757  
+
+<|ref|>text<|/ref|><|det|>[[90, 118, 910, 135]]<|/det|>
+70. doi:10.7554/elife.62757  
+
+<|ref|>text<|/ref|><|det|>[[90, 136, 916, 172]]<|/det|>
+86. Schmittgen TD, Livak KJ. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4).  
+
+<|ref|>text<|/ref|><|det|>[[90, 219, 896, 271]]<|/det|>
+Acknowledgements: We thank all the authors and members in LBD for their contribution and helpful discussions during the course of our study.  
+
+<|ref|>text<|/ref|><|det|>[[90, 283, 917, 466]]<|/det|>
+Authors contribution: N.A.J. and P.H.S-M. designed, performed the experiments and wrote the manuscript; L.C. and E.S. contributed making the sRNAs libraries and sequencing; J.P.T and M.C. contributed on the bioinformatics analysis of smallRNAs; E.L.G. and K.B. contributed on the RNA methylation analysis; J.I.F. and L.A.P. contributed on the fish handling, sperm acquisition and eggs fertilization. All the authors contributed on the final manuscript edition.  
+
+<|ref|>text<|/ref|><|det|>[[90, 480, 905, 600]]<|/det|>
+Funding: This work was supported by the Agencia Nacional de Promoción Científica y Tecnológica (PICT 2018- 1879 to P.H.S-M.). J.P.T. would like to acknowledge support from Universidad de la República, Uruguay (CSIC I+D_2020_433). Work in the Greer lab was supported by an NIH grant (DP2AG055947) to E.L.G.  
+
+<|ref|>text<|/ref|><|det|>[[90, 612, 696, 632]]<|/det|>
+Competing interests: The authors declare no competing interests.  
+
+<|ref|>text<|/ref|><|det|>[[90, 644, 895, 696]]<|/det|>
+Materials & correspondence: Correspondence and material requests should be send to strobl@intech.gov.ar.  
+
+<|ref|>text<|/ref|><|det|>[[90, 708, 916, 795]]<|/det|>
+Data availability: Trimmed sequencing data that was generated in this study for the initial bioinformatics analysis have been submitted to the NCBI Sequence Read Archive under BioProject ID PRJNA857097 (http://www.ncbi.nlm.nih.gov/bioproject/857097).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[186, 111, 773, 528]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[91, 536, 875, 621]]<|/det|>
+<center>Figure S1: (A) Histogram showing the percentage of fertilization utilizing the sperm from control and MTX treated males. (B) Kaplan Meier plot displaying the survival of fertilized eggs until the beginning of hatch. Histogram showing the percentage of hatching embryos (C) and the day of hatching (D) on control and MTX group. Statistics for A and C were generated by contingency table followed by Chi-square test. Statistics for B and D were generated by using the Long-rank (Matel-Cox) test, Gehan-Breslow-Wilcoxon test. ns: \(P > 0.05\) . </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[90, 85, 880, 640]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[95, 650, 900, 728]]<|/det|>
+<center>Figure S2: (A) Histogram displaying percentage of 5' halves relative to their corresponding 3' halves from tRNAs not affected (tRNA-Pro<sup>UGG</sup>, -Arg<sup>UCU</sup>, -Val<sup>AAC</sup>) or having a reduction (tRNA-Ser<sup>GCU</sup>) on MTX treatment. (B) Histogram showing the read coverage for tRNA-Glu<sup>CUC</sup> and -Ser<sup>GCU</sup>. (C) Histogram showing the lack of length variation of mapped rRNAs on control and MTX treated males. See also supplementary table 5.</center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 83, 312, 101]]<|/det|>
+## Supplementary tables:  
+
+<|ref|>text<|/ref|><|det|>[[90, 113, 909, 193]]<|/det|>
+Supplementary table 1: Relative abundance of sncRNA (tRNA, miRNA, snoRNA, snRNA, miscRNA) for control and MTX groups. Values are indicated for each replicate. Relative abundance of tRNAs, mean values are indicated.  
+
+<|ref|>text<|/ref|><|det|>[[90, 204, 866, 253]]<|/det|>
+Supplementary table 2: List of differentially expressed sncRNAs comparing MTX and control groups.  
+
+<|ref|>text<|/ref|><|det|>[[90, 264, 866, 312]]<|/det|>
+Supplementary table 3: List of differentially expressed 3' and 5' tsRNAs comparing MTX and control group.  
+
+<|ref|>text<|/ref|><|det|>[[90, 324, 880, 373]]<|/det|>
+Supplementary table 4: Proportion of 5' halves for the most abundant tRNAs, overall coverage length for tRNAs, and coverage- length for tRNA- AspGUC, - GlyGCC, - LysCUU.  
+
+<|ref|>text<|/ref|><|det|>[[90, 384, 881, 434]]<|/det|>
+Supplementary table 5: Proportion of 5' halves for tRNA- SerGcU, - ProUGG, - ArgUCU, - ValAAC. Size distribution of rRNAs reads, and coverage- length for tRNA- GluCUU, - SerGcU.  
+
+<|ref|>text<|/ref|><|det|>[[90, 446, 856, 467]]<|/det|>
+Supplementary table 6: List of primers and oligo adapters for RT- qPCR and smallRNA- seq.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 43, 310, 71]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 93, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 131, 319, 283]]<|/det|>
+Supplementarytable1. xlsx Supplementarytable2. xlsx Supplementarytable3. xlsx Supplementarytable4. xlsx SupplementaryTable5. xlsx Supplementarytable6. docx
+
+<--- Page Split --->

preprint/preprint__9828c81619ba2f2920e2f04445ff9b9dc5c80df021f1f0112ee62d50f62d662e/images_list.json

@@ -0,0 +1,77 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1 Location of the archaeological site of Dispilio and detailed view of the analysed trench. a: map of S-E Europe marking the location of the enlarged area in b.; b: Location of Dispilio and other Neolithic sites within \\(\\sim 100 \\mathrm{km}\\) with reported good wood preservation and similar chronological placement, therefore with high potential for dendrochronological cross-dating with Dispilio (1-Anarghiri III; 2-Anarghiri IXb; 3-Crkveni Livadi; 4-Dispilio; 5-Dunavec; 6-Limnochori II; 7-Lin 3; 8-Maliq; 9-Ohridati/Penelopa; 10-Ustie na Drim, 11-Sovjan; QGIS 3.16, EPSG 32634; Lake Maliq according to Fouache et al. (2010)) c: drone photograph of the site of Dispilio and its surroundings, the dendrochronologically analysed East Sector marked in the foreground; d: close-up of the East Sector before sampling of wooden elements in 2019, vertical elements are seen sticking out of the ground, each marked with a unique white label. (a.,b.-A. Maczkowski; c.-M. Hostettler; d.-Dispilio Excavation Archive)",
+    "footnote": [],
+    "bbox": [
+      [
+        130,
+        98,
+        872,
+        631
+      ]
+    ],
+    "page_idx": 4
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2, Archaeological finds from Neolithic Dispilio. a: almost completely preserved ornate anthropomorphic vessel from Late Neolithic, many similar ones have been recovered from the site, scale in cm; b: bone spear/harpoon tip with preserved hafting adhesives, scale in cm; c.: an assemblage of Late Neolithic personal adornments (a.,b.,c.,-Dispilio Excavation Archive)",
+    "footnote": [],
+    "bbox": [
+      [
+        111,
+        473,
+        907,
+        670
+      ]
+    ],
+    "page_idx": 5
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3, Scatter plot of \\(\\Delta^{14}\\mathrm{C}\\) data from Dispilio against reference from Brehm et al. (2022) \\(^{17}\\) , and best last ring fit for the dated wood samples \\(\\left(\\chi^{2}\\right)\\) . a: Measured \\(^{14}\\mathrm{C}\\) concentrations represented as \\(\\Delta^{14}\\mathrm{C}\\) , vertical bars represent 1s uncertainties (Supplementary Table T1); samples marked with \"DISP-\" refer to measurements on wood samples obtained in this study, other labels represent data from BR22 \\(^{17}\\) - Bristlecone pine \\(^{14}\\mathrm{C}\\) data are shifted forward by 1 year from the original Brehm et al. publication, following a correction to the dating of the master bristlecone chronology (Supplementary Material S3.2); shaded band represents IntCal20 \\(^{66}\\) . Panels below, b, c: chi-squared tests of Dispilio measurements against the average from BR22 \\(^{17}\\) for wood samples DISP-10070 and -10063 (b, \\(\\chi^{2}\\) crit. value=9.49), and DISP-10206 and -10611 (c, \\(\\chi^{2}\\) crit. value=15.51). Figure produced in \\(R^{74}\\) , code and source data available in Supplementary Material 4.",
+    "footnote": [],
+    "bbox": [
+      [
+        110,
+        87,
+        880,
+        710
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4, Bar chart of tree-ring chronologies, felling dates, and site plan development. a: bar plot of Dispilio oak and juniper chronologies; each horizontal bar represents individual wood sample in its dendrochronologically cross-dated position, bar length corresponds to its span in years (i.e., number of tree-rings). Red stars indicate wood samples sampled for annual \\(^{14}\\mathrm{C}\\) ; b.: schematic plan of the East Sector (see also Fig. 1c-d); each symbol represents one vertical wooden element, different shapes and colours correspond to a same felling phase spread over 1-2 years; additionally, colour-shaded polygons outline the groups of same symbols (same felling-phase elements), however they do not represent definite structure plans.",
+    "footnote": [],
+    "bbox": [
+      [
+        112,
+        88,
+        797,
+        830
+      ]
+    ],
+    "page_idx": 10
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5 Wiggle-matching of different sets of annual \\(^{14}\\mathrm{C}\\) data from Dispilio modelled in OxCal v4.4, against IntCal20<sup>66</sup>, and IntCal20plus. IntCal20plus has the non-annual IntCal20 data for a 82-year period around the 5259 BC Miyake event replaced by annual average of Brehm et al. (2022)<sup>17</sup> annual data. Dotted blue lines represent actual felling dates determined through dendrochronology and Miyake event-matching. Acronyms in brackets next to sample name refer to AMS lab that furnished the measurements. Data for figure obtained from OxCal<sup>65,66</sup>. Figure produced in \\(R^{74}\\) , code and data in Supplementary Material 4.",
+    "footnote": [],
+    "bbox": [
+      [
+        111,
+        87,
+        879,
+        485
+      ]
+    ],
+    "page_idx": 12
+  }
+]

preprint/preprint__9828c81619ba2f2920e2f04445ff9b9dc5c80df021f1f0112ee62d50f62d662e/preprint__9828c81619ba2f2920e2f04445ff9b9dc5c80df021f1f0112ee62d50f62d662e.mmd

@@ -0,0 +1,311 @@
+
+# Absolutely dating the European Neolithic through a rapid 14C excursion  
+
+Andrej Maczkowski  
+
+andrej.maczkowski@unibe.ch  
+
+University of Bern https://orcid.org/0000- 0003- 3081- 3769  
+
+Charlotte Pearson University of Arizona  
+
+John Francuz University of Bern  
+
+Tryfon Giagkoulis University of Thessaloniki  
+
+Sonke Szidat  
+
+Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern https://orcid.org/0000- 0002- 1824- 6207  
+
+Lukas Wacker  
+
+Swiss Federal Institute of Technology (ETH) https://orcid.org/0000- 0002- 8215- 2678  
+
+Matthias Bolliger University of Bern  
+
+Kostas Kotsakis University of Thessaloniki  
+
+Albert Hafner University of Bern  
+
+Article  
+
+Keywords:  
+
+Posted Date: October 20th, 2023  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 3419721/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: There is NO Competing Interest.
+
+<--- Page Split --->
+
+Version of Record: A version of this preprint was published at Nature Communications on May 20th, 2024. See the published version at https://doi.org/10.1038/s41467-024-48402-1.
+
+<--- Page Split --->
+
+# Absolutely dating the European Neolithic through a rapid \(^{14}\mathrm{C}\) excursion  
+
+2 3 Andrej Maczkowski 1, \(2^{*}\) 4 Charlotte Pearson 3 5 John Francuz 1 6 Tryfon Giagkoulis 4 7 Sonke Szidat 5, 2 8 Lukas Wacker 7 9 Matthias Bolliger 1, 2, 6 10 Kostas Kotsakis 4 11 Albert Hafner 1, 2 12 13 Affiliations 14 1 Institute of Archaeological Sciences, University of Bern, Switzerland 15 2 Oeschger Centre for Climate Change Research, University of Bern, Switzerland 16 3 Laboratory of Tree-Ring Research, University of Arizona, USA 17 4 School of History and Archaeology, University of Thessaloniki, Greece 18 5 Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Switzerland 19 6 Laboratory for Dendrochronology, Archaeological Service Canton of Bern, Switzerland 20 7 Laboratory for Ion Beam Physics, ETH Zurich, Switzerland  
+
+## Abstract  
+
+The discovery of abrupt radiocarbon ( \(^{14}\mathrm{C}\) ) excursions (Solar Energetic Particle events, or Miyake events) in sequences of radiocarbon measurements from calendar dated tree- rings, has yielded new opportunities to assign absolute, calendar dates to undated wood samples from widely ranging contexts in history and prehistory. We report on an important tree- ring and \(^{14}\mathrm{C}\) - dating based study, which secures the Neolithic site of Dispilio, Northern Greece, a key site for the Aegean Neolithic, in absolute, calendar- dated time using the Miyake event of 5259 BC. The last ring of the 303- year- long juniper tree- ring chronology from Dispilio is dated to 5140 BC. Dispilio is thus the first prehistoric site absolutely dated through a \(^{14}\mathrm{C}\) signature (Miyake event), but also the first absolutely, calendar- year dated prehistoric site in the wider Mediterranean region.
+
+<--- Page Split --->
+
+## Introduction  
+
+The Neolithic period in western Eurasia marks one of the most important transitions in human social, economic, and technological history. This transition, lasting several millennia, is chiefly characterized by the appearance and gradual adoption of agriculture and animal husbandry, accompanied with increasing social and material culture complexity. The beginning of the Neolithic in Western Eurasia is dated to before \(\sim 9500\) BC in the Levant<sup>1</sup>, while its appearance on the Aegean coasts and continental Europe is dated to around \(\sim 6500\) BC<sup>2- 5</sup>. The earliest Neolithic sites on the continent are in Southeastern Europe, and their precise dating is essential for our understanding of the Neolithic transitions in Europe and critical to assessments of the environmental footprint of the new farming subsistence practices. However, the temporal resolution of archaeological and environmental proxies in the region is highly variable, producing significant discrepancies between various chronological and terminological systems that deal with the periodisation of the Neolithic<sup>6</sup>. Here we present the absolute dating of the Neolithic site of Dispilio in Northern Greece, via a combination of tree- ring dating (dendrochronology) and rapid <sup>14</sup>C excursions. This new data may serve as the basis for absolute dendrochronological dating of other sites from the Neolithic period in the region (Fig. 1).  
+
+Tree- rings enable high- resolution dating, the possibility of annually resolved climatic reconstruction and multidisciplinary chronological synchronization to (at best) a single growth season of a specific calendar dated year<sup>7</sup>. Until now, dendrochronological dating was possible only against reference tree- ring chronologies, which are continuous, unbroken sequences of tree- ring width records extending from the present back to the past. In this way, calendar dated tree- ring years can be assigned based on the known date of modern material, and then extended backwards through time using climatically constrained, region specific, tree- ring growth patterns. Long- term concentrated efforts in search for old wood samples has resulted in the construction of long tree- ring records extending for many thousands of years and widely applied to dating<sup>8- 10</sup>, and in some cases paleoclimatic analyses<sup>11,12</sup> of past human and environmental interactions. These records are however geographically limited and rare, and many prehistoric tree- ring chronologies are only approximately constrained on a calendar time- scale through conventional <sup>14</sup>C wiggle- matching and have no absolute calendar anchor.  
+
+This limitation can now be overcome by a new hybrid form of dendrochronological and single year radiocarbon analyses. Annual measurements of <sup>14</sup>C in dendrochronologically dated Holocene tree- rings have revealed the existence of rapid short- term spikes in atmospheric <sup>14</sup>C concentration in the past<sup>13,14</sup>. These <sup>14</sup>C spikes – also called Miyake or SEP (solar energetic particle) events – are uniquely suitable for absolute dating of any wooden objects with detectable annual rings<sup>15,16</sup>. The discovery of these short- term events has also led to a proliferation of annual <sup>14</sup>C measurements on single tree- rings, now spanning several millennia<sup>17- 19</sup>. The mechanisms behind these <sup>14</sup>C events are still debated<sup>20,21</sup>. However, a consensus explanation is that they are a result of coronal mass ejections on the Sun<sup>20,22- 24</sup> manifested as a surge of SEPs colliding with the Earth's atmosphere, in turn increasing the production of cosmogenic radionuclides<sup>17,24</sup>. To date, there are only five events<sup>13,14,17,25</sup> with an atmospheric <sup>14</sup>C increase ≥1% within 2 years<sup>17</sup>. Of these, the two most recently discovered events are in the first half of the Holocene – 7176 BC and 5259 BC<sup>17</sup> – offering for the first time the possibility for absolute annual dating of wood from the European Neolithic and Mesolithic using annual <sup>14</sup>C measurements.
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1 Location of the archaeological site of Dispilio and detailed view of the analysed trench. a: map of S-E Europe marking the location of the enlarged area in b.; b: Location of Dispilio and other Neolithic sites within \(\sim 100 \mathrm{km}\) with reported good wood preservation and similar chronological placement, therefore with high potential for dendrochronological cross-dating with Dispilio (1-Anarghiri III; 2-Anarghiri IXb; 3-Crkveni Livadi; 4-Dispilio; 5-Dunavec; 6-Limnochori II; 7-Lin 3; 8-Maliq; 9-Ohridati/Penelopa; 10-Ustie na Drim, 11-Sovjan; QGIS 3.16, EPSG 32634; Lake Maliq according to Fouache et al. (2010)) c: drone photograph of the site of Dispilio and its surroundings, the dendrochronologically analysed East Sector marked in the foreground; d: close-up of the East Sector before sampling of wooden elements in 2019, vertical elements are seen sticking out of the ground, each marked with a unique white label. (a.,b.-A. Maczkowski; c.-M. Hostettler; d.-Dispilio Excavation Archive) </center>  
+
+In temperate climates archaeological wood, and organic materials in general, can be preserved only in very stable conditions – such as constant low- oxygen waterlogged sediments at wetland archaeological sites \(27 - 29\) . While excavated wetland sites are very numerous and often excavated in Central Europe, several wetland sites have also been found and excavated in Southeastern Europe, notably in the south- western part of the Balkans \(30 - 36\) . Dendrochronological work on these sites led to the construction of several tree- ring width chronologies, which were fixed in time by means of \(^{14}\mathrm{C}\) modelling (wiggle- matching) \(37,38\) . The archaeological site of Dispilio on the shores of Lake Kastoria in Northern Greece is a premier prehistoric wetland site in the region. Numerous lines of evidence have yielded detailed results on the
+
+<--- Page Split --->
+
+georachaeology \(^{35}\) , palynology \(^{39,40}\) anthracology \(^{41,42}\) , woodworking technology \(^{43}\) , and material culture \(^{44,45}\) . The approximate calendar-age chronology of the site has been established through radiocarbon dates, mostly performed on charcoal samples \(^{35,46}\) . The calibrated date-ranges point to settlement phases between the later Middle Neolithic ("5600 cal BC \(^{47}\) ) and the Bronze Age ("2100 cal BC \(^{46}\) ). The excavations at Dispilio have also yielded a great number of wood remains, with over 1200 mapped construction elements in the Eastern Sector to date (Fig 1c). Yet despite the extensive remains of wooden construction elements, no systematic sampling and no tree-ring based chronological studies via dendrochronology have yet been conducted at the site. The value of developing a precise and accurate calendar- dated chronological sequence using these wooden remains is further enhanced by the fact that the site of Dispilio with more than 1700 complete ceramic vessels (Fig. 2) boasts one of the largest complete Neolithic ceramic assemblages in Europe. Tree-ring dating at Dispilio can therefore be used, via the existing ceramics network, to underpin and improve the relative chronology of the entire region.  
+
+In 2019 a large- scale fieldwork campaign took place at Dispilio's Eastern Sector (Fig. 1d), during which over 900 wooden construction elements (piles) were mapped, of which 787 were sampled for the first dendrochronological analysis. The dendrochronological results provided an oak chronology spanning 120 years, and an overlapping juniper chronology spanning 303 years. This record could not be dated dendrochronologically however, because despite the existence of several millennia- long tree- ring chronologies in the Eastern Mediterranean \(^{11,48,49}\) , none extend back for 7500 years. Here we overcome this limitation by using the combination of dendrochronological and single year radiocarbon analysis, thus providing the first absolute dating of a Neolithic site in the wider Mediterranean region.  
+
+![](images/Figure_2.jpg)
+
+<center>Figure 2, Archaeological finds from Neolithic Dispilio. a: almost completely preserved ornate anthropomorphic vessel from Late Neolithic, many similar ones have been recovered from the site, scale in cm; b: bone spear/harpoon tip with preserved hafting adhesives, scale in cm; c.: an assemblage of Late Neolithic personal adornments (a.,b.,c.,-Dispilio Excavation Archive) </center>  
+
+## Results  
+
+## Dendrochronology  
+
+Of the total wood samples from the archaeological site of Dispilio in 2019 (n=787), 23% were cross- dated into two master tree- ring width (TRW) chronologies. Wood anatomical species determination revealed that the majority of the wooden piles came from oak (Quercus spp., 21%) and juniper (Juniperus spp., 62%) wood. The third most abundant species are pines (Pinus spp., 17%), which were not suitable for dendrochronological cross- dating given the low number of annual rings on most pine samples. The majority of the pine samples could be classified as belonging to the subgenus Pinus (cf. Pinus
+
+<--- Page Split --->
+
+nigra/sylvestris) with several pieces belonging to the subgenus Strobus (cf. Pinus peuce). Due to the wood- anatomical intra- species similarity of junipers \(^{50,51}\) , and of deciduous oaks from the subgenus Quercus \(^{52}\) , a definitive species- level identification was not possible. Based on modern tree species in the region \(^{41,53,54}\) , Dispilio oak wood samples most likely come from Q. frainetto, Q. petraea, and/or Q. pubescens wood, and the junipers are most likely Juniperus excelsa, J. foetedissima, and/or J. deltoides (for the latter cf. J oxycedrus).  
+
+The oak TRW chronology produced was 120- years- long composed of 58 wood samples (Fig 4). It consists of tree- ring sequences with an average segment length of 66 years. Some sapwood was present on most of the oak samples \((n = 45)\) , however the last growth ring (or "waney- edge"), which is important for archaeological interpretation, was conserved on only 4 pieces either as a result of the lower durability of oak sapwood or its intentional removal. The mean inter- series correlation (leave- one- out principle \(^{55}\) ) of the oak tree- ring sequences is 0.51.  
+
+A 303- years- long juniper TRW- chronology was also constructed consisting of 118 tree- ring sequences and an average segment length of 86 years (Fig 4). The mean inter- series correlation (leave- one- out principle \(^{55}\) ) of the juniper chronology is 0.62. Juniper wood, owing to its chemical \(^{56}\) and physical \(^{57}\) properties has a higher resistance to degradation. These qualities made juniper wood the material of choice for construction purposes in many ancient societies in the Eastern Mediterranean \(^{58 - 60}\) . The preservation of juniper wood in Dispilio is also exceptional and the waney- edge on junipers is quite common, enabling an annually resolved reconstruction of the building phases and occupation duration on the site (Fig 4b).  
+
+All samples with a preserved waney edge had a last growth ring terminating with latewood, thus implying a felling date during the dormant period of the trees between late summer and early spring. The juniper and oak tree- ring chronologies have robust dendrochronological dating against each other (t- value = 4.9 \(^{61}\) and = 5.1 \(^{62}\) ; GLK = 63% \(^{63}\) ) over a period of 108 years where sample replication is >4, further supported by \(^{14}\) C wiggle- matching (Supplementary Material S1)  
+
+## Tree-ring \(^{14}\) C cosmogenic signature  
+
+Conventional radiocarbon wiggle- matching models \(^{64,65}\) based on several blocks of 1- 11 tree- rings modelled against the atmospheric data for the Northern Hemisphere (IntCal20 \(^{66}\) ) produced the initial modelled age- ranges for the tree- ring chronologies. Preliminary annual sampling at test positions on the juniper tree- ring chronology indicated that the last ring of this chronology dated between 5233 and 5137 cal BC (at 95% probability). On this basis, a suite of additional single year \(^{14}\) C measurements were made to pinpoint the exact years surrounding the 5259 BC Miyake event. Four wood samples from the juniper chronology were selected covering the part of the chronology where the 5259 BC Miyake event should be located (Fig. 3a). We present here the final 115 \(^{14}\) C measurements (Supplementary Table T1) performed to locate the 5259 BC Miyake event in all 4 wood samples from the Dispilio juniper tree- ring chronology (Fig. 3a). The \(^{14}\) C measurements were performed at the Laboratory for the Analysis of Radiocarbon with AMS at the University of Bern (LARA) \(^{67}\) and the Laboratory of Ion Beam Physics at ETH Zürich (ETH) \(^{68,69}\) . An average year- to- year increase (sensu Miyake et al. \(^{13}\) ) of \(\sim 15.8\%\) in \(\Delta^{14}\) C was detected in all samples in the exact same dendrochronologically cross- dated tree- rings corresponding to the relative ring 184 of the Dispilio juniper chronology. This increase varies from the lowest of \(\sim 11.1\%\) \(\Delta^{14}\) C in DISP- 10070, to \(\sim 13.1\%\) in DISP- 10206, to \(\sim 14.8\%\) in DISP- 10063, to \(\sim 18.6\%\) in DISP- 10611 (Fig. 3a, Supplementary Table T1).  
+
+To compare the \(^{14}\) C results from Dispilio with the published reference data for the 5259 BC event, a mean- value annually resolved reference curve (RC) was established from the dataset in Brehm et al. (2022 –
+
+<--- Page Split --->
+
+henceforth referred to as 'BR22'17). A common approach for verifying the position of Miyake events is wiggle- matching using a goodness- of- fit \(\chi^2\) test15,70,71 against a reference, so that the \(\chi^2\) value becomes minimal for the correct placement of the sample's waney- edge64. The lowest \(\chi^2\) values are reached when the end- dates of the samples are placed at 5240 BC for DISP- 10070 and DISP- 10063 (Fig. 3b), 5153 BC for DISP- 10206, and 5155 BC for DISP- 10611 (Fig. 3c), corresponding to their cross- dated position along the tree- ring chronology. The 5259 BC event signal is clearly identified in all wood samples (Fig. 3a).  
+
+In order to test how close conventional radiocarbon wiggle- matching would be relative to the absolute calendar dating supplied by the Miyake event, the annual data from all the wood samples were wiggle- matched against the IntCal20 calibration curve66 using the \(^{14}\mathrm{C}\) calibration software OxCal 4.4 \(^{64,65}\) . In none of the cases does the \(95\%\) probability end- date range include the actual felling date when IntCal20 is used (Fig. 5, Supplementary Material S4). Longer series of \(^{14}\mathrm{C}\) dates which span some years before and after the event (Fig. 3a, Fig 5), as from wood samples DISP- 10611 and - 10206, yield end- dates which are only \(\sim 15 - 20\) cal years older, while shorter series, wood samples DISP- 10070 and - 10063, result in end- dates over \(\sim 40\) cal years younger than the actual felling dates (Fig. 5). It has been noted previously \(^{72}\) that IntCal20 is poorly replicated during the 53rd- 52nd century BC. Notably, the 53rd century BC is represented by only 16 measurements, of which 14 are decadal and bi- decadal (i.e. blocks of 10- 20 tree- rings), with only two 4- and 5- year blocks \(^{66,73}\) (see Supplementary Material S2.8). The variability in the calibrated end- date ranges suggests that IntCal20 might produce misleading results when wiggle- matching annual data coming from the period in question. The annual \(^{14}\mathrm{C}\) dates were also wiggle- matched against a modified IntCal20 – IntCal20plus – where the default IntCal20 multiple- year blocks of BP (Before Present) data for the 82 years period around the event were substituted with the average of the annual BR22 dataset. Calibrating against this dataset predictably yields the accurate and more precise end- date ranges at \(95\%\) probability for all wood samples (Fig. 5).
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3, Scatter plot of \(\Delta^{14}\mathrm{C}\) data from Dispilio against reference from Brehm et al. (2022) \(^{17}\) , and best last ring fit for the dated wood samples \(\left(\chi^{2}\right)\) . a: Measured \(^{14}\mathrm{C}\) concentrations represented as \(\Delta^{14}\mathrm{C}\) , vertical bars represent 1s uncertainties (Supplementary Table T1); samples marked with "DISP-" refer to measurements on wood samples obtained in this study, other labels represent data from BR22 \(^{17}\) - Bristlecone pine \(^{14}\mathrm{C}\) data are shifted forward by 1 year from the original Brehm et al. publication, following a correction to the dating of the master bristlecone chronology (Supplementary Material S3.2); shaded band represents IntCal20 \(^{66}\) . Panels below, b, c: chi-squared tests of Dispilio measurements against the average from BR22 \(^{17}\) for wood samples DISP-10070 and -10063 (b, \(\chi^{2}\) crit. value=9.49), and DISP-10206 and -10611 (c, \(\chi^{2}\) crit. value=15.51). Figure produced in \(R^{74}\) , code and source data available in Supplementary Material 4. </center>  
+
+The growing season of trees is influenced by many factors and can vary between and among species as a function of cambial age, temperature, water, slope, aspect, soil etc. Personal observations of growth termination in modern oaks and junipers in the region have revealed that latewood can be completed in both genera in the beginning of September (Supplementary Materials S2.6- S2.7). While cell- wall
+
+<--- Page Split --->
+
+thickening in temperate conifers continues for several weeks after the cessation of cell- wall enlargement \(^{75}\) , the amount of cellulose carbon that would be deposited during this last stage of latewood formation constitutes a small percentage of the whole tree- ring \(^{76}\) . Considering the robustness of the \(^{14}\mathrm{C}\) signal in the Displilio junipers tree- rings (Fig. 3) it is unlikely that it only represents the \(^{14}\mathrm{C}\) incorporated at the end of the cell- wall thickening stage. Consequently, it can be stated that the \(^{14}\mathrm{C}\) signal of the 5259 BC event in the in deciduous junipers was incorporated in the same growing season characteristic for deciduous species, i.e. spring to late summer/early autumn 5259 BC.  
+
+According to the dendrochronologically cross- dated position of all wood samples, the ring in which the Miyake event is detected corresponds to relative year number 184 of the 303- year- long juniper TRW chronology. This allows us to set the absolute end- date of the whole Displilio juniper tree- ring chronology at 5140 BC. Furthermore, the identification of the event in DISP- 10070 and - 10063 confirms the correct placement of the better- replicated earlier half of the chronology (Fig. 4a.). Given the dendrochronological cross- dating between the juniper and oak chronologies, also the latter is absolutely dated, placing its last ring at 5311 BC (Fig. 4a.).  
+
+## Site plan and felling phases  
+
+By considering the latest juniper felling dates together with the earliest secure felling dates from the oak chronology it is possible to establish a minimum duration of construction activities of 188 years between 5328 and 5140 BC, with intermittent periods of wood felling/construction, which do not necessarily reflect a continuous, uninterrupted occupation at the same location. Such a chronological resolution for a settlement phase duration on a prehistoric site in the Eastern Mediterranean has not been established to date. Plotting of groups of cross- dated wood samples with felling dates within 1- 2 years of one another using a GIS software revealed blueprints representing different structures (Fig. 4b). Identification of building outlines was possible only for groups that are composed of a substantial number of cross- dated samples. The structures seem to be oriented along the lakeshore. Of particular note is the concentration of building activities in the eastern part of the Eastern Sector. In this part, building activities on the same spot outline an area with a felling date in 5294 BC, and a felling phase which ends in 5257 BC (Fig. 4a, b). A felling phase ending in 5320 BC precedes the group of 5294 BC, however due to the suboptimal preservation of oak samples only two of this group have preserved waney edge. These are complemented by several oak samples dated between 5328 BC and 5320 BC with at least 20 sapwood rings indicating the proximity of the waney edge. The mapping of the dendrochronological results further implies that building practices in some cases either included short term storage (1- 2 years) of timber or consisted of a construction period spread over several years.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4, Bar chart of tree-ring chronologies, felling dates, and site plan development. a: bar plot of Dispilio oak and juniper chronologies; each horizontal bar represents individual wood sample in its dendrochronologically cross-dated position, bar length corresponds to its span in years (i.e., number of tree-rings). Red stars indicate wood samples sampled for annual \(^{14}\mathrm{C}\) ; b.: schematic plan of the East Sector (see also Fig. 1c-d); each symbol represents one vertical wooden element, different shapes and colours correspond to a same felling phase spread over 1-2 years; additionally, colour-shaded polygons outline the groups of same symbols (same felling-phase elements), however they do not represent definite structure plans. </center>
+
+<--- Page Split --->
+
+## Discussion  
+
+According to the archaeo- chronological periodisation in the region, for which there is no universal absolute timeframe<sup>6</sup>, the occupation phases of Dispilio discussed here would fall at the later Middle Neolithic and/or Late Neolithic. The absolute dating and duration of the Middle/Late Neolithic occupation phase in Dispilio is unique in the context of the Balkans, but also in the wider Eastern Mediterranean Neolithic. The site also provides sufficiently replicated dendrochronological information to allow independent controls for settlement duration estimates. The felling dates in the excavated sector indicate activity over a period of at least 188 years, with indications from oak sapwood estimates to extend this backwards by a further 30 years. Of particular interest is the succession of 2 construction phases in the western half of the analysed trench and 3 construction phases in its eastern half (Fig. 4a, b). Although the nature of these structural outlines (Fig. 4b) is not clear at present, a timespan between the construction episodes of 29 years in the western half (5311 and 5282 BC), and 35- 37 years in the eastern half (5320, 5294 and 5257 BC) is consistent with the few available estimates of house lifespans in Neolithic S- E Europe<sup>77,78</sup>. However, determining whether these contemporary structure outlines with same felling dates correspond to one or multiple buildings will require further detailed multidisciplinary work. Intermittent periods without felling dates may simply be a result of preservation or the limited size of the excavated area, but may also reflect a hiatus in occupation or indicate a non- perennial character of the settlement. Detection of annual or decadal- scale hiatuses is extremely difficult in archaeological stratigraphy, with settlement phase duration usually derived from <sup>14</sup>C sequence models based on organic samples from consecutive stratigraphical units. This approach can lead to interpretations of centuries- long settlement continuities<sup>4,79</sup>. Such interpretations may underestimate settlement discontinuities of durations shorter than the associated precision of <sup>14</sup>C measurements and calibration. This underlines the importance of the annually resolved data from Dispilio.
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 5 Wiggle-matching of different sets of annual \(^{14}\mathrm{C}\) data from Dispilio modelled in OxCal v4.4, against IntCal20<sup>66</sup>, and IntCal20plus. IntCal20plus has the non-annual IntCal20 data for a 82-year period around the 5259 BC Miyake event replaced by annual average of Brehm et al. (2022)<sup>17</sup> annual data. Dotted blue lines represent actual felling dates determined through dendrochronology and Miyake event-matching. Acronyms in brackets next to sample name refer to AMS lab that furnished the measurements. Data for figure obtained from OxCal<sup>65,66</sup>. Figure produced in \(R^{74}\) , code and data in Supplementary Material 4. </center>  
+
+The last centuries of the \(6^{\text{th}}\) millennium BC mark an important change within the Neolithic period in the southern Balkans. It is a period of a steep increase in the number and size of settlements, associated with a demographic boom<sup>6,80–82</sup>. Anthropogenic influence on the local environment becomes more pronounced during this period<sup>83,84</sup>, as documented also in Dispilio<sup>39,40</sup>. Diversity increased in all aspects of human behaviour, from pottery production techniques and styles<sup>85</sup>, architecture<sup>81</sup>, settlement organisation<sup>81,86,87</sup> to the first signs of metallurgy<sup>88</sup>. Evidence from this transitional period also points to a shifting social focus from the collective to the domestic<sup>89,90</sup>. In this setting, high- resolution chronological data can improve our understanding of societal changes, human land use, and intensifying influence on the local and regional environment. For instance, the preference of settling in the proximity of wetlands has been documented in the Early Neolithic<sup>3,91</sup>, a practice continuing in subsequent Neolithic subperiods<sup>32,91</sup>. Wetland and shoreline locations would have represented ideal catchment areas for the Neolithic subsistence, providing various soil types that could be exploited for cultivating crops with different requirements, serve as pasture lands, or supply aquatic resources as a dietary complement<sup>91</sup>. A number of wetland sites with similar chronology to Dispilio (2<sup>nd</sup> half of the \(6^{\text{th}}\) millennium BC) have been documented or excavated in existing or former lakes in the region, some of them yielding large amounts of well- preserved wooden construction elements (Fig. 1b,<sup>32–34,92–94</sup>). Although the dating of these sites has much lower chronological resolution than at Dispilio, some of them would have been in use for centuries before and/or after the \(54^{\text{th}}\) - 52nd century BC phases in Dispilio. It is highly likely that it will be possible to cross- date the tree- ring widths of the wood remains from these peripheral sites with the now absolutely dated tree- ring chronologies from
+
+<--- Page Split --->
+
+Dispilio, and thus extend the absolutely dated chronological network for the region well beyond the \(6^{\text{th}}\) millennium BC.  
+
+Beyond the chronological significance, absolutely dated tree- ring records are one of the most utilized proxies for high- resolution climate reconstructions offering unique insights into the relationship between humans and climate. Precipitation is a limiting factor for most low and mid- altitudes trees in the Eastern Mediterranean. In fact, it has been shown that modern juniper \(^{49}\) and oak \(^{11,95}\) tree- ring sequences are good predictors of precipitation in the Eastern Mediterranean. Precipitation was a crucial factor in early agriculture which mainly consisted of rain- fed \(^{96}\) and flood- water \(^{97}\) farming. Preliminary observations of the Dispilio TRW chronologies imply a period of suppressed growth in both the juniper and oak tree- ring sequences for a period of around 20 years between 5360 and 5340 BC. Such suppressed growth period can be associated with decrease in precipitation, which may significantly influence the water table of small water bodies such as Lake Kastoria. A short- term Mid/Late Neolithic eutrophication of the lake previously inferred from increased presence of green algae \(^{35}\) could potentially be correlated with this tree- ring width suppression. Although the Neolithic tree- ring sequences from Dispilio are relatively short if compared to modern tree- ring proxies used in climate reconstructions, they still may provide valuable absolutely dated, annually resolved information on environmental conditions during the Neolithic in Kastoria Basin and the surrounding region.  
+
+Finally, the results from this study underline the value that single year measurements of radiocarbon in tree- rings can have for radiocarbon calibration and dendrochronological dating. Significant advances in AMS technology \(^{68}\) , have made it possible to create long and continuous time- series of annual radiocarbon that are constantly improving the accuracy of the radiocarbon calibration process. More than this though, the utilization of SEP events in anchoring regional timelines through hybrid tree- ring and radiocarbon studies is once again demonstrated. The \(^{14}\text{C}\) - anchored Dispilio tree- ring chronologies now provide a calendar dated reference for dendrochronological dating of other sites from the time period. This provides the opportunity to extend calendar dated chronologies across the region further back into prehistory. Such high- resolution dating, especially in cases where it can be coupled with stratigraphic information or used to derive climatic indicators, will elucidate a more nuanced understanding of deterministic interpretations of the environmental influence on societies in the past (e.g. for the 6.2 ka BC cooling event). This study demonstrates how the discovery of the new SEP events in this time period creates new possibilities in prehistoric archaeology and offers the construction of historical- timescale narratives for societies and their environments from the very distant past.  
+
+## Materials and Methods  
+
+## Wood samples  
+
+The wood material analysed in this study was sampled in August and September 2019 from wooden piles remains at the archaeological site of Dispilio, near Kastoria, Greece (40.485444 N, 21.289694 E; h=627 masl). The site is one of the best- known prehistoric sites in the country and has been investigated, almost continuously, since 1992. Excavations and sampling that took place on the site were performed in full compliance with the regulations of the Greek Ministry of Culture concerning archaeological material. Whole cross- section discs (n=787) were sampled from the wooden remains with handsaws and chainsaws during the 2019 fieldwork campaign. The wood samples documentation, cleaning, preparation, and sealing in plastic bags with water, took place on- site during the 2019 field campaign. Dendrochronological
+
+<--- Page Split --->
+
+measurement took place initially on- site and continued at the University of Bern. Tree- ring width (TRW) measurements were performed according to standard dendrochronological procedures \(^{98,99}\) , by means of a measuring table under a binocular stereo microscope. TRWs were recorded with a precision of 0.01 mm. Two to four radii were measured per sample and averaged together to represent the sample. Descriptive dendrochronological statistics were performed in the dplR package in R \(^{55,74,100}\) . The TRW measurements of DISP- 10611, - 10206, 10070, and - 10063 are available in the Supplementary Material S3.3.  
+
+Wood taxonomy was determined based on stem wood anatomy. Each measured wood sample was sectioned with a razor blade and cell arrangements in the transversal, radial, and tangential sections were identified and compared with references in wood- anatomical atlases \(^{51,52,101}\) . Given the wood anatomical similarity of different deciduous oak species from the subgenus Quercus \(^{52}\) , and considering the high dendrofloristic diversity of oaks in the region \(^{53,54}\) it is not possible to distinguish them to species level. However, it is likely that several deciduous oak species from the subgenus Quercus are represented, notably \(Q\) . frainetto, \(Q\) . petraea, and/or \(Q\) . pubescens. Oak trees from the subgenus Cerris are one of the more abundant groups of oaks in the region, however no wood samples from Dispilio could be assigned to this group which is anatomically characterised by larger and solitary latewood pores. Similarly, wood anatomical differentiation between different juniper species is not possible \(^{50,51,101}\) . Considering todays distribution of tree- like junipers in the region, the most likely species utilized in Dispilio are Juniperus excelsa, \(J\) . foetidissima, and/or \(J\) . deltoides Adams (cf. \(J\) . oxycedrus L.). While majority of the pine samples exhibited denticulate walls on end- tracheids, a characteristic of the pine subgenus Pinus (cf. Pinus nigra/sylvestris- type), several pine wood samples could be identified as members of the Subgenus Strobus (cf. \(P\) . peuce) based on the presence of smooth- walled end- tracheids.  
+
+Local climate in the Kastoria Basin can be defined as continental to sub- Mediterranean, with temperate weather, continental winters, and warm and dry summers. The yearly average precipitation of \(\sim 600 \text{mm}\) increases with altitude, with the wettest months being November and December, while July and August are the driest and hottest months. Yearly average temperature is \(\sim 12.5^{\circ} \text{C}\) . Main climate classes according to the Köppen system \(^{102}\) are Cfa, Cfb, Csa.  
+
+## Sample preparation and radiocarbon measurement  
+
+Individual tree- rings were dissected by hand under a binocular microscope with a one- sided razor blade (Supplementary Material S2.5). Whole rings were used for all \(^{14} \text{C}\) measurements (Supplementary Table S2.1). About 30- 70 mg of material were sampled per ring, depending on its width. Earlywood comprises ca. 80- 90% of a juniper tree- ring. Since most of the of the ring- structure of junipers growing on mesic sites is completed by the end of September \(^{103}\) (see also Supplementary Material S2.6- S2.7), the tree- ring structural carbon concentration should reflect temperate spring- to- late summer carbon uptake.  
+
+Wiggle- matching of several \(^{14} \text{C}\) dates provided the initial estimate of the 40- rings segment of the tree- ring chronology where the event will be located. A "buffer zone" of 15 rings at each limit was added to the estimate, and 70 individual rings were sampled centred around the estimated "event ring" from the first wood sample that was analysed (DISP- 10206, Supplementary Material S2.1). The \(^{14} \text{C}\) content of every \(4^{\text{th}}\) sampled ring was subsequently measured until the \(^{14} \text{C}\) spike was located, after which the \(^{14} \text{C}\) in 20 consecutive annual rings around the event was measured. The "event ring" on all the other wood samples (DISP- 10611, - 10070, - 10063, Supplementary Material S2.2- S2.4) was identified according to the samples' cross- dating position along the tree- ring chronology.
+
+<--- Page Split --->
+
+Cellulose from wood samples analysed at the Laboratory for the Analysis of Radiocarbon with AMS at the University of Bern (LARA)67 was extracted following the BABAB method104 including the modifications of Sookdeo et al. (2020) at \(70^{\circ}C\) for all steps. Samples were submerged in a 1M NaOH overnight and treated in 1M HCl followed by 1M NaOH in a shaker for one hour each. Bleaching of the samples was performed on addition of 5 mL water, a few drops of 1M HCl to reach pH 2- 3 and 100 mg NaClO2 by shaking for at least two hours or until the colour of the wood samples turned white. Drying of the material was achieved by lyophilisation overnight. Samples were measured using the LARA MICADAS AMS system. DISP- 10070, - 10206 and a first run of - 10611 was analysed together with three oxalic acid II (SRM 4990C, NIST) standards and three chemical blanks. Later, a second run of DISP- 10611 and - 10063 was dated together with five oxalic acid II standards and four chemical blanks that were used for blank subtraction, standard normalization, and correction for isotope fractionations as well as two IAEA- C5, two IAEA- C7, two 1515 CE reference samples and two cellulose blanks as secondary standards and blanks, respectively. For details, see Supplementary Material S3.1 and Supplementary Table T1.  
+
+For the analyses performed at ETHZ, the tree- ring samples were prepared in 15 ml glass test tubes together with four wood blanks (2 BC and 2 KB) and 2 1515 CE reference samples each weighing \(30 - 60mg^{68}\) . In a slightly modified procedure following104, samples were first soaked in 5 ml 1M NaOH overnight at \(70^{\circ}C\) in an oven. Then the samples were treated with 1M HCl and 1M NaOH for 1 hour each at \(70^{\circ}C\) in a heat block, before they were bleached at a pH of 2- 3 with 0.35M NaClO2 at \(70^{\circ}C\) for 2 h. The remaining white holocellulose was then freeze- dried overnight. About 2.5 mg dried holo- cellulose was wrapped in cleaned Al capsules and converted to graphite using the automated graphitization line AGE- 3. A measurement set was made up of the tree- ring samples, three oxalic acid one (OX1) and four oxalic acid two (OX2) standards, two cellulose blanks, two chemical blanks, and two 1515 CE reference samples and measured in the MICADAS accelerator mass spectrometer.  
+
+## Radiocarbon matching and modelling  
+
+The new \(^{14}\mathrm{C}\) measurements presented in this study were matched to the constructed reference curve \(^{17}\) (see also Supplementary Material S4) using a common \(\chi^2\) test approach so that the \(\chi^2\) value becomes minimal for the correct placement of the sample's waney- edge \(^{15,64,71}\) :  
+
+\[\chi^2 (x) = \sum_{i = 1}^{n}\frac{(R_i - C_{(x - r_i)})^2}{\delta R_i^2 + \delta C_{(x - r_i)}^2}\]  
+
+Where \(R_{i}\pm \delta R_{i}\) represent the new \(^{14}\mathrm{C}\) measurements, and \(C_{(x - r_i)}\pm \delta C_{(x - r_i)}\) represent the reference curve \(^{14}\mathrm{C}\) concentrations in the year \((x - r_i)\) ; \(r_i\) stands for the tree ring number starting with 0, representing the last growth ring of the tree (waney- edge).  
+
+The Bayesian wiggle- matching was performed in the software OxCal 4.4 with the inbuilt D_Sequence command against the atmospheric data from IntCal20 \(^{65,66}\) , for the CQL code see Supplementary Material S1 and S4.  
+
+The year- to- year increase in \(\Delta^{14}\mathrm{C}\) presented in the Results section was calculated as a difference between the values in 5260 BC and 5259 BC (sensu Miyake et al. \(^{13}\) ). For a detailed discussion on the magnitude and \(^{14}\mathrm{C}\) production during the 5259 BC Miyake event see \(^{17}\) , and \(^{105}\) .
+
+<--- Page Split --->
+
+## Data uncertainty  
+
+Data uncertaintyThe genus Juniperus is known to produce intra- annual density fluctuation ('false rings') or have 'missing rings' \(^{106}\) in parts of the stem. Missing rings are very often a product of the stem growth habit of junipers, so- called 'lobate growth', which consists of higher cambial activity and faster growth in certain areas of the stem, resulting in an undulating cross- section of the stem in older trees, where the less active areas may not produce rings in certain years. However, missing rings or measuring false rings can be accounted for when sufficient numbers of wood samples with complete stem cross- sections are available, as in Dispilio. The correct location of the "event ring" on all wood samples based on their cross- dated position is further supporting a correct ring count. Moreover, the dendrochronological cross- dating of the first half of the juniper chronology against the oak chronology serves as an additional control for the correct ring count, considering that oak trees almost never have missing rings \(^{107}\) .  
+
+## Data availability  
+
+Data availabilitySupplementary Material, including code, text, figures, and datasets referred to and presented in this paper are available at the following repository: 10.5281/zenodo.8407222.  
+
+## Acknowledgments  
+
+AcknowledgmentsThe 2019 fieldwork and the subsequent dendrochronological and radiocarbon analyses were conducted in the framework of the ERC project 'Exploring the dynamics and causes of prehistoric land use change in the cradle of European farming' (EXPLO). This project is financially supported by the European Union's Horizon 2020 research and innovation programme, under the grant agreement No 810586 (project EXPLO, explo project.eu).  
+
+We would like to thank all the Archaeology students involved in the fieldwork and sample curation from the Universities of Thessaloniki and Bern, and the staff of the Ephorate of Antiquities of Kastoria.  
+
+## Contributions  
+
+ContributionsA.M., together with C.P., A.H. conceived and designed the study. K.K. & T.G. led the fieldwork, while A.M. and J.F. participated in part of it. J.F. & A.M., together with M.B., performed the dendrochronological and wood- anatomical analyses. A.M. sampled individual tree- rings. S.S. and L.W. performed and provided the 14C measurements. A.M., & C.P., drafted the manuscript, and all authors edited and contributed to the manuscript. A.H. and K.K. obtained funding.  
+
+## References  
+
+1. Grosman, L. The Natufian Chronological Scheme – New Insights and their Implications. in Natufian Foragers in the Levant: Terminal Pleistocene Social Changes in Western Asia (eds. Bar-Yosef, O. & Valla, F. R.) 622–637 (Berghahn Books, 2013). doi:10.2307/j.ctv8bt33h.41.  
+
+2. Maniatis, Y. Radiocarbon dating of the major cultural phases in prehistoric Macedonia: Recent developments (In Greek, English summary). in A Century of Research in Prehistoric Macedonia 1912-2012, International Conference Proceedings, Archaeological Museum of Thessaloniki, 22-24 November 2012 (eds. Stefani, E., Merousis, N. & Dimoula, A.) 205–222 (Archaeological Museum of Thessaloniki, 2014).
+
+<--- Page Split --->
+
+450 3. Karamitrou-Mentessidi, G., Efstratidou, N., Kaczanowska, M. & Kozłowski, J. Early Neolithic settlement of Mavropigi in western Greek Macedonia. Eurasian Prehistory 12, 47–115 (2015).  
+
+452 4. Maniatis, Y. & Adaktylou, F. Revenia-Korinos: one of the earliest Neolithic settlements in North Greece as evidenced by radiocarbon dating. Radiocarbon 63, 1025–1051 (2021).  
+
+454 5. Horejs, B. et al. The Aegean in the Early 7th Millennium BC: Maritime Networks and Colonization. J. World Prehistory 28, 289–330 (2015).  
+
+456 6. Tsirtsoni, Z. Chapter 1. The chronological framework in Greece and Bulgaria between the late 6th and the early 3rd millennium BC, and the "Balkans 4000" project. in The Human Face of Radiocarbon (ed. Tsirtsoni, Z.) 13–39 (MOM Éditions, 2016). doi:10.4000/books.momeditions.503.  
+
+460 7. Schweingruber, F. H. Tree Rings: Basics and Applications of Dendrochronology. (Kluwer Academic Publishers, 1988). doi:10.1007/978- 94- 009- 1273- 1.  
+
+462 8. Friedrich, M. et al. The 12,460-Year Hohenheim Oak and Pine Tree-Ring Chronology from Central Europe—A Unique Annual Record for Radiocarbon Calibration and Paleoenvironment Reconstruction. Radiocarbon 46, 1111–1122 (2004).  
+
+465 9. Salzer, M. W. & Hughes, M. K. Bristlecone pine tree rings and volcanic eruptions over the last 5000 yr. Quat. Res. 67, 57–68 (2007).  
+
+467 10. Nicolussi, K. et al. A 9111 year long conifer tree-ring chronology for the European Alps: a base for environmental and climatic investigations. The Holocene 19, 909–920 (2009).  
+
+469 11. Griggs, C., Degaetano, A. & Newton, M. A regional high-frequency reconstruction of May – June precipitation in the north Aegean from oak tree rings, A. D. 1089 – 1989. Int. J. Climatol. 27, 1075–1089 (2007).  
+
+472 12. Nicolussi, K., Matuschik, I. & Tegel, W. Klimavariabilität und Siedlungsdynamik am Beispiel der Feuchtbodensiedlungen im Raum Oberschwaben, Bodensee und Nordostschweiz 4400- 3400 BC. in Dendro – Chronologie – Typologie – Ökologie. Festschrift für André Billamboz zum 65. Geburtstag (eds. Bleicher, N., Schlichtherle, H., Gassman, P. & Martinelli, N.) 61–77 (2013).  
+
+476 13. Miyake, F., Nagaya, K., Masuda, K. & Nakamura, T. A signature of cosmic-ray increase in 774-775 from tree rings in Japan. Nature 486, 240–242 (2012).  
+
+478 14. Miyake, F., Masuda, K. & Nakamura, T. Another rapid event in the carbon-14 content of tree rings. Nat. Commun. 4, 1748 (2013).  
+
+480 15. Wacker, L. et al. Radiocarbon Dating to a Single Year by Means of Rapid Atmospheric 14C Changes. Radiocarbon 56, 573–579 (2014).  
+
+482 16. Dee, M. W. & Pope, B. J. S. Anchoring historical sequences using a new source of astrochronological tie-points. Proc. R. Soc. A Math. Phys. Eng. Sci. 472, 20160263 (2016).  
+
+484 17. Brehm, N. et al. Tree-rings reveal two strong solar proton events in 7176 and 5259 BCE. Nat. Commun. 13, 1–8 (2022).  
+
+486 18. Pearson, C. et al. Annual Variation in Atmospheric C between 1700 BC and 1480 BC. Radiocarbon 62, 939–952 (2020).  
+
+488 19. Büntgen, U. et al. Tree rings reveal globally coherent signature of cosmogenic radiocarbon events
+
+<--- Page Split --->
+
+489 in 774 and 993 CE. Nat. Commun. 9, 3605 (2018).
+490 20. Park, J., Southon, J., Fahrni, S., Creasman, P. P. & Mewaldt, R. Relationship between solar activity and Δ14C peaks in AD 775, AD 994, and 660 BC. Radiocarbon 59, 1147–1156 (2017).
+491 21. Cliver, E. W., Schrijver, C. J., Shibata, K. & Usoskin, I. G. Extreme solar events. Living Rev. Sol. Phys. 19, 2 (2022).
+492 22. Usoskin, I. & Miyake, F. Introduction. in Extreme Solar Particle Storms - The hostile Sun (eds. Miyake, F., Usoskin, I. & Poluianov, S.) 1–3 (IOP Publishing, 2019). doi:10.1088/2514- 3433/ab404ach1.
+493 23. Uusitalo, J. et al. Solar superstorm of AD 774 recorded subannually by Arctic tree rings. Nat. Commun. 9, 3495 (2018).
+494 24. Mekhaldi, F. et al. Multiradionuclide evidence for the solar origin of the cosmic-ray events of AD 774/5 and 993/4. Nat. Commun. 6, 8611 (2015).
+495 25. Sakurai, H. et al. Prolonged production of 14C during the “660 BCE solar proton event from Japanese tree rings. Sci. Rep. 10, 660 (2020).
+496 26. Fouache, E. et al. Palaeogeographical reconstructions of Lake Maliq (Korça Basin, Albania) between 14,000 BP and 2000 BP. J. Archaeol. Sci. 37, 525–535 (2010).
+497 27. Tegel, W., Muigg, B., Skiadaresis, G., Vanmoerkerke, J. & Seim, A. Dendroarchaeology in Europe. Front. Ecol. Evol. 10, 1–31 (2022).
+498 28. Mazzucco, N. et al. Multiproxy study of 7500-year-old wooden sickles from the Lakeshore Village of La Marmotta, Italy. Sci. Rep. 12, 14976 (2022).
+499 29. Bleicher, N. et al. Middens, currents and shorelines: Complex depositional processes of waterlogged prehistoric lakeside settlements based on the example of Zurich-Parkhaus Opéra (Switzerland). J. Archaeol. Sci. 97, 26–41 (2018).
+500 30. Hafner, A. et al. First Absolute Chronologies of Neolithic and Bronze Age Settlements at Lake Ohrid based on Dendrochronology and Radiocarbon Dating. J. Archaeol. Sci. Reports 38, 1–30 (2021).
+501 31. Reich, J. et al. A novel Structure from Motion-based approach to underwater pile field documentation. J. Archaeol. Sci. Reports 39, 103120 (2021).
+502 32. Chrysostomou, P., Jagoulis, T. & Mäder, A. The ‘Culture of Four Lakes’: Prehistoric lakeside settlements (6th - 2nd mill. BC) in the Aminedon Basin, Western Macedonia, Greece. Archäologie Schweiz 38, 24–32 (2015).
+503 33. Westphal, T., Tegel, W., Heussner, K. U., Lera, P. & Rittershofer, K.-F. Erste dendrochronologische Datierungen historischer Hölzer in Albanien. Archaeol. Anz. (2010) doi:https://publications.dainst.org/journals/aa/82/4764.
+504 34. Naumov, G. Prähistorische Pfahlbauten im Ohrid-See, Republik Mazedonien. Plattform. Jahrb. des Vereins für Pfahlbau- und Heimatkd. e.V. 23/24, 201, 10–19 (2016).
+505 35. Karkanas, P. et al. Palaeoenvironments and site formation processes at the Neolithic lakeside settlement of Dispilio, Kastoria, Northern Greece. Geoarchaeology 26, 83–117 (2011).
+506 36. Δισηπλό. 7500 χρόνια μετά /Dispilio. 7500 years after (in Greek). (University Studio Press, 2002).
+
+<--- Page Split --->
+
+37. Bolliger, M. et al. Dendroarchaeology at Lake Ohrid: 5th and 2nd millennia BCE tree-ring 29 chronologies from the waterlogged site of Ploča Mičov Grad, North Macedonia. 530 Dendrochronologia 79, 126095 (2023). 31 38. Maczkowski, A. et al. The Early Bronze Age dendrochronology of Sovjan (Albania): A first tree-ring 32 sequence of the 24th – 22nd c. BC for the southwestern Balkans. Dendrochronologia 66, 125811 33 (2021). 34 39. Kouli, K. & Dermitzakis, M. D. 11. Lake Orestiás (Kastoria, northern Greece). Grana 49, 154–156 35 (2010). 36 40. Kouli, K. Plant landscape and land use at the Neolithic lake settlement of Dispilió (Macedonia, 37 northern Greece). Plant Biosyst. - An Int. J. Deal. with all Asp. Plant Biol. 149, 195–204 (2015). 38 41. Ntinou, M. La paleovegetación en el norte de Grecia desde el Tardigliacar hasta el Atlántico : 39 formaciones vegetales, recursos y usos. (2002). 40 42. Ntinou, M. Палаюперибаллов каи авбрцвихес бразтнриотчес: Н авбраколойа сто лмвайо 41 Neolibikо окизмо сто Диспнли Касторцас / Palaeoenvironment and human activities: Wood 42 charcoal analysis at the neolithic lake-settlement at Dispilio, Kastoria (in Greek). Anaskamma 04, 43 45–60 (2010). 43. Chatzitoulousis, S. Н техволойа тоц ёулоо сто веолбико лмваю окизмо тоц Диспнлои 45 Касторцас / Woodworking technology at the Neolithic lakeside settlement of Dispilio, Kastoria (in 46 Greek, English summary). Anaskamma 10, 93–123 (2009). 44. Sofronidou, M. & Dimitriadis, S. Continuity and discontinuity in the pottery of the early phases of 48 neolithic Dispilio (In Greek, English summary). in A Century of Research in Prehistoric Macedonia 49 1912-2012, International Conference Proceedings, Archaeological Museum of Thessaloniki, 22-24 50 November 2012 (eds. Stefani, E., Merousis, N. & Dimoula, A.) 537–548 (Archaeological Museum of 51 Thessaloniki, 2014). 45. Voulgari, E. The anthropomorphism of human-like pots: in Bodies of Clay (eds. Schwarzberg, H. & 53 Becker, V.) 23–43 (Oxbow Books, 2017). doi:10.2307/j.ctt1vgw6th.6. 46. Facorellis, Y., Sofronidou, M. & Hourmouziadis, G. Radiocarbon Dating of the Neolithic Lakeside 55 Settlement of Dispilio, Kastoria, Northern Greece. Radiocarbon 56, 511–528 (2014). 47. Maczkowski, A., Bolliger, M. & Francuz, J. Wetland dendrochronology. An overview of the 57 prehistoric chronologies from the southwestern Balkans. in Prehistoric Wetland Sites of Southern 58 Europe. Archaeology, Chronology, Palaeoecology and Bioarchaeology (eds. Ballmer, A., Hafner, A. 59 & Willy, T.) (Cham: Springer (in press), 2023). 48. Pearson, C. et al. Securing timelines in the ancient Mediterranean using multiproxy annual tree- 61 ring data. Proc. Natl. Acad. Sci. 201917445 (2020) doi:10.1073/pnas.1917445117. 49. Touchan, R., Akkemik, Ü., Hughes, M. K. & Erkan, N. May–June precipitation reconstruction of 63 southwestern anatolia, Turkey during the last 900 years from tree rings. Quat. Res. 68, 196–202 64 (2007). 50. Schweingruber, F. H. Trees and wood in dendrochronology : morphological, anatomical, and tree- 66 ring analytical characteristics of trees frequently used in dendrochronology. (Springer-Verlag, 67 1993). 51. Akkemik, Ü. & Yaman, B. Wood Anatomy of Eastern Mediterranean Species. (Kessel Publishing
+
+<--- Page Split --->
+
+569 House, 2012). 570 52. Schweingruber, F. H. Mikroskopische Holzanatomie. Formenspektren mitteleuropäischer Stamm- 571 und Zweighölzer zur Bestimmung von rezentem und subfossilem Material. Anatomie 572 microscopique du bois. Microscopic wood anatomy. (Eidgenössische Forschungsanstalt für Wald, 573 Schnee und Landschaft, 1990). 574 53. Matevski, V. et al. Forest vegetation of the Galičica mountain range in Macedonia = Шумската 575 вегетација на планината Галичица во Македонија = Gozdna vegetacija gorovja Galičica v 576 Makedoniji. (Biološki institut Jovana Hadžija ZRC SAZU; Makedonska akademija na naukite i 577 umetnostite;, 2011). doi:https://doi.org/10.3986/9789610502906. 578 54. Dida, M. State of Forest Tree Genetic Resources in Albania. 579 http://www.fao.org/3/j2108e/j2108e00.htm#TopOfPage (2003). 580 55. Bunn, A. et al. dplR: Dendrochronology Program Library in R; version 1.7.1. https://CRAN.R- 581 project.org/package=dplR. (2020). 582 56. Adamopoulos, S. & Koch, G. Wood structure and topochemistry of Juniperus Excelsa. IAWA J. 32, 583 67–76 (2011). 584 57. Hänninen, T. et al. Ultrastructural evaluation of compression wood-like properties of common 585 juniper (Juniperus communis L.). Holzforschung 66, 389–395 (2012). 586 58. Kuniholm, P. I., Newton, M. W. & Liebhart, R. F. Dendrochronology at Gordion. in The New 587 Chronology of Iron Age Gordion (eds. Rose, C. B. & Darbyshire, G.) 79–122 (University of 588 Pennsylvania Press, 2011). doi:https://doi.org/10.9783/9781934536551. 589 59. Manning, S. W. et al. Integrated tree-ring-radiocarbon high-resolution timeframe to resolve 590 earlier second millennium bce mesopotamian chronology. PLoS One 11, 1–27 (2016). 591 60. Abdrabou, A., Zidan, E., Nishisaka, A., Kurokochi, H. & Yoshimura, S. King Khufu’s Second Boat: 592 Scientific Identification of Wood Species for Deckhouse, Canopy, and Forecastle. Forests 13, 2118 593 (2022). 594 61. Hollstein, E. Mitteleuropäische Eichenchronologie : Trierer dendrochronologische Forschungen zur 595 Archäologie und Kunstgeschichte. (Mainz am Rhein : von Zabern, 1980). 596 62. Baillie, M. & Pilcher, J. A Simple Crossdating Program for Tree-Ring Research. Tree-ring Bull. 7–14 597 (1973). 598 63. Eckstein, D. & Bauch, J. Beitrag zur Rationalisierung eines dendrochronologischen Verfahrens und 599 zur Analyse seiner Aussagesicherheit. Forstwissenschaftliches Cent. 88, 230–250 (1969). 600 64. Bronk Ramsey, C., van der Plicht, J. & Weninger, B. ‘Wiggle Matching’ Radiocarbon Dates. 601 Radiocarbon 43, 381–389 (2001). 602 65. Ramsey, C. B. Bayesian analysis of radiocarbon dates. Radiocarbon 51, 337–360 (2009). 603 66. Reimer, P. J. et al. The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal 604 kBP). Radiocarbon 62, 725–757 (2020). 605 67. Szidat, S. et al. 14 C Analysis and Sample Preparation at the New Bern Laboratory for the Analysis 606 of Radiocarbon with AMS (LARA) . Radiocarbon 56, 561–566 (2014). 607 68. Sookdeo, A. et al. Quality Dating: A Well-Defined Protocol Implemented at ETH for High-Precision
+
+<--- Page Split --->
+
+608 14C-Dates Tested on Late Glacial Wood. Radiocarbon 62, 891- 899 (2020). 609 69. Wacker, L., Bollhalder, S., Sookdeo, A. & Synal, H.- A. Re- evaluation of the New Oxalic Acid 610 standard with AMS. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. 611 Atoms 455, 178- 180 (2019). 612 70. Kuitems, M. et al. Radiocarbon- based approach capable of subannual precision resolves the 613 origins of the site of Por- Bajin. 0, 8- 11 (2020). 614 71. Kuitems, M. et al. Evidence for European presence in the Americas in ad 1021. 601, (2022). 615 72. Bayliss, A. et al. IntCal20 Tree Rings: An Archaeological Swot Analysis. Radiocarbon 62, 1045- 1078 616 (2020). 617 73. Blaauw, M. IntCal: Radiocarbon Calibration Curves. R package version 0.3.1. https://CRAN.R- 618 project.org/package=IntCal. (2022). 619 74. R Core Team. R: A Language and Environment for Statistical Computing. https://www.r- 620 project.org/. https://www.r- project.org/ (2020). 621 75. Rathgeber, C. B. K., Cuny, H. E. & Fonti, P. Biological basis of tree- ring formation: A crash course. 622 Front. Plant Sci. 7, 1- 7 (2016). 623 76. Cuny, H. E. & Rathgeber, C. B. K. Xylogenesis: Coniferous Trees of Temperate Forests Are Listening 624 to the Climate Tale during the Growing Season But Only Remember the Last Words! Plant Physiol. 625 171, 306- 317 (2016). 626 77. Tasić, N. et al. Interwoven Strands for Refining the Chronology of the Neolithic Tell of Vinča- Belo 627 Brdo, Serbia. Radiocarbon 58, 795- 831 (2016). 628 78. Bayliss, A. & Whittle, A. What kind of history in prehistory? in Time and History in Prehistory (eds. 629 Souvatzi, S., Baysal, A. & Baysal, E.) 123- 146 (Routledge, 2018). doi:10.4324/9781315531854- 8. 630 79. Stratouli, G. et al. New excavations in Northwestern Greece: The Neolithic settlement of Avgi, 631 Kastoria. J. Greek Archaeol. 5, 63- 134 (2020). 632 80. Weiberg, E. et al. Long- term trends of land use and demography in Greece: A comparative study. 633 The Holocene 29, 742- 760 (2019). 634 81. Kotsos, S. Settlements and Landscape in the Neolithic of the Southern and Central Balkans. in 635 Making spaces into places : the north Aegean, the Balkans and western Anatolia in the Neolithic 636 (eds. Tasić, N. N., Urem- Kotsou, D. & Burić, M.) 105- 123 (2020). 637 82. Elezi, G. Sociocultural dimensions of production, use, and circulation of Late Neolithic pottery from 638 southern Balkans. (University of California, PhD Thesis, 2020). 639 83. Lespez, L. et al. The lowest levels at Dikili Tash, northern Greece: A missing link in the Early 640 Neolithic of Europe. Antiquity 87, 30- 45 (2013). 641 84. Gassner, S. et al. 20,000 years of interactions between climate, vegetation and land use in 642 Northern Greece. Veg. Hist. Archaeobot. 29, 75- 90 (2020). 643 85. Bonga, L. Late Neolithic pottery from mainland Greece, ca. 5,300- 4,300 B.C. (Temple University, 644 2013). 645 86. Halstead, P. Farming, material culture, and ideology: repackaging the Neolithic of Greece (and
+
+<--- Page Split --->
+
+646 Europe). Dyn. Neolit. Eur. Stud. honour Andrew Sherratt 131- 151 (2011). 647 87. Kotsakis, K. What Tells Can Tell: Social Space and Settlement in the Greek Neolithic. in Neolithic 648 Society in Greece (ed. Halstead, P.) 66- 76 (1999). 649 88. Malamidou, D., Tsirtsoni, Z. & Vaxevanopoulos, M. The emergence of metal use in Greek Eastern 650 Macedonia during the Neolithic period (late 6th- 5th millennia BC). Doc. Praehist. 49, 2- 21 (2022). 651 89. Kotsakis, K. Domesticating the periphery. New research into the Neolithic of Greece. Pharos 20, 652 41- 73 (2014). 653 90. Pappa, M. et al. The Neolithic site of Makriyalos, northern Greece: reconstruction of social and 654 economic structure of the settlement through comparative study of the finds. in Diet, Economy 655 and Society in the Ancient Greek World - Towards a Better Integration of Archaeology and Science. 656 Proceedings of the International Conference held at the Netherlands Institute at Athens on 22- 24 657 March 2010 (eds. Voutsaki, S. & Valamoti, S. M.) 77- 88 (Peeters, 2013). 658 91. Gkouma, M. & Karkanas, P. The physical environment in Northern Greece at the advent of the 659 Neolithic. Quat. Int. 496, 14- 23 (2018). 660 92. Giagkoulis, T. The Pile- field of the Neolithic Lakeside Settlement Anarghiri IXb (Aminedon, 661 Western Macedonia, Greece) and the Non- Residential Wooden Structures on the Periphery of the 662 Habitation. in Settling Waterscapes in Europe. The Archaeology of Neolithic and Bronze Age Pile- 663 Dwellings (eds. Hafner, A., Dolbunova, E., Mazurkevich, A., Pranckenaite, E. & Hinz, M.) 137- 155 664 (Propylaeum, 2020). doi:https://doi.org/10.11588/propylaeum.714. 665 93. Korkuti, M. Neolithikum und Chalkolithikum in Albanien = Neolithique et chalcolithique en Albania. 666 (Phillip von Zabern Verlag, 1995). 667 94. Gori, M. Along the Rivers and Through the Mountains: A revised chrono- cultural framework for 668 the south- western Balkans during the late 3rd and early 2nd millennium BCE. 669 Universitätsforschungen Zur Prähistorischen Archäologie, Band 268. (Dr. Rudolf Habelt GmbH, 670 2015). 671 95. Hughes, M. K. et al. Aegean Tree- Ring Signature Years Explained. Tree- Ring Res. 57, 67- 73 (2001). 672 96. Halstead, P. & Isaakidou, V. Pioneer farming in earlier Neolithic Greece. in Farmers at the Frontier 673 (eds. Gron, K. J., Sørensen, L. & Rowley- Conwy, P.) 77- 100 (Oxbow Books, 2020). 674 doi:10.2307/j.ctv13gvh1g.7. 675 97. Van Andel, T. H. & Runnels, C. N. The earliest farmers in Europe. Antiquity 69, 481- 500 (1995). 676 98. Speer, J. H. Fundamentals of Tree Ring Research. (University of Arizona Press, 2010). 677 99. Baillie, M. A Slice Through Time. Dendrochronology and Precision Dating. (Routledge, 1995). 678 100. Bunn, A. G. A dendrochronology program library in R (dplR). Dendrochronologia 26, 115- 124 679 (2008). 680 101. Crivellaro, A. & Schweingruber, F. H. Atlas of Wood, Bark and Pith Anatomy of Eastern 681 Mediterranean Trees and Shrubs : with a Special Focus on Cyprus. (Springer- Verlag Berlin 682 Heidelberg, 2013). doi:10.1007/978- 3- 642- 37235- 3. 683 102. Beck, H. E. et al. Present and future Köppen- Geiger climate classification maps at 1- km resolution. 684 Sci. Data 5, 180214 (2018).
+
+<--- Page Split --->
+
+685 Camarero, J. J., Olano, J. M. & Parras, A. Plastic bimodal xylogenesis in conifers from continental Mediterranean climates. New Phytol. 185, 471–480 (2010).  
+
+687 Némec, M., Wacker, L., Hajdas, I. & Gägeler, H. Alternative methods for cellulose preparation for ams measurement. Radiocarbon 52, 1358–1370 (2010).  
+
+689 Zhang, Q. et al. Modelling cosmic radiation events in the tree-ring radiocarbon record Subject Areas: Author for correspondence: (2022).  
+
+691 Esper, J. Long-term tree-ring variations in Juniperus at the upper timber-line in the Karakorum (Pakistan). Holocene 10, 253–260 (2000).  
+
+693 Haneca, K., Katarina Čufar & Beeckman, H. Oaks, tree- rings and wooden cultural heritage: a review of the main characteristics and applications of oak dendrochronology in Europe. J. Archaeol. Sci. 36, 1–11 (2009).
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- SupTabT1Dispilio5259.xlsx- SupMatS1S3Dispilio5259.pdf- SupMatS4Dispilio5259.rar- CompetingintereststatementAM.pdf- EditorialpolicychecklistAM.pdf- RreportingsummaryAM.pdf
+
+<--- Page Split --->

preprint/preprint__9828c81619ba2f2920e2f04445ff9b9dc5c80df021f1f0112ee62d50f62d662e/preprint__9828c81619ba2f2920e2f04445ff9b9dc5c80df021f1f0112ee62d50f62d662e_det.mmd

@@ -0,0 +1,404 @@
+<|ref|>title<|/ref|><|det|>[[42, 107, 928, 175]]<|/det|>
+# Absolutely dating the European Neolithic through a rapid 14C excursion  
+
+<|ref|>text<|/ref|><|det|>[[44, 195, 200, 213]]<|/det|>
+Andrej Maczkowski  
+
+<|ref|>text<|/ref|><|det|>[[55, 222, 345, 240]]<|/det|>
+andrej.maczkowski@unibe.ch  
+
+<|ref|>text<|/ref|><|det|>[[50, 269, 570, 289]]<|/det|>
+University of Bern https://orcid.org/0000- 0003- 3081- 3769  
+
+<|ref|>text<|/ref|><|det|>[[44, 294, 238, 333]]<|/det|>
+Charlotte Pearson University of Arizona  
+
+<|ref|>text<|/ref|><|det|>[[44, 340, 214, 379]]<|/det|>
+John Francuz University of Bern  
+
+<|ref|>text<|/ref|><|det|>[[44, 385, 280, 425]]<|/det|>
+Tryfon Giagkoulis University of Thessaloniki  
+
+<|ref|>text<|/ref|><|det|>[[44, 431, 171, 449]]<|/det|>
+Sonke Szidat  
+
+<|ref|>text<|/ref|><|det|>[[50, 454, 825, 495]]<|/det|>
+Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern https://orcid.org/0000- 0002- 1824- 6207  
+
+<|ref|>text<|/ref|><|det|>[[44, 501, 171, 519]]<|/det|>
+Lukas Wacker  
+
+<|ref|>text<|/ref|><|det|>[[50, 523, 797, 543]]<|/det|>
+Swiss Federal Institute of Technology (ETH) https://orcid.org/0000- 0002- 8215- 2678  
+
+<|ref|>text<|/ref|><|det|>[[44, 548, 210, 586]]<|/det|>
+Matthias Bolliger University of Bern  
+
+<|ref|>text<|/ref|><|det|>[[44, 592, 280, 632]]<|/det|>
+Kostas Kotsakis University of Thessaloniki  
+
+<|ref|>text<|/ref|><|det|>[[44, 638, 210, 678]]<|/det|>
+Albert Hafner University of Bern  
+
+<|ref|>text<|/ref|><|det|>[[44, 722, 103, 740]]<|/det|>
+Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 760, 137, 778]]<|/det|>
+Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 797, 329, 816]]<|/det|>
+Posted Date: October 20th, 2023  
+
+<|ref|>text<|/ref|><|det|>[[44, 835, 475, 855]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 3419721/v1  
+
+<|ref|>text<|/ref|><|det|>[[42, 873, 911, 914]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 933, 534, 953]]<|/det|>
+Additional Declarations: There is NO Competing Interest.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 77, 911, 120]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on May 20th, 2024. See the published version at https://doi.org/10.1038/s41467-024-48402-1.
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[66, 102, 692, 123]]<|/det|>
+# Absolutely dating the European Neolithic through a rapid \(^{14}\mathrm{C}\) excursion  
+
+<|ref|>text<|/ref|><|det|>[[66, 130, 901, 660]]<|/det|>
+2 3 Andrej Maczkowski 1, \(2^{*}\) 4 Charlotte Pearson 3 5 John Francuz 1 6 Tryfon Giagkoulis 4 7 Sonke Szidat 5, 2 8 Lukas Wacker 7 9 Matthias Bolliger 1, 2, 6 10 Kostas Kotsakis 4 11 Albert Hafner 1, 2 12 13 Affiliations 14 1 Institute of Archaeological Sciences, University of Bern, Switzerland 15 2 Oeschger Centre for Climate Change Research, University of Bern, Switzerland 16 3 Laboratory of Tree-Ring Research, University of Arizona, USA 17 4 School of History and Archaeology, University of Thessaloniki, Greece 18 5 Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Switzerland 19 6 Laboratory for Dendrochronology, Archaeological Service Canton of Bern, Switzerland 20 7 Laboratory for Ion Beam Physics, ETH Zurich, Switzerland  
+
+<|ref|>sub_title<|/ref|><|det|>[[66, 673, 185, 689]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[63, 694, 888, 860]]<|/det|>
+The discovery of abrupt radiocarbon ( \(^{14}\mathrm{C}\) ) excursions (Solar Energetic Particle events, or Miyake events) in sequences of radiocarbon measurements from calendar dated tree- rings, has yielded new opportunities to assign absolute, calendar dates to undated wood samples from widely ranging contexts in history and prehistory. We report on an important tree- ring and \(^{14}\mathrm{C}\) - dating based study, which secures the Neolithic site of Dispilio, Northern Greece, a key site for the Aegean Neolithic, in absolute, calendar- dated time using the Miyake event of 5259 BC. The last ring of the 303- year- long juniper tree- ring chronology from Dispilio is dated to 5140 BC. Dispilio is thus the first prehistoric site absolutely dated through a \(^{14}\mathrm{C}\) signature (Miyake event), but also the first absolutely, calendar- year dated prehistoric site in the wider Mediterranean region.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[112, 124, 217, 140]]<|/det|>
+## Introduction  
+
+<|ref|>text<|/ref|><|det|>[[111, 144, 886, 400]]<|/det|>
+The Neolithic period in western Eurasia marks one of the most important transitions in human social, economic, and technological history. This transition, lasting several millennia, is chiefly characterized by the appearance and gradual adoption of agriculture and animal husbandry, accompanied with increasing social and material culture complexity. The beginning of the Neolithic in Western Eurasia is dated to before \(\sim 9500\) BC in the Levant<sup>1</sup>, while its appearance on the Aegean coasts and continental Europe is dated to around \(\sim 6500\) BC<sup>2- 5</sup>. The earliest Neolithic sites on the continent are in Southeastern Europe, and their precise dating is essential for our understanding of the Neolithic transitions in Europe and critical to assessments of the environmental footprint of the new farming subsistence practices. However, the temporal resolution of archaeological and environmental proxies in the region is highly variable, producing significant discrepancies between various chronological and terminological systems that deal with the periodisation of the Neolithic<sup>6</sup>. Here we present the absolute dating of the Neolithic site of Dispilio in Northern Greece, via a combination of tree- ring dating (dendrochronology) and rapid <sup>14</sup>C excursions. This new data may serve as the basis for absolute dendrochronological dating of other sites from the Neolithic period in the region (Fig. 1).  
+
+<|ref|>text<|/ref|><|det|>[[110, 409, 886, 628]]<|/det|>
+Tree- rings enable high- resolution dating, the possibility of annually resolved climatic reconstruction and multidisciplinary chronological synchronization to (at best) a single growth season of a specific calendar dated year<sup>7</sup>. Until now, dendrochronological dating was possible only against reference tree- ring chronologies, which are continuous, unbroken sequences of tree- ring width records extending from the present back to the past. In this way, calendar dated tree- ring years can be assigned based on the known date of modern material, and then extended backwards through time using climatically constrained, region specific, tree- ring growth patterns. Long- term concentrated efforts in search for old wood samples has resulted in the construction of long tree- ring records extending for many thousands of years and widely applied to dating<sup>8- 10</sup>, and in some cases paleoclimatic analyses<sup>11,12</sup> of past human and environmental interactions. These records are however geographically limited and rare, and many prehistoric tree- ring chronologies are only approximately constrained on a calendar time- scale through conventional <sup>14</sup>C wiggle- matching and have no absolute calendar anchor.  
+
+<|ref|>text<|/ref|><|det|>[[110, 639, 886, 876]]<|/det|>
+This limitation can now be overcome by a new hybrid form of dendrochronological and single year radiocarbon analyses. Annual measurements of <sup>14</sup>C in dendrochronologically dated Holocene tree- rings have revealed the existence of rapid short- term spikes in atmospheric <sup>14</sup>C concentration in the past<sup>13,14</sup>. These <sup>14</sup>C spikes – also called Miyake or SEP (solar energetic particle) events – are uniquely suitable for absolute dating of any wooden objects with detectable annual rings<sup>15,16</sup>. The discovery of these short- term events has also led to a proliferation of annual <sup>14</sup>C measurements on single tree- rings, now spanning several millennia<sup>17- 19</sup>. The mechanisms behind these <sup>14</sup>C events are still debated<sup>20,21</sup>. However, a consensus explanation is that they are a result of coronal mass ejections on the Sun<sup>20,22- 24</sup> manifested as a surge of SEPs colliding with the Earth's atmosphere, in turn increasing the production of cosmogenic radionuclides<sup>17,24</sup>. To date, there are only five events<sup>13,14,17,25</sup> with an atmospheric <sup>14</sup>C increase ≥1% within 2 years<sup>17</sup>. Of these, the two most recently discovered events are in the first half of the Holocene – 7176 BC and 5259 BC<sup>17</sup> – offering for the first time the possibility for absolute annual dating of wood from the European Neolithic and Mesolithic using annual <sup>14</sup>C measurements.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[130, 98, 872, 631]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 648, 884, 762]]<|/det|>
+<center>Figure 1 Location of the archaeological site of Dispilio and detailed view of the analysed trench. a: map of S-E Europe marking the location of the enlarged area in b.; b: Location of Dispilio and other Neolithic sites within \(\sim 100 \mathrm{km}\) with reported good wood preservation and similar chronological placement, therefore with high potential for dendrochronological cross-dating with Dispilio (1-Anarghiri III; 2-Anarghiri IXb; 3-Crkveni Livadi; 4-Dispilio; 5-Dunavec; 6-Limnochori II; 7-Lin 3; 8-Maliq; 9-Ohridati/Penelopa; 10-Ustie na Drim, 11-Sovjan; QGIS 3.16, EPSG 32634; Lake Maliq according to Fouache et al. (2010)) c: drone photograph of the site of Dispilio and its surroundings, the dendrochronologically analysed East Sector marked in the foreground; d: close-up of the East Sector before sampling of wooden elements in 2019, vertical elements are seen sticking out of the ground, each marked with a unique white label. (a.,b.-A. Maczkowski; c.-M. Hostettler; d.-Dispilio Excavation Archive) </center>  
+
+<|ref|>text<|/ref|><|det|>[[112, 775, 884, 920]]<|/det|>
+In temperate climates archaeological wood, and organic materials in general, can be preserved only in very stable conditions – such as constant low- oxygen waterlogged sediments at wetland archaeological sites \(27 - 29\) . While excavated wetland sites are very numerous and often excavated in Central Europe, several wetland sites have also been found and excavated in Southeastern Europe, notably in the south- western part of the Balkans \(30 - 36\) . Dendrochronological work on these sites led to the construction of several tree- ring width chronologies, which were fixed in time by means of \(^{14}\mathrm{C}\) modelling (wiggle- matching) \(37,38\) . The archaeological site of Dispilio on the shores of Lake Kastoria in Northern Greece is a premier prehistoric wetland site in the region. Numerous lines of evidence have yielded detailed results on the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 87, 884, 306]]<|/det|>
+georachaeology \(^{35}\) , palynology \(^{39,40}\) anthracology \(^{41,42}\) , woodworking technology \(^{43}\) , and material culture \(^{44,45}\) . The approximate calendar-age chronology of the site has been established through radiocarbon dates, mostly performed on charcoal samples \(^{35,46}\) . The calibrated date-ranges point to settlement phases between the later Middle Neolithic ("5600 cal BC \(^{47}\) ) and the Bronze Age ("2100 cal BC \(^{46}\) ). The excavations at Dispilio have also yielded a great number of wood remains, with over 1200 mapped construction elements in the Eastern Sector to date (Fig 1c). Yet despite the extensive remains of wooden construction elements, no systematic sampling and no tree-ring based chronological studies via dendrochronology have yet been conducted at the site. The value of developing a precise and accurate calendar- dated chronological sequence using these wooden remains is further enhanced by the fact that the site of Dispilio with more than 1700 complete ceramic vessels (Fig. 2) boasts one of the largest complete Neolithic ceramic assemblages in Europe. Tree-ring dating at Dispilio can therefore be used, via the existing ceramics network, to underpin and improve the relative chronology of the entire region.  
+
+<|ref|>text<|/ref|><|det|>[[110, 317, 886, 464]]<|/det|>
+In 2019 a large- scale fieldwork campaign took place at Dispilio's Eastern Sector (Fig. 1d), during which over 900 wooden construction elements (piles) were mapped, of which 787 were sampled for the first dendrochronological analysis. The dendrochronological results provided an oak chronology spanning 120 years, and an overlapping juniper chronology spanning 303 years. This record could not be dated dendrochronologically however, because despite the existence of several millennia- long tree- ring chronologies in the Eastern Mediterranean \(^{11,48,49}\) , none extend back for 7500 years. Here we overcome this limitation by using the combination of dendrochronological and single year radiocarbon analysis, thus providing the first absolute dating of a Neolithic site in the wider Mediterranean region.  
+
+<|ref|>image<|/ref|><|det|>[[111, 473, 907, 670]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[111, 680, 886, 723]]<|/det|>
+<center>Figure 2, Archaeological finds from Neolithic Dispilio. a: almost completely preserved ornate anthropomorphic vessel from Late Neolithic, many similar ones have been recovered from the site, scale in cm; b: bone spear/harpoon tip with preserved hafting adhesives, scale in cm; c.: an assemblage of Late Neolithic personal adornments (a.,b.,c.,-Dispilio Excavation Archive) </center>  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 771, 175, 786]]<|/det|>
+## Results  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 795, 251, 811]]<|/det|>
+## Dendrochronology  
+
+<|ref|>text<|/ref|><|det|>[[111, 817, 886, 926]]<|/det|>
+Of the total wood samples from the archaeological site of Dispilio in 2019 (n=787), 23% were cross- dated into two master tree- ring width (TRW) chronologies. Wood anatomical species determination revealed that the majority of the wooden piles came from oak (Quercus spp., 21%) and juniper (Juniperus spp., 62%) wood. The third most abundant species are pines (Pinus spp., 17%), which were not suitable for dendrochronological cross- dating given the low number of annual rings on most pine samples. The majority of the pine samples could be classified as belonging to the subgenus Pinus (cf. Pinus
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 87, 886, 198]]<|/det|>
+nigra/sylvestris) with several pieces belonging to the subgenus Strobus (cf. Pinus peuce). Due to the wood- anatomical intra- species similarity of junipers \(^{50,51}\) , and of deciduous oaks from the subgenus Quercus \(^{52}\) , a definitive species- level identification was not possible. Based on modern tree species in the region \(^{41,53,54}\) , Dispilio oak wood samples most likely come from Q. frainetto, Q. petraea, and/or Q. pubescens wood, and the junipers are most likely Juniperus excelsa, J. foetedissima, and/or J. deltoides (for the latter cf. J oxycedrus).  
+
+<|ref|>text<|/ref|><|det|>[[111, 208, 886, 317]]<|/det|>
+The oak TRW chronology produced was 120- years- long composed of 58 wood samples (Fig 4). It consists of tree- ring sequences with an average segment length of 66 years. Some sapwood was present on most of the oak samples \((n = 45)\) , however the last growth ring (or "waney- edge"), which is important for archaeological interpretation, was conserved on only 4 pieces either as a result of the lower durability of oak sapwood or its intentional removal. The mean inter- series correlation (leave- one- out principle \(^{55}\) ) of the oak tree- ring sequences is 0.51.  
+
+<|ref|>text<|/ref|><|det|>[[111, 328, 886, 456]]<|/det|>
+A 303- years- long juniper TRW- chronology was also constructed consisting of 118 tree- ring sequences and an average segment length of 86 years (Fig 4). The mean inter- series correlation (leave- one- out principle \(^{55}\) ) of the juniper chronology is 0.62. Juniper wood, owing to its chemical \(^{56}\) and physical \(^{57}\) properties has a higher resistance to degradation. These qualities made juniper wood the material of choice for construction purposes in many ancient societies in the Eastern Mediterranean \(^{58 - 60}\) . The preservation of juniper wood in Dispilio is also exceptional and the waney- edge on junipers is quite common, enabling an annually resolved reconstruction of the building phases and occupation duration on the site (Fig 4b).  
+
+<|ref|>text<|/ref|><|det|>[[111, 466, 886, 558]]<|/det|>
+All samples with a preserved waney edge had a last growth ring terminating with latewood, thus implying a felling date during the dormant period of the trees between late summer and early spring. The juniper and oak tree- ring chronologies have robust dendrochronological dating against each other (t- value = 4.9 \(^{61}\) and = 5.1 \(^{62}\) ; GLK = 63% \(^{63}\) ) over a period of 108 years where sample replication is >4, further supported by \(^{14}\) C wiggle- matching (Supplementary Material S1)  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 568, 370, 585]]<|/det|>
+## Tree-ring \(^{14}\) C cosmogenic signature  
+
+<|ref|>text<|/ref|><|det|>[[111, 590, 886, 865]]<|/det|>
+Conventional radiocarbon wiggle- matching models \(^{64,65}\) based on several blocks of 1- 11 tree- rings modelled against the atmospheric data for the Northern Hemisphere (IntCal20 \(^{66}\) ) produced the initial modelled age- ranges for the tree- ring chronologies. Preliminary annual sampling at test positions on the juniper tree- ring chronology indicated that the last ring of this chronology dated between 5233 and 5137 cal BC (at 95% probability). On this basis, a suite of additional single year \(^{14}\) C measurements were made to pinpoint the exact years surrounding the 5259 BC Miyake event. Four wood samples from the juniper chronology were selected covering the part of the chronology where the 5259 BC Miyake event should be located (Fig. 3a). We present here the final 115 \(^{14}\) C measurements (Supplementary Table T1) performed to locate the 5259 BC Miyake event in all 4 wood samples from the Dispilio juniper tree- ring chronology (Fig. 3a). The \(^{14}\) C measurements were performed at the Laboratory for the Analysis of Radiocarbon with AMS at the University of Bern (LARA) \(^{67}\) and the Laboratory of Ion Beam Physics at ETH Zürich (ETH) \(^{68,69}\) . An average year- to- year increase (sensu Miyake et al. \(^{13}\) ) of \(\sim 15.8\%\) in \(\Delta^{14}\) C was detected in all samples in the exact same dendrochronologically cross- dated tree- rings corresponding to the relative ring 184 of the Dispilio juniper chronology. This increase varies from the lowest of \(\sim 11.1\%\) \(\Delta^{14}\) C in DISP- 10070, to \(\sim 13.1\%\) in DISP- 10206, to \(\sim 14.8\%\) in DISP- 10063, to \(\sim 18.6\%\) in DISP- 10611 (Fig. 3a, Supplementary Table T1).  
+
+<|ref|>text<|/ref|><|det|>[[111, 875, 886, 911]]<|/det|>
+To compare the \(^{14}\) C results from Dispilio with the published reference data for the 5259 BC event, a mean- value annually resolved reference curve (RC) was established from the dataset in Brehm et al. (2022 –
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 87, 884, 198]]<|/det|>
+henceforth referred to as 'BR22'17). A common approach for verifying the position of Miyake events is wiggle- matching using a goodness- of- fit \(\chi^2\) test15,70,71 against a reference, so that the \(\chi^2\) value becomes minimal for the correct placement of the sample's waney- edge64. The lowest \(\chi^2\) values are reached when the end- dates of the samples are placed at 5240 BC for DISP- 10070 and DISP- 10063 (Fig. 3b), 5153 BC for DISP- 10206, and 5155 BC for DISP- 10611 (Fig. 3c), corresponding to their cross- dated position along the tree- ring chronology. The 5259 BC event signal is clearly identified in all wood samples (Fig. 3a).  
+
+<|ref|>text<|/ref|><|det|>[[110, 208, 886, 519]]<|/det|>
+In order to test how close conventional radiocarbon wiggle- matching would be relative to the absolute calendar dating supplied by the Miyake event, the annual data from all the wood samples were wiggle- matched against the IntCal20 calibration curve66 using the \(^{14}\mathrm{C}\) calibration software OxCal 4.4 \(^{64,65}\) . In none of the cases does the \(95\%\) probability end- date range include the actual felling date when IntCal20 is used (Fig. 5, Supplementary Material S4). Longer series of \(^{14}\mathrm{C}\) dates which span some years before and after the event (Fig. 3a, Fig 5), as from wood samples DISP- 10611 and - 10206, yield end- dates which are only \(\sim 15 - 20\) cal years older, while shorter series, wood samples DISP- 10070 and - 10063, result in end- dates over \(\sim 40\) cal years younger than the actual felling dates (Fig. 5). It has been noted previously \(^{72}\) that IntCal20 is poorly replicated during the 53rd- 52nd century BC. Notably, the 53rd century BC is represented by only 16 measurements, of which 14 are decadal and bi- decadal (i.e. blocks of 10- 20 tree- rings), with only two 4- and 5- year blocks \(^{66,73}\) (see Supplementary Material S2.8). The variability in the calibrated end- date ranges suggests that IntCal20 might produce misleading results when wiggle- matching annual data coming from the period in question. The annual \(^{14}\mathrm{C}\) dates were also wiggle- matched against a modified IntCal20 – IntCal20plus – where the default IntCal20 multiple- year blocks of BP (Before Present) data for the 82 years period around the event were substituted with the average of the annual BR22 dataset. Calibrating against this dataset predictably yields the accurate and more precise end- date ranges at \(95\%\) probability for all wood samples (Fig. 5).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[110, 87, 880, 710]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[111, 718, 886, 829]]<|/det|>
+<center>Figure 3, Scatter plot of \(\Delta^{14}\mathrm{C}\) data from Dispilio against reference from Brehm et al. (2022) \(^{17}\) , and best last ring fit for the dated wood samples \(\left(\chi^{2}\right)\) . a: Measured \(^{14}\mathrm{C}\) concentrations represented as \(\Delta^{14}\mathrm{C}\) , vertical bars represent 1s uncertainties (Supplementary Table T1); samples marked with "DISP-" refer to measurements on wood samples obtained in this study, other labels represent data from BR22 \(^{17}\) - Bristlecone pine \(^{14}\mathrm{C}\) data are shifted forward by 1 year from the original Brehm et al. publication, following a correction to the dating of the master bristlecone chronology (Supplementary Material S3.2); shaded band represents IntCal20 \(^{66}\) . Panels below, b, c: chi-squared tests of Dispilio measurements against the average from BR22 \(^{17}\) for wood samples DISP-10070 and -10063 (b, \(\chi^{2}\) crit. value=9.49), and DISP-10206 and -10611 (c, \(\chi^{2}\) crit. value=15.51). Figure produced in \(R^{74}\) , code and source data available in Supplementary Material 4. </center>  
+
+<|ref|>text<|/ref|><|det|>[[111, 842, 886, 914]]<|/det|>
+The growing season of trees is influenced by many factors and can vary between and among species as a function of cambial age, temperature, water, slope, aspect, soil etc. Personal observations of growth termination in modern oaks and junipers in the region have revealed that latewood can be completed in both genera in the beginning of September (Supplementary Materials S2.6- S2.7). While cell- wall
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 87, 885, 215]]<|/det|>
+thickening in temperate conifers continues for several weeks after the cessation of cell- wall enlargement \(^{75}\) , the amount of cellulose carbon that would be deposited during this last stage of latewood formation constitutes a small percentage of the whole tree- ring \(^{76}\) . Considering the robustness of the \(^{14}\mathrm{C}\) signal in the Displilio junipers tree- rings (Fig. 3) it is unlikely that it only represents the \(^{14}\mathrm{C}\) incorporated at the end of the cell- wall thickening stage. Consequently, it can be stated that the \(^{14}\mathrm{C}\) signal of the 5259 BC event in the in deciduous junipers was incorporated in the same growing season characteristic for deciduous species, i.e. spring to late summer/early autumn 5259 BC.  
+
+<|ref|>text<|/ref|><|det|>[[111, 225, 885, 354]]<|/det|>
+According to the dendrochronologically cross- dated position of all wood samples, the ring in which the Miyake event is detected corresponds to relative year number 184 of the 303- year- long juniper TRW chronology. This allows us to set the absolute end- date of the whole Displilio juniper tree- ring chronology at 5140 BC. Furthermore, the identification of the event in DISP- 10070 and - 10063 confirms the correct placement of the better- replicated earlier half of the chronology (Fig. 4a.). Given the dendrochronological cross- dating between the juniper and oak chronologies, also the latter is absolutely dated, placing its last ring at 5311 BC (Fig. 4a.).  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 365, 314, 381]]<|/det|>
+## Site plan and felling phases  
+
+<|ref|>text<|/ref|><|det|>[[110, 387, 886, 697]]<|/det|>
+By considering the latest juniper felling dates together with the earliest secure felling dates from the oak chronology it is possible to establish a minimum duration of construction activities of 188 years between 5328 and 5140 BC, with intermittent periods of wood felling/construction, which do not necessarily reflect a continuous, uninterrupted occupation at the same location. Such a chronological resolution for a settlement phase duration on a prehistoric site in the Eastern Mediterranean has not been established to date. Plotting of groups of cross- dated wood samples with felling dates within 1- 2 years of one another using a GIS software revealed blueprints representing different structures (Fig. 4b). Identification of building outlines was possible only for groups that are composed of a substantial number of cross- dated samples. The structures seem to be oriented along the lakeshore. Of particular note is the concentration of building activities in the eastern part of the Eastern Sector. In this part, building activities on the same spot outline an area with a felling date in 5294 BC, and a felling phase which ends in 5257 BC (Fig. 4a, b). A felling phase ending in 5320 BC precedes the group of 5294 BC, however due to the suboptimal preservation of oak samples only two of this group have preserved waney edge. These are complemented by several oak samples dated between 5328 BC and 5320 BC with at least 20 sapwood rings indicating the proximity of the waney edge. The mapping of the dendrochronological results further implies that building practices in some cases either included short term storage (1- 2 years) of timber or consisted of a construction period spread over several years.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[112, 88, 797, 830]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[112, 840, 886, 923]]<|/det|>
+<center>Figure 4, Bar chart of tree-ring chronologies, felling dates, and site plan development. a: bar plot of Dispilio oak and juniper chronologies; each horizontal bar represents individual wood sample in its dendrochronologically cross-dated position, bar length corresponds to its span in years (i.e., number of tree-rings). Red stars indicate wood samples sampled for annual \(^{14}\mathrm{C}\) ; b.: schematic plan of the East Sector (see also Fig. 1c-d); each symbol represents one vertical wooden element, different shapes and colours correspond to a same felling phase spread over 1-2 years; additionally, colour-shaded polygons outline the groups of same symbols (same felling-phase elements), however they do not represent definite structure plans. </center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[113, 124, 202, 140]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[110, 144, 886, 549]]<|/det|>
+According to the archaeo- chronological periodisation in the region, for which there is no universal absolute timeframe<sup>6</sup>, the occupation phases of Dispilio discussed here would fall at the later Middle Neolithic and/or Late Neolithic. The absolute dating and duration of the Middle/Late Neolithic occupation phase in Dispilio is unique in the context of the Balkans, but also in the wider Eastern Mediterranean Neolithic. The site also provides sufficiently replicated dendrochronological information to allow independent controls for settlement duration estimates. The felling dates in the excavated sector indicate activity over a period of at least 188 years, with indications from oak sapwood estimates to extend this backwards by a further 30 years. Of particular interest is the succession of 2 construction phases in the western half of the analysed trench and 3 construction phases in its eastern half (Fig. 4a, b). Although the nature of these structural outlines (Fig. 4b) is not clear at present, a timespan between the construction episodes of 29 years in the western half (5311 and 5282 BC), and 35- 37 years in the eastern half (5320, 5294 and 5257 BC) is consistent with the few available estimates of house lifespans in Neolithic S- E Europe<sup>77,78</sup>. However, determining whether these contemporary structure outlines with same felling dates correspond to one or multiple buildings will require further detailed multidisciplinary work. Intermittent periods without felling dates may simply be a result of preservation or the limited size of the excavated area, but may also reflect a hiatus in occupation or indicate a non- perennial character of the settlement. Detection of annual or decadal- scale hiatuses is extremely difficult in archaeological stratigraphy, with settlement phase duration usually derived from <sup>14</sup>C sequence models based on organic samples from consecutive stratigraphical units. This approach can lead to interpretations of centuries- long settlement continuities<sup>4,79</sup>. Such interpretations may underestimate settlement discontinuities of durations shorter than the associated precision of <sup>14</sup>C measurements and calibration. This underlines the importance of the annually resolved data from Dispilio.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[111, 87, 879, 485]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[112, 496, 884, 567]]<|/det|>
+<center>Figure 5 Wiggle-matching of different sets of annual \(^{14}\mathrm{C}\) data from Dispilio modelled in OxCal v4.4, against IntCal20<sup>66</sup>, and IntCal20plus. IntCal20plus has the non-annual IntCal20 data for a 82-year period around the 5259 BC Miyake event replaced by annual average of Brehm et al. (2022)<sup>17</sup> annual data. Dotted blue lines represent actual felling dates determined through dendrochronology and Miyake event-matching. Acronyms in brackets next to sample name refer to AMS lab that furnished the measurements. Data for figure obtained from OxCal<sup>65,66</sup>. Figure produced in \(R^{74}\) , code and data in Supplementary Material 4. </center>  
+
+<|ref|>text<|/ref|><|det|>[[112, 579, 884, 926]]<|/det|>
+The last centuries of the \(6^{\text{th}}\) millennium BC mark an important change within the Neolithic period in the southern Balkans. It is a period of a steep increase in the number and size of settlements, associated with a demographic boom<sup>6,80–82</sup>. Anthropogenic influence on the local environment becomes more pronounced during this period<sup>83,84</sup>, as documented also in Dispilio<sup>39,40</sup>. Diversity increased in all aspects of human behaviour, from pottery production techniques and styles<sup>85</sup>, architecture<sup>81</sup>, settlement organisation<sup>81,86,87</sup> to the first signs of metallurgy<sup>88</sup>. Evidence from this transitional period also points to a shifting social focus from the collective to the domestic<sup>89,90</sup>. In this setting, high- resolution chronological data can improve our understanding of societal changes, human land use, and intensifying influence on the local and regional environment. For instance, the preference of settling in the proximity of wetlands has been documented in the Early Neolithic<sup>3,91</sup>, a practice continuing in subsequent Neolithic subperiods<sup>32,91</sup>. Wetland and shoreline locations would have represented ideal catchment areas for the Neolithic subsistence, providing various soil types that could be exploited for cultivating crops with different requirements, serve as pasture lands, or supply aquatic resources as a dietary complement<sup>91</sup>. A number of wetland sites with similar chronology to Dispilio (2<sup>nd</sup> half of the \(6^{\text{th}}\) millennium BC) have been documented or excavated in existing or former lakes in the region, some of them yielding large amounts of well- preserved wooden construction elements (Fig. 1b,<sup>32–34,92–94</sup>). Although the dating of these sites has much lower chronological resolution than at Dispilio, some of them would have been in use for centuries before and/or after the \(54^{\text{th}}\) - 52nd century BC phases in Dispilio. It is highly likely that it will be possible to cross- date the tree- ring widths of the wood remains from these peripheral sites with the now absolutely dated tree- ring chronologies from
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 884, 123]]<|/det|>
+Dispilio, and thus extend the absolutely dated chronological network for the region well beyond the \(6^{\text{th}}\) millennium BC.  
+
+<|ref|>text<|/ref|><|det|>[[111, 135, 886, 409]]<|/det|>
+Beyond the chronological significance, absolutely dated tree- ring records are one of the most utilized proxies for high- resolution climate reconstructions offering unique insights into the relationship between humans and climate. Precipitation is a limiting factor for most low and mid- altitudes trees in the Eastern Mediterranean. In fact, it has been shown that modern juniper \(^{49}\) and oak \(^{11,95}\) tree- ring sequences are good predictors of precipitation in the Eastern Mediterranean. Precipitation was a crucial factor in early agriculture which mainly consisted of rain- fed \(^{96}\) and flood- water \(^{97}\) farming. Preliminary observations of the Dispilio TRW chronologies imply a period of suppressed growth in both the juniper and oak tree- ring sequences for a period of around 20 years between 5360 and 5340 BC. Such suppressed growth period can be associated with decrease in precipitation, which may significantly influence the water table of small water bodies such as Lake Kastoria. A short- term Mid/Late Neolithic eutrophication of the lake previously inferred from increased presence of green algae \(^{35}\) could potentially be correlated with this tree- ring width suppression. Although the Neolithic tree- ring sequences from Dispilio are relatively short if compared to modern tree- ring proxies used in climate reconstructions, they still may provide valuable absolutely dated, annually resolved information on environmental conditions during the Neolithic in Kastoria Basin and the surrounding region.  
+
+<|ref|>text<|/ref|><|det|>[[111, 420, 886, 673]]<|/det|>
+Finally, the results from this study underline the value that single year measurements of radiocarbon in tree- rings can have for radiocarbon calibration and dendrochronological dating. Significant advances in AMS technology \(^{68}\) , have made it possible to create long and continuous time- series of annual radiocarbon that are constantly improving the accuracy of the radiocarbon calibration process. More than this though, the utilization of SEP events in anchoring regional timelines through hybrid tree- ring and radiocarbon studies is once again demonstrated. The \(^{14}\text{C}\) - anchored Dispilio tree- ring chronologies now provide a calendar dated reference for dendrochronological dating of other sites from the time period. This provides the opportunity to extend calendar dated chronologies across the region further back into prehistory. Such high- resolution dating, especially in cases where it can be coupled with stratigraphic information or used to derive climatic indicators, will elucidate a more nuanced understanding of deterministic interpretations of the environmental influence on societies in the past (e.g. for the 6.2 ka BC cooling event). This study demonstrates how the discovery of the new SEP events in this time period creates new possibilities in prehistoric archaeology and offers the construction of historical- timescale narratives for societies and their environments from the very distant past.  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 721, 306, 737]]<|/det|>
+## Materials and Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 745, 223, 760]]<|/det|>
+## Wood samples  
+
+<|ref|>text<|/ref|><|det|>[[111, 768, 886, 913]]<|/det|>
+The wood material analysed in this study was sampled in August and September 2019 from wooden piles remains at the archaeological site of Dispilio, near Kastoria, Greece (40.485444 N, 21.289694 E; h=627 masl). The site is one of the best- known prehistoric sites in the country and has been investigated, almost continuously, since 1992. Excavations and sampling that took place on the site were performed in full compliance with the regulations of the Greek Ministry of Culture concerning archaeological material. Whole cross- section discs (n=787) were sampled from the wooden remains with handsaws and chainsaws during the 2019 fieldwork campaign. The wood samples documentation, cleaning, preparation, and sealing in plastic bags with water, took place on- site during the 2019 field campaign. Dendrochronological
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 87, 884, 197]]<|/det|>
+measurement took place initially on- site and continued at the University of Bern. Tree- ring width (TRW) measurements were performed according to standard dendrochronological procedures \(^{98,99}\) , by means of a measuring table under a binocular stereo microscope. TRWs were recorded with a precision of 0.01 mm. Two to four radii were measured per sample and averaged together to represent the sample. Descriptive dendrochronological statistics were performed in the dplR package in R \(^{55,74,100}\) . The TRW measurements of DISP- 10611, - 10206, 10070, and - 10063 are available in the Supplementary Material S3.3.  
+
+<|ref|>text<|/ref|><|det|>[[111, 208, 885, 480]]<|/det|>
+Wood taxonomy was determined based on stem wood anatomy. Each measured wood sample was sectioned with a razor blade and cell arrangements in the transversal, radial, and tangential sections were identified and compared with references in wood- anatomical atlases \(^{51,52,101}\) . Given the wood anatomical similarity of different deciduous oak species from the subgenus Quercus \(^{52}\) , and considering the high dendrofloristic diversity of oaks in the region \(^{53,54}\) it is not possible to distinguish them to species level. However, it is likely that several deciduous oak species from the subgenus Quercus are represented, notably \(Q\) . frainetto, \(Q\) . petraea, and/or \(Q\) . pubescens. Oak trees from the subgenus Cerris are one of the more abundant groups of oaks in the region, however no wood samples from Dispilio could be assigned to this group which is anatomically characterised by larger and solitary latewood pores. Similarly, wood anatomical differentiation between different juniper species is not possible \(^{50,51,101}\) . Considering todays distribution of tree- like junipers in the region, the most likely species utilized in Dispilio are Juniperus excelsa, \(J\) . foetidissima, and/or \(J\) . deltoides Adams (cf. \(J\) . oxycedrus L.). While majority of the pine samples exhibited denticulate walls on end- tracheids, a characteristic of the pine subgenus Pinus (cf. Pinus nigra/sylvestris- type), several pine wood samples could be identified as members of the Subgenus Strobus (cf. \(P\) . peuce) based on the presence of smooth- walled end- tracheids.  
+
+<|ref|>text<|/ref|><|det|>[[112, 491, 885, 582]]<|/det|>
+Local climate in the Kastoria Basin can be defined as continental to sub- Mediterranean, with temperate weather, continental winters, and warm and dry summers. The yearly average precipitation of \(\sim 600 \text{mm}\) increases with altitude, with the wettest months being November and December, while July and August are the driest and hottest months. Yearly average temperature is \(\sim 12.5^{\circ} \text{C}\) . Main climate classes according to the Köppen system \(^{102}\) are Cfa, Cfb, Csa.  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 594, 490, 610]]<|/det|>
+## Sample preparation and radiocarbon measurement  
+
+<|ref|>text<|/ref|><|det|>[[112, 617, 885, 726]]<|/det|>
+Individual tree- rings were dissected by hand under a binocular microscope with a one- sided razor blade (Supplementary Material S2.5). Whole rings were used for all \(^{14} \text{C}\) measurements (Supplementary Table S2.1). About 30- 70 mg of material were sampled per ring, depending on its width. Earlywood comprises ca. 80- 90% of a juniper tree- ring. Since most of the of the ring- structure of junipers growing on mesic sites is completed by the end of September \(^{103}\) (see also Supplementary Material S2.6- S2.7), the tree- ring structural carbon concentration should reflect temperate spring- to- late summer carbon uptake.  
+
+<|ref|>text<|/ref|><|det|>[[112, 737, 885, 883]]<|/det|>
+Wiggle- matching of several \(^{14} \text{C}\) dates provided the initial estimate of the 40- rings segment of the tree- ring chronology where the event will be located. A "buffer zone" of 15 rings at each limit was added to the estimate, and 70 individual rings were sampled centred around the estimated "event ring" from the first wood sample that was analysed (DISP- 10206, Supplementary Material S2.1). The \(^{14} \text{C}\) content of every \(4^{\text{th}}\) sampled ring was subsequently measured until the \(^{14} \text{C}\) spike was located, after which the \(^{14} \text{C}\) in 20 consecutive annual rings around the event was measured. The "event ring" on all the other wood samples (DISP- 10611, - 10070, - 10063, Supplementary Material S2.2- S2.4) was identified according to the samples' cross- dating position along the tree- ring chronology.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 87, 886, 325]]<|/det|>
+Cellulose from wood samples analysed at the Laboratory for the Analysis of Radiocarbon with AMS at the University of Bern (LARA)67 was extracted following the BABAB method104 including the modifications of Sookdeo et al. (2020) at \(70^{\circ}C\) for all steps. Samples were submerged in a 1M NaOH overnight and treated in 1M HCl followed by 1M NaOH in a shaker for one hour each. Bleaching of the samples was performed on addition of 5 mL water, a few drops of 1M HCl to reach pH 2- 3 and 100 mg NaClO2 by shaking for at least two hours or until the colour of the wood samples turned white. Drying of the material was achieved by lyophilisation overnight. Samples were measured using the LARA MICADAS AMS system. DISP- 10070, - 10206 and a first run of - 10611 was analysed together with three oxalic acid II (SRM 4990C, NIST) standards and three chemical blanks. Later, a second run of DISP- 10611 and - 10063 was dated together with five oxalic acid II standards and four chemical blanks that were used for blank subtraction, standard normalization, and correction for isotope fractionations as well as two IAEA- C5, two IAEA- C7, two 1515 CE reference samples and two cellulose blanks as secondary standards and blanks, respectively. For details, see Supplementary Material S3.1 and Supplementary Table T1.  
+
+<|ref|>text<|/ref|><|det|>[[110, 336, 886, 519]]<|/det|>
+For the analyses performed at ETHZ, the tree- ring samples were prepared in 15 ml glass test tubes together with four wood blanks (2 BC and 2 KB) and 2 1515 CE reference samples each weighing \(30 - 60mg^{68}\) . In a slightly modified procedure following104, samples were first soaked in 5 ml 1M NaOH overnight at \(70^{\circ}C\) in an oven. Then the samples were treated with 1M HCl and 1M NaOH for 1 hour each at \(70^{\circ}C\) in a heat block, before they were bleached at a pH of 2- 3 with 0.35M NaClO2 at \(70^{\circ}C\) for 2 h. The remaining white holocellulose was then freeze- dried overnight. About 2.5 mg dried holo- cellulose was wrapped in cleaned Al capsules and converted to graphite using the automated graphitization line AGE- 3. A measurement set was made up of the tree- ring samples, three oxalic acid one (OX1) and four oxalic acid two (OX2) standards, two cellulose blanks, two chemical blanks, and two 1515 CE reference samples and measured in the MICADAS accelerator mass spectrometer.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 529, 390, 546]]<|/det|>
+## Radiocarbon matching and modelling  
+
+<|ref|>text<|/ref|><|det|>[[111, 552, 886, 606]]<|/det|>
+The new \(^{14}\mathrm{C}\) measurements presented in this study were matched to the constructed reference curve \(^{17}\) (see also Supplementary Material S4) using a common \(\chi^2\) test approach so that the \(\chi^2\) value becomes minimal for the correct placement of the sample's waney- edge \(^{15,64,71}\) :  
+
+<|ref|>equation<|/ref|><|det|>[[375, 614, 620, 677]]<|/det|>
+\[\chi^2 (x) = \sum_{i = 1}^{n}\frac{(R_i - C_{(x - r_i)})^2}{\delta R_i^2 + \delta C_{(x - r_i)}^2}\]  
+
+<|ref|>text<|/ref|><|det|>[[110, 686, 886, 743]]<|/det|>
+Where \(R_{i}\pm \delta R_{i}\) represent the new \(^{14}\mathrm{C}\) measurements, and \(C_{(x - r_i)}\pm \delta C_{(x - r_i)}\) represent the reference curve \(^{14}\mathrm{C}\) concentrations in the year \((x - r_i)\) ; \(r_i\) stands for the tree ring number starting with 0, representing the last growth ring of the tree (waney- edge).  
+
+<|ref|>text<|/ref|><|det|>[[110, 754, 886, 808]]<|/det|>
+The Bayesian wiggle- matching was performed in the software OxCal 4.4 with the inbuilt D_Sequence command against the atmospheric data from IntCal20 \(^{65,66}\) , for the CQL code see Supplementary Material S1 and S4.  
+
+<|ref|>text<|/ref|><|det|>[[110, 819, 886, 874]]<|/det|>
+The year- to- year increase in \(\Delta^{14}\mathrm{C}\) presented in the Results section was calculated as a difference between the values in 5260 BC and 5259 BC (sensu Miyake et al. \(^{13}\) ). For a detailed discussion on the magnitude and \(^{14}\mathrm{C}\) production during the 5259 BC Miyake event see \(^{17}\) , and \(^{105}\) .
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[113, 90, 253, 106]]<|/det|>
+## Data uncertainty  
+
+<|ref|>text<|/ref|><|det|>[[111, 111, 886, 293]]<|/det|>
+Data uncertaintyThe genus Juniperus is known to produce intra- annual density fluctuation ('false rings') or have 'missing rings' \(^{106}\) in parts of the stem. Missing rings are very often a product of the stem growth habit of junipers, so- called 'lobate growth', which consists of higher cambial activity and faster growth in certain areas of the stem, resulting in an undulating cross- section of the stem in older trees, where the less active areas may not produce rings in certain years. However, missing rings or measuring false rings can be accounted for when sufficient numbers of wood samples with complete stem cross- sections are available, as in Dispilio. The correct location of the "event ring" on all wood samples based on their cross- dated position is further supporting a correct ring count. Moreover, the dendrochronological cross- dating of the first half of the juniper chronology against the oak chronology serves as an additional control for the correct ring count, considering that oak trees almost never have missing rings \(^{107}\) .  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 310, 249, 326]]<|/det|>
+## Data availability  
+
+<|ref|>text<|/ref|><|det|>[[111, 331, 886, 367]]<|/det|>
+Data availabilitySupplementary Material, including code, text, figures, and datasets referred to and presented in this paper are available at the following repository: 10.5281/zenodo.8407222.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 384, 266, 401]]<|/det|>
+## Acknowledgments  
+
+<|ref|>text<|/ref|><|det|>[[111, 405, 886, 496]]<|/det|>
+AcknowledgmentsThe 2019 fieldwork and the subsequent dendrochronological and radiocarbon analyses were conducted in the framework of the ERC project 'Exploring the dynamics and causes of prehistoric land use change in the cradle of European farming' (EXPLO). This project is financially supported by the European Union's Horizon 2020 research and innovation programme, under the grant agreement No 810586 (project EXPLO, explo project.eu).  
+
+<|ref|>text<|/ref|><|det|>[[111, 507, 886, 542]]<|/det|>
+We would like to thank all the Archaeology students involved in the fieldwork and sample curation from the Universities of Thessaloniki and Bern, and the staff of the Ephorate of Antiquities of Kastoria.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 560, 228, 576]]<|/det|>
+## Contributions  
+
+<|ref|>text<|/ref|><|det|>[[111, 581, 880, 671]]<|/det|>
+ContributionsA.M., together with C.P., A.H. conceived and designed the study. K.K. & T.G. led the fieldwork, while A.M. and J.F. participated in part of it. J.F. & A.M., together with M.B., performed the dendrochronological and wood- anatomical analyses. A.M. sampled individual tree- rings. S.S. and L.W. performed and provided the 14C measurements. A.M., & C.P., drafted the manuscript, and all authors edited and contributed to the manuscript. A.H. and K.K. obtained funding.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 718, 205, 733]]<|/det|>
+## References  
+
+<|ref|>text<|/ref|><|det|>[[111, 768, 886, 820]]<|/det|>
+1. Grosman, L. The Natufian Chronological Scheme – New Insights and their Implications. in Natufian Foragers in the Levant: Terminal Pleistocene Social Changes in Western Asia (eds. Bar-Yosef, O. & Valla, F. R.) 622–637 (Berghahn Books, 2013). doi:10.2307/j.ctv8bt33h.41.  
+
+<|ref|>text<|/ref|><|det|>[[111, 830, 886, 914]]<|/det|>
+2. Maniatis, Y. Radiocarbon dating of the major cultural phases in prehistoric Macedonia: Recent developments (In Greek, English summary). in A Century of Research in Prehistoric Macedonia 1912-2012, International Conference Proceedings, Archaeological Museum of Thessaloniki, 22-24 November 2012 (eds. Stefani, E., Merousis, N. & Dimoula, A.) 205–222 (Archaeological Museum of Thessaloniki, 2014).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[52, 88, 876, 122]]<|/det|>
+450 3. Karamitrou-Mentessidi, G., Efstratidou, N., Kaczanowska, M. & Kozłowski, J. Early Neolithic settlement of Mavropigi in western Greek Macedonia. Eurasian Prehistory 12, 47–115 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[52, 130, 864, 165]]<|/det|>
+452 4. Maniatis, Y. & Adaktylou, F. Revenia-Korinos: one of the earliest Neolithic settlements in North Greece as evidenced by radiocarbon dating. Radiocarbon 63, 1025–1051 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[52, 174, 875, 209]]<|/det|>
+454 5. Horejs, B. et al. The Aegean in the Early 7th Millennium BC: Maritime Networks and Colonization. J. World Prehistory 28, 289–330 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[52, 217, 876, 286]]<|/det|>
+456 6. Tsirtsoni, Z. Chapter 1. The chronological framework in Greece and Bulgaria between the late 6th and the early 3rd millennium BC, and the "Balkans 4000" project. in The Human Face of Radiocarbon (ed. Tsirtsoni, Z.) 13–39 (MOM Éditions, 2016). doi:10.4000/books.momeditions.503.  
+
+<|ref|>text<|/ref|><|det|>[[52, 295, 875, 330]]<|/det|>
+460 7. Schweingruber, F. H. Tree Rings: Basics and Applications of Dendrochronology. (Kluwer Academic Publishers, 1988). doi:10.1007/978- 94- 009- 1273- 1.  
+
+<|ref|>text<|/ref|><|det|>[[52, 339, 875, 391]]<|/det|>
+462 8. Friedrich, M. et al. The 12,460-Year Hohenheim Oak and Pine Tree-Ring Chronology from Central Europe—A Unique Annual Record for Radiocarbon Calibration and Paleoenvironment Reconstruction. Radiocarbon 46, 1111–1122 (2004).  
+
+<|ref|>text<|/ref|><|det|>[[52, 400, 848, 435]]<|/det|>
+465 9. Salzer, M. W. & Hughes, M. K. Bristlecone pine tree rings and volcanic eruptions over the last 5000 yr. Quat. Res. 67, 57–68 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[52, 444, 875, 479]]<|/det|>
+467 10. Nicolussi, K. et al. A 9111 year long conifer tree-ring chronology for the European Alps: a base for environmental and climatic investigations. The Holocene 19, 909–920 (2009).  
+
+<|ref|>text<|/ref|><|det|>[[52, 488, 866, 540]]<|/det|>
+469 11. Griggs, C., Degaetano, A. & Newton, M. A regional high-frequency reconstruction of May – June precipitation in the north Aegean from oak tree rings, A. D. 1089 – 1989. Int. J. Climatol. 27, 1075–1089 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[52, 549, 870, 620]]<|/det|>
+472 12. Nicolussi, K., Matuschik, I. & Tegel, W. Klimavariabilität und Siedlungsdynamik am Beispiel der Feuchtbodensiedlungen im Raum Oberschwaben, Bodensee und Nordostschweiz 4400- 3400 BC. in Dendro – Chronologie – Typologie – Ökologie. Festschrift für André Billamboz zum 65. Geburtstag (eds. Bleicher, N., Schlichtherle, H., Gassman, P. & Martinelli, N.) 61–77 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[52, 628, 875, 664]]<|/det|>
+476 13. Miyake, F., Nagaya, K., Masuda, K. & Nakamura, T. A signature of cosmic-ray increase in 774-775 from tree rings in Japan. Nature 486, 240–242 (2012).  
+
+<|ref|>text<|/ref|><|det|>[[52, 672, 842, 707]]<|/det|>
+478 14. Miyake, F., Masuda, K. & Nakamura, T. Another rapid event in the carbon-14 content of tree rings. Nat. Commun. 4, 1748 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[52, 715, 820, 750]]<|/det|>
+480 15. Wacker, L. et al. Radiocarbon Dating to a Single Year by Means of Rapid Atmospheric 14C Changes. Radiocarbon 56, 573–579 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[52, 759, 803, 794]]<|/det|>
+482 16. Dee, M. W. & Pope, B. J. S. Anchoring historical sequences using a new source of astrochronological tie-points. Proc. R. Soc. A Math. Phys. Eng. Sci. 472, 20160263 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[52, 803, 840, 838]]<|/det|>
+484 17. Brehm, N. et al. Tree-rings reveal two strong solar proton events in 7176 and 5259 BCE. Nat. Commun. 13, 1–8 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[52, 847, 875, 882]]<|/det|>
+486 18. Pearson, C. et al. Annual Variation in Atmospheric C between 1700 BC and 1480 BC. Radiocarbon 62, 939–952 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[52, 891, 878, 909]]<|/det|>
+488 19. Büntgen, U. et al. Tree rings reveal globally coherent signature of cosmogenic radiocarbon events
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[52, 88, 880, 925]]<|/det|>
+489 in 774 and 993 CE. Nat. Commun. 9, 3605 (2018).
+490 20. Park, J., Southon, J., Fahrni, S., Creasman, P. P. & Mewaldt, R. Relationship between solar activity and Δ14C peaks in AD 775, AD 994, and 660 BC. Radiocarbon 59, 1147–1156 (2017).
+491 21. Cliver, E. W., Schrijver, C. J., Shibata, K. & Usoskin, I. G. Extreme solar events. Living Rev. Sol. Phys. 19, 2 (2022).
+492 22. Usoskin, I. & Miyake, F. Introduction. in Extreme Solar Particle Storms - The hostile Sun (eds. Miyake, F., Usoskin, I. & Poluianov, S.) 1–3 (IOP Publishing, 2019). doi:10.1088/2514- 3433/ab404ach1.
+493 23. Uusitalo, J. et al. Solar superstorm of AD 774 recorded subannually by Arctic tree rings. Nat. Commun. 9, 3495 (2018).
+494 24. Mekhaldi, F. et al. Multiradionuclide evidence for the solar origin of the cosmic-ray events of AD 774/5 and 993/4. Nat. Commun. 6, 8611 (2015).
+495 25. Sakurai, H. et al. Prolonged production of 14C during the “660 BCE solar proton event from Japanese tree rings. Sci. Rep. 10, 660 (2020).
+496 26. Fouache, E. et al. Palaeogeographical reconstructions of Lake Maliq (Korça Basin, Albania) between 14,000 BP and 2000 BP. J. Archaeol. Sci. 37, 525–535 (2010).
+497 27. Tegel, W., Muigg, B., Skiadaresis, G., Vanmoerkerke, J. & Seim, A. Dendroarchaeology in Europe. Front. Ecol. Evol. 10, 1–31 (2022).
+498 28. Mazzucco, N. et al. Multiproxy study of 7500-year-old wooden sickles from the Lakeshore Village of La Marmotta, Italy. Sci. Rep. 12, 14976 (2022).
+499 29. Bleicher, N. et al. Middens, currents and shorelines: Complex depositional processes of waterlogged prehistoric lakeside settlements based on the example of Zurich-Parkhaus Opéra (Switzerland). J. Archaeol. Sci. 97, 26–41 (2018).
+500 30. Hafner, A. et al. First Absolute Chronologies of Neolithic and Bronze Age Settlements at Lake Ohrid based on Dendrochronology and Radiocarbon Dating. J. Archaeol. Sci. Reports 38, 1–30 (2021).
+501 31. Reich, J. et al. A novel Structure from Motion-based approach to underwater pile field documentation. J. Archaeol. Sci. Reports 39, 103120 (2021).
+502 32. Chrysostomou, P., Jagoulis, T. & Mäder, A. The ‘Culture of Four Lakes’: Prehistoric lakeside settlements (6th - 2nd mill. BC) in the Aminedon Basin, Western Macedonia, Greece. Archäologie Schweiz 38, 24–32 (2015).
+503 33. Westphal, T., Tegel, W., Heussner, K. U., Lera, P. & Rittershofer, K.-F. Erste dendrochronologische Datierungen historischer Hölzer in Albanien. Archaeol. Anz. (2010) doi:https://publications.dainst.org/journals/aa/82/4764.
+504 34. Naumov, G. Prähistorische Pfahlbauten im Ohrid-See, Republik Mazedonien. Plattform. Jahrb. des Vereins für Pfahlbau- und Heimatkd. e.V. 23/24, 201, 10–19 (2016).
+505 35. Karkanas, P. et al. Palaeoenvironments and site formation processes at the Neolithic lakeside settlement of Dispilio, Kastoria, Northern Greece. Geoarchaeology 26, 83–117 (2011).
+506 36. Δισηπλό. 7500 χρόνια μετά /Dispilio. 7500 years after (in Greek). (University Studio Press, 2002).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[52, 88, 880, 925]]<|/det|>
+37. Bolliger, M. et al. Dendroarchaeology at Lake Ohrid: 5th and 2nd millennia BCE tree-ring 29 chronologies from the waterlogged site of Ploča Mičov Grad, North Macedonia. 530 Dendrochronologia 79, 126095 (2023). 31 38. Maczkowski, A. et al. The Early Bronze Age dendrochronology of Sovjan (Albania): A first tree-ring 32 sequence of the 24th – 22nd c. BC for the southwestern Balkans. Dendrochronologia 66, 125811 33 (2021). 34 39. Kouli, K. & Dermitzakis, M. D. 11. Lake Orestiás (Kastoria, northern Greece). Grana 49, 154–156 35 (2010). 36 40. Kouli, K. Plant landscape and land use at the Neolithic lake settlement of Dispilió (Macedonia, 37 northern Greece). Plant Biosyst. - An Int. J. Deal. with all Asp. Plant Biol. 149, 195–204 (2015). 38 41. Ntinou, M. La paleovegetación en el norte de Grecia desde el Tardigliacar hasta el Atlántico : 39 formaciones vegetales, recursos y usos. (2002). 40 42. Ntinou, M. Палаюперибаллов каи авбрцвихес бразтнриотчес: Н авбраколойа сто лмвайо 41 Neolibikо окизмо сто Диспнли Касторцас / Palaeoenvironment and human activities: Wood 42 charcoal analysis at the neolithic lake-settlement at Dispilio, Kastoria (in Greek). Anaskamma 04, 43 45–60 (2010). 43. Chatzitoulousis, S. Н техволойа тоц ёулоо сто веолбико лмваю окизмо тоц Диспнлои 45 Касторцас / Woodworking technology at the Neolithic lakeside settlement of Dispilio, Kastoria (in 46 Greek, English summary). Anaskamma 10, 93–123 (2009). 44. Sofronidou, M. & Dimitriadis, S. Continuity and discontinuity in the pottery of the early phases of 48 neolithic Dispilio (In Greek, English summary). in A Century of Research in Prehistoric Macedonia 49 1912-2012, International Conference Proceedings, Archaeological Museum of Thessaloniki, 22-24 50 November 2012 (eds. Stefani, E., Merousis, N. & Dimoula, A.) 537–548 (Archaeological Museum of 51 Thessaloniki, 2014). 45. Voulgari, E. The anthropomorphism of human-like pots: in Bodies of Clay (eds. Schwarzberg, H. & 53 Becker, V.) 23–43 (Oxbow Books, 2017). doi:10.2307/j.ctt1vgw6th.6. 46. Facorellis, Y., Sofronidou, M. & Hourmouziadis, G. Radiocarbon Dating of the Neolithic Lakeside 55 Settlement of Dispilio, Kastoria, Northern Greece. Radiocarbon 56, 511–528 (2014). 47. Maczkowski, A., Bolliger, M. & Francuz, J. Wetland dendrochronology. An overview of the 57 prehistoric chronologies from the southwestern Balkans. in Prehistoric Wetland Sites of Southern 58 Europe. Archaeology, Chronology, Palaeoecology and Bioarchaeology (eds. Ballmer, A., Hafner, A. 59 & Willy, T.) (Cham: Springer (in press), 2023). 48. Pearson, C. et al. Securing timelines in the ancient Mediterranean using multiproxy annual tree- 61 ring data. Proc. Natl. Acad. Sci. 201917445 (2020) doi:10.1073/pnas.1917445117. 49. Touchan, R., Akkemik, Ü., Hughes, M. K. & Erkan, N. May–June precipitation reconstruction of 63 southwestern anatolia, Turkey during the last 900 years from tree rings. Quat. Res. 68, 196–202 64 (2007). 50. Schweingruber, F. H. Trees and wood in dendrochronology : morphological, anatomical, and tree- 66 ring analytical characteristics of trees frequently used in dendrochronology. (Springer-Verlag, 67 1993). 51. Akkemik, Ü. & Yaman, B. Wood Anatomy of Eastern Mediterranean Species. (Kessel Publishing
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[52, 88, 880, 923]]<|/det|>
+569 House, 2012). 570 52. Schweingruber, F. H. Mikroskopische Holzanatomie. Formenspektren mitteleuropäischer Stamm- 571 und Zweighölzer zur Bestimmung von rezentem und subfossilem Material. Anatomie 572 microscopique du bois. Microscopic wood anatomy. (Eidgenössische Forschungsanstalt für Wald, 573 Schnee und Landschaft, 1990). 574 53. Matevski, V. et al. Forest vegetation of the Galičica mountain range in Macedonia = Шумската 575 вегетација на планината Галичица во Македонија = Gozdna vegetacija gorovja Galičica v 576 Makedoniji. (Biološki institut Jovana Hadžija ZRC SAZU; Makedonska akademija na naukite i 577 umetnostite;, 2011). doi:https://doi.org/10.3986/9789610502906. 578 54. Dida, M. State of Forest Tree Genetic Resources in Albania. 579 http://www.fao.org/3/j2108e/j2108e00.htm#TopOfPage (2003). 580 55. Bunn, A. et al. dplR: Dendrochronology Program Library in R; version 1.7.1. https://CRAN.R- 581 project.org/package=dplR. (2020). 582 56. Adamopoulos, S. & Koch, G. Wood structure and topochemistry of Juniperus Excelsa. IAWA J. 32, 583 67–76 (2011). 584 57. Hänninen, T. et al. Ultrastructural evaluation of compression wood-like properties of common 585 juniper (Juniperus communis L.). Holzforschung 66, 389–395 (2012). 586 58. Kuniholm, P. I., Newton, M. W. & Liebhart, R. F. Dendrochronology at Gordion. in The New 587 Chronology of Iron Age Gordion (eds. Rose, C. B. & Darbyshire, G.) 79–122 (University of 588 Pennsylvania Press, 2011). doi:https://doi.org/10.9783/9781934536551. 589 59. Manning, S. W. et al. Integrated tree-ring-radiocarbon high-resolution timeframe to resolve 590 earlier second millennium bce mesopotamian chronology. PLoS One 11, 1–27 (2016). 591 60. Abdrabou, A., Zidan, E., Nishisaka, A., Kurokochi, H. & Yoshimura, S. King Khufu’s Second Boat: 592 Scientific Identification of Wood Species for Deckhouse, Canopy, and Forecastle. Forests 13, 2118 593 (2022). 594 61. Hollstein, E. Mitteleuropäische Eichenchronologie : Trierer dendrochronologische Forschungen zur 595 Archäologie und Kunstgeschichte. (Mainz am Rhein : von Zabern, 1980). 596 62. Baillie, M. & Pilcher, J. A Simple Crossdating Program for Tree-Ring Research. Tree-ring Bull. 7–14 597 (1973). 598 63. Eckstein, D. & Bauch, J. Beitrag zur Rationalisierung eines dendrochronologischen Verfahrens und 599 zur Analyse seiner Aussagesicherheit. Forstwissenschaftliches Cent. 88, 230–250 (1969). 600 64. Bronk Ramsey, C., van der Plicht, J. & Weninger, B. ‘Wiggle Matching’ Radiocarbon Dates. 601 Radiocarbon 43, 381–389 (2001). 602 65. Ramsey, C. B. Bayesian analysis of radiocarbon dates. Radiocarbon 51, 337–360 (2009). 603 66. Reimer, P. J. et al. The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal 604 kBP). Radiocarbon 62, 725–757 (2020). 605 67. Szidat, S. et al. 14 C Analysis and Sample Preparation at the New Bern Laboratory for the Analysis 606 of Radiocarbon with AMS (LARA) . Radiocarbon 56, 561–566 (2014). 607 68. Sookdeo, A. et al. Quality Dating: A Well-Defined Protocol Implemented at ETH for High-Precision
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[55, 88, 884, 920]]<|/det|>
+608 14C-Dates Tested on Late Glacial Wood. Radiocarbon 62, 891- 899 (2020). 609 69. Wacker, L., Bollhalder, S., Sookdeo, A. & Synal, H.- A. Re- evaluation of the New Oxalic Acid 610 standard with AMS. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. 611 Atoms 455, 178- 180 (2019). 612 70. Kuitems, M. et al. Radiocarbon- based approach capable of subannual precision resolves the 613 origins of the site of Por- Bajin. 0, 8- 11 (2020). 614 71. Kuitems, M. et al. Evidence for European presence in the Americas in ad 1021. 601, (2022). 615 72. Bayliss, A. et al. IntCal20 Tree Rings: An Archaeological Swot Analysis. Radiocarbon 62, 1045- 1078 616 (2020). 617 73. Blaauw, M. IntCal: Radiocarbon Calibration Curves. R package version 0.3.1. https://CRAN.R- 618 project.org/package=IntCal. (2022). 619 74. R Core Team. R: A Language and Environment for Statistical Computing. https://www.r- 620 project.org/. https://www.r- project.org/ (2020). 621 75. Rathgeber, C. B. K., Cuny, H. E. & Fonti, P. Biological basis of tree- ring formation: A crash course. 622 Front. Plant Sci. 7, 1- 7 (2016). 623 76. Cuny, H. E. & Rathgeber, C. B. K. Xylogenesis: Coniferous Trees of Temperate Forests Are Listening 624 to the Climate Tale during the Growing Season But Only Remember the Last Words! Plant Physiol. 625 171, 306- 317 (2016). 626 77. Tasić, N. et al. Interwoven Strands for Refining the Chronology of the Neolithic Tell of Vinča- Belo 627 Brdo, Serbia. Radiocarbon 58, 795- 831 (2016). 628 78. Bayliss, A. & Whittle, A. What kind of history in prehistory? in Time and History in Prehistory (eds. 629 Souvatzi, S., Baysal, A. & Baysal, E.) 123- 146 (Routledge, 2018). doi:10.4324/9781315531854- 8. 630 79. Stratouli, G. et al. New excavations in Northwestern Greece: The Neolithic settlement of Avgi, 631 Kastoria. J. Greek Archaeol. 5, 63- 134 (2020). 632 80. Weiberg, E. et al. Long- term trends of land use and demography in Greece: A comparative study. 633 The Holocene 29, 742- 760 (2019). 634 81. Kotsos, S. Settlements and Landscape in the Neolithic of the Southern and Central Balkans. in 635 Making spaces into places : the north Aegean, the Balkans and western Anatolia in the Neolithic 636 (eds. Tasić, N. N., Urem- Kotsou, D. & Burić, M.) 105- 123 (2020). 637 82. Elezi, G. Sociocultural dimensions of production, use, and circulation of Late Neolithic pottery from 638 southern Balkans. (University of California, PhD Thesis, 2020). 639 83. Lespez, L. et al. The lowest levels at Dikili Tash, northern Greece: A missing link in the Early 640 Neolithic of Europe. Antiquity 87, 30- 45 (2013). 641 84. Gassner, S. et al. 20,000 years of interactions between climate, vegetation and land use in 642 Northern Greece. Veg. Hist. Archaeobot. 29, 75- 90 (2020). 643 85. Bonga, L. Late Neolithic pottery from mainland Greece, ca. 5,300- 4,300 B.C. (Temple University, 644 2013). 645 86. Halstead, P. Farming, material culture, and ideology: repackaging the Neolithic of Greece (and
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[55, 87, 883, 911]]<|/det|>
+646 Europe). Dyn. Neolit. Eur. Stud. honour Andrew Sherratt 131- 151 (2011). 647 87. Kotsakis, K. What Tells Can Tell: Social Space and Settlement in the Greek Neolithic. in Neolithic 648 Society in Greece (ed. Halstead, P.) 66- 76 (1999). 649 88. Malamidou, D., Tsirtsoni, Z. & Vaxevanopoulos, M. The emergence of metal use in Greek Eastern 650 Macedonia during the Neolithic period (late 6th- 5th millennia BC). Doc. Praehist. 49, 2- 21 (2022). 651 89. Kotsakis, K. Domesticating the periphery. New research into the Neolithic of Greece. Pharos 20, 652 41- 73 (2014). 653 90. Pappa, M. et al. The Neolithic site of Makriyalos, northern Greece: reconstruction of social and 654 economic structure of the settlement through comparative study of the finds. in Diet, Economy 655 and Society in the Ancient Greek World - Towards a Better Integration of Archaeology and Science. 656 Proceedings of the International Conference held at the Netherlands Institute at Athens on 22- 24 657 March 2010 (eds. Voutsaki, S. & Valamoti, S. M.) 77- 88 (Peeters, 2013). 658 91. Gkouma, M. & Karkanas, P. The physical environment in Northern Greece at the advent of the 659 Neolithic. Quat. Int. 496, 14- 23 (2018). 660 92. Giagkoulis, T. The Pile- field of the Neolithic Lakeside Settlement Anarghiri IXb (Aminedon, 661 Western Macedonia, Greece) and the Non- Residential Wooden Structures on the Periphery of the 662 Habitation. in Settling Waterscapes in Europe. The Archaeology of Neolithic and Bronze Age Pile- 663 Dwellings (eds. Hafner, A., Dolbunova, E., Mazurkevich, A., Pranckenaite, E. & Hinz, M.) 137- 155 664 (Propylaeum, 2020). doi:https://doi.org/10.11588/propylaeum.714. 665 93. Korkuti, M. Neolithikum und Chalkolithikum in Albanien = Neolithique et chalcolithique en Albania. 666 (Phillip von Zabern Verlag, 1995). 667 94. Gori, M. Along the Rivers and Through the Mountains: A revised chrono- cultural framework for 668 the south- western Balkans during the late 3rd and early 2nd millennium BCE. 669 Universitätsforschungen Zur Prähistorischen Archäologie, Band 268. (Dr. Rudolf Habelt GmbH, 670 2015). 671 95. Hughes, M. K. et al. Aegean Tree- Ring Signature Years Explained. Tree- Ring Res. 57, 67- 73 (2001). 672 96. Halstead, P. & Isaakidou, V. Pioneer farming in earlier Neolithic Greece. in Farmers at the Frontier 673 (eds. Gron, K. J., Sørensen, L. & Rowley- Conwy, P.) 77- 100 (Oxbow Books, 2020). 674 doi:10.2307/j.ctv13gvh1g.7. 675 97. Van Andel, T. H. & Runnels, C. N. The earliest farmers in Europe. Antiquity 69, 481- 500 (1995). 676 98. Speer, J. H. Fundamentals of Tree Ring Research. (University of Arizona Press, 2010). 677 99. Baillie, M. A Slice Through Time. Dendrochronology and Precision Dating. (Routledge, 1995). 678 100. Bunn, A. G. A dendrochronology program library in R (dplR). Dendrochronologia 26, 115- 124 679 (2008). 680 101. Crivellaro, A. & Schweingruber, F. H. Atlas of Wood, Bark and Pith Anatomy of Eastern 681 Mediterranean Trees and Shrubs : with a Special Focus on Cyprus. (Springer- Verlag Berlin 682 Heidelberg, 2013). doi:10.1007/978- 3- 642- 37235- 3. 683 102. Beck, H. E. et al. Present and future Köppen- Geiger climate classification maps at 1- km resolution. 684 Sci. Data 5, 180214 (2018).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[56, 88, 879, 130]]<|/det|>
+685 Camarero, J. J., Olano, J. M. & Parras, A. Plastic bimodal xylogenesis in conifers from continental Mediterranean climates. New Phytol. 185, 471–480 (2010).  
+
+<|ref|>text<|/ref|><|det|>[[56, 135, 880, 170]]<|/det|>
+687 Némec, M., Wacker, L., Hajdas, I. & Gägeler, H. Alternative methods for cellulose preparation for ams measurement. Radiocarbon 52, 1358–1370 (2010).  
+
+<|ref|>text<|/ref|><|det|>[[56, 177, 856, 212]]<|/det|>
+689 Zhang, Q. et al. Modelling cosmic radiation events in the tree-ring radiocarbon record Subject Areas: Author for correspondence: (2022).  
+
+<|ref|>text<|/ref|><|det|>[[56, 220, 857, 255]]<|/det|>
+691 Esper, J. Long-term tree-ring variations in Juniperus at the upper timber-line in the Karakorum (Pakistan). Holocene 10, 253–260 (2000).  
+
+<|ref|>text<|/ref|><|det|>[[56, 263, 840, 314]]<|/det|>
+693 Haneca, K., Katarina Čufar & Beeckman, H. Oaks, tree- rings and wooden cultural heritage: a review of the main characteristics and applications of oak dendrochronology in Europe. J. Archaeol. Sci. 36, 1–11 (2009).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[42, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[42, 92, 768, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[59, 130, 400, 283]]<|/det|>
+- SupTabT1Dispilio5259.xlsx- SupMatS1S3Dispilio5259.pdf- SupMatS4Dispilio5259.rar- CompetingintereststatementAM.pdf- EditorialpolicychecklistAM.pdf- RreportingsummaryAM.pdf
+
+<--- Page Split --->

preprint/preprint__983b461104a4714a4a7a2ab18500d0bb6f3e3f74a3dc5b5a4fe9b5961806ae4b/images_list.json

@@ -0,0 +1,62 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1. Proteomic analysis of cell-secreted proteins demonstrates a rich extracellular signaling environment along human fibroblast reprogramming. A) Schematics of the in-scale conventional and microfluidic setup (top), and comparison of reprogramming efficiency therein (bottom). Wilcoxon's test was used to assess differences among the conditions, \"P < 0.01. B) Experimenta design for proteomic experimental data collection. Proteomic data were obtained by tandem mass spectrometry analysis of conditioned media along the same reprogramming experiments. C) Principal component analysis of the 4542 proteins detected in at least one time point. Each sample of proteomic data refers to medium conditioned over a 48-hour period. D) Enrichment analysis within the Reactome database of the 555 proteins identified as secreted (Supplementary Table 1). Edges connecting different categories reproduce Reactome hierarchy relationship. Complete results are reported in supplemental Table 2. E) Hierarchical clustering of proteins identified in this study and belonging to the core ECM components at specific stages of embryo development. F) Hierarchical clustering of secreted proteins from the following enriched signalling pathways (according to Reactome database): Signaling by Interleukins (R-HSA-449147), Regulation of Insulin-like Growth Factor transport and uptake by Insulin-like Growth Factor Binding Proteins (R-HSA-381426), Signaling by PDGF (R-HSA-186797), Signaling by MET (R-HSA-6806834), Signaling by WNT (R-HSA-195721); playing a role as senescence-associated secreted proteins (51, and as ligands (53). Collagens and Laminins were excluded from these protein sets as they were plotted in (E). Only selected protein names discussed within the text or identified as most specific cluster markers in the subsequent single-cell RNA-seq analysis are shown.",
+    "footnote": [],
+    "bbox": [
+      [
+        170,
+        155,
+        660,
+        648
+      ]
+    ],
+    "page_idx": 22
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2. Single-cell RNA-seq analysis of human reprogramming cells unveils a dynamic somatic subpopulation involved in the expression of signalling and ECM related genes. A) Schematic representation of the experimental design for single-cell RNA-seq data collection. Human BJ fibroblasts were grown in Pluriton medium and daily transfected with OSKML mRNA. Starting from day 9, cells were grown in IPS Brew Medium till day 15. Samples were collected by stopping parallel experiments at day 0, 3 and every 48 hours. B) Force-Directed Layout Embedding (FLE) map showing the distribution of cells across time-points and C) identified clusters. D) Time-points enrichment for each cluster (left) and heatmap of Z-scored normalized counts, averaged by clusters, for key reprogramming related genes (right). NA cluster not shown. E) GSEA results for each cluster. Only significant results are shown. NES, Normalized enrichment score.",
+    "footnote": [],
+    "bbox": [
+      [
+        146,
+        105,
+        677,
+        408
+      ]
+    ],
+    "page_idx": 23
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3. Trajectory inference reveals different fates during reprogramming. Gene expression-based interaction analysis suggests an existing crosstalk between somatic and reprogramming cells through known and novel ligand-receptor couples. A) Matrisome and Late pluripotency enrichment scores shown along the FLE map. B) Monocle3 (black line) and WOT (colored dots) trajectory inferences are displayed on the FLE graph. Arrows point to the starting point (blue) and 4 end points (red) of the inferred trajectories. A representative scheme of the trajectories is shown on the top-right. C) Enrichment Score graph relative to the GSEA of SR2 cluster for senescence-associated secreted proteins (SASP50-53. Black lines on the x axis represent a match between the ranked list and the genes analyzed. NES, Normalized enrichment score. FDR, False Discovery Rate. D) Venn diagram representing the intersection between SASP genes and SR2 cluster marker genes and their relative gene expression, shown in A) heatmap of Z-scored normalized counts, averaged by clusters. Genes with * have been detected in secretome analysis. F) Schematic representation of ligand-receptor interactions hypothesized during reprogramming. Fibroblasts (D0, left) develop two fates: a somatic secretory phenotype (bottom) and induced pluripotency (top). Black arrows show the directionality of the examined interaction. G) Heatmap of z-scored standardized interaction scores for top ligand-receptor pairs. Selection criteria is described in Methods. H) Log2 proteomic expression relative to D1-D2 of SASP-related ligands among the top pairs at each time-point. I-K) HGF and MET gene expression profiles (log2 CPM) are shown in different reprogramming systems. I) In our data, they are displayed on the FLE map as fold change relative to HGF and averaged across the time course (bottom-left). J) In Liu et al., 2020, they are shown as averaged across their identified clusters and K) in Caccharielli et al., 2015, they are shown as mouse and human mean normalized expression at sampling day 8 (** BH-adjusted p-value < 0.01). L-N) NRG1 and ERBB3 gene expression profiles are shown in different reprogramming systems. L) In our data, they are displayed on the FLE map as fold change relative to NRG1 and averaged across the time course (bottom-left). M) In Liu et al., 2020, they are shown as averaged across their identified clusters and N) in Caccharielli et al., 2015, they are shown as mean FPKM across the time-course.",
+    "footnote": [],
+    "bbox": [
+      [
+        152,
+        106,
+        680,
+        504
+      ]
+    ],
+    "page_idx": 24
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4. Perturbation of STAT3 pathway components affect the efficiency of reprogramming. A) A schematic representation of HGF/c-MET/STAT3 signalling pathway. HGF binds the receptor c-MET, inducing its kinase catalytic activity. After trans-phosphorylation, c-MET starts a phosphorylation cascade that is in common with the LIF and IL6 pathways. It includes the Janus kinase 1 (JAK1) and ends in the phosphorylation and dimerization of STAT3. Active phosphorylated dimers can translocate to the cell nucleus where they act as transcription activators of target genes, such as pluripotency-related genes (Created with BioRender.com). B) STAT3 target expression correlates with MET transcription. In the FLE graph, green dots represent cells with positive enrichment scores for STAT3 target genes (Methods). Bigger circles summarize averaged HGF (left) and MET (right) gene expression in identified clusters. Significant inter-cluster HGF-MET interactions are displayed (arrows). Arrow thickness relates to the strength of the interaction. C) Top, representative images of expression of nuclear STAT3 and c-MET during reprogramming performed in microfluidics at day 6. Bottom, correlation between the expression intensity of nuclear STAT3, c-MET, and cell size obtained from experimental data shown on top. Data from n=61 cells (n=3 independent experiments). D) Left, reprogramming efficiency in microfluidics measured as the relative area occupied by NANOG+ colonies in cells upon inhibition of c-Met and JAK1 kinases using small molecules at day 12, compared to the ones treated with the vehicle (n=6 for vehicle, n=12 for JAK1 and n=7 for c-MET). ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***p < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. E) Left, reprogramming efficiency in microfluidics upon knock-down of STAT3 using siRNAs at day 12 (n=8 for scramble siRNA, n=11 for siSTAT3); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. F) Bottom, reprogramming efficiency in standard 24-well plates upon addition of HGF, IL-6 and soluble IL6 receptor (sIL6R), or NRG1 at day 9 (n=14 for control, n=19 for HGF, n=5 for IL6 + sIL6R, n=16 for NRG1); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001. Top, representative quantification pictures in standard 24-well plates assessed by immunostaining of NANOG and TRA-1-60. G) Bottom, reprogramming efficiency in standard 24-well plates upon temporally modulate addition of HGF, IL-6 and soluble IL6 receptor (sIL6R), and NRG1 at day 9 (n=14 for control, n=6 for HGF in the early phase and NRG1 in the late phase, n=4 for HGF + IL6 + sIL6R in the early phase and NRG1 in the late phase, n=4 for HGF+IL6 and sIL6R + NRG1 for the entire process); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001, ***P < 0.001. Top, representative quantification pictures in standard 24-well plates assessed by immunostaining of NANOG and TRA-1-60.",
+    "footnote": [],
+    "bbox": [
+      [
+        155,
+        108,
+        675,
+        360
+      ]
+    ],
+    "page_idx": 25
+  }
+]

preprint/preprint__983b461104a4714a4a7a2ab18500d0bb6f3e3f74a3dc5b5a4fe9b5961806ae4b/preprint__983b461104a4714a4a7a2ab18500d0bb6f3e3f74a3dc5b5a4fe9b5961806ae4b.mmd

@@ -0,0 +1,400 @@
+
+# Cellular population dynamics shape the route to human pluripotency  
+
+Nicola Elvassore ( \(\boxed{ \begin{array}{r l} \end{array} }\) nicola.elvassore@unipd.it) University of Padua https://orcid.org/0000- 0002- 7029- 6287  
+
+Onelia Gagliano University of Padova https://orcid.org/0000- 0002- 2571- 3176  
+
+Francesco Panariello Department of Medical Biotechnology and Translational Medicine, University of Milan https://orcid.org/0000- 0002- 0351- 5689  
+
+Camilla Luni University of Bologna https://orcid.org/0000- 0002- 1211- 9629  
+
+Antonio Grimaldi Telethon Institute of Genetics and Medicine  
+
+Silvia Angiolillo University of Padua  
+
+Wei Qin University of Padova  
+
+Anna Manfredi Next Generation Diagnostic, srl  
+
+Patrizia Annunziata Next Generation Diagnostic, srl  
+
+Shaked Slovin Telethon Institute of Genetics and Medicine  
+
+Lorenzo Vaccaro Telethon Institute of Genetics and Medicine  
+
+Sara Riccardo Next Generation Diagnostic, srl  
+
+Valentina Bouche Telethon Institute of Genetics and Medicine  
+
+Manuela Dionisi Next Generation Diagnostic, srl  
+
+Marcello Salvi Next Generation Diagnostic, srl  
+
+Sebastian Martewicz ShanghaiTech University https://orcid.org/0000- 0001- 6663- 301X
+
+<--- Page Split --->
+
+Manli Hu Shanghai Institute for Advanced Immunochemical Studies  
+
+Meihua Cui Shanghai Institute for Advanced Immunochemical Studies  
+
+Hannah Stuart University of Padova  
+
+Cecilia Laterza University of Padova https://orcid.org/0000- 0002- 3964- 3375  
+
+Giacomo Baruzzo University of Padova https://orcid.org/0000- 0001- 6129- 5007  
+
+Geoffrey Schiebinger University of British Columbia https://orcid.org/0000- 0002- 8290- 7997  
+
+Barbara Di Camillo University of Padova, Department of Information Engineering https://orcid.org/0000- 0001- 8415- 4688  
+
+Davide Cacchiarelli  
+
+Davide CacchiarelliTelethon Institute of Genetics and Medicine, Department of Translational Medicine, University of Naples Federico II, Scuola Superiore Meridionale (SSM), Università degli Studi di Napoli "Fede  
+
+## Article  
+
+Keywords:  
+
+Posted Date: November 10th, 2022  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 301720/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: Yes there is potential Competing Interest. O.G, Ca.L. and N.E. are co- inventors on patent applications describing the reprogramming and differentiation processes in microfluidics, application number PD2013A000220, IT UA20162645 and 102016000039189 and PCT/IB2017/052167. O.G. and N.E. are co- founders of Onyel Biotech Srl. Davide Cacchiarelli is founder, shareholder, and consultant of Next Generation Diagnostic srl.  
+
+Version of Record: A version of this preprint was published at Nature Communications on May 17th, 2023. See the published version at https://doi.org/10.1038/s41467- 023- 37270- w.
+
+<--- Page Split --->
+
+# Cellular population dynamics shape the route to human pluripotency  
+
+Onelia Gagliano \(^{1,2,8}\) , Francesco Panariello \(^{3,8}\) , Camilla Luni \(^{4,5,8}\) , Antonio Grimaldi \(^{3,8}\) , Silvia Angiolillo \(^{1,2}\) , Wei Qin \(^{1,2,4}\) , Anna Manfredi \(^{3,6}\) , Patrizia Annunziata \(^{3,6}\) , Shaked Slovin \(^{3}\) , Lorenzo Vaccaro \(^{3,8}\) , Sara Riccardo \(^{3,6}\) , Valentina Bouche \(^{3}\) , Manuela Dionisi \(^{3,6}\) , Marcello Salvi \(^{3,6}\) , Sebastian Martewicz \(^{4}\) , Manli Hu \(^{4}\) , Meihua Cui \(^{4}\) , Hannah Stuart \(^{1,2}\) , Cecilia Laterza \(^{1,2}\) , Giacomo Baruzzo \(^{7}\) , Geoffrey Schiebinger \(^{8}\) , Barbara Di Camillo \(^{7,9,10}\) , Davide Cacchiarelli \(^{3,11,*}\) , Nicola Elvassore \(^{1,2,4,12,*}\) .  
+
+\(^{1}\) Dept. of Industrial Engineering, University of Padova, Padova, Italy \(^{2}\) Veneto Institute of Molecular Medicine (VIMM), Padova, Italy \(^{3}\) Telethon Institute of Genetics and Medicine (TIGEM), Armenise/Harvard Laboratory of Integrative Genomics, Pozzuoli, Italy \(^{4}\) Shanghai Institute for Advanced Immunochemical Studies (SIAIS), ShanghaiTech University, Shanghai, China \(^{5}\) Dept. of Civil, Chemical, Environmental and Materials Engineering (DICAM), University of Bologna, Bologna, Italy \(^{6}\) Next Generation Diagnostic srl, Pozzuoli, Italy \(^{7}\) Dept. of Information Engineering, University of Padova, Padova, Italy \(^{8}\) Dept. of Mathematics, University of British Columbia, Vancouver, Canada \(^{9}\) Department of Comparative Biomedicine and Food Science, University of Padova, Padova, Italy \(^{10}\) CRIBI Biotechnology Center, University of Padova, Padova, Italy \(^{11}\) Department of Translational Medicine, University of Naples Federico II, Naples, Italy \(^{12}\) Stem Cell and Regenerative Medicine Section, GOS Institute of Child Health, University College London, London, UK \(^{8}\) equal contribution: Onelia Gagliano, Francesco Panariello, Camilla Luni & Antonio Grimaldi \(^{*}\) equal lead contribution: Davide Cacchiarelli, Nicola Elvassore  
+
+correspondence to: d.cacchiarelli@tigem.it; nicola.elvassore@unipd.it;  
+
+## Abstract  
+
+Human cellular reprogramming to induced pluripotency is still an inefficient process and this has long hindered the study of the role of critical intermediate stages. We take advantage of high efficiency reprogramming in microfluidics and temporal multi- omics to identify and resolve distinct sub- populations and their interactions. The combination of secretome analysis and single- cell transcriptomics shows functional extrinsic pathways of protein communication between reprogramming sub- populations and the re- shaping of a permissive extracellular environment. We pinpointed the HGF/MET/STAT3 axis as a potent enhancer of reprogramming, which acts via HGF accumulation within the confined system of microfluidics, and in conventional dishes needs to be supplied exogenously to enhance efficiency.  
+
+Our data integrate the notion of human cellular reprogramming as a transcription factor- driven process with the concept that it is deeply dependent on extracellular context and cell population determinants.
+
+<--- Page Split --->
+
+## Introduction  
+
+The discovery of human induced pluripotent stem cells (hiPSCs) has emphasized the function of transcription factors in controlling cell identity, overlooking the role of cell- extrinsic signals. The reprogramming of somatic cells into hiPSCs is paradigmatic of a transcription factor- driven change of cell identity in three distinct and well- defined phases: cells exit a somatic state, transition through a transgene- dependent promiscuous transcriptional and epigenetic state, finally establish a self- renewing pluripotent identity. Several studies have established hallmarks and roadmaps of hiPSC formation2- 6, and new technological advancements, such as single- cell analyses, further refined our understanding of the reprogramming process7,8. These works better characterized initial and final stages of human reprogramming in detail but connected with intermediate stages via hypothetical and more uncertain trajectories. Whilst there is a body of literature describing reprogramming trajectory in mouse9- 11, the fine dynamics of human reprogramming intermediates, which constitute the bottleneck of the process, remain largely unexplored due to the complexity of recognizing and selecting rare phenotypes that will evolve into a hiPSC fate.  
+
+It is thought that individual cells during reprogramming evolve with a smooth progression under a selective pressure that results into dominant "elite" clones12- 15. It has been recently suggested that reprogramming of murine cells may also depend on population dynamics through cell- nonautonomous mechanisms in a context- dependent manner, i.e. mediated by cell- secreted factors10,16,17. Consistently with this hypothesis, we recently reported that the efficiency of reprogramming of human somatic cells to hiPSCs can be dramatically improved in a microfluidic confined environment18, which- enhances the accumulation of secreted factors19,20 and sustains the acquisition of both primed18,21 and naive human pluripotency22.  
+
+In line with this, we hypothesize that during human cellular reprogramming, specific subpopulations control fate decisions towards pluripotency by cell- extrinsic factors. We envision that the communication between distinct intermediate sub- populations and their shared extracellular environment lying in- between contributes to shaping the route to pluripotency.  
+
+We take advantage of reprogramming in microfluidics to have a high efficiency within a confined environment (Fig. 1A), where secreted signals are accumulated, and distinctive intermediate subpopulations can be effectively captured and characterized.  
+
+An integrated temporal multi- omic profiling during reprogramming reveals finely regulated dynamics of secreted proteins accumulating in the extracellular space and a cellular heterogeneity arising during intermediate stages of reprogramming. We investigate how these complex population dynamics are modulated by extrinsic signals and how this facilitates the prompt acquisition of pluripotency.  
+
+## Results  
+
+Secretoe analysis during reprogramming. To investigate the dynamic changes of cell- secreted proteins during reprogramming of human fibroblasts, tandem mass spectrometry (LC- MS/MS) was performed on conditioned media pooled from microfluidic channels every 2 days19- 21 (Fig. 1B). High efficiency reprogramming of human fibroblasts was achieved in microfluidics with daily transfections of non- modified messenger RNAs (mRNAs) encoding for OCT4, SOX2, KLF4, MYC, LIN28, NANOG18 (Extended Data Fig. 1A and Methods). A medium that contains only 3 types of proteins (FGF2, INS, TF)18 preserved the high efficiency of microfluidic reprogramming (Extended Data Fig. 1B) while enabling high- resolution and accurate detection of cell- secreted proteins (Extended Data Fig. 1C- E). Protein tagging allowed us to obtain a relative quantification of each protein along the process (Extended Data Fig. 1A).  
+
+We quantified 4542 proteins, identified in either two (19%) or three replicates (81%). A principal component analysis showed that the samples follow a reprogramming temporal trajectory with high reproducibility between replicates (Fig. 1C). To get rid of intracellular proteins potentially released by dead cells, we specifically selected 555 proteins known to be secreted (Supplementary Table 1). We classified the identified categories into the two broad groups of extracellular matrix (ECM)- and soluble signal- related functional annotations (Fig. 1D and Supplementary Table 2).  
+
+Embryonic ECM accumulates during reprogramming. Many ECM- related categories were highly significant, including ECM deposition, degradation and remodeling, and both integrin- and non
+
+<--- Page Split --->
+
+integrin- mediated cell- ECM interactions (Fig. 1D left). A previous RNAi screen also identified the critical role of cell adhesion in human reprogramming, highlighting the role of intercellular factors needed for filament assembly, branching, and disassembly<sup>5</sup>.  
+
+In our data, we found an overall increasing trend of ECM- related protein accumulation, with different ECM components exhibiting distinct dynamics (Fig. 1E). These dynamic changes started already at days 3- 4 (SPP1, COL4A1/2, SPARC), in some cases at days 5- 6 (LAMC1), or even later (COL18A1). We wondered whether the observed global changes somehow resembled embryo development stages. To address this question, we selected the ECM proteins in our data that were previously reported to be expressed at mRNA level at different stages of human embryo development<sup>23</sup>. The concentration dynamics of these proteins in our system showed the progressive establishment of an ECM that recapitulates the one deposited at the stage of the late inner cell mass (Fig. 1E, Supplementary Table 3). In conclusion, our data support the idea that during reprogramming, not only fibroblasts are converted to a primed pluripotent phenotype, but also the extracellular context is shaped accordingly.  
+
+Dynamics of extrinsic regulatory signals during reprogramming. Our secreted proteins were enriched in several other processes, demonstrating that this extracellular environment is rich in regulatory signals. Fig. 1D (right) shows a selection of signalling pathways enriched within the Reactome database (see Supplementary Table 2 for full results). Among receptor tyrosine kinase pathways, PDGF and WNT have already been shown to be implicated in embryo development and reprogramming<sup>24,25</sup>. We also identified the MET pathway as a link between cell- cell communication via soluble environment, and cell- ECM interaction via PTK2 (also known as FAK) adhesion. Moreover, the regulation of insulin- like growth factor (IGF) pathway through IGF binding proteins (IGFBP) was significantly enriched, in line with previous studies<sup>26</sup>.  
+
+Looking at the temporal profiles of enriched signalling pathway proteins and ligands (Fig. 1F, Supplementary Table 4), we found a progressive accumulation of proteins that were previously shown to play a role in mouse cell- non- autonomous reprogramming regulation: some senescence- associated secreted proteins (SASP), such as CXCL1 (also known as Gro- \(\alpha\) ), CXCL8, CCL2, IL6<sup>27</sup>; YAP- target CCN1, also known as CYR6128; inflammatory cytokines, such as IL6/11/19, CSF1/2/3, LIF17. We found that JAK- STAT pathway, downstream of interleukin signalling, was also significantly differentially expressed at transcriptomic level between freshly- derived microfluidic hiPSC colonies and the same colonies after 3- passage expansion in conventional wells<sup>21</sup> (Extended Data Fig. 1F). We conclude that secreted proteins follow precise dynamics during reprogramming and encompass a number of potential regulators of autocrine/paracrine signalling, including those involved in ECM- mediated and soluble communication.  
+
+Resolving cell population heterogeneity during reprogramming. In order to identify cellular subpopulations arising during human somatic cell reprogramming and their role in secretion of signalling molecules, we performed high- throughput single- cell RNA sequencing (scRNA- seq). Cells were collected before the first transfection (D0), 3 days after transfection (D3) and then every 2 days (D5- D15) during the high- efficiency reprogramming of human fibroblasts in microfluidics (Fig. 2A). We generated scRNA- seq libraries from independent captures for at least two replicates per time- point, collecting altogether more than 40,000 single- cell transcriptomes. Dataset quality control, filtering and down- sampling to 2,500 cells per time point produced 20,000 high- quality single- cell transcriptome profiles, with a median of 5464 genes detected in each single cell for a total of 12,932 total detected genes (Extended Data Fig. 2A- B and Methods). Data dimensionality was reduced using the Force- Directed Layout Embedding (FLE) algorithm, which we previously described and applied to mouse reprogramming<sup>10</sup>. The resulting FLE diagram (Fig. 2B) illustrates the expression profile of each cell as a point in a Euclidean space where cells are grouped based on their transcriptional similarity. We observed high homogeneity of the fibroblasts population at day 0 (D0) and higher heterogeneity thereafter. To characterize this heterogeneity, we clustered cells using an unsupervised community detection algorithm<sup>29</sup>, that resulted in 12 clusters (Fig. 2C). We then took advantage of our formerly defined reprogramming- associated gene signatures<sup>2</sup> to annotate these clusters (Extended Data Fig. 2C, Fig.2E left). 7 clusters showed high expression of somatic genes ("Somatic- Related" clusters, SR), whereas 4 clusters were highly enriched by the developmental signature ("Developmental- Related" clusters, DR). Finally, a residual cluster was not enriched by
+
+<--- Page Split --->
+
+either of those signatures, thus we named it "NA". As expected, SR clusters included non- transfected fibroblasts (SR1) and cells captured at earlier days (SR2- 5), while DR clusters were enriched by cells collected at later time points (from D9 to D15) and highly cycling (Fig. 2D and Extended Data Fig. 2C- D). However, more than \(97\%\) of SR6 and SR7 cells were sampled from day 11 (Fig. 2D) and were characterized by low but detectable expression of embryonic genes (e.g. POU5F1, LEFTY2) and were negative for NANOG, indicating reshaping of fibroblast identity but at the same time inefficient acquisition of pluripotency. Furthermore, these cells are in the G0/G1 phase of the cell cycle, thus confirming their somatic nature and suggesting peculiar identity in the reprogramming timeline (Fig. 2D and Extended Data Fig. 2D). Despite their developmental features, DR4 cells also did not express NANOG, while showing high and very specific transcriptional levels of mesendoderm genes (e.g. CER1, EOMES), suggesting a possible similarity with a differentiating stage. Whilst DR clusters appear to contain the productively reprogramming cells, the role of the SR clusters is less clear (Fig. 2E left). To address the role of SR clusters we perform Gene Set Enrichment Analysis (GSEA) using the secreted proteins previously identified and some gene signatures that were found enriched in the proteomic analysis (Fig. 2E right and Supplementary Table 5). Surprisingly, the secreted proteins detected by mass spectrometry appear to be transcribed by the cells in the SR clusters, except for SR3 that might not be involved in the secretory phenotype. These results highlight the presence of an unproductive somatic fate, whose role is to express and secrete those factors that we found to be shaping the extracellular environment during reprogramming and that have been found to characterize later stages of embryonic development.  
+
+Signalling contributions from different cellular subpopulations. Among all the gene sets analyzed, Matrisome \(^{30}\) and Late pluripotency \(^{2}\) associated genes were found to best describe the phenotype of D13- 15 endpoints (Fig. 3A). Therefore, we decided to computationally investigate the routes linking such states to the somatic start- point by applying Waddington Optimal Transport (WOT) \(^{10}\) (Fig. 3B and Extended Data Fig. 3A). Results showed a common path until day 5 (D5), after which cells started to exhibit different trajectories (Fig. 3B, Extended Data Fig. 3B). We validated these findings through an unsupervised pseudotime- based approach using Monocle \(^{31,32}\) , which not only confirmed the bifurcation at day 7 (D7) leading to endpoints inside SR7 matrisomal and DR3 pluripotent clusters, but also introduced two additional outcomes inside DR4 and SR2, respectively (Fig. 3B and Extended Data 3C). While the mesendodermal nature of DR4 was previously assessed, we focused on the characterization of SR2. GSEA using common pathways (Methods) revealed the enrichment for terms related to signalling molecules (Supplementary Table 6), therefore, we hypothesized that this cluster might be implicated in the secretion of the ligands detected in the medium. Indeed, most of them were significantly enriched, with SASP having the highest enrichment score (Fig. 3C and Extended Data Fig. 3D). We found SASP genes are highly expressed and specific of this cluster (Supplementary Table 7), such as cytokines (CXCL1, IL1B, CXCL8), metalloproteases (MMP1, MMP3), HGF and its activators, PLAU and PLAUR (Fig. 3D- E). Notably, almost all of them were detected by LC- MS/MS with some (CXCL1, CXCL8, CCL2, SPP1, PLAU) being the first to be accumulated in the medium (Fig. 1F).  
+
+In conclusion, we were able to define human somatic reprogramming as a process consisting of two major outcomes, matrisomal and pluripotent, deriving from the same starting cells which bifurcate around day 7 (D7). Moreover, among matrisomal somatic cells, we identified and characterized an early sub- population of cells which contributes to the expression and secretion of SASP- related signalling molecules.  
+
+Reprogramming fates interact through different ligand- receptor pairs. To rationally understand whether somatic subpopulations arising during reprogramming are actively involved in the population cross- talk with productive reprogramming intermediates, we developed a ligand- receptor interaction analysis from the cells laying on the somatic trajectory towards the reprogramming ones (Fig. 3F). Using the previously identified secreted proteins (Fig. 1) that fall in the list of experimentally validated ligand- receptor couples \(^{33}\) , we restricted the number of putative interactors involved in subpopulation crosstalk to a set of 82 pairs (Extended Data Fig. 3E, Supplementary Table 8 and Methods). We were able to identify a standardized interaction score (sIS) by leveraging the gene expression trends of ligands along the matrisome route and of receptors along the path to pluripotency (Methods).
+
+<--- Page Split --->
+
+The results showed that almost every ligand-receptor pair had a significant sIS in at least one time- point (Extended Data Fig. 3F and Supplementary Table 9). Moreover, when looking at the couples with the greatest scores, we observed many ligands involved in signalling cascades which are already known to be associated with pluripotency maintenance, such as Wnt, Tgfβ and Inhb signalling<sup>25,34,35</sup> (Fig. 3G). These results were overall confirmed by a complementary unbiased approach, based on an alternative interaction score computed as a function of the absolute ligand and receptor expression levels and their \(\log_2\) fold change with respect to the average expression level across all time points (Extended Data Fig. 3G, Supplementary Table 10 and Methods).  
+
+Among these interactors, 8 ligands were related to SASP, of these 4 were soluble and highly dynamic in both transcriptomic and proteomic data: SPP1, INHBA, NRG1 and HGF (Fig. 3H). As INHBA is a known pluripotency regulator<sup>34</sup>, and SPP1 is the major HGF- regulated gene<sup>36</sup>, we focused our analyses on HGF and NRG1. The HGF- MET interaction occurred at early time- points of the reprogramming (Fig. 3I) with HGF expressed by cluster SR2 and SR5 and its receptor MET expressed by cluster DR1. Both HGF and MET were highly expressed in the early intermediate stages and decreased in the later time points, suggesting a role in the reprogramming intermediates. Thus, we explored whether the same HGF- MET dynamics was present in a conventional (i.e., Petri dish) human reprogramming approach<sup>7</sup> and not strictly related to the microfluidic environment. scRNA- seq data exploration, using authors- defined clusters<sup>7</sup>, showed that the cluster noRepro1, enriched for SR signatures (Extended Data Fig. 3H), expressed high levels of HGF (Fig. 3J), whereas MET expression was observed in the mixed intermediate cluster, overlooked by the authors (Fig. 3J). Remarkably, the analysis of RNAseq data from reprogramming of secondary human fibroblasts cultured on mouse embryonic fibroblast feeder (MEF)<sup>2</sup>, showed the expression of HGF only from MEFs while MET was upregulated in human cells undergoing reprogramming at day 8 (OSKM - Fig. 3K). These results showed a common behaviour of HGF vs MET expression in the early phase of the reprogramming, being expressed by matrisome producing/supporting cells and reprogramming intermediates respectively, regardless of reprogramming approach and culture system. On the other hand, the NRG1- ERBB3 interaction showed higher sIS between clusters along the same developmental trajectory in a sequential fashion: NRG1 is expressed by DR clusters at earlier stages (until D9), while its receptor, Erb- B2 Receptor Tyrosine Kinase 3 (ERBB3), is expressed by late DR clusters (starting from D7) (Fig. 3L). The same information can be retrieved from Liu et al., 2020<sup>7</sup> and Cacchiarelli et al., 2015<sup>2</sup>, observing the sequential expression of NRG1 then ERBB3 only along the reprogramming intermediates, with NRG1 decreasing halfway during reprogramming route, ERBB3 increasing from halfway, and a central timeframe of co- presence (Fig. 3M- N). Therefore, as NRG1- ERBB3 expression occurs only along the reprogramming trajectory, we did not get significant results when comparing MEFs versus human reprogramming intermediates from our human secondary system (data not shown)<sup>2</sup>. Altogether, these findings suggest a crosstalk between cell subpopulations, with an active role of non- pluripotent cells in supporting the route of other cells to pluripotency. We demonstrated that such non- pluripotent cells can be part of the same (i.e. NRG1 and ERBB3 both expressed during DR trajectory to pluripotency) or different trajectories (i.e. HGF ligand expressed by SR trajectory towards matrisome vs MET receptor expressed by DR trajectory towards pluripotency).  
+
+HGF- MET crosstalk functionally sustains the acquisition of pluripotency through STAT3. Considering the results from the ligand- receptor analyses, we then asked whether the HGF- MET interaction has a functional role in the progression of intermediate states towards pluripotency. HGF is a growth factor involved in many cell functions and it is mostly secreted by mesenchymal cells, while acting on epithelial ones<sup>37</sup>. In our reprogramming, it is biologically active as its activator complex PLAU/PLAUR was also found in the secreted medium (Fig. 1F). On the other hand, MET is a tyrosine kinase receptor activated by its ligand HGF. This binding induces MET catalytic activity and results in downstream initiation of multiple pathways, including STAT3 direct phosphorylation or via Janus kinase 1 (JAK1 - Fig. 4A). This activation axis is shared with other two ligands (i.e. LIF and IL6), known to be involved in murine pluripotency (Fig. 4A)<sup>16,38</sup>. To test the STAT3 pathway involvement in our reprogramming setup, we investigated its activation throughout the reprogramming process in microfluidics.
+
+<--- Page Split --->
+
+First, HGF and MET were differentially expressed by SR (higher HGF) and DR (higher MET) clusters and came up as early interactors in a cluster- based interaction analysis (circles and arrows in Fig. 4B). Furthermore, STAT3 nuclear target transcriptional enrichment<sup>39</sup> revealed their activation from day 5, along the reprogramming route (green dotted cells in Fig. 4B), in agreement with MET signalling activity. Finally, at the protein level, we observed STAT3 nuclear localization (indicative of STAT3 activation) during intermediate days (D4, D7) and at the end of the process (D12 - Extended Data Fig. 4A).  
+
+To give further evidence, we then investigated the localization of MET and STAT3 at day 6. We found that cells of smaller size (undergoing the mesenchymal- to- epithelial transition) show the highest intensity of both c- MET and nuclear STAT3 (Fig. 4C). Finally, we separately inhibited two kinases along the STAT3 axis, MET and JAK1, using small molecules and assessed reprogramming efficiency by immunostaining analysis of NANOG at day 12. Consistent with our hypothesis, we observed a significant loss of reprogramming efficiency upon inhibition of STAT3 (Fig. 4D). These data were confirmed by a direct knock- down of STAT3 mRNA using specific siRNA, that efficiently reduced both STAT3 nuclear localization in all cells at day 6 (Supplementary Fig. 4B) and reprogramming efficiency at day 12 (Fig. 4E).  
+
+Lastly, we tested whether the addition of signalling molecules was capable of further improving reprogramming yield in conventional culture systems that are otherwise far less efficient than microfluidic systems (Fig. 1A). For this purpose, we selected molecules that were found dynamically in the secretome analysis or involved in cell- cell interactions (e.g. HGF, IL6 and NRG1). Because reprogramming efficiency in microfluidic has reached saturation, we did not observe any significant difference, neither in efficiency nor in STAT3 nuclear localization, upon addition of signalling molecules (data not shown).  
+
+However, in conventional culture (i.e. Petri dishes), we saw a significant increase of about 2- fold in reprogramming efficiency in terms of relative TRA- 1- 60+/NANOG<sup>+</sup> area when medium was supplemented with either HGF, IL6 and its soluble receptor (sIL6R) to activate STAT3 signaling<sup>16</sup>, and NRG1 throughout the reprogramming process (Fig. 4F). Consistent with the idea that multiple signals are involved in the first phase of reprogramming and the second phase of hiPSCs stabilization, secretome and single- cell RNA sequencing data showed more accumulation of HGF and IL6 in early phases of the reprogramming process, while NRG1 came out at later stages. To mimic this timing, we added HGF alone or with IL6/sIL6R in the first half, and NRG1 in the second half. This resulted in a further increase in the reprogramming efficiency up to three folds (Fig. 4G). However, when supplementing the medium with HGF, IL- 6, sIL- 6R and NRG1 together, we were able to reach the highest efficiency (i.e. 5- fold over controls), thus suggesting that the combination of specific signalling pathways further boosts hiPSCs formation (Fig. 4G).  
+
+## Discussion  
+
+Our integrative approach of secretome and single- cell transcriptomic analyses revealed a previously unappreciated crosstalk between subpopulations during the intermediate stages of human reprogramming. Whilst population heterogeneity was also described in recent papers, both in mouse<sup>9- 11</sup> and human<sup>4,7,8</sup>, these works reported the formation of distinctive cell clusters and diversification of pluripotent trajectories, viewing the unproductive/refractory subpopulations as a "problem" or limitation in the process. Instead, here we highlight the crucial role of reprogramming intermediates and the positive contribution of non- pluripotent clusters as actively supporting and shaping the route of the reprogramming cells towards a hiPSC identity.  
+
+The efficiency of human somatic cell reprogramming heavily relies on the successful transient accessibility and overcoming of specific intermediate stages but, given the generally low reprogramming efficiency, these stages have been hard to identify. Few strategies were previously adopted to capture human intermediate reprogramming- committed subpopulations such as cell sorting<sup>3,4</sup> and secondary reprogramming systems<sup>2</sup>.  
+
+Supported by the microfluidic culture system, we took a step further through the unbiased identification of the reprogramming subpopulation trajectories and interactions based on an integrative secreted proteome and scRNA- seq analysis. The former identified a number of secreted cytokines, growth factors and ECM- related proteins actually present in the extracellular space during reprogramming and contributing to establish an environmental signaling resembling the early
+
+<--- Page Split --->
+
+embryo basal lamina. scRNA- seq identified two main trajectories during reprogramming, with one almost exclusively responsible for secretory activity and one committed to reprogram. It was probably the reduced secretory activity of nascent hiPSCs or their low abundance that led previous works to overlook the role of the extracellular environment, failing to recognize nascent hiPSCs as a secretome target<sup>4</sup>. Recently, a few works suggested the potential for cross- population signalling in mouse reprogramming<sup>9- 11</sup> including the role of SASP and senescence<sup>27</sup>, but until now the molecular mechanisms and rationale behind human non- cell autonomous signaling remained unclear.  
+
+In this study, scRNA- seq could identify the putative subpopulation interaction dynamics during microfluidic reprogramming. In particular, the identification of the two distinctive trajectories, somatic secretory and reprogramming, was instrumental for scoring the putative ligand- receptor association responsible for the unidirectional support of the developmental trajectory towards pluripotency. Secretome analysis, performed here for the first time, could further reduce the dimensionality of the interactions, restricting them to those whose soluble ligand was actually detected as secreted at protein level. Only four ligands passed these restrictive selection criteria: INHBA, SPP1, NRG1 and HGF. INHBA was previously described<sup>34</sup>, SPP1 is downstream of the HGF pathway<sup>36</sup>, thus we focused on NRG1 and HGF, not previously implicated as reprogramming regulators. Interestingly, NRG1 signalling occurred within the reprogramming trajectory, while HGF involved population cross-talk from the secretory somatic to the reprogramming trajectory.  
+
+HGF is part of SASP, however it was not measured in Mosteiro et al., 2016<sup>27</sup> who instead identified IL6 in mouse cell reprogramming. Both HGF and IL6 signaling have STAT3 as a common effector, although via different receptors<sup>40</sup>, and other works reported a positive correlation between STAT3 activity and in vivo reprogramming efficiency<sup>16,41</sup>. In our human reprogramming systems<sup>2,18</sup>, IL6 was present both at transcriptional and proteomic level, however we could not detect its receptor, IL6R, in any subpopulation at any stage. Indeed, we were able to enhance reprogramming efficiency with IL6 only upon providing a soluble form of IL6R. The axis HGF/MET/STAT3 was first reported in cancer stemness and promotes the expression of pluripotent genes<sup>40</sup>. HGF- MET was demonstrated to take part in a mesenchymal- epithelial cross- talk<sup>42</sup>.  
+
+somatic trajectory, while its receptor, MET, especially present along the reprogramming one (Supplementary Fig. 4C- D).  
+
+We performed extensive experimental validation both in microfluidics and in conventional culture systems. Our loss of function data clearly show that MET activation and STAT3 signalling play an important role in preserving the efficiency of reprogramming, supporting the idea that HGF/MET/STAT3 may have a crucial role in the phenotypic conversion of developmental subpopulation towards pluripotency. Our gain of function experiments within the conventional culture system (i.e., Petri dish) support our hypothesis of the role of miniaturization in concentrating endogenous HGF and show the possibility of scaling up our findings for wider applicability. Whilst a positive role of STAT3 signalling has been extensively characterised during maintenance and induction of mouse naive pluripotency<sup>47</sup>, STAT3 signalling pathway is not active in primed human hiPSCs. It is therefore particularly striking that we find transient STAT3 activity to be of benefit during human reprogramming to primed hiPSC identity, and highlights that we must consider the environmental niche requirements of the intermediate states, which may differ from those of the endpoint target identity.  
+
+In our work, we followed an unbiased approach that supports the idea that the route to pluripotency can be broadened by cell- non- autonomous mechanisms. Paracrine signalling is established by highly regulated dynamics with multi- factorial contribution. We showed the use of HGF for gain of function during reprogramming in a conventional culture system, but this efficiency was amenable to further enhancement when multifactorial contributions were used. In particular, we used IL6 and soluble IL6R for a more effective downstream activation of STAT3. Moreover, we found that NRG1
+
+<--- Page Split --->
+
+contribute to enhance efficiency of hiPSC formation consistently with previous works, which upon binding ERBB2/ERBB3 receptors activates MAPK/ERK pathway and showed improved maintenance and passage of hiPSCs<sup>48,49</sup>.  
+
+## Conclusion  
+
+This work reports an overview of the environment- mediated subpopulation cross- talk during reprogramming and identifies some specific critical players. Important implications of our work are related to in vivo reprogramming, where environmental factors cannot be controlled but may affect potential applications. Moreover, strategies to reprogram in vitro fibroblasts from any donor with high efficiency are down the road and unlock the possibilities of using hiPSC as modeling systems for a large number of patients, including their use as diagnostic tools in predicting patient- specific genotype- phenotype associations in disease.  
+
+## Material and methods  
+
+## Cell culture  
+
+BJ cells (Miltenyi Biotec, 130- 096- 726), human newborn skin fibroblasts, were cultured with complete Dulbecco's modified Eagle's medium (DMEM, Thermo Fisher, 41965 or 11965), supplemented with \(10\%\) fetal bovine serum (FBS, Thermo Fisher, 10270106 or 10099- 141). Cells were maintained at \(37^{\circ}C\) in the presence of \(5\%\) \(\mathrm{CO_2}\) and periodically tested for mycoplasma contamination.  
+
+## Reprogramming of human fibroblasts to hiPSC colonies  
+
+We generated hiPSCs from human foreskin BJ fibroblasts using microfluidic technology as previously described (Gagliano et al 2019). For proteomic analysis, a total of \(10\mathrm{mRNA}\) transfections were performed using StemRNA- NM reprogramming kit (Stemgent, 00- 0076) and StemMACS mRNA transfection kit (Miltenyi, 130- 104- 463), in E7 medium, made from E6 medium (Thermo Fisher, A1516401) supplemented with \(100\mathrm{ng / mL}\) FGF2 (Peprotech, 100- 18B- 1000), switched to E8 medium (Stem Cell Technologies, 05990) from day 11. Whereas, for single- cell RNA- seq, \(8\mathrm{mRNA}\) transfections were performed in supplemented Pluriton medium (Stemgent, 00- 0070), switched to StemMACS iPSBrew XF medium (Miltenyi Biotec, 130- 104- 368) from day 9. Validation experiments were performed either in microfluidics according to single- cell RNA- seq protocol or in standard 24- well plates according to manufacturer's instructions; they were performed under suboptimal conditions to enhance reprogramming efficiency differences, and medium was supplemented with HGF \(100\mathrm{ng / mL}\) (Peprotech, 100- 39), IL- 6 \(50\mathrm{ng / mL}\) (Peprotech, 200- 06), IL- 6r \(10\mathrm{ng / mL}\) (Peprotech, 200- 06R), NRG1 \(100\mathrm{ng / mL}\) (R&D, 396- HB), during the whole process duration, according to the specified perturbation conditions using both Pluriton medium and Nutristem hPSC XF Medium (Biological Industries, 06- 5100- 01- 1A) supplemented with \(20\mathrm{ng / mL}\) FGF2. The loss of function experiments have been made in microfluidics supplementing the medium with Jak Inhibitor I 1uL (Millipore, 420097) and c- METi \(600\mathrm{uM}\) (Selleck, PF- 02341066) from day 1 to day 6. In STAT3 knock- out experiments, siRNA STAT3 10 uM (Qiagen, 1027416) or MOCK siRNA 10 uM (Qiagen, 1027284) was added in the transfection mix from day 1 to day 6. In all cases, the whole process was performed in a hypoxia incubator ( \(5\%\) \(\mathrm{O_2}\) , \(5\%\) \(\mathrm{CO_2}\) ) at \(37^{\circ}C\) .  
+
+## Sample preparation for LC-MS/MS  
+
+During reprogramming, at every medium change or reprogramming transfection, medium was collected in three replicates, pooling together the conditioned medium from the same 40 channels for each replicate. The media were stored at \(- 80^{\circ}C\) until prepared for proteomic analysis. After thawing, media from four collections (two consecutive days) were pooled together. For example, sample D1- D2 was conditioned by the cells within the microfluidic chamber from day 1 to day 3 mornings. 3kDa cut- off centrifugation membranes (Amicon Ultra \(0.5\mathrm{mL}\) , Ultracel 3K, Merck) were used for filter- aided sample preparation (FASP)<sup>55</sup>. Proteins were concentrated by centrifugation for 20 min at \(4^{\circ}C\) and \(14,000\mathrm{g}\) , then washed twice with a \(50\mathrm{mM}\) triethylammonium bicarbonate (TEAB, Thermo Scientific) buffer containing \(8\mathrm{M}\) urea (Sigma- Aldrich). Protein content was quantified by Pierce BCA Protein Assay Kit (Thermo Scientific). Each sample proteins were reduced for 60 min at \(56^{\circ}C\) with \(100\mathrm{mM}\) DTT (Sigma- Aldrich), and alkylated for 30 min at room temperature in the dark
+
+<--- Page Split --->
+
+with 55 mM iodoacetamide (Sigma-Aldrich). Samples were washed with 50 mM TEAB for three times. An equal amount of proteins for each sample was digested by trypsin (Promega) at \(37^{\circ}C\) for 16 h. Digested peptides were desalted by C- 18 spin column (Pierce) and vacuum dried. Then, labeling by 6- plex Tandem Mass Tag (TMT6, ThermoScientific) was performed according to manufacturer's instructions using 50 \(\mu g\) of peptides from each sample. The six- time point samples of each of the three replicates were pooled, then desalted and vacuum dried.  
+
+## Mass spectrometry analysis.  
+
+Mass spectrometry analysis. 25 pre- fractions were collected on UPLC (Agilent 1290) with high pH C18 column (2.1 mm x 30 mm). Before MS analysis, peptides were resuspended in \(10 \mu L\) of \(0.1\%\) formic acid. Thermo Fusion Mass Spectrometer coupled with Thermo EasyNLC1000 Liquid Chromatography was used to get peptides profiles. 90 min of LC- MS gradients were generated by mixing buffer A (0.1% formic acid in water) with buffer B (0.1% formic acid in 80% ACN in water) by different proportions. Using NSI as the ion source and Orbitrap as the detector, the mass scan Range was at 300- 1800 m/z, and the resolution was set to 120K. The MS/MS was isolated by Quadrupole and detected by Ion trap, whose resolution was set to 60K. The activation type was HCD.  
+
+## Proteomic bioinformatic analysis  
+
+Proteomic bioinformatic analysis  Peak list files were searched against UniProt human reference proteome (UP000005640) by MaxQuant (v. 1.6.3.4) \(^{56}\) . TMT6 modification and carbamidomethyl on cysteine were set as fixed modifications. The oxidation of methionine, acetylation of protein N- terminus, and phosphorylation (STY) were set as variable modifications. Peptide- spectrum matches (PSMs) were adjusted to 1% and then assembled further to a final protein- level false discovery rate (FDR) of 1%. Proteins not identified in at least 2 replicates in at least one time point were excluded from further analysis. Common contaminants (keratins and Bos taurus proteins) were also filtered out, for a final number of 4542 proteins identified. Missing values were imputed by the mean value of the other two replicates. TMT intensities were normalized according to BCA quantification to obtain a relative quantification proportional to protein concentration in culture. The distributions of the three replicates of TMT intensities were scaled by their respective medians. A principal component analysis (PCA) was performed in MATLAB R2017a (The Mathworks) using mean- centered TMT intensities. A list of secreted proteins was manually annotated by integrating the following resources: secreted proteins predicted by MDSEC as reported in Protein Atlas database \(^{57}\) (http://www.proteinatlas.org), secreted proteins from Table S1 in Gonzalez et al., 2010 \(^{58}\) ; a list of ligands from Gene Ontology- Molecular Function categories "cytokine activity", "growth factor activity", and "hormone activity", and senescence- associated secreted proteins (SASP) annotated from literature \(^{50 - 53}\) . Of the proteins identified in this study, only those secreted according to the criteria above were further studied, in order to avoid the proteins possibly derived from cell death. Proteins whose concentration was maximal only at the first time point (D1- D2 sample) were excluded from further analysis, as potential residual proteins from FBS used during fibroblast expansion. Functional enrichment analysis of Reactome pathways was performed using ReactomePA \(^{59}\) Bioconductor package. Reactome hierarchy was visualized using ClueGO \(^{60}\) within Cytoscape \(^{61}\) . Genes specific of different human embryonic stages were derived from a published single- cell RNA- seq study \(^{23}\) , of these core ECM genes were selected based on the annotations in Naba et al., 2012 \(^{30}\) . Proteins playing a role as ligands were taken from Ramilowski et al., 2015 \(^{33}\) . Hierarchical clustering with heat map data visualization was performed in MATLAB 2017a, using Euclidean distance and complete linkage.  
+
+## Sample preparation for single-cell RNA-seq  
+
+For each time- point, cells were detached using TrypLE- express (ThermoFisher, Gibco 12604). Harvested cells were then centrifuged at 300 g and resuspended at the final cell density of 100 cells/mL using a solution containing 40% KnockOut Serum Replacement (KSR, ThermoFisher, Gibco 10828) in DMEM. For each timepoint, two replicates were produced, each containing cells from 4 independent chips that were pooled together then divided in aliquots containing 5,000- 80,000 cells. Samples were cryopreserved in DMEM supplemented with 40 % KSR and 15 % DMSO and stored in liquid nitrogen.  
+
+scRNA- seq libraries were generated using one or two samples for each replicate. Briefly, each cryopreserved aliquot was thawed at 37 °C until a tiny ice crystal remained in solution. Then each
+
+<--- Page Split --->
+
+sample was diluted under gentle shaking by dropwise adding 10 volumes of DMEM supplemented with \(40\%\) KSR. Cells were washed twice using a washing buffer containing \(8\%\) MACS Running Buffer (Miltenyi, 130- 091- 221) in PBS. Cells were then resuspended in the washing buffer and filtered through a \(40\mu \mathrm{m}\) cell strainer (Biosigma, 010198Z). Cell viability and concentration were checked by visual inspection using Trypan Blue (Logos Biosystems, L12002).  
+
+Single- cell RNA seq libraries were produced according to 10X Single Cell 3' v2.0 standard protocol and sequenced on Novaseq 6000 (Illumina).  
+
+## Single-cell RNA-seq data pre-processing  
+
+scRNA- seq data pre- processing was performed using the cellranger software (v 2.2). Fastq files were generated using the Cellranger pipeline mkfastq using 10X standard Chromium barcode sequences. Alignment, filtering, barcode and UMI counting were performed using the Cellranger count pipeline. Human pre- built genome index has been applied (hg38 genome reference and GRCh38 annotation, including protein coding, linc and antisense RNAs). Each feature- barcode matrix from each independent sample was merged to build up the final dataset, containing 33,694 genes and 44,197 cells, then subjected to cells and genes filtering. Cells having less than 1,000 detected genes and with the mitochondrial associated reads percentage greater than \(10\%\) were filtered out. Furthermore, in order to have a homogenous sampling for each reprogramming day, the cell dataset was randomly subsampled to 2,500 cells per time point. The final dataset retained only those genes expressed in at least \(5\%\) of all the cells, leading to 12,932 total genes. Gene expression values were normalized to CPM (counts per million) and transformed to the \(\log_2\) scale using a pseudocount of 1. Finally, cell- cycle scores and, consequently, phases were assigned to each cell by Seurat's (v.3.1.5) CellCycleScoring function.  
+
+## Single-cell RNA-seq data visualization and clustering  
+
+To better visualize and characterize single cell data, high dimensionality was reduced. First, we computed the neighborhood graph using the function compute_neighborhood_graph from the Python package wot (v 1.0.5)10, using 50 neighbors and choosing the first 100 PCA components and the first 20 diffusion map components. The resulting 120 components were used as input to initialize the Force- Directed Layout Embedding (FLE) algorithm, using forceatlas2 (v 1.0.3) with 1000 iterations and reducing the space to 2 dimensions (FLE1 - FLE2). The same components were also applied to perform an unsupervised graph- based algorithm (louvain) using the FindNeighbours and FindClusters (resolution \(= 0.6\) ) functions in the Seurat (v.3.1.5) package. This step resulted in the identification of 12 clusters, annotated based on the enrichment of somatic and developmental signatures at the single- cell level (SR = somatic related; DR = developmental related; NA = not assigned) and ordered by their composition in terms of time- points.  
+
+## Single-cell RNA-seq differential gene expression and gene sets enrichment  
+
+Differentially expressed genes among clusters were identified using the FindAllMarkers function from Seurat (v.3.1.5), taking just LFC (log2 fold change) more than 0.25. For each gene, significance was assessed with the Wilcoxon rank- sum test P values, adjusted for multiple testing using the Benjamini- Hochberg correction to retrieve the false discovery rate (FDR). Only genes with FDR < 0.01 were considered. As expected, many gene markers were shared by clusters from the same group (SR or DR) because of the continuous nature of data. We therefore decided to select unique markers and to take duplicated markers once, preferring the cluster where the LFC was the highest. To perform enrichment of gene signatures in clusters, we used pre- ranked Gene Set Enrichment Analysis (GSEA) from fgsea (v 1.14.0)62 R package. Pre- ranked lists for each cluster were generated by assigning to each gene its LFC relative to the average expression across all the other clusters. Common pathways were defined as belonging to several databases, i.e. Hallmark63, KEGG, Biocarta, Reactome and Gene Ontology Biological Process.  
+
+Enrichment scores (ES) of gene signatures at the single cell level were obtained by computing the z- score for each gene across the data sheet. After truncating these scores at 5 or - 5, the enrichment score was defined by the average z- score over all genes in the gene set.  
+
+## Single-cell RNA-seq trajectory inference
+
+<--- Page Split --->
+
+To infer the reprogramming trajectory, two different approaches were used: wot (v 1.0.5) \(^{10}\) and Monocle3 (v 0.2.3.0) \(^{32}\) . The former applies the Mass Optimal Transport theory to the gene expression space to infer, for each cell in a given sample, the most probable ascending and descending cells in the previous and following timepoints. First, birth- death rates were computed for each cell by applying a logistic function to the enrichment scores for Cell- cycle \(^{64}\) and Apoptosis (R- HSA- 109581, hsa04210, HALLMARK_APOPTOSIS in Liberzon et al., 2015 \(^{63}\) ). \(\beta\) and \(\delta\) logistic functions were optimized (center = - 0.1 and center = 0.15, respectively). Second, transport maps were generated in batch for each pair of subsequent time- points using the functions wot.ot.OTModel (epsilon = 0.2) and compute_all_transport_maps. Finally, trajectories were inferred using population_from_cell_sets and trajectories functions starting from D15 cells that showed high enrichment (> 2) for the signatures Matrisome \(^{30}\) and Late pluripotency \(^{2}\) . For each timepoint, cells having a trajectory probability greater than the mean were considered to belong to the trajectory. Monocle 3, on the other hand, learns a trajectory graph looking at the gene expression changes required for each cell to move from a state to another during a dynamic biological process. In particular, UMAP coordinates in Monocle 3 were replaced with the FLE ones, in order to obtain an FLE- based Monocle trajectory. Furthermore, cluster_cells and learn_graph were performed by tuning the parameters \(k\) (30) and ncenter (96), respectively.  
+
+## Single-cell RNA-seq interaction analyses  
+
+Interaction analyses have been performed on a set of 82 ligand- receptor pairs obtained as follows. A putative list of 3333 couples has been generated from the ligands identified in the secretome analysis with every possible receptor. Afterwards, receptors have been filtered out in case they were not defined as receptor on BioGrid or they did not belong to any of these GO terms: GO- CC:0009897, GO- CC:0098802 and GO:0004714. The resulting list of 1082 pairs was then filtered based on the expression of both ligand and receptor in at least one cell (491). Finally, we selected only those pairs that were experimentally validated \(^{33}\) .  
+
+Interaction scores between trajectories throughout the time- course were evaluated as shown in Schiebinger et al., 2019 \(^{10}\) (Approach 1). Top interactors were selected by ordering the results by standardized interaction score (sIS). Then, the highest ligand- receptor pair for each day was assessed. All the unique couples with a sIS comprised between the first and the last day- specific occurrence was taken.  
+
+HGF/MET cluster- to- cluster interaction scores were computed as the product between the average gene expression value of MET in a cluster and the value of HGF in another. Significance was assessed with empirical p- value, generating a null distribution of 1000 permutations on the association between cells and clusters.  
+
+Matrisome to late- pluripotency interaction was also evaluated using a different, independent approach (Approach 2). First, each ligand and each receptor in a trajectory was scored based on the probability of observing expression values higher than the ones observable by chance from the expression levels of random genes in the same trajectory and time point. Second, ligand- receptor pairs scores were computed as the minimum (i.e. a fuzzy logical AND operator) between the ligand score and the receptor score. An empirical p- value was also computed doing the above procedure multiple times on a randomly permuted gene expression level matrix (i.e. permuting multiple times the gene expression levels of each cell independently) and then measuring the percentage of interaction sub- scores higher than the obtained one. The score and the corresponding p- value was computed as a function of both the absolute ligand and receptor expression levels (as explained above) and, similarly, for their log2 fold change in a specific trajectory/time point, with respect to their average expression across the entire data matrix.  
+
+## STAT3 targets expression  
+
+STAT3 targets were identified using a ChIP- seq dataset on HUS64 human embryonic stem cells \(^{39}\) . In particular, STAT3 target genes were defined as genes with STAT3 significant peaks at \(\pm 3000\) bp from the transcription start site. For each cell, the STAT3 pathway enrichment was computed from the scaled gene expression matrix as the average value for all the STAT3 targets. For each enrichment value, the corresponding p- value was calculated by performing a hypergeometric test and using a random gene list to obtain the null distribution.
+
+<--- Page Split --->
+
+## Bulk RNA-seq analysis of reprogramming data  
+
+Bulk RNA-seq analysis of reprogramming dataTo analyze the relationship between mouse feeders and human reprogramming cells at day 8, we re- analyzed bulk RNA- seq data from Cacchiarelli et al., 2015<sup>2</sup>. Fastqs have been trimmed using Trim Galore (https://github.com/FelixKrueger/TrimGalore) for quality and adapters removal. Then, reads have been mapped with TopHat (v. 2.1.0)<sup>65</sup> and Bowtie2 (v. 2.3.2)<sup>66</sup> with default parameters against an hybrid build of the human (hg38) and mouse (mm10) genomes. Reads aligned to the mouse reference were few (alignment rate < 20%), but it was consistent with the purified nature of the samples, where mouse cells should just represent contamination. Finally, read quantification was performed with HTSeq (v. 0.9.1)<sup>67</sup> on GENCODE human (GRCh38) and mouse (mm10) genome annotations, including protein coding, linc and antisense RNAs. The final count matrix was created by merging mouse and human genes by orthology and differential expression analysis was performed between human and mouse (feeders) samples using DESeq<sup>68</sup>.  
+
+## Immunofluorescence staining  
+
+Immunofluorescence stainingFor immunofluorescence staining, cells were fixed in \(4\%\) paraformaldehyde for \(10\mathrm{min}\) at room temperature, then permeabilized with \(0.1\%\) Triton X- 100 for \(10\mathrm{min}\) , blocked in blocking solution (DPBS with \(10\%\) horse serum and \(0.1\%\) Triton X- 100 for intracellular targets) for \(45\mathrm{min}\) , followed by overnight incubation with primary antibodies. The following antibodies were used for immunofluorescence: rabbit anti- NANOG (Cell Signaling, 4903), mouse anti TRA1- 60 (Millipore, MAB4360), mouse anti- STAT3 (Cell Signaling, 9139), goat anti- HGFR/c- MET (R&D, AF276). Alexa488 or Alexa594 conjugated rabbit or mouse secondary antibodies (1:200) were used (Life Technologies, A21202; A21207). The nuclei were stained with Hoechst 33342 (Life Technologies). Images were acquired on a confocal TCS SP5 microscope (Leica) at \(40\mathrm{x}\) magnification and on a fluorescence microscope DM6B (Leica) at 5 and \(10\mathrm{x}\) magnification.  
+
+## Assessment of reprogramming efficiency  
+
+Assessment of reprogramming efficiencyReprogramming efficiency was quantified after immunostaining with TRA1- 60 and NANOG markers. When the efficiency of reprogramming was too high to allow counting single colonies, it was quantified as relative TRA1- 60<sup>+</sup> and NANOG<sup>+</sup> cell area divided by the total area occupied by the cells.  
+
+## Microarray data analysis  
+
+Microarray data analysisPreviously published microarray data<sup>21</sup> were analyzed by the Quantitative Set Analysis for Gene Expression (QuSAGE)<sup>69</sup> Bioconductor package within MSig DB – Hallmark gene set collection<sup>63</sup>. Results were plotted by MATLAB R2017a.  
+
+## Statistical analysis  
+
+Data variability is presented as boxplots. The number of replicates and the tests used to assess statistical differences are reported within each figure caption.  
+
+## Acknowledgements  
+
+AcknowledgementsThis research was supported by the OAK Foundation Award (W1095/OCAY- 14- 191 to N.E.), the Telethon Foundation Award (GGP15275 and GGP20105 to N.E.), and STARS Wild Card Grants (STARS- WiC award to N.E.) and Cariparo Foundation. This work was supported by Fondazione Telethon Core Grant, Armenise- Harvard Foundation Career Development Award, European Research Council (grant agreement 759154, CellKarma), and the Rita- Levi Montalcini program from MIUR (to D.C.). Ca.L., S.M., M.H., M.C. and N.E. were supported by grant F- 0301- 15- 009 by ShanghaiTech University. Ca.L. was supported by the Natural Science Foundation of China (31601178). This work was partially funded by PROACTIVE 2017 “From Single- Cell to Multi- Cells Information Systems Analysis” (C92F17003530005, Department of Information Engineering, University of Padova) and “Research Grant (type B) – B senior initiative” (Department of Information Engineering, University of Padova). We are grateful to SIAIS Analytical Platform (Dr. Wenzhang Chen and Dr. Wei Zhu) at ShanghaiTech University for mass spectrometry analyses.
+
+<--- Page Split --->
+
+## Author Contributions  
+
+O.G., F.P., Ca.L., D.C. and N.E. designed the study. Ca.L. and N.E. designed the proteomic experiments. W.Q. and M.H. performed microfluidic platform production for proteomic experiments. Ca.L. performed reprogramming for proteomic analyses, bioinformatic analysis on proteomic and microarray data. Ca.L., S.M., and M.C. performed proteomic sample preparation. O.G. designed and optimized microfluidic reprogramming and performed microfluidic reprogramming experiments for scRNA- seq analysis and for LoF and GoF. S.A. and W.Q. helped in reprogramming experiments. A.G. and A.M. pre- processed samples for scRNA- seq and performed library preparation. F.P. analyzed scRNA- seq data. F.P. and G.S. performed trajectory inference analysis. O.G. and S.A. performed immunofluorescence stainings, image acquisition and analysis. P.A., L.V., S.S. and M.S. helped in analyzing bulk RNA- seq data. S.R. and M.D. sequenced scRNA- seq libraries. V.B. helped in sample management. Ce.L. helped image acquisition and analysis. H.S. helped in the design of the loss of function experiments. G.B. and B.D.C. performed interaction score analyses. O.G., F.P., Ca.L., A.G., D.C. and N.E. critically discussed the data and wrote the manuscript. D.C. and N.E. supervised the project.  
+
+## Competing Interest Statement  
+
+O.G, Ca.L. and N.E. are co- inventors on patent applications describing the reprogramming and differentiation processes in microfluidics, application number PD2013A000220, IT UA20162645 and 102016000039189 and PCT/IB2017/052167. O.G. and N.E. are co- founders of Onyel Biotech Srl. Davide Cacchiarelli is founder, shareholder, and consultant of Next Generation Diagnostic srl.  
+
+## Data and material availability  
+
+The authors declare that data supporting the findings of this study are available within the article, its Supplementary Information, attached files or from the authors upon reasonable request. Only Sc RNA- seq and proteomic data will be deposited on public repositories before publication.  
+
+## Code availability  
+
+The authors declare that data analysis in this study was performed with bioinformatic algorithms already publicly available.
+
+<--- Page Split --->
+
+Figure 1. Proteomic analysis of cell- secreted proteins demonstrates a rich extracellular signalling environment along human fibroblast reprogramming. A) Schematics of the in- scale conventional and microfluidic setup (top), and comparison of reprogramming efficiency therein (bottom). Wilcoxon's test was used to assess differences among the conditions, \(^{***}P < 0.001\) . B) Experimental design for proteomic experimental data collection. Proteomic data were obtained by tandem mass spectrometry analysis of conditioned media along the same reprogramming experiments. C) Principal component analysis of the 4542 proteins detected in at least one time point. Each sample of proteomic data refers to medium conditioned over a 48- hour period. D) Enrichment analysis within the Reactome database of the 555 proteins identified as secreted (Supplementary Table 1). Edges connecting different categories reproduce Reactome hierarchy relationships. Complete results are reported in Supplementary Table 2. E) Hierarchical clustering of proteins identified in this study and belonging to the core ECM components \(^{30}\) at specific stages of embryo development \(^{23}\) . F) Hierarchical clustering of secreted proteins from the following enriched signalling pathways (according to Reactome database): Signaling by Interleukins (R- HSA- 449147), Regulation of Insulin- like Growth Factor transport and uptake by Insulin- like Growth Factor Binding Proteins (R- HSA- 381426), Signaling by PDGF (R- HSA- 186797), Signaling by MET (R- HSA- 6806834), Signaling by WNT (R- HSA- 195721); playing a role as senescence- associated secreted proteins \(^{50 - 53}\) , and as ligands \(^{33}\) . Collagens and Laminins were excluded from these protein sets as they were plotted in (E). Only selected protein names discussed within the text or identified as most specific cluster markers in the subsequent single- cell RNA- seq analysis are shown.  
+
+Figure 2. Single- cell RNA- seq analysis of human reprogramming cells unveils a dynamic somatic subpopulation involved in the expression of signalling and ECM related genes. A) Schematic representation of the experimental design for single- cell RNA- seq data collection. Human BJ fibroblasts were grown in Pluriton medium and daily transfected with OSKML mRNA. Starting from day 9, cells were grown in IPS Brew Medium till day 15. Samples were collected by stopping parallel experiments at day 0, 3 and every 48 hours. B) Force- Directed Layout Embedding (FLE) map showing the distribution of cells across time- points and C) identified clusters. D) Time- points enrichment for each cluster (left) and heatmap of Z- scored normalized counts, averaged by clusters, for key reprogramming related genes (right). NA cluster not shown. E) GSEA results for each cluster. Only significant results are shown. NES, Normalized enrichment score.  
+
+Figure 3. Trajectory inference reveals different fates during reprogramming. Gene expression- based interaction analysis suggests an existing crosstalk between somatic and reprogramming cells through known and novel ligand- receptor couples. A) Matrisome and Late pluripotency enrichment scores shown along the FLE map. B) Monocle3 (black line) and WOT (colored dots) trajectory inferences are displayed on the FLE graph. Arrows point to the starting point (blue) and 4 end points (red) of the inferred trajectories. A representative scheme of the trajectories is shown on the top- right. C) Enrichment Score graph relative to the GSEA of SR2 cluster for senescence- associated secreted proteins geneset (SASP) \(^{50 - 53}\) . Black lines on the x axis represent a match between the ranked list and the geneset analyzed. NES, Normalized enrichment score. FDR, False Discovery Rate. D) Venn diagram representing the intersection between SASP geneset and SR2 cluster marker genes and their relative gene expression, shown in a E) heatmap of Z- scored normalized counts, averaged by clusters. Genes with \* have been detected in secretome analysis. F) Schematic representation of ligand- receptor interactions hypothesized during reprogramming. Fibroblasts (D0, left) develop two fates: a somatic secretory phenotype (bottom) and induced pluripotency (top). Black arrows show the directionality of the examined interaction. G) Heatmap of z- scored standardized interaction scores for top ligand- receptor pairs. Selection criteria is described in Methods. H) Log2 proteomic expression relative to D1- D2 of SASP- related ligands among the top pairs at each time- point. I- K) HGF and MET gene expression profiles (log2 CPM) are shown in different reprogramming systems. I) In our data, they are displayed on the FLE map as fold change relative to HGF and averaged across the time course (bottom- left). J) In Liu et al., 2020 \(^{7}\) , they are shown as averaged across their identified clusters and K) in Cacchiarelli et al., 2015 \(^{2}\) , they are shown as mouse and human mean normalized expression at sampling day 8 (\*\* BH- adjusted p- value < 0.01). L- N) NRG1 and ERBB3 gene expression profiles are shown in different reprogramming systems. L) In our data, they are displayed on the FLE map as fold change relative
+
+<--- Page Split --->
+
+to NRG1 and averaged across the time course (bottom-left). M) In Liu et al., 2020<sup>7</sup>, they are shown as averaged across their identified clusters and N) in Cacchiarelli et al., 2015<sup>2</sup>, they are shown as mean FPKM across the time-course.  
+
+Figure 4. Perturbation of STAT3 pathway components affect the efficiency of reprogramming. A) A schematic representation of HGF/c- MET/STAT3 signalling pathway. HGF binds the receptor c- MET, inducing its kinase catalytic activity. After trans- phosphorylation, c- MET starts a phosphorylation cascade that is in common with the LIF and IL6 pathways. It includes the Janus kinase 1 (JAK1) and ends in the phosphorylation and dimerization of STAT3. Active phosphorylated dimers can translocate to the cell nucleus where they act as transcription activators of target genes, such as pluripotency- related genes (Created with BioRender.com). B) STAT3 target expression correlates with MET transcription. In the FLE graph, green dots represent cells with positive enrichment scores for STAT3 target genes (Methods). Bigger circles summarize averaged HGF (left) and MET (right) gene expression in identified clusters. Significant inter- cluster HGF- MET interactions are displayed (arrows). Arrow thickness relates to the strength of the interaction. C) Top, representative images of expression of nuclear STAT3 and c- MET during reprogramming performed in microfluidics at day 6. Bottom, correlation between the expression intensity of nuclear STAT3, c- MET, and cell size obtained from experimental data shown on top. Data from n=61 cells (n=3 independent experiments). D) Left, reprogramming efficiency in microfluidics measured as the relative area occupied by NANOG<sup>+</sup> colonies in cells upon inhibition of c- Met and JAK1 kinases using small molecules at day 12, compared to the ones treated with the vehicle (n=6 for vehicle, n=12 for JAKi and n=7 for c- METi); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions, ***P < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. E) Left, reprogramming efficiency in microfluidics upon knock- down of STAT3 using siRNAs at day 12 (n=8 for scramble siRNA, n=11 for siSTAT3); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions, ***P < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. F) Bottom, reprogramming efficiency in standard 24- well plates upon addition of HGF, IL-6 and soluble IL6 receptor (sIL6R), or NRG1 at day 9 (n=14 for control, n=19 for HGF, n=5 for IL6 + sIL6R, n=16 for NRG1); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions, **P < 0.01, ***P < 0.001. Top, representative quantification pictures in standard 24- well plates assessed by immunostaining of NANOG and TRA-1-60. G) Bottom, reprogramming efficiency in standard 24- well plates upon temporally modulate addition of HGF, IL6 and soluble IL6 receptor (sIL6R), and NRG1 at day 9 (n=14 for control, n=6 for HGF in the early phase and NRG1 in the late phase, n=4 for HGF + IL6 + sIL6R in the early phase and NRG1 in the late phase, n=4 for HGF+ IL6 and sIL6R + NRG1 for the entire process); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions, **P < 0.01, ***P < 0.001. Top, representative quantification pictures in standard 24- well plates assessed by immunostaining of NANOG and TRA-1-60.  
+
+Extended Data Figure 1. A) Schematic representation of the experimental design for the proteomic study. B) Reprogramming efficiency within the same microfluidic channels used for proteomic analyses. C) Results of protein quantification by BCA for samples analyzed by LC- MS/MS. Each value refers to samples obtained by pooling together medium collected from 40 microfluidic channels during 48- hour conditioning (4 time points of collection). Error bar is mean±standard deviation (n=3). Significant differences were evaluated by one- way ANOVA with Tukey's post- test (*p<0.05, **p<0.01, ***p<0.001). D) Visualization of proteomic data correlation between replicates. Each dot represents an identified protein. Log2 relative quantification is shown on the axes (a.u.). E) Comparison of the concentration profile of IL6 by mass spectrometry and ELISA, performed on the same conditioned media collected during the proteomic experiment described in (A). F) Analysis of microarray data in Luni et al., 2016<sup>21</sup> by Quantitative Set Analysis for Gene Expression (QuSAGE) within Hallmark gene set in Msig DB collection. Microarray analysis was performed on mRNA extracted from 4 single hiPSC colonies derived in microfluidics (p0) and then expanded for 3 passages in a conventional multi- well plate (p3). Differences show that JAK- STAT pathway is the
+
+<--- Page Split --->
+
+most significantly different between the two culture systems, suggesting that microfluidic environment may affect this pathway.  
+
+Extended Data Figure 2. A) Left, scatter plot representing the number of reads (x axis) over the number of detected genes (y axis) for each cell. Color gradient shows the percentage of reads associated with mitochondrial genes. The dotted line has been put at 1,000 detected genes, used for filtering. Right, schematic representation of cells/genes filtering from raw data to the final dataset. B) Heatmaps of pearson correlation coefficient for each replicate, divided by each time- point. Correlation has been evaluated by comparing the distribution of each replicate in the clusters identified in Fig. 2C. C) Somatic and Developmental signatures enrichment scores shown along the FLE map. D) Cell cycle phases assigned to each cell are shown along the FLE map.  
+
+Extended Data Figure 3. A) Workflow for the detection of trajectories starting from Matrisome and Pluripotency D15 enriched cells. Cells assigned to each trajectory are colored and divided by time- point, as described in Methods. B) Trend of pearson correlation coefficients evaluated between trajectories probability values for Matrisome and Pluripotency, assigned to each cell and divided by time- point. C) Monocle3 (black line) trajectory inferences are displayed on the FLE graph. Dots (cells) are colored according to Monocle3 pseudotime. D) GSEA has been performed on SR2 cluster using signaling- related genesets used in Fig. 1F. The results are shown as a barplot, displaying FDR (x axis) and NES (colors). NES, Normalized Enrichment Score. FDR, False Discovery Rate. E) Schematic representation of ligand- receptor pairs selection for interaction score analyses, as described in Methods. F) Heatmap of z- scored standardized interaction scores for all the ligand- receptor pairs analyzed. G) Venn diagram representing the intersection between the number of ligand- receptor pairs with at least one significant interaction score in the reported (orange - Approach 1) or alternative (green - Approach 2) approaches. H) GSEA has been performed on the noRepro1 cluster from Liu et al., 2020 using secretome- related genesets used in Fig. 2E right. The results are shown as a barplot, displaying FDR (x axis) and NES (colors). NES, Normalized Enrichment Score. FDR, False Discovery Rate.  
+
+Extended Data Figure 4. A) Time course of STAT3 activation during the microfluidic reprogramming process. Representative images showing a first wave around day 4 to day 7, and then a second wave at the end of the process, when it is active just in the hiPSC colonies. Scale bar, \(50 \mu \mathrm{m}\) . Representative images from \(n = 2\) independent experiments. B) Left, quantification of mean STAT3 nuclear intensity versus mean nuclei intensity, stained with Hoechst ( \(n = 25\) for scramble siRNA, \(n = 37\) for siSTAT3); Tukey's multiple comparisons test was used to assess differences among the conditions, \(^{***}P < 0.001\) . Right, representative images. Scale bar, \(50 \mu \mathrm{m}\) . C) Schematic representation of primitive streak formation (left - Adapted from Boccaccio and Comoglio, 2006<sup>54</sup>) and primitive streak components (right - Created with BioRender.com). Black arrows show the origin of each component based on average gene expression between SR or DR cells. C, cleavage. D) Heatmap of Z- scored normalized counts, averaged by day, of genes encoding for primitive streak components shown in (C).  
+
+## Supplementary Table 1.  
+
+Supplementary Table 1. Proteomic data of the secreted proteins identified in this study.  
+
+## Supplementary Table 2.  
+
+Supplementary Table 2. Complete results of the functional enrichment analysis reported in Figure 1D.  
+
+## Supplementary Table 3.  
+
+Supplementary Table 3. Complete data from Figure 1E.  
+
+## Supplementary Table 4.  
+
+Supplementary Table 4. Complete data from Figure 1F.  
+
+## Supplementary Table 5.  
+
+Supplementary Table 5. List of the gene sets used to perform gene set enrichment analysis in Figure 2E.
+
+<--- Page Split --->
+
+853 Supplementary Table 6. 855 Positive and significant results of gene set enrichment analysis on SR2 cluster for Common 856 pathways. 857 Supplementary Table 7. 859 List of uniquely assigned marker genes divided by cluster. 860 Supplementary Table 8. 862 Complete results from Interaction Score Analysis for the selected 82 ligand-receptor pairs. 863 Supplementary Table 9. 865 Complete data from Figure 3G and Extended Data Figure S3C.
+
+<--- Page Split --->
+
+References1. Takahashi, K. et al. Induction of Pluripotent Stem Cells from Adult Human Fibroblasts by Defined Factors. Cell 131, 861- 872 (2007).2. Cacchiarelli, D. et al. Integrative Analyses of Human Reprogramming Reveal Dynamic Nature of Induced Pluripotency. Cell 162, 412- 424 (2015).3. Takahashi, K. et al. Induction of pluripotency in human somatic cells via a transient state resembling primitive streak-like mesendoderm. Nat. Commun. 5, 1- 9 (2014).4. Xing, Q. R. et al. Diversification of reprogramming trajectories revealed by parallel single-cell transcriptome and chromatin accessibility sequencing. Sci. Adv. 6, (2020).5. Qin, J. & Gronenborn, A. M. Weak protein complexes: Challenging to study but essential for life. FEBS Journal vol. 281 1948- 1949 (2014).6. Toh, C. X. D. et al. RNAi Reveals Phase-Specific Global Regulators of Human Somatic Cell Reprogramming. Cell Rep. 15, 2597- 2607 (2016).7. Liu, X. et al. Reprogramming roadmap reveals route to human induced trophoblast stem cells. Nature 586, 101- 107 (2020).8. Guo, L. et al. Resolving Cell Fate Decisions during Somatic Cell Reprogramming by Single-Cell RNA-Seq A generic bifurcation model for cell fate decisions. Mol. Cell 73, 815- 829 (2019).9. Tran, K. A. et al. Defining Reprogramming Checkpoints from Single-Cell Analyses of Induced Pluripotency. Cell Rep. 27, 1726- 1741.e5 (2019).10. Schiebinger, G. et al. Optimal- Transport Analysis of Single- Cell Gene Expression Identifies Developmental Trajectories in Reprogramming. Cell 176, 928- 943.e22 (2019).11. Zhao, T. et al. Single- Cell RNA- Seq Reveals Dynamic Early Embryonic- like Programs during Chemical Reprogramming. Cell Stem Cell 23, 31- 45.e7 (2018).12. Yamanaka, S. Elite and stochastic models for induced pluripotent stem cell generation. Nature vol. 460 49- 52 (2009).13. Hanna, J. et al. Direct cell reprogramming is a stochastic process amenable to acceleration. Nature 462, 595- 601 (2009).14. Buganim, Y. et al. Single- cell expression analyses during cellular reprogramming reveal an early stochastic and a late hierarchic phase. Cell 150, 1209- 1222 (2012).15. Shakiba, N. et al. Cell competition during reprogramming gives rise to dominant clones. Science (80- ). 364, (2019).16. Mosteiro, L., Pantoja, C., de Martino, A. & Serrano, M. Senescence promotes in vivo reprogramming through p16 INK4a and IL- 6. Aging Cell 17, (2018).17. Mahmoudi, S. et al. Heterogeneity in old fibroblasts is linked to variability in reprogramming and wound healing. Nature 574, 553- 558 (2019).18. Gagliano, O. et al. Microfluidic reprogramming to pluripotency of human somatic cells. Nat. Protoc. 14, 722- 737 (2019).19. Hu, J. et al. Embryonic germ cell extracts erase imprinted genes and improve the efficiency of induced pluripotent stem cells. Sci. Rep. 8, 10955 (2018).20. Michielin, F. et al. The Microfluidic Environment Reveals a Hidden Role of Self- Organizing Extracellular Matrix in Hepatic Commitment and Organoid Formation of hiPSCs. Cell Rep. 33, (2020).21. Luni, C. et al. High- efficiency cellular reprogramming with microfluidics. Nat. Methods (2016) doi:10.1038/nmethod.3832.22. Giulitti, S. et al. Direct generation of human naive induced pluripotent stem cells from somatic cells in microfluidics. Nat. Cell Biol. 21, 275- 286 (2019).23. Boroviak, T. et al. Single cell transcriptome analysis of human, marmoset and mouse embryos reveals common and divergent features of preimplantation development. Dev. 145, (2018).24. Andrae, J., Gallini, R. & Betsholtz, C. Role of platelet- derived growth factors in physiology and medicine. Genes and Development vol. 22 1276- 1312 (2008).25. Marson, A. et al. Wnt Signaling Promotes Reprogramming of Somatic Cells to Pluripotency. Cell Stem Cell vol. 3 132- 135 (2008).26. Phanstiel, D. H. et al. Proteomic and phosphoproteomic comparison of human ES and iPS cells. Nat. Methods 8, 821- 827 (2011).
+
+<--- Page Split --->
+
+927 27. Mosteiro, L. et al. Tissue damage and senescence provide critical signals for cellular reprogramming in vivo. Science (80- ). 354, (2016). 924 28. Hartman, A. A. et al. YAP Non-cell-autonomously Promotes Pluripotency Induction in Mouse Cells. Stem Cell Reports 14, 730- 743 (2020). 926 29. Blondel, V. D., Guillaume, J. L., Lambiotte, R. & Lefebvre, E. Fast unfolding of communities in large networks. J. Stat. Mech. Theory Exp. 2008, 10008 (2008). 928 30. Naba, A. et al. The matrisome: In silico definition and in vivo characterization by proteomics of normal and tumor extracellular matrices. Mol. Cell. Proteomics 11, (2012). 929 31. Trapnell, C. et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat. Biotechnol. 32, 381- 386 (2014). 930 32. Cao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. Nature 566, 496- 502 (2019). 931 33. Ramilowski, J. A. et al. A draft network of ligand-receptor-mediated multicellular signalling in human. Nat. Commun. 6, 1- 12 (2015). 932 34. James, D., Levine, A. J., Besser, D. & Hemmati- Brivanlou, A. TGFβ/activin/nodal signaling is necessary for the maintenance of pluripotency in human embryonic stem cells. Development 132, 1273- 1282 (2005). 933 35. Tesar, P. J. et al. New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature 448, 196- 199 (2007). 934 36. Medico, E. et al. Osteopontin is an autocrine mediator of hepatocyte growth factor- induced invasive growth. Cancer Res. 61, 5861- 5868 (2001). 935 37. O'Leary, N. A. et al. Reference sequence (RefSeq) database at NCBI: Current status, taxonomic expansion, and functional annotation. Nucleic Acids Res. 44, D733- D745 (2016). 936 38. Graf, U., Casanova, E. A. & Cinelli, P. The role of the leukemia inhibitory factor (LIF) - Pathway in derivation and maintenance of murine pluripotent stem cells. Genes vol. 2 280- 297 (2011). 937 39. Tsankov, A. M. et al. Transcription factor binding dynamics during human ES cell differentiation. Nature 518, 344- 349 (2015). 938 40. Boccaccio, C. et al. Induction of epithelial tubules by growth factor HGF depends on the STAT pathway. Nature 391, 285- 288 (1998). 939 41. Khourieh, J. et al. A deep intronic splice mutation of STAT3 underlies hyper IgE syndrome by negative dominance. Proc. Natl. Acad. Sci. U. S. A. 116, 16463- 16472 (2019). 940 42. Prat, M. et al. The receptor encoded by the human C-MET oncogene is expressed in hepatocytes, epithelial cells and solid tumors. Int. J. Cancer 49, 323- 328 (1991). 941 43. Andermarcher, E., Surani, M. A. & Gherardi, E. Co- expression of the HGF/SF and c- met genes during early mouse embryogenesis precedes reciprocal expression in adjacent tissues during organogenesis. Dev. Genet. 18, 254- 266 (1996). 942 44. Sonnenberg, E., Meyer, D., Weidner, K. M. & Birchmeier, C. Scatter factor/hepatocyte growth factor and its receptor, the c- met tyrosine kinase, can mediate a signal exchange between mesenchyme and epithelia during mouse development. J. Cell Biol. 123, 223- 235 (1993). 943 45. Nakaya, Y., Sukowati, E. W., Alev, C., Nakazawa, F. & Sheng, G. Involvement of dystroglycan in epithelial- mesenchymal transition during chick gastrulation. Cells Tissues Organs 193, 64- 73 (2010). 944 46. Mogi, K. & Toyoizumi, R. Invasion by matrix metalloproteinase- expressing cells is important for primitive streak formation in early chick blastoderm. Cells Tissues Organs 192, 1- 16 (2010). 945 47. Chen, H. et al. Reinforcement of STAT3 activity reprogrammes human embryonic stem cells to naive- like pluripotency. Nat. Commun. 6, 1- 17 (2015). 946 48. Kuo, H. H. et al. Negligible- Cost and Weekend- Free Chemically Defined Human iPSC Culture. Stem Cell Reports 14, 256- 270 (2020). 947 49. Marinho, P. A., Chailangkarn, T. & Muotri, A. R. Systematic optimization of human pluripotent stem cells media using Design of Experiments. Sci. Rep. 5, 1- 13 (2015). 948 50. Coppé, J. P., Desprez, P. Y., Krtolica, A. & Campisi, J. The senescence- associated secretory phenotype: The dark side of tumor suppression. Annual Review of Pathology: Mechanisms of Disease vol. 5 99- 118 (2010).
+
+<--- Page Split --->
+
+978 51. Coppé, J.- P. et al. Senescence-Associated Secretory Phenotypes Reveal Cell- Nonautonomous Functions of Oncogenic RAS and the p53 Tumor Suppressor. PLoS Biol. 980 6, e301 (2008). 981 52. Acosta, J. C. et al. A complex secretory program orchestrated by the inflammasome controls 982 paracrine senescence. Nat. Cell Biol. 15, 978- 990 (2013). 983 53. Lopes- Paciencia, S. et al. The senescence- associated secretory phenotype and its 984 regulation. Cytokine 117, 15- 22 (2019). 985 54. Boccaccio, C. & Comoglio, P. M. Invasive growth: A MET- driven genetic programme for 986 cancer and stem cells. Nature Reviews Cancer vol. 6 637- 645 (2006). 987 55. Wisniewski, J. R., Zougman, A., Nagaraj, N. & Mann, M. Universal sample preparation 988 method for proteome analysis. Nat. Methods 6, 359- 362 (2009). 989 56. Tyanova, S., Temu, T. & Cox, J. The MaxQuant computational platform for mass 990 spectrometry- based shotgun proteomics. Nat. Protoc. 11, 2301- 2319 (2016). 991 57. Uhlén, M. et al. Tissue- based map of the human proteome. Science (80- ). 347, (2015). 992 58. Gonzalez, R. et al. Screening the mammalian extracellular proteome for regulators of 993 embryonic human stem cell pluripotency. Proc. Natl. Acad. Sci. U. S. A. 107, 3552- 3557 994 (2010). 995 59. Yu, G. & He, Q. Y. ReactomePA: An R/Bioconductor package for reactome pathway 996 analysis and visualization. Mol. Biosyst. 12, 477- 479 (2016). 997 60. Bindea, G. et al. ClueGO: A Cytoscape plug- in to decipher functionally grouped gene 998 ontology and pathway annotation networks. Bioinformatics 25, 1091- 1093 (2009). 999 61. Shannon, P. et al. Cytoscape: A software Environment for integrated models of biomolecular 1000 interaction networks. Genome Res. 13, 2498- 2504 (2003). 1001 62. Korotkevich, G. et al. Fast gene set enrichment analysis. bioRxiv 060012 (2016) 1002 doi:10.1101/060012. 1003 63. Liberzon, A. et al. The Molecular Signatures Database Hallmark Gene Set Collection. Cell 1004 Syst. 1, 417- 425 (2015). 1005 64. Kowalczyk, M. S. et al. Single- cell RNA- seq reveals changes in cell cycle and differentiation 1006 programs upon aging of hematopoietic stem cells. Genome Res. 25, 1860- 1872 (2015). 1007 65. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: Discovering splice junctions with RNA- 1008 Seq. Bioinformatics 25, 1105- 1111 (2009). 1009 66. Langmead, B. & Salzberg, S. L. Fast gapped- read alignment with Bowtie 2. Nat. Methods 9, 1010 357- 359 (2012). 1011 67. Anders, S., Pyl, P. T. & Huber, W. HTSeq- a Python framework to work with high- throughput 1012 sequencing data. Bioinformatics 31, 166- 169 (2015). 1013 68. Love, M., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for 1014 RNA- seq data with DESeq2. bioRxiv 2832 (2014) doi:10.1101/002832. 1015 69. Yaari, G., Bolen, C. R., Thakar, J. & Kleinstein, S. H. Quantitative set analysis for gene 1016 expression: A method to quantify gene set differential expression including gene- gene 1017 correlations. Nucleic Acids Res. 41, (2013).
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1. Proteomic analysis of cell-secreted proteins demonstrates a rich extracellular signaling environment along human fibroblast reprogramming. A) Schematics of the in-scale conventional and microfluidic setup (top), and comparison of reprogramming efficiency therein (bottom). Wilcoxon's test was used to assess differences among the conditions, "P < 0.01. B) Experimenta design for proteomic experimental data collection. Proteomic data were obtained by tandem mass spectrometry analysis of conditioned media along the same reprogramming experiments. C) Principal component analysis of the 4542 proteins detected in at least one time point. Each sample of proteomic data refers to medium conditioned over a 48-hour period. D) Enrichment analysis within the Reactome database of the 555 proteins identified as secreted (Supplementary Table 1). Edges connecting different categories reproduce Reactome hierarchy relationship. Complete results are reported in supplemental Table 2. E) Hierarchical clustering of proteins identified in this study and belonging to the core ECM components at specific stages of embryo development. F) Hierarchical clustering of secreted proteins from the following enriched signalling pathways (according to Reactome database): Signaling by Interleukins (R-HSA-449147), Regulation of Insulin-like Growth Factor transport and uptake by Insulin-like Growth Factor Binding Proteins (R-HSA-381426), Signaling by PDGF (R-HSA-186797), Signaling by MET (R-HSA-6806834), Signaling by WNT (R-HSA-195721); playing a role as senescence-associated secreted proteins (51, and as ligands (53). Collagens and Laminins were excluded from these protein sets as they were plotted in (E). Only selected protein names discussed within the text or identified as most specific cluster markers in the subsequent single-cell RNA-seq analysis are shown. </center>  
+
+## Figure 1  
+
+See image above for figure legend.
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2. Single-cell RNA-seq analysis of human reprogramming cells unveils a dynamic somatic subpopulation involved in the expression of signalling and ECM related genes. A) Schematic representation of the experimental design for single-cell RNA-seq data collection. Human BJ fibroblasts were grown in Pluriton medium and daily transfected with OSKML mRNA. Starting from day 9, cells were grown in IPS Brew Medium till day 15. Samples were collected by stopping parallel experiments at day 0, 3 and every 48 hours. B) Force-Directed Layout Embedding (FLE) map showing the distribution of cells across time-points and C) identified clusters. D) Time-points enrichment for each cluster (left) and heatmap of Z-scored normalized counts, averaged by clusters, for key reprogramming related genes (right). NA cluster not shown. E) GSEA results for each cluster. Only significant results are shown. NES, Normalized enrichment score. </center>  
+
+## Figure 2  
+
+See image above for figure legend.
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3. Trajectory inference reveals different fates during reprogramming. Gene expression-based interaction analysis suggests an existing crosstalk between somatic and reprogramming cells through known and novel ligand-receptor couples. A) Matrisome and Late pluripotency enrichment scores shown along the FLE map. B) Monocle3 (black line) and WOT (colored dots) trajectory inferences are displayed on the FLE graph. Arrows point to the starting point (blue) and 4 end points (red) of the inferred trajectories. A representative scheme of the trajectories is shown on the top-right. C) Enrichment Score graph relative to the GSEA of SR2 cluster for senescence-associated secreted proteins (SASP50-53. Black lines on the x axis represent a match between the ranked list and the genes analyzed. NES, Normalized enrichment score. FDR, False Discovery Rate. D) Venn diagram representing the intersection between SASP genes and SR2 cluster marker genes and their relative gene expression, shown in A) heatmap of Z-scored normalized counts, averaged by clusters. Genes with * have been detected in secretome analysis. F) Schematic representation of ligand-receptor interactions hypothesized during reprogramming. Fibroblasts (D0, left) develop two fates: a somatic secretory phenotype (bottom) and induced pluripotency (top). Black arrows show the directionality of the examined interaction. G) Heatmap of z-scored standardized interaction scores for top ligand-receptor pairs. Selection criteria is described in Methods. H) Log2 proteomic expression relative to D1-D2 of SASP-related ligands among the top pairs at each time-point. I-K) HGF and MET gene expression profiles (log2 CPM) are shown in different reprogramming systems. I) In our data, they are displayed on the FLE map as fold change relative to HGF and averaged across the time course (bottom-left). J) In Liu et al., 2020, they are shown as averaged across their identified clusters and K) in Caccharielli et al., 2015, they are shown as mouse and human mean normalized expression at sampling day 8 (** BH-adjusted p-value < 0.01). L-N) NRG1 and ERBB3 gene expression profiles are shown in different reprogramming systems. L) In our data, they are displayed on the FLE map as fold change relative to NRG1 and averaged across the time course (bottom-left). M) In Liu et al., 2020, they are shown as averaged across their identified clusters and N) in Caccharielli et al., 2015, they are shown as mean FPKM across the time-course. </center>  
+
+## Figure 3  
+
+See image above for figure legend.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4. Perturbation of STAT3 pathway components affect the efficiency of reprogramming. A) A schematic representation of HGF/c-MET/STAT3 signalling pathway. HGF binds the receptor c-MET, inducing its kinase catalytic activity. After trans-phosphorylation, c-MET starts a phosphorylation cascade that is in common with the LIF and IL6 pathways. It includes the Janus kinase 1 (JAK1) and ends in the phosphorylation and dimerization of STAT3. Active phosphorylated dimers can translocate to the cell nucleus where they act as transcription activators of target genes, such as pluripotency-related genes (Created with BioRender.com). B) STAT3 target expression correlates with MET transcription. In the FLE graph, green dots represent cells with positive enrichment scores for STAT3 target genes (Methods). Bigger circles summarize averaged HGF (left) and MET (right) gene expression in identified clusters. Significant inter-cluster HGF-MET interactions are displayed (arrows). Arrow thickness relates to the strength of the interaction. C) Top, representative images of expression of nuclear STAT3 and c-MET during reprogramming performed in microfluidics at day 6. Bottom, correlation between the expression intensity of nuclear STAT3, c-MET, and cell size obtained from experimental data shown on top. Data from n=61 cells (n=3 independent experiments). D) Left, reprogramming efficiency in microfluidics measured as the relative area occupied by NANOG+ colonies in cells upon inhibition of c-Met and JAK1 kinases using small molecules at day 12, compared to the ones treated with the vehicle (n=6 for vehicle, n=12 for JAK1 and n=7 for c-MET). ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***p < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. E) Left, reprogramming efficiency in microfluidics upon knock-down of STAT3 using siRNAs at day 12 (n=8 for scramble siRNA, n=11 for siSTAT3); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001. Right, representative quantification pictures in microfluidic channels assessed by immunostaining of NANOG. F) Bottom, reprogramming efficiency in standard 24-well plates upon addition of HGF, IL-6 and soluble IL6 receptor (sIL6R), or NRG1 at day 9 (n=14 for control, n=19 for HGF, n=5 for IL6 + sIL6R, n=16 for NRG1); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001. Top, representative quantification pictures in standard 24-well plates assessed by immunostaining of NANOG and TRA-1-60. G) Bottom, reprogramming efficiency in standard 24-well plates upon temporally modulate addition of HGF, IL-6 and soluble IL6 receptor (sIL6R), and NRG1 at day 9 (n=14 for control, n=6 for HGF in the early phase and NRG1 in the late phase, n=4 for HGF + IL6 + sIL6R in the early phase and NRG1 in the late phase, n=4 for HGF+IL6 and sIL6R + NRG1 for the entire process); ANOVA followed by Tukey's multiple comparisons test was used to assess differences among the conditions. ***P < 0.001, ***P < 0.001. Top, representative quantification pictures in standard 24-well plates assessed by immunostaining of NANOG and TRA-1-60. </center>  
+
+## Figure 4  
+
+See image above for figure legend.  
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.
+
+<--- Page Split --->
+
+SupplementaryTables19.xlsx
+
+<--- Page Split --->

preprint/preprint__98621a0a9a75bfd4307471acfa62e2ac69cd7da3d48c9482423c2bb4b8db561a/images_list.json

@@ -0,0 +1,85 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1",
+    "footnote": [],
+    "bbox": [
+      [
+        44,
+        99,
+        720,
+        808
+      ]
+    ],
+    "page_idx": 32
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 34
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3",
+    "footnote": [],
+    "bbox": [
+      [
+        55,
+        55,
+        682,
+        760
+      ]
+    ],
+    "page_idx": 34
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4",
+    "footnote": [],
+    "bbox": [
+      [
+        50,
+        55,
+        648,
+        747
+      ]
+    ],
+    "page_idx": 36
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5",
+    "footnote": [],
+    "bbox": [
+      [
+        55,
+        260,
+        933,
+        836
+      ]
+    ],
+    "page_idx": 38
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Figure 6",
+    "footnote": [],
+    "bbox": [
+      [
+        163,
+        312,
+        835,
+        805
+      ]
+    ],
+    "page_idx": 39
+  }
+]

preprint/preprint__98621a0a9a75bfd4307471acfa62e2ac69cd7da3d48c9482423c2bb4b8db561a/preprint__98621a0a9a75bfd4307471acfa62e2ac69cd7da3d48c9482423c2bb4b8db561a.mmd

@@ -0,0 +1,428 @@
+
+# Structural differences between the closely related RNA helicases, UAP56 and URH49 fashions distinct functional apo-complexes  
+
+Seiji Masuda ( smasuda@nara.kindai.ac.jp )  
+
+Kindai University  
+
+Ken- ichi Fujita Fujita Health University  
+
+Misa Ito Kyoto University  
+
+Midori Irie Kyoto University  
+
+Kotaro Harada Kyoto University  
+
+Naoko Fujiwara Kyoto University  
+
+Yuya Ikeda Kyoto University  
+
+Hanae Yoshioka Kyoto University  
+
+Tomohiro Yamazaki Kyoto University  
+
+Masaki Kojima Tokyo University of Pharmacy and Life Sciences  
+
+Bunzo Mikami Kyoto University  
+
+Akila Mayeda  
+
+Institute for Comprehensive Medical Science, Fujita Health University https://orcid.org/0000- 0002-  
+
+9562- 550X  
+
+Article  
+
+Keywords:  
+
+Posted Date: May 12th, 2023
+
+<--- Page Split --->
+
+DOI: https://doi.org/10.21203/rs.3.rs- 2819840/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+## Additional Declarations:  
+
+There is NO Competing Interest.  
+
+Table 1 is available in the Supplementary Files section  
+
+Version of Record: A version of this preprint was published at Nature Communications on January 15th, 2024. See the published version at https://doi.org/10.1038/s41467- 023- 44217- 8.
+
+<--- Page Split --->
+
+1 Structural differences between the closely related RNA helicases, UAP56 and 2 URH49 fashions distinct functional apo-complexes 3 4 Ken-ichi Fujita1,2\\*, Misa Ito1, Midori Irie1, Kotaro Harada1, Naoko Fujiwara1,3, Yuya 5 Ikeda1, Hanae Yoshioka1, Tomohiro Yamazaki1,3, Masaki Kojima4, Bunzo Mikami5,6, 6 Akila Mayeda2, Seiji Masuda1,7,8,9,10\\* 7 8 1 Division of Integrated Life Sciences, Graduate School of Biostudies, Kyoto 9 University, Kyoto, Kyoto, 606-8502, Japan. 10 2 Division of Gene Expression Mechanism, Center for Medical Science, Fujita Health 11 University, Toyoake, Aichi 470-1192, Japan 12 3 Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka 565-0871, 13 Japan. 14 4 School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, Hachioji, 15 Tokyo, 192-0392, Japan. 16 5 Research Institute for Sustainable Humano sphere, Kyoto University 17 6 Institute of Advanced Energy, Kyoto University 18 7 Department of Food Science and Nutrition, Faculty of Agriculture Kindai University, 19 Nara, Nara 631-8505, Japan 20 8 Agricultural Technology and Innovation Research Institute, Kindai University, Nara, 21 Nara, 631-8505, Japan 22 9 Antiaging center, Kindai University, Higashiosaka, Osaka 577-8502, Japan 23 10 Lead Contact 24 \* To whom correspondence should be addressed. Tel: +81-742-43-1713; Email: 25 smasuda@nara.kindai.ac.jp 26 27 ORCID 28 Ken-ichi Fujita; 0000-0002-3104-5274, Naoko Fujiwara; 0000-0003-2366-9076, 29 Tomohiro Yamazaki; 0000-0003-0866-5173, Masaki Kojima; 0009-0001-9190-4923, 30 Bunzo Mikami; 0000-0003-0638-8619, Akila Mayeda; 0000-0002-9562-550X, Seiji 31 Masuda;. 0000-0003-0295-6789 32
+
+<--- Page Split --->
+
+## Summary  
+
+Messenger RNA export is a regulated pathway that control gene expression and specific physiological events. In humans, closely related RNA helicases, UAP56 and URH49 shape selective mRNA export pathways through distinct apo- complex formation and remodeling to similar ATP- complex to achieve precise gene regulation. The difference in apo- complex is the key to functional divergence. However, the profile of the apo- complex formed by URH49 (named apo- AREX complex) and why UAP56 and URH49 exhibit distinct complex formation remain unknown. Here, we investigated unidentified apo- AREX components. RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM were physically and functionally associated with URH49, indicating the key components of the apo- AREX complex. Integrating analysis of crystal structure and complex formation of UAP56/URH49 chimera mutants demonstrated that their structural features contribute to respective complex formation. This study provides insights into the specific function of two helicases and into how two helicases diverged from a single ancestral gene, Sub2.
+
+<--- Page Split --->
+
+## Introduction  
+
+During the expression of protein- coding genes, pre- mRNAs are transcribed in the nucleus and undergo several RNA processing steps, including capping, splicing, and polyadenylation. Subsequently, the mature mRNA is exported to the cytoplasm for translation. These processes are coupled with one another through appropriate assembly and remodeling of mRNA- protein (mRNP) complexes to achieve accurate gene expression<sup>1</sup>.  
+
+A key player integrating transcription and mRNA export is the evolutionarily conserved ATP- dependent multi- subunit Transcription- EXport (TREX) complex. The human ATP- bound TREX complex consists of the THO subcomplex, comprising THOC1, THOC2, THOC3, THOC5, THOC6, and THOC7, and several affiliated proteins: ALYREF, CIP29, CHTOP, PDIP3, ZC3H11A, and DEAD- box RNA helicase  
+
+UAP56<sup>2,3</sup> (and refs therein<sup>4,5,6</sup>). The TREX components are recruited onto transcribing RNA polymerase II (Pol II) and loaded onto spliced- mRNA in a splicing- dependent manner, which is crucial for subsequent export<sup>5,7</sup>.  
+
+Perhaps the most crucial factor in the assembly of the TREX complex is UAP56 (Sub2 in yeast). During splicing, UAP56 is loaded onto pre- mRNA through the interaction with U2AF65<sup>8</sup> and in turn it regulates spliceosome assembly<sup>9,10</sup>. UAP56 interacts with the THO subcomplex in an ATP- independent manner<sup>11</sup>. When ATP binds UAP56, it recruits CIP29, ALYREF, CHTOP, PDIP3, and ZC11A into the TREX complex<sup>2,3,6</sup>. Thus, the TREX complex exists in two states that remodel depending on whether ATP binds to UAP56. We term the ATP- unbound form as the apo- TREX complex and the ATP- bound one as the ATP- TREX complex to distinguish both complexes<sup>11</sup> (Fig.1A). The formation of the ATP- TREX complex drives the export of bound mRNA because ALYREF, CHTOP, and THOC5 act as adaptors of the NXF1- NXT1 heterodimer which functions in the final step of the mRNA export<sup>2,12,13</sup>.  
+
+In mammals, UAP56 has a paralogue that is \(90\%\) identical, URH49<sup>14</sup>. Furthermore, we have previously shown that UAP56 and URH49 form distinct apoproteins. UAP56 forms the apo- TREX complex, and URH49 forms the apo- Alternative- mRNA- EXport (AREX) complex. Unlike the apo- TREX complex, the apo- AREX complex contains CIP29 and it does not contain the THO subcomplex<sup>15</sup>. Like the apo- TREX complex, the apo- AREX complex is remodeled to ATP- complex when ATP is loaded onto URH49, and accesses NXF1- NXT1 heterodimer for mRNA
+
+<--- Page Split --->
+
+export<sup>11</sup>. Irrespective of whether the precursor is an apo- TREX or an apo- AREX complex, ATP- complexes resemble each other and are called the ATP- TREX complex.  
+
+In addition, each helicase selectively exports a specific subset of mRNAs. UAP56 and URH49 selectively regulate distinct subsets of key mitotic regulators<sup>15</sup>. URH49 is also required for the gene expression involved in cytokinesis<sup>11</sup>. Besides, both helicases and the components of their respective complexes are required for a variety of physiologically important roles in lifelong cell differentiation<sup>16,17,18</sup>. Consequently, abnormalities of their mRNA export pathways including disruption of their expression have been associated with serious diseases such as cancer and neurodegenerative disorders<sup>7,19</sup>. Thus, the evolutionarily diversified mRNA export pathways formed by UAP56 and URH49 contribute to fine- tuned gene expression and are required for various physiological events.  
+
+Therefore, elucidation of the functional machinery of UAP56 and URH49 and their differences is important, not only for a better understanding of gene regulation in higher organisms but also for an understanding of a variety of diseases caused by disruption of these two helicases. DEAD- box family helicases generally bind RNA in a sequence- independent manner, and target recognition is primarily provided via partner proteins<sup>20</sup>. Thus, identifying the compositions of the apo- TREX and the apo- AREX complexes, and elucidating the molecular basis of the involvement of UAP56 and URH49 in complex formation, may be the key to understanding their function. However, the factor(s) of the apo- AREX complex are unknown except for CIP29, which is also in the ATP- TREX complex (Fig.1A). Importantly, the mechanisms by which UAP56 and URH49 form distinct complexes, despite their extensive homology, remains unknown.  
+
+In this study, we first used tandem- immunoprecipitation and mass spectrometry to investigate the factors of the apo- AREX complex. Then, we determined the reason why UAP56 and URH49 form different apo- complexes.
+
+<--- Page Split --->
+
+## Result  
+
+## Identification of the novel apo-AREX components  
+
+To analyze the composition of the apo- AREX complex, we performed immunoprecipitation using nuclear extract prepared from Flp- In T- REx 293 cells expressing either FLAG- UAP56 or FLAG- URH49 in the ATP- depleted condition. FLAG- UAP56 or FLAG- URH49 coprecipitated different components: FLAG- UAP56 was associated with the apo- TREX components (THOC1, THOC2, and THOC5) and FLAG- URH49 precipitated the apo- AREX component CIP29 (Fig.1B). These interactions are consistent with previously reported different interactions of endogenous UAP56 and URH49<sup>11,15</sup>. We then added ATP and found that the ATP- TREX components (THOC1, THOC2, THOC5, ALYREF, and CIP29) interacted with both FLAG- UAP56 and FLAG- URH49. We refer to the ATP- TREX complex containing UAP56 as the ATP- TREX (UAP56) complex and the ATP- TREX complex containing URH49 as the ATP- TREX (URH49) complex. This remodeling was also observed in the presence of ADP or AMP- PNP, indicating that the ATP binding, but not the ATP hydrolysis is sufficient to exert the complex remodeling (Fig.1B, Extended Data Fig.1A). In the FLAG- URH49 precipitate, we detected many candidates for the novel apo- AREX components. To identify these factors as authentic apo- AREX components, we performed tandem purification of the apo- AREX complex with the nuclear extract expressing both known apo- AREX components, FLAG- URH49 and HA- CIP29, and identified isolated factors by LC- MS/MS (Fig.1C, Extended Data Fig.1B- D).  
+
+Among the coimmunoprecipitated factors of FLAG- URH49 and FLAG- UAP56, we observed enrichment of splicing- associated factors, indicating that involvements of both helicases with the splicing process (Extended Data Fig.2A- C see also Table S1). Moreover, various RNA- binding proteins are found to bind to URH49, but not UAP56. RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM were reliably detected from the tandem immune- precipitate as well as from the FLAG- URH49 precipitate, but not in the FLAG- UAP56 precipitate (Fig.1C, Extended Data Fig.2B). RUVBL1 and RUVBL2 form heterodimers, as do ILF2 and ILF3<sup>21,22</sup>. HNRNPM interacts with ILF2, ILF3, and other factors<sup>23</sup>. Thus, we focused on these factors, and interactions between these factors and URH49 were confirmed by immunoblotting of FLAG- URH49 precipitate (Fig.1D).
+
+<--- Page Split --->
+
+To confirm the physical association of the apo- AREX candidates with URH49, we generated cell lines stably expressing FLAG- RUVBL1, RUVBL2, ILF2, ILF3, or HNRNPM, respectively. Anti- FLAG- precipitates from nuclear extracts expressing each apo- AREX candidate efficiently captured URH49 whereas there was no obvious binding to THOC1 and ALYREF, components of the apo- and the ATP- TREX complex, was observed (Fig.1E- G). Please see Extended Data Fig.2D in which complex each factor is present. The interactions between RUVBL1 and RUVBL2, as well as between ILF2 and ILF3 were sustained in the presence of ATP, as previously reported<sup>21,22</sup>. In contrast, these factors dissociated from URH49 upon ATP addition. These results indicate that these factors interact with URH49 as the apo- AREX complex but do not interact with the ATP- TREX complex (Extended Data Fig.2D).  
+
+## The novel apo-AREX components are associated with URH49-target mRNA processing and export  
+
+Next, we evaluated the functional significance of novel apo- AREX candidates in mRNA processing and export. In addition to ILF3 (their alternative name is NF110), NF90, a truncated isoform of ILF3, was produced from ILF3 gene. This factor also interacted with ILF2 as in the case of ILF3<sup>21</sup>. A previous study reported that the use of an ILF3- specific siRNA can deplete ILF3 expression without affecting NF90 expression, while the knockdown of ILF2 downregulated the expression of both ILF3 and NF90<sup>21</sup>. The depletion of either RUVBL1 or RUVBL2 causes a co- depletion of the other<sup>24</sup>. Thus, we depleted RUVBL1, ILF3, HNRNPM, and CIP29, a known apo- AREX component by siRNA- mediated knockdown (Extended Data Fig.3A). Depletion of either factor induced the nuclear accumulation of poly(A)<sup>+</sup> RNAs, which co- localized with nuclear speckles (Fig.2A- C). This observation is probably reflecting the perturbed mRNA splicing and export by knocking down either factor as shown by previous reports<sup>25,26,27</sup>. Similar results were observed with other cell lines and with other siRNAs of any of the factors (Extended Data Fig.3B- D). These observations indicated that our apo- AREX candidates function in mRNA processing and export.  
+
+To further clarify the function of these factors in the apo- AREX complex, we depleted each apo- AREX candidate and assessed its effect on the cytoplasmic mRNA expression of UAP56 or URH49 targets. Depletion of each factor specifically reduced the expression of the URH49 targets but did not cause a reduction of the UAP56 targets,
+
+<--- Page Split --->
+
+indeed in some cases it upregulated them (Fig.2D). Such upregulation of UAP56 targets was also observed when URH49 and CIP29 were depleted<sup>11</sup>, suggesting that depletion of the apo- AREX component probably enhances the UAP56 export pathway as a compensatory mechanism. These results indicated that each factor functions as the apo- AREX complex, and specifically regulates URH49- target mRNA export.  
+
+## A single amino acid difference between UAP56 and URH49 impacts apo-complex formation and function  
+
+Subsequently, we investigated how UAP56 and URH49 form distinct apo- complexes despite their high homology. DEAD- box RNA helicase contains a conserved core region with two domains (N- domain and C- domain, respectively), a linker region between them, and terminal regions<sup>20</sup> (Fig.3A, Extended Data Fig.4). We hypothesized that differences in a specific region(s) between UAP56 and URH49 are important to form their distinct apo- complexes. To identify the region(s) determining the distinct apo- complexes formed, we generated plasmids expressing mutants in which various regions of UAP56 and URH49 were swapped and examined apo- complex formation (Fig.3A, Extended Data Fig.5). The N- and C- terminal regions are relatively different compared to core regions of UAP56 and URH49. However, swapping of either terminal region did not affect the apo- complex formation (Extended Data Fig.5A). In contrast, mutants swapped of each N- domain (described as “UAP56 N- core” and “URH49 N- core”) dramatically altered apo- complex formation (Extended Data Fig.5B- C), suggesting the N- domain determines which apo- complex forms.  
+
+In humans, UAP56 and URH49 have twelve amino acid differences in the N- domain (Fig.3A, Extended Data Fig.4). Subsequently, we analyzed twelve point mutants in which different individual amino acids are swapped. Strikingly we found that the URH49 C223V mutant specifically switched the complex formation from the apo- AREX complex to the apo- TREX complex (Fig.3B, Extended Data Fig.5D). UAP56 V224C, the mutant corresponding to URH49 C223V, did not alter the apo- complex formation (Extended Data Fig.5D). To further examine the possibility that other amino acid differences besides UAP56- V224 and URH49- C223 also contribute to their distinct complex formation, we generated the UAP56 mutant described as “UAP56 N- core C224V”, in which the N- domains of UAP56 other than UAP56- V224 were replaced with the N- domains of URH49. This mutant lost the ability to form the apo
+
+<--- Page Split --->
+
+AREX complex but did form the apo- TREX complex (Extended Data Fig.5E). These results indicate that the difference between UAP56- V224 and URH49- C223 is the crucial determinant of apo- complex formation.  
+
+It has been reported that UAP56 and URH49 export not only mRNAs but also circular RNAs which are generated via "back- splicing"28. In that report, UAP56 plays a role in long circular RNAs export while URH49 exports short circular RNAs. The four different amino acids that are located in their N- domains between UAP56 and URH49 determine their specificity for circular RNAs. However, we did not find a difference in complex formation between the apo- TREX and the apo- AREX in the mutants with these four amino acid substitutions (Extended Data Fig.5F). Thus, the mechanism of mRNA export appears to be different from that of circular RNA export.  
+
+We next examined whether alteration of the apo- complex formation affects mRNA export activities of the two helicases. The depletion of UAP56 or URH49 induced bulk nuclear poly(A) \(^+\) RNA accumulation, respectively15,29. The forced expression of siRNA- resistant UAP56 rescued the nuclear poly(A) \(^+\) RNA accumulation induced by endogenous UAP56 knockdown, but did not rescue the nuclear poly(A) \(^+\) RNA accumulation provoked by the disruption of URH49, and vice versa (Fig.3C- D, Extended Data Fig.6). This result reflects that UAP56 and URH49 export distinct subsets of bulk mRNA substrates and do not the other15. URH49 C223V and URH49 chimera mutant, URH49 N- core which form the apo- TREX complex, could specifically rescue the nuclear poly(A) \(^+\) RNA accumulation caused by the knockdown of endogenous UAP56. In addition, UAP56 chimera mutant UAP56 N- core, which form the apo- AREX complex could rescue the nuclear poly(A) \(^+\) RNA accumulation induced by the disruption of endogenous URH49. These data clearly demonstrate that mRNA export selectivity was controlled at the apo- complex formation step. Taken together, the formation of distinct apo- complex due to the difference in a single amino acid between UAP56 and URH49 has a key role in the selective mRNA export by the two helicases.  
+
+## UAP56 and URH49 form different apo-structures, but with similar ADP binding structures  
+
+DEAD- box helicases have similar structural features20. In the apo- state, DEAD- box family proteins adopt a variety of open structures with the configuration of N- domain and C- domain different for each member. The ATP- binding triggers the
+
+<--- Page Split --->
+
+rearrangement of the N- domain and the C- domain into similar closed- structure via interactions of ATP with both domains. These structural features determine what kind of complex forms according to their apo- and ATP- binding state. In fact, the remodeling of the apo- AREX complex to the ATP- TREX complex (URH49) dramatically altered the protein composition between the two complexes (Fig. 1D- G). These results led us to hypothesize that the structures of UAP56 and URH49 in the apo- and ATP- binding states dictate their apo- and ATP- complex formation. Supplying ADP caused the complex remodeling of UAP56 and URH49 as well as the addition of AMP- PNP, a non- hydrolysable analog of ATP (Extended Data Fig.1A), indicating that the ADP- bound structures of two helicases resemble the ATP- bound structures. Thus, we compared the structural features of UAP56 and URH49 in the apo- and the ADP- bound states by limited proteolysis.  
+
+The sites of the primary amino acid sequence predicted to be digested by trypsin were the same in both helicases. However, the digested fragments and thus the structures in the apo- state differed between UAP56 and URH49 (Fig.4A, B). As for UAP56, the digested products in the presence of ADP were similar to those in the absence of ADP. Previously, crystal structures of both apo- and ADP- bound form UAP56ΔN42, which lacked N terminal 42 residues of UAP56, were reported<sup>30</sup>. In that study, UAP56ΔN42 exhibits a relatively closed conformation in the apo- state compared to other DEAD- box proteins<sup>30</sup>. And ADP binding induces a slight structural rearrangement only around the ATP- binding pocket without the configurational change of their N- and C- domains. Our observations seem to reflect these findings. On the contrary, the digestion pattern of URH49 in the presence of ADP differed from that in the absence of ADP, and changed to that of UAP56 upon the addition of ADP. This indicates that unlike UAP56, URH49 undergoes the significant conformational change upon binding of ADP.  
+
+We also generated the UAP56ΔN42 and URH49ΔN41 which is the URH49 mutant corresponding to UAP56ΔN42. Their digested fragments were different under the apo- condition and became similar in the presence of ADP as well as in the case of UAP56 and URH49 (Fig.4C, D). These results indicate that UAP56ΔN42 has a similar structure to UAP56 and URH49ΔN41 to URH49. We, next, estimated the cleavage sites of the UAP56ΔN42 and URH49ΔN41 by detecting peptides using LC- MS/MS analysis (described A1- 4 and R1- 4 in Fig.4C, D). The A1 and A2 fragments generated in the
+
+<--- Page Split --->
+
+absence of ADP covered peptides from the N- domain and linker region of UAP56ΔN42. The peptide composition of R1 fragment was similar to that of A1 fragment while the peptide composition of R2 fragment contained the C- domain and the linker region, and was completely different from that of A2 fragment (Fig.4C, E). These results indicated that the sensitive site of UAP56 digestion by trypsin was different from that of URH49, probably based on their distinct structures in the absence of ATP (Fig.4E bottom, Extended Data Fig.7A). The A3 and 4 fragments generated in the presence of ADP had the same digestion pattern with the R3 and 4 fragments (Fig.4D, F), indicating that URH49 underwent a significant structural change by the loading of ADP and UAP56 did not.  
+
+To further confirm that URH49 underwent the structural rearrangement upon ADP- binding, we employed mutants lacking the ATP binding activities: UAP56ΔN42 K95N and URH49ΔN41 K94N<sup>11</sup>. These mutants had the same digestion pattern as one another in the presence of ADP (Extended Data Fig.7B, C). From these results, we concluded that the two helicases form different apo- structures, but were remodeled to similar structures on ADP binding. Importantly, the digested pattern of URH49 C223VΔN41 was similar to that of UAP56ΔN42 (Fig.4C). This indicates that V224 of UAP56 and C223 of URH49 play important roles in forming their different apo- structures, and raises the possibility that the structural feature of UAP56 and URH49 were associated with their apo- and ADP- complex formation.  
+
+## The Different Structures of apo-UAP56 and apo-URH49  
+
+To analyze the difference between both apo- structures, we solved the crystal structure of URH49ΔN41 by x- ray diffraction (8IJU) and compared it with the published apo- UAP56ΔN42 structure (1XTI)<sup>30</sup>. The folds of two N- and C- domains in URH49ΔN41 are essentially the same as the apo- UAP56ΔN42 structure<sup>30</sup>. However, the N- and C- domains of URH49ΔN41 were located in distinct positions from the respective domains of UAP56ΔN42 (Extended Data Fig.8A, Table 1). The crystal of URH49ΔN41 contained SO<sub>4</sub><sup>2-</sup> and polyethylene glycol (PEG) around the interspace between N- and C- domains and the ATP- binding pocket (Extended Data Fig.8B). This raised the possibility that the structure of apo- URH49ΔN41 containing SO<sub>4</sub><sup>2-</sup> and PEG differs from that of UAP56ΔN42 because of its interaction with these compounds.
+
+<--- Page Split --->
+
+To exclude this possibility, we generated URH49ΔN41 apo-structure models lacking \(\mathrm{SO_4^{2 - }}\) and PGE by molecular dynamics analysis<sup>31</sup>. Among these models, the Fr48 model is the representative conformation without thermodynamical destabilization (Extended Data Fig.8C, D). This structural model showed essentially the same structure to that of URH49ΔN41 containing \(\mathrm{SO_4^{2 - }}\) and PEG (Extended Data Fig.8A, E)). Thus, we concluded that \(\mathrm{SO_4^{2 - }}\) and PGE did not significantly affect the overall structure of URH49ΔN41 and continued our analysis of this model structure as the authentic apo- URH49ΔN41.  
+
+The apo- URH49ΔN41 model exhibited three different structural features compared to the apo- UAP56ΔN42 structure (Fig.5A). Firstly, although the amino acid sequence of the linker is the same in both helicases, the linker of UAP56 did not have any secondary structure, while this region of URH49 showed a clearly oriented \(\alpha\) - helical structure. This structural difference in the linker part was also implicated by the finding that the R2 fragment containing the linker and C domain was derived exclusively from URH49ΔN41 but not from UAP56ΔN42 (Fig.4E). Secondly, the relative orientation of the N- and C- domains differed between the apo- UAP56��N42 and the apo- URH49ΔN41. The overall folds of N- or C- domains were similar to each other, which characters are conserved in the DEAD- box families<sup>30</sup> (Fig.5B, Extended Data Fig.8E). The distinct spatial arrangement of the N- and C- domains in DEAD- box family proteins affects their complex formation<sup>20</sup>, implying that these differences may contribute to their unique apo- complex formation. We did not find any clear differences in the spatial arrangement of the residue V224 of UAP56 and the residue C223 of URH49 within their apo- conformation (Extended Data Fig.8F). Therefore, we described details about this point in the Discussion section. Finally, the loop structures in the C- domain formed by residues 344- 354 of UAP56 and residues 343- 353 of URH49 (hereafter referred to as C- domain loop) are positioned differently from each other (Fig.5C). In addition, the C- domain loop of URH49 covers its own ATP binding pocket. Consistent with this observation, URH49 had a lower ATP- binding and ATP- dependent helicase activity than UAP56 (Extended Data Fig.8G, H), implying that the C- domain loop prevented ATP from binding to the apo- URH49. Therefore, the URH49 C- domain loop likely gives URH49 less affinity for ATP in the apo- state, allowing it to maintain a different conformation from the apo- state of UAP56.
+
+<--- Page Split --->
+
+## Evolutionary diversified apo-structures from Sub2 to UAP56 and URH49  
+
+Finally, we investigated whether the structural features of either UAP56 or URH49 observed in their apo states were conserved in yeast Sub2, the ancestor gene of UAP56. Firstly, we performed the limited proteolysis of Sub2ΔN59, a mutant corresponding to UAP56ΔN42, in ATP depleted conditions. The digestion pattern of Sub2ΔN59 was similar to that of UAP56ΔN42 (Extended Data Fig.8I). Secondly, we compared the structural difference of C- domains in UAP56, URH49, and Sub2 extracted from the co- crystal structure of Sub2- THO \(^{32}\) . The location of the C- domain loop of Sub2 was similar to that of UAP56 (Fig.5D). These data suggested that the structure of UAP56 is evolutionarily conserved with Sub2, while URH49 has diversified from UAP56 during evolution to form a different apo- structure.
+
+<--- Page Split --->
+
+## Discussion  
+
+In this study, we uncover unknown apo- AREX components and the molecular basis for their distinct complex formation, which is crucial for the functional divergence of both helicases playing distinct roles in mRNA processing and export.  
+
+## The apo-AREX complexes regulate gene expression of URH49 targets genes  
+
+With the exception of CIP29, the detail of the apo- AREX composition were not determined. Here, we identified RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM as factors which interact with URH49 by two- step affinity purification based on the apo- AREX complex under the ATP depletion condition. Depletion of each apo- AREX component induced the accumulation of poly(A) \(^+\) RNA in nuclear speckles. mRNAs with retained introns are tethered in nuclear speckles and thus inefficiently exported to the cytoplasm \(^{25,26}\) . This observation has led to the idea that the apo- AREX complex may have a link to upstream mRNA processing such as splicing as well as the apo- TREX complex does \(^{5,33}\) .  
+
+In addition, all of the newly identified apo- AREX components have other previously describe roles in nuclear RNA dynamics. RUVBL1 and RUVBL2 form heterodimers and function in chromatin remodeling as INO80 and SRCAP complexes \(^{22}\) . RUVBL1 and RUVBL2, as TIP160 complex components, are also involved in the regulation of transcription via histone acetylation at promoters \(^{22}\) . HNRNPM, which belongs to the hnRNPs family, together with various interacting factors, contributes to many aspects of RNA metabolism \(^{23,34,35}\) . ILF2 and ILF3 also form heterodimers and function in RNA splicing as the LASR complex with numerous proteins including HNRNPM \(^{23,35}\) . Moreover, CIP29 contains an evolutionarily conserved DNA- binding motif, SAF domain, and binds to DNA, which led to the speculation that CIP29 functions in transcription \(^{36,37}\) . These findings suggested that the novel factors identified in this study may form multiple complexes including the apo- AREX complex and are widely involved in RNA metabolism from chromatin regulation to splicing. Further studies will uncover how two closely related complexes recognize their target transcripts, which will reveal the individual function of these two complexes in coupling the processes from mRNA transcription to export.
+
+<--- Page Split --->
+
+## The apo-AREX complexes is implicated in cell proliferation and cancer progression  
+
+UAP56 is continuously expressed during the cell cycle, while URH49 is expressed specifically during the proliferation phase and not during the quiescent phase<sup>14</sup>. URH49 are required for gene expression of subsets of key regulators of mitosis<sup>15</sup> and cytokinesis<sup>11</sup>. The apo- AREX components were also required for the expression of representative targets regulated by URH49. These results indicate that the apo- AREX complex is implicated in cell proliferation via regulation of its target gene expression.  
+
+Aberrant expression of UAP56 and URH49 is involved in tumorigenesis and cancer progression. UAP56 is upregulated in colorectal and ovarian cancers, and is associated with their progressions<sup>38,39</sup>. The association between URH49 and cancer has been reported more often observed than that of UAP56. In Cancer Genome Atlas (TCGA), pan- cancer cohort analysis showed that URH49 is upregulated in 18 cancer types than normal tissue<sup>40</sup>. Actually, aberrant up- regulation of URH49 is observed in various cancer tissues and cancer cell lines, and is positively correlated with advanced clinical stage and poor prognosis<sup>40,41,42</sup>. URH49 is important for the genes expression involved in cell proliferation and promotes malignancy of these cancer<sup>40,42</sup>. CIP29 is highly expressed in various cancer and is associated with cancer malignancy<sup>7,43</sup>. Overexpression of RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM are observed in various cancers with their progression (RUVBL1 and RUVBL2:22,44, ILF2 and ILF3:45,46,47, HNRNPM:48,49). Thus, it is possible that the apo- AREX components are important regulators of gene expression in cancer. Therefore, agents which impair the apo- AREX complex expressions and/or activity could be potential targets for cancer therapy. Indeed, YM155, an inhibitor of ILF3, and CB- 6644, an inhibitor of RUVBL1 and RUVBL2 sub- complex, exhibit anticancer activities<sup>50,51</sup>. The research focusing on the regulation of the apo- AREX complex may contribute to a therapeutic benefit in cancer.  
+
+## Complex formation and structure of UAP56 and URH49  
+
+We observed that UAP56 and URH49 have different structural features in the apo- state and remodeled to similar structures upon the ADP binding. The structure of UAP56 and URH49 in apo- state showed the distinct configuration of the N- and C-
+
+<--- Page Split --->
+
+domain probably due to different linker structures. UAP56 interacts with the THO subcomplex via their N- and C- domains within the reconstituted UAP56- THO subcomplex<sup>52</sup>. These interactions were also conserved in the crystal structure of yeast Sub2- Tho<sup>32,53</sup>. These findings suggest that the spatial arrangement of N- and C- domains of UAP56 is important for the formation of the apo- and ATP- TREX (UAP56) complex. Therefore, the difference in the configurations between UAP56 and URH49 structures may play a pivotal role in their unique complex formation.  
+
+The loop structure within the C- domain of apo- URH49ΔN41 covers the ATP- binding pocket in their N- domain and prevents its own ATP binding. Although several DEAD- box proteins inhibit their ATP binding by intramolecular interactions<sup>20</sup>, the arrangement of the loop in the C- domain is not observed among other DEAD- box proteins, suggesting that URH49 has evolutionarily acquired a unique structure to repress ATP binding. Consistent with our findings, Sub2 exhibited sufficient helicase activity as well as UAP56<sup>54</sup>. In addition, CIP29 stimulates ATPase activity of URH49, followed by ATP binding in URH49<sup>55</sup>, implying that URH49 forms the apo- AREX complex as a steady state, then, remodels to the ATP- TREX complex in the cell.  
+
+Orthologs of UAP56 and URH49 are present in vertebrates, while only the ortholog of UAP56 is present in insects, implying that the ortholog of URH49 diversified during the evolution between vertebrates and invertebrates. Furthermore, amino acids corresponding to human UAP56- V224 and URH49- C223 are already present in orthologs of UAP56 and URH49. Thus, the amino acid substitution between UAP56- V224 and URH49- C223 probably occurred after diversification. But the precise effect of the differences between UAP56- V224 and URH49- C223 on their apo- conformation is not yet fully understood this time. Because UAP56- V224 and URH49- C223 are present inside N- domains of the apo- UAP56ΔN42 and URH49ΔN41, substitutions of these amino acids do not appear to affect the surface structures. One attractive possibility that needs to be verified later is that these substitutions affected the linker structure of UAP56 and URH49, resulting in the different orientations of the N- and C- domains between the two helicases.  
+
+Integrating our findings into previous observations, we proposed the following model that UAP56 and URH49 form the apo- TREX and - AREX complex based on their apo- conformation (Fig.6). The binding of ATP into UAP56 and URH49 promotes the conformational change to a highly similar closed conformation, triggering
+
+<--- Page Split --->
+
+the remodeling of the respective apo-complex to the ATP- TREX complex. It has been implicated that UAP56 and URH49 function in selective mRNA export by forming respective complex formation. Therefore, we provided the possibility that diversified apo- structures of UAP56 and URH49 derived from Sub2 have contributed to the organization of gene regulation in humans. Further progress in genome analysis of other species is expected to advance our understanding of the diversification of UAP56 and URH49.  
+
+Interestingly, amino acids in UAP56 and URH49 required for selective circular RNA export are not linked to the apo- complex formations. ATPase activities of UAP56 and URH49 are not required for circular RNA export<sup>28</sup>. While ATP loading and ATPase activity are known to be essential for their function of mRNA export<sup>11,56</sup>. In addition, there is no significant difference in length between the mRNAs selectively exported by the two helicases and these pre- mRNAs<sup>11</sup>. Thus, the underlying mechanism of circular RNA export needs to be investigated separately from that of mRNA export including whether the complex formation is required for circular RNA export.  
+
+During evolution, many RNA- binding proteins have functionally diversified to execute well- tuned gene expression contributing to the complexity of living organisms. These include UAP56 and URH49 which have diversified to form different complexes and function in selective mRNA processing and export<sup>15</sup>. In addition to both helicases, several key mRNA processing and export factors such as NXFs, NXTs, DDX19s, and SR proteins have evolutionarily diversified from yeast to human. Some factors have gained different target specificities from their originated paralogs, but the molecular mechanism behind these differences is mostly unknown<sup>57</sup>. Further elucidation of the diversification of mRNA export- related proteins will uncover the mechanistic insight into the accurate gene expression through mRNA processing and export in humans and how it has developed during evolution.  
+
+## ACKNOWLEDGMENTS  
+
+We thank Mr. Yuzo Watanabe for help with the LC- MS/MS analysis and our lab members for their constructive discussions. Diffraction data were collected at the BL26B1 and BL44XU stations of SPring- 8 (Hyogo, Japan) with the approval of JASRI (proposal nos. 2015A1063, 2015B2063 and 2017B6750). This work was supported in part by "Grants- in- Aid" from JSPS KAKENHI (Grant Numbers 26292053, 17K19232,
+
+<--- Page Split --->
+
+19K22280, 19H02884, 21K19078 and 22H02264 to S.M, 19K15807 to K.F). This work was also supported in part by "Grants- in- Aid" from The Sasakawa Scientific Research Grant from The Japan Science Society to K.F.  
+
+## AUTHOR CONTRIBUTIONS  
+
+K.F. and S.M. conceived and designed this study; K.F. performed the experiments and analyses, organized the data and drafted the manuscript; M.I, M.I, K.H, Y.I, H.Y, and T.Y performed biological experiments; M.K. performed MD analysis; B.M. performed crystal structure analysis; K.F., N.F, A.M, and S.M. analyzed the results and wrote the paper. All authors reviewed the final manuscript.  
+
+## DECLARATION OF INTERESTS  
+
+The authors declare no competing interests.  
+
+## CONTACT FOR REAGENT AND RESOURCE SHARING  
+
+For more information and requests regarding resources and reagents, please contact the Lead Contact, Seiji Masuda (smasuda@nara.kindai.ac.jp).  
+
+## METHOD DETAILS  
+
+## Cell culture and establishment of stable cell line  
+
+Flp- In T- REx 293, U2OS, MCF7, A549 cells were maintained in Dulbecco's Modified Eagle's Medium (Fujifile Wako, Tokyo, Japan) supplemented with \(10\%\) heat- inactivated fetal bovine serum at \(37^{\circ}\mathrm{C}\) . Flp- In T- REx 293 cells stably expressing 3x FLAG- tagged protein were obtained by the transfection of pcDNA5 3x FLAG- tagged protein expression vector with pOG44, respectively.  
+
+## Reagents and antibodies, preparation of serum  
+
+4', 6- diamidino- 2- phenylindole (DAPI) was purchased from Fujifilm Wako. Antibodies were obtained as follows: FLAG M2 mouse monoclonal antibody, mouse anti- \(\beta\) - actin antibody, rabbit anti- HNRNPM antibody and mouse anti- SRRM2 antibody (Sigma- Aldrich Japan, Tokyo, Japan), HA (12CA5) mouse monoclonal antibody (GeneTex, Irvine, CA), mouse anti- GAPDH antibody (Fujifilm Wako), antibodies against THOC1, THOC2, THOC5, ALYREF, CIP29, UAP56 and URH49 have been described previously<sup>11</sup>. Anti- RUVBL1, anti- RUVBL2, anti- ILF2, and anti- ILF3 sera
+
+<--- Page Split --->
+
+were prepared from immunized rats as described previously<sup>58</sup> in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Animal Committee in Kyoto University (Animal experiments were approved by the Committee on the Ethics of Animal Experiments of Kyoto University, Experiment permission number: Lif- K17002). The antibodies used in this study are listed in supplemental Tables S3.  
+
+## Plasmids, primers and siRNAs  
+
+To construct the following plasmids, the fragments were obtained by PCR amplification with the addition of restriction enzyme sites or infusion enzyme sites at both ends. pcDNA5- 3xFLAG and pcDNA5- HA vectors were generated as described previously<sup>11,59</sup>. 3xFLAG- UAP56, 3xFLAG- URH49, 3x FLAG- CIP29, 3xFLAG- RUVBL1, 3xFLAG- RUVBL2, 3xFLAG- ILF2, 3xFLAG- ILF3, and 3xFLAG- HNRNPM expression vectors were generated by the insertion of the respective open reading frame into pcDNA5- 3xFLAG, respectively. The HA- CIP29 expression vector was generated by the insertion of the CIP29 open reading frame into pcDNA5- HA. To construct GST- UAP56 and GST- URH49, and their derivative mutant expression plasmids, respective open reading frames of UAP56, URH49 and their mutants were inserted into pGEX6p2. The GST- Sub2Δ59 (60- 446 amino acids) expression plasmid was constructed by inserting the respective region into pGEX6p2. MBP- RUVBL1 (250- 456 amino acids), MBP- RUVBL2 (1- 225 amino acids), MBP- ILF2 (240- 390 amino acids), and MBP- ILF3 (280- 355 amino acids) expression plasmids were constructed by inserting their respective region into pMALc2X. To construct mutants of FLAG- UAP56, or - URH49 expression plasmids, overlap extension PCR was performed to induce the mutation. The construction of the plasmids was confirmed by sequencing. The primers and siRNAs used in this study are listed in supplemental Tables S4 and Tables S5.  
+
+## Plasmids or siRNA transfection  
+
+Transient transfection of siRNA and plasmids was performed using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's instructions.  
+
+## Total, cytoplasmic and nuclear RNA isolation
+
+<--- Page Split --->
+
+Total RNA was isolated by Sepasol- RNA I super G (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions. For cytoplasmic RNA preparation, the cells were treated with lysis buffer (20 mM Tris- HCl (pH 8.0), 200 mM NaCl, 1 mM MgCl2, 1% NP- 40) on ice for 5 min. The cytoplasmic fraction was isolated by brief spin while the nuclear fraction was prepared from the precipitate.  
+
+## Quantitative and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR)  
+
+Quantitative RT- PCR (RT- qPCR) was performed with TB Green Premix Ex Taq II (TakaraBio, Tokyo, Japan) and analyzed by Thermal Cycler Dice real time system II (TakaraBio). PGK1 was used for standardization. The quantity of each mRNA was calculated by threshold cycle (Ct) values. The relative expression of each mRNA was evaluated by the values of \(2^{\circ}[\mathrm{Ct(TBP)} - \mathrm{Ct(eachmRNA)}]\) . Primer sets are listed in Table S6.  
+
+## Immunoprecipitation, immunoblotting, LC-MS/MS analysis and Silver staining  
+
+Immunoprecipitation, immunoblotting, LC- MS/MS analysis and Silver stainingPreparation of nuclear extract and immunoprecipitation were performed as described previously<sup>11</sup>. Briefly, nuclear extract was incubated for 30 min at \(20^{\circ}\mathrm{C}\) for the depletion of ATP and centrifuged to recover the supernatant. Then, RNaseA (100 ng/μL), ATP (500 μM), MgCl<sub>2</sub> (3.2 mM) and creatine phosphate (20 mM) were added to the nuclear extract, and the reaction mixture was incubated for 30 min at \(30^{\circ}\mathrm{C}\) . In the ATP (- ) condition, ATP, MgCl<sub>2</sub>, and creatine phosphate were omitted. After a brief spin, the clear supernatant was mixed with anti DYKDDDDK tag antibody beads (Fujifilm Wako) or anti HA antibody beads (Fujifilm Wako) and rotated overnight at \(4^{\circ}\mathrm{C}\) . The beads were extensively washed with PBS containing 0.1% TritonX100, 0.2 mM PMSF and 0.5 mM DTT to remove nonspecifically bound proteins. The proteins attached to the beads were dissolved in SDS sample buffer (250 mM Tris- HCl, 1% sodium lauryl sulfate), 0.002% bromophenol blue and 40% Glycerol for 10 min at \(37^{\circ}\mathrm{C}\) . The eluate was recovered to a new tube and DTT was added to 10 mM, and boiled for 2 min. For tandem immunoprecipitation, the proteins attached to the beads were eluted with FLAG peptide (M&S TechnoSystems Inc, Osaka, Japan) or HA peptide (MBL, Nagoya, Japan).
+
+<--- Page Split --->
+
+Samples were separated by SDS- polyacrylamide gel electrophoresis (SDS- PAGE) and blotted onto polyvinylidene difluoride membrane (Pall, Ann Arbor, MI). The blotted membrane was blocked with PBS containing \(0.1\%\) polyoxyethylene sorbitan monolaurate (Tween20) and \(5\%\) skim milk for \(1\mathrm{h}\) and reacted with primary antibodies at \(4^{\circ}\mathrm{C}\) overnight with gentle rotation. The membrane was extensively washed with PBS containing \(0.1\%\) Tween20. Secondary antibody conjugated with horseradish peroxidase was reacted with the membrane by rotating for \(2\mathrm{h}\) . After the extensive washing, the membrane was reacted with a chemiluminescence reagent (Millipore, Darmstadt, Germany). Signals were detected with LAS 4000 mini (GE Healthcare Japan, Tokyo, Japan).  
+
+The LC- MS/MS analysis was performed by Q Exactive Plus (Thermo Fisher Scientific, Waltham, MA). As outputs of the LC- MS/MS analysis, prot score was calculated using Mascot software (Matrix Science, London, UK). For factors with different prot_acc but the same GeneName, only the largest prot_score is listed. For comparison, protein scores were calculated by subtracting the prot_score of the control from each data. Gene ontology (GO) was analyzed using Database for Annotation, Visualization and Integrated Discovery (DAVID: version \(6.7)^{60}\) .  
+
+For silver staining, proteins were separated with SuperSep™ Ace, \(5\% - 20\%\) , 17- well (Fujifile Wako). Silver staining was performed as described previously.  
+
+## Immunofluorescence staining  
+
+Cells ( \(5 \times 10^{4}\) cells/mL) on glass coverslips in a 12- well plate were cultured for \(24\mathrm{h}\) and transfected with siRNA or plasmid. After a \(48\mathrm{h}\) incubation, cells were fixed in \(4\%\) formaldehyde in PBS, permeabilized with \(0.1\%\) Triton X- 100 in PBS, and blocked with \(6\%\) bovine serum albumin (BSA) in PBS. The coverslips were reacted with primary antibodies in \(2\%\) BSA in PBS, secondary antibody conjugated with Alexa- 488 or Alexa- 594 (Molecular Probes, Eugene, OR) and DAPI to counterstain the nuclei. Fluorescence images were obtained with a fluorescent microscopy, Axioplan 2 (Carl Zeiss, Germany) or FV10i (Olympus, Tokyo, Japan), a laser scanning confocal microscopy, using the \(x60\) objective lens. Line Plot analysis was performed using FV10- ASW v4.1 software (Olympus)  
+
+## RNA-fluorescence in situ hybridization (FISH)
+
+<--- Page Split --->
+
+RNA- FISH was performed as described previously<sup>11</sup>. Briefly, cells \((5 \times 10^{4}\) cells/mL) were inoculated on glass coverslips in a 12- well plate, cultured for 24 h and transfected with siRNA or plasmid. After 24 to 48 h incubation, cells were fixed with \(10\%\) formaldehyde in PBS for 20 min and permeabilized in \(0.1\%\) Triton X- 100 in PBS for 10 min. The coverslip was washed three times with PBS for 10 min, and once with \(2\times\) Standard Saline Citrate (SSC) for 5 min. Cells were prehybridized with ULTRAhyb- Oligo Hybridization Buffer (Ambion, Austin, TX) for 1 h at \(42^{\circ}\mathrm{C}\) in a humidified chamber. Then, they were treated with 10 pmol Alexa Fluor 594- labeled oligo- dT<sub>45</sub> probe (Molecular Probes) overnight. Cells were washed for 20 min at \(42^{\circ}\mathrm{C}\) with \(2\times\) SSC, \(0.5\times \mathrm{SSC}\) , and \(0.1\times \mathrm{SSC}\) . Nuclei were counterstained with DAPI. Fluorescent images were obtained with Axioplan 2. Poly (A) \(^+\) RNA signal intensities in the nucleus and the cell were calculated with ImageJ software (https://imagej.nih.gov/ij/).  
+
+## Protein expression and purification  
+
+GST- fusion proteins were produced in \(E\) . coli BL21 strain. The production of recombinant protein was induced by the addition of \(0.05\mathrm{mM}\) IPTG at \(18^{\circ}\mathrm{C}\) overnight. Cells were pelleted by centrifugation at \(6000\times \mathrm{g}\) for \(10\mathrm{min}\) . The pellet was resuspended in PBS containing \(0.2\mathrm{mM}\) phenyl methyl sulfonyl fluoride (PMSF) and \(1\mathrm{mM}\) dithiothreitol (DTT), and then, sonicated 30 seconds four times on ice. The clear lysate was obtained by centrifugation at \(8000\times \mathrm{g}\) for \(15\mathrm{min}\) and transferred to a new tube. Glutathione- fixed beads (GE Healthcare) were added to the clear lysate and rotated for \(3\mathrm{h}\) at \(4^{\circ}\mathrm{C}\) . After the extensive washing with PBS containing \(0.2\mathrm{mM}\) PMSF and \(1\mathrm{mM}\) DTT, precision protease (GE Healthcare) was added to remove the GST- tag and rotated overnight at \(4^{\circ}\mathrm{C}\) . The eluate containing the GST- tag removed protein was further purified on a gel filtration column, HiPrep 16/60 Sephacryl S- 100 HiResolution (GE Healthcare). The purity and concentration of recombinant protein were confirmed by SDS- PAGE followed by Coomassie Brilliant Blue R- 250 (Nacalai Tesque) staining.  
+
+## Limited proteolysis  
+
+Recombinant protein was incubated with 1/100 (weight ratio) of trypsin (Promega Japan) at \(25^{\circ}\mathrm{C}\) for 0, 30, 120, or 300 min. In the ADP condition, ADP (1 mM) and \(\mathrm{MgCl_2}\) (10 mM) were added to the reaction mixture. The digestion was stopped by adding an equal volume of SDS sample buffer. Samples were boiled for 2 min, then,
+
+<--- Page Split --->
+
+separated by SDS- PAGE and stained with Coomassie Brilliant Blue R- 250. The LC- MS/MS analysis for some of the separated bands was performed by Q Exactive Plus.  
+
+## ATP binding assay  
+
+Recombinant protein was incubated with ATP beads (Jena Bioscience, Jena, Germany) and PBS containing \(0.1\%\) Triton X- 100, \(0.2 \mathrm{mM}\) PMSF, and \(1 \mathrm{mM}\) DTT at \(4^{\circ} \mathrm{C}\) for \(1 \mathrm{~h}\) , and then washed with PBS buffer. Input sample and bound proteins with ATP beads were eluted with SDS sample buffer and separated by SDS- PAGE, then stained with Coomassie Brilliant Blue R- 250 (Nacalai Tesque).  
+
+## Helicase assay  
+
+Helicase assays were performed as described previously with a slight modification \(^{62}\) . Substrate sequences were as follows; \(5^{\prime}\) duplex RNA was UGCUUGCUUUACGGUGCUAGUUUUGUUUGUUUGAUUUCGCCC, and \(5^{\prime}\) duplex DNA was GTAAAGCAAGCTTGAGT. Underlining indicates the region of the duplex. DNA was labeled by T4 polynucleotide kinase (Toyobo, Osaka, Japan) in reaction buffer with \([\gamma - ^{32}\mathrm{P}]\) ATP (GE healthcare) and purified on a Sephadex G- 25 column (GE healthcare). To make duplex, RNA and \(^{32}\mathrm{P}\) labeled DNA were annealed in annealing buffer (20 mM Tris- Cl, pH 8.0, 200 mM potassium acetate, 0.1 mM EDTA). Unwinding reaction buffer (50 nM duplex, \(1 \mu \mathrm{M}\) cold trap oligo DNA, 20mM HEPES- KOH pH 7.9, 50 mM potassium acetate, \(0.1 \mathrm{mM}\) MgCl₂, \(2 \mathrm{mM}\) DTT, \(0.01\%\) BSA, \(4 \mathrm{U} / \mu \mathrm{l}\) RNase inhibitor (Toyobo), \(1 \mathrm{mM}\) ATP, and \(50 \mathrm{ng}\) of GST fusion protein) containing duplex was preincubated without ATP at \(37^{\circ} \mathrm{C}\) for 5 minutes and incubated with ATP at \(37^{\circ} \mathrm{C}\) for 30 minutes. The reaction was stopped by the addition of proteinase K buffer (1 mg/ml proteinase K, \(10 \mathrm{mM}\) Tris- Cl, pH 8.0, \(12.5 \mathrm{mM}\) EDTA, \(150 \mathrm{mM}\) sodium chloride, \(1\%\) SDS) and incubated at \(37^{\circ} \mathrm{C}\) for 25 minutes. Sample was loaded to a \(15 \%\) polyacrylamide gel. Gels were exposed to an imaging plate (Fujifilm Wako and image data were obtained using a BAS2000 (Fujifilm Wako).  
+
+## Crystallization and Crystal structural analysis  
+
+URH49ΔN41 was concentrated at \(5 \mathrm{mg} / \mathrm{ml}\) and crystallized by the sitting- drop vapor diffusion. Briefly, \(1 \mu \mathrm{l}\) of a protein solution was mixed with \(1 \mu \mathrm{l}\) of a mother liquid containing \(0.2 \mathrm{mM}\) NaPO₄, pH 8.5, \(30 \%\) (w/v) PEG3350 at \(20^{\circ} \mathrm{C}\) . The diffraction of
+
+<--- Page Split --->
+
+the crystals was confirmed by an in- house Bruker Hi- star detector after flash- cooling in a cold nitrogen gas stream (100 K) with \(25\%\) (v/v) ethylene glycol. The diffraction images were collected at \(100\mathrm{K}\) (in a cold nitrogen gas stream) on a Rayonix MX225 CCD detector (Rayonix, Evanston, IL) with a wavelength of \(1.0\mathrm{\AA}\) at BL26B2 in SPring- 8 (Hyogo, Japan). The resulting data sets were processed, merged, and scaled using XDS \(^{63}\) . The structure was solved by molecular replacement with UAP56ΔN42 (Protein Data Bank entry 1XTI) using a search mode by Molrep implemented in CCP4i software \(^{64}\) . The model was refined using PHENIX 1.20.1 software \(^{65}\) , rebuilt using COOT \(0.8.9^{66}\) and further modified based on sigma- weighted (2|Fo|-|Fc|) and (|Fo|-|Fc|) electron density maps. Protein structure images were depicted using PyMOL software (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC).  
+
+## Molecular Dynamics (MD) analysis  
+
+MD simulation was performed for the apo- form of URH49ΔN41 using Desmond Molecular Dynamics System, version 5.2 (D. E. Shaw Research, New York, NY) \(^{31}\) . The atomic coordinates of \(\mathrm{SO_4^{2 - }}\) and PGE were removed from the crystal structure of their complex with URH49ΔN41 to generate the initial structure for the simulation. First, the structure was preprocessed with Protein Preparation Wizard of Maestro (version 11.4), the GUI for Desmond, to assign bond orders, add hydrogens, and create disulfide bonds. Then, it was solvated in a box with a buffer distance of \(10\mathrm{\AA}\) to the boundary. Afterwards, solvation was performed in a box with a buffer distance to the boundary of \(10\mathrm{\AA}\) . Sodium and chloride ions were added to neutralize the entire solvated system. OPLS_2005 force field \(^{67}\) and SPC model \(^{68}\) were used for the protein and water molecules, respectively. After relaxing the system according to the Maestro's default relaxation protocol, an MD run was performed in the constant- NPT ensemble at \(300\mathrm{K}\) and \(1.013\mathrm{bar}\) for \(1\mu \mathrm{s}\) . The coordinates were recorded every \(1\mathrm{ns}\) to yield 1,001 snapshots. Otherwise, the default setting in Desmond was adopted. The resulting MD trajectory was equidistantly divided into 101 frames so that each frame could contain ten consecutive snapshots. Then the Root Mean Square Deviation (RMSD) values between main chains of arbitrary two frames were calculated to generate an RMSD matrix. In the matrix, frames with an RMSD less than \(2\mathrm{\AA}\) were assigned to belong to
+
+<--- Page Split --->
+
+the same cluster. In each cluster, the frame with the minimal RMSDs to the other members was considered a representative structure of the cluster.  
+
+## Quantification and statistical analysis  
+
+RT- qPCR results were quantified using Thermal Cycler Dice real time system II (TakaraBio). FISH data was quantified using ImageJ software. Immunofluorescence staining data were quantified using FV10- ASW v4.1 software (Olympus). LC- MS/MS data were analyzed using Mascot software (Matrix Science, London, UK). The statistical significance for two- group and multiple comparisons was tested using R software<sup>69</sup>, as indicated in the legend of each Fig.. Non- adjusted (two- group comparison) and adjusted (multiple comparisons) P- values are indicated in each Fig.. In box plots, the first and third quartiles are indicated by both ends of the box, the median is indicated by a vertical line in the box, and the minimum and maximum excluding outliers are the ends of the whiskers. The outliers are indicated with open circles.  
+
+## Data and code availability  
+
+Atomic coordinates and structure factors for the reported crystal structures have been deposited with the Protein Data bank under accession number 8IJU (URH49A41).
+
+<--- Page Split --->
+
+Reference1. Singh, G., Pratt, G., Yeo, G. W. & Moore, M. J. The clothes make the mRNA: Past and present trends in mRNAFashion. Annu. Rev. Biochem. 84, 325–354 (2015).2. Chang, C. Te et al. Chtop is a component of the dynamic TREX mRNA export complex. EMBO J. 32, 473–486 (2013).3. Folco, E. G., Lee, C. S., Dufu, K., Yamazaki, T. & Reed, R. The proteins PDIP3 and ZC11A associate with the human TREX complex in an ATP-dependent manner and function in mRNA export. PLoS One 7, 1–7 (2012).4. Stäßer, K. et al. TREX is a conserved complex coupling transcription with messenger RNA export. Nature 417, 304–308 (2002).5. Masuda, S. et al. Recruitment of the human TREX complex to mRNA during splicing. Genes Dev. 19, 1512–1517 (2005).6. Dufu, K. et al. ATP is required for interactions between UAP56 and two conserved mRNA export proteins, Aly and CIP29, to assemble the TREX complex. Genes Dev. 24, 2043–2053 (2010).7. Heath, C. G., Viphakone, N. & Wilson, S. A. The role of TREX in gene expression and disease. Biochem. J. 473, 2911–2935 (2016).8. Fleckner, J., Zhang, M., Valcarcel, J. & Green, M. R. U2AF65 recruits a novel human DEAD box protein required for the U2 snRNP-branchpoint interaction. Genes Dev. 11, 1864–1872 (1997).9. Shen, J., Zhang, L. & Zhao, R. Biochemical characterization of the ATPase and helicase activity of UAP56, an essential pre-mRNA splicing and mRNA export factor. J. Biol. Chem. 282, 22544–22550 (2007).10. Shen, H. et al. Distinct activities of the DExD/H-box splicing factor hUAP56 facilitate stepwise assembly of the spliceosome. Genes Dev. 22, 1796–1803 (2008).11. Fujita, K. ichi et al. URH49 exports mRNA by remodeling complex formation and mediating the NXF1-dependent pathway. Biochim. Biophys. Acta - Gene Regul. Mech. 1863, 1–14 (2020).12. Katahira, J., Inoue, H., Hurt, E. & Yoneda, Y. Adaptor Aly and co-adaptor Thoc5 function in the Tap-p15-mediated nuclear export of HSP70 mRNA. EMBO J. 28, 556–567 (2009).
+
+<--- Page Split --->
+
+763 13. Viphakone, N. et al. TREX exposes the RNA-binding domain of Nxf1 to enable mRNA export. Nat. Commun. 3, (2012).765 14. Pryor, A. et al. Growth-regulated expression and G0-specific turnover of the mRNA that encodes URH49, a mammalian DExH/D box protein that is highly related to the mRNA export protein UAP56. Nucleic Acids Res. 32, 1857–1865 (2004).769 15. Yamazaki, T. et al. The Closely Related RNA helicases, UAP56 and URH49, Preferentially Form Distinct mRNA Export Machineries and Coordinately Regulate Mitotic Progression. Mol. Biol. Cell 21, 2953–2965, (2010).772 16. Mancini, A., Koch, A., Whetton, A. D. & Tamura, T. The M-CSF receptor substrate and interacting protein FMIP is governed in its subcellular localization by protein kinase C-mediated phosphorylation, and thereby potentiates M-CSF-mediated differentiation. Oncogene 23, 6581–6589 (2004).776 17. Wang, L. et al. The THO complex regulates pluripotency gene mRNA export and controls embryonic stem cell self-renewal and somatic cell reprogramming. Cell Stem Cell 13, 676–690 (2013).779 18. Maeder, C. I. et al. The THO Complex Coordinates Transcripts for Synapse Development and Dopamine Neuron Survival. Cell 174, 1436–1449 (2018).781 19. Galarza-Muñoz, G. et al. Human Epistatic Interaction Controls IL7R Splicing and Increases Multiple Sclerosis Risk. Cell 169, 72–84 (2017).783 20. Ozgur, S. et al. The conformational plasticity of eukaryotic RNA-dependent ATPases. FEBS J. 282, 850–863 (2015).785 21. Guan, D. et al. Nuclear Factor 45 (NF45) Is a Regulatory Subunit of Complexes with NF90/110 Involved in Mitotic Control. Mol. Cell. Biol. 28, 4629–4641 (2008).788 22. Jha, S. & Dutta, A. RVB1/RVB2: Running Rings around Molecular Biology. Mol. Cell 34, 521–533 (2009).790 23. Damianov, A. et al. Rbfox Proteins Regulate Splicing as Part of a Large Multiprotein Complex LASR. Cell 165, 606–619 (2016).792 24. Gentili, C. et al. Chromosome missegregation associated with RUVBL1 deficiency. PLoS One 10, 1–19 (2015).
+
+<--- Page Split --->
+
+794 25. Johnson, C. et al. Tracking COL1A1 RNA in osteogenesis imperfecta: Splice- 795 defective transcripts initiate transport from the gene but are retained within the 796 SC35 domain. J. Cell Biol. 150, 417-431 (2000). 797 26. Kaida, D. et al. Spliceostatin A targets SF3b and inhibits both splicing and 798 nuclear retention of pre-mRNA. Nat. Chem. Biol. 3, 576-83 (2007). 799 27. Ilik, I. A. et al. Son and srrr2 are essential for nuclear speckle formation. Elife 9, 800 1-48 (2020). 801 28. Huang, C., Liang, D., Tatomer, D. C. & Wilusz, J. E. A length-dependent 802 evolutionarily conserved pathway controls nuclear export of circular RNAs. 803 Genes Dev. 32, 639-644 (2018). 804 29. Kapadia, F., Pryor, A., Chang, T. H. & Johnson, L. F. Nuclear localization of 805 poly(A)+ mRNA following siRNA reduction of expression of the mammalian 806 RNA helicases UAP56 and URH49. Gene 384, 37-44 (2006). 807 30. Shi, H., Cordin, O., Minder, C. M., Linder, P. & Xu, R. M. Crystal structure of 808 the human ATP-dependent splicing and export factor UAP56. Proc. Natl. Acad. 809 Sci. U. S. A. 101, 17628-17633 (2004). 810 31. Bowers, K. J. et al. Scalable Algorithms for Molecular Dynamics Simulations on 811 Commodity Clusters. in Proceedings of the 2006 ACM/IEEE Conference on 812 Supercomputing 43-43 (2006). doi:10.1109/sc.2006.54. 813 32. Ren, Y., Schmiege, P. & Blobel, G. Structural and biochemical analyses of the 814 DEAD-box ATPase Sub2 in association with THO or Yra1. Elife 6, 1-17 (2017). 815 33. Luo, M. L. et al. Pre-mRNA splicing and mRNA export linked by direct 816 interactions between UAP56 and Aly. Nature 413, 644-647 (2001). 817 34. Geuens, T., Bouhy, D. & Timmerman, V. The hnRNP family: insights into their 818 role in health and disease. Hum. Genet. 135, 851-867 (2016). 819 35. Ying, Y. et al. Splicing Activation by Rbfox Requires Self-Aggregation through 820 Its Tyrosine-Rich Domain. Cell 170, 312-323 (2017). 821 36. Aravind, L. & Koonin, E. V. SAP - A putative DNA-binding motif involved in 822 chromosomal organization. Trends Biochem. Sci. 25, 112-114 (2000). 823 37. Matsuda-Hashii, Y. et al. A novel partner gene CIP29 containing a SAP domain 824 with MLL identified in infantile myelomonocytic leukemia [4]. Leukemia 18, 825 1546-1548 (2004).
+
+<--- Page Split --->
+
+38. He, C. et al. The DDX39B/FUT3/TGFβR-I axis promotes tumor metastasis and EMT in colorectal cancer. Cell Death Dis. 12, (2021).
+39. Zhang, H. et al. DDX39B contributes to the proliferation of colorectal cancer through direct binding to CDK6/CCND1. Cell Death Discov. 8, (2022).
+40. Bao, Y. et al. DDX39 as a predictor of clinical prognosis and immune checkpoint therapy efficacy in patients with clear cell renal cell carcinoma. Int. J. Biol. Sci. 17, 3158–3172 (2021).
+41. Kuramitsu, Y. et al. Up-regulation of DDX39 in human malignant pleural mesothelioma cell lines compared to normal pleural mesothelial cells. Anticancer Res. 33, 2557–2560 (2013).
+42. Zhang, T. et al. DDX39 promotes hepatocellular carcinoma growth and metastasis through activating Wnt/β-catenin pathway article. Cell Death Dis. 9, (2018).
+43. Kang, G. J. et al. SARNP, a participant in mRNA splicing and export, negatively regulates E-cadherin expression via interaction with pinin. J. Cell. Physiol. 235, 1–13 (2019).
+44. Grigoletto, A., Lestienne, P. & Rosenbaum, J. The multifaceted proteins Reptin and Pontin as major players in cancer. Biochim. Biophys. Acta - Rev. Cancer 1815, 147–157 (2011).
+45. Castella, S., Bernard, R., Corno, M., Fradin, A. & Larcher, J. C. Ilf3 and NF90 functions in RNA biology. Wiley Interdiscip. Rev. RNA 6, 243–256 (2015).
+46. Cheng, S. et al. Expression and critical role of interleukin enhancer binding factor 2 in Hepatocellular Carcinoma. Int. J. Mol. Sci. 17, 1373 (2016).
+47. Marchesini, M. et al. ILF2 Is a Regulator of RNA Splicing and DNA Damage Response in 1q21-Amplified Multiple Myeloma. Cancer Cell 32, 88–100 (2017).
+48. Xu, Y. et al. Cell type-restricted activity of hnRNPM promotes breast cancer metastasis via regulating alternative splicing. Genes Dev. 28, 1191–1203 (2014).
+49. Yang, W. H., Ding, M. J., Cui, G. Z., Yang, M. & Dai, D. L. Heterogeneous nuclear ribonucleoprotein M promotes the progression of breast cancer by regulating the axin/β-catenin signaling pathway. Biomed. Pharmacother. 105, 848–855 (2018).
+50. Cheng, C. C. et al. EGFR-mediated interleukin enhancer-binding factor 3 contributes to formation and survival of cancer stem-like tumorspheres as a
+
+<--- Page Split --->
+
+859 therapeutic target against EGFR- positive non- small cell lung cancer. Lung 860 Cancer 116, 80- 89 (2018). 861 51. Assimon, V. A. et al. CB- 6644 Is a Selective Inhibitor of the RUVBL1/2 862 Complex with Anticancer Activity. ACS Chem. Biol. 14, 236- 244 (2019). 863 52. Pühringer, T. et al. Structure of the human core transcription- export complex 864 reveals a hub for multivalent interactions. Elife 9, 1- 21 (2020). 865 53. Schuller, S. K. et al. Structural insights into the nucleic acid remodeling 866 mechanisms of the yeast tho- SUB2 complex. Elife 9, 1- 51 (2020). 867 54. Saguez, C. et al. Mutational analysis of the yeast RNA helicase Sub2p reveals 868 conserved domains required for growth, mRNA export, and genomic stability. 869 Rna 19, 1363- 1371 (2013). 870 55. Sugiura, T., Sakurai, K. & Nagano, Y. Intracellular characterization of DDX39, a 871 novel growth- associated RNA helicase. Exp. Cell Res. 313, 782- 790 (2007). 872 56. Taniguchi, I. & Ohno, M. ATP- Dependent Recruitment of Export Factor 873 Aly/REF onto Intronless mRNAs by RNA Helicase UAP56. Mol. Cell. Biol. 28, 601- 608 (2008). 875 57. Siddiqui, N. & Borden, K. L. B. B. mRNA export and cancer. Wiley Interdiscip. 876 Rev. RNA 3, 13- 25 (2012). 877 58. Okamura, M. et al. Depletion of mRNA export regulator DBP5/ DDX19, GLE1 878 or IPPK that is a key enzyme for the production of IP6resulting in differentially 879 altered cytoplasmic mRNA expression and specific cell defect. PLoS One 13, 1- 880 24 (2018). 881 59. Fujiwara, N. et al. MPP6 stimulates both RRP6 and DIS3 to degrade a specified 882 subset of MTR4- sensitive substrates in the human nucleus. Nucleic Acids Res. 883 50, 8779- 8806 (2022). 884 60. Huang, D. W., Sherman, B. T. & Lempicki, R. A. Bioinformatics enrichment 885 tools: Paths toward the comprehensive functional analysis of large gene lists. 886 Nucleic Acids Res. 37, 1- 13 (2009). 887 61. Kurata, M. et al. Food- Derived Compounds Apigenin and Luteolin Modulate 888 mRNA Splicing of Introns with Weak Splice Sites. iScience 22, 336- 352 (2019). 889 62. Rocak, S., Emery, B., Tanner, N. K. & Linder, P. Characterization of the ATPase 890 and unwinding activities of the yeast DEAD- box protein Has1p and the analysis 891 of the roles of the conserved motifs. Nucleic Acids Res. 33, 999- 1009 (2005).
+
+<--- Page Split --->
+
+892 63. Kabsch, W. XDS. Acta Crystallogr. D. Biol. Crystallogr. 66, 125- 32 (2010).893 64. Winn, M. D. et al. Overview of the CCP4 suite and current developments. Acta Crystallogr. Sect. D Biol. Crystallogr. 67, 235- 242 (2011).895 65. Liebschner, D. et al. Macromolecular structure determination using X- rays, neutrons and electrons: Recent developments in Phenix. Acta Crystallogr. Sect. D Struct. Biol. 75, 861- 877 (2019).896 66. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. Sect. D Biol. Crystallogr. 66, 486- 501 (2010).897 67. Banks, J. L. et al. Integrated Modeling Program, Applied Chemical Theory (IMPACT). J. Comput. Chem. 26, 1752- 1780 (2005).898 68. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F. & Hermans, J. Interaction Models for Water in Relation to Protein Hydration. in Intermolecular Forces 331- 342 (1981). doi:10.1007/978- 94- 015- 7658- 1_21.899 69. Andy Bunn, M. K. A language and environment for statistical computing. R Found. Stat. Comput. 10, 11- 18 (2017).
+
+<--- Page Split --->
+
+## Figures  
+
+![](images/Figure_1.jpg)
+
+<center>Figure 1 </center>  
+
+Identification of novel apo- AREX components Immunoprecipitation using anti- DYKDDDDK (equivalent to FLAG) tag antibody beads to pull down nuclear extract of Flp- In T- REx 293 cells stably expressing FLAG tagged each protein. Each precipitated sample was separated and detected by silver staining (left) or
+
+<--- Page Split --->
+
+immunoblotting (right) with the indicated antibodies. (A) The model of the apo- TREX, - AREX and the ATP- TREX complex formations of UAP56 and URH49. The dotted circles refer to apo- AREX complex components that were unidentified prior to this study. (B) FLAG- UAP56 and - URH49 differ in apo- complex formation but are similar in ATP- complex formation. In the right panel, known apo TREX, - AREX, and ATP- TREX components were detected by immunoblotting. Single asterisk represents precipitated FLAG- UAP56 or - URH49, double asterisk represents IgG light chain. (C) Identification of novel apo- AREX components by tandemimmunoprecipitation. Precipitated proteins by FLAG immunoprecipitation or tandemimmunoprecipitation (first: HA, second: FLAG) are detected respectively. Identified proteins are shown on the right side. Details of proteins identified are indicated in Extended Data Fig.2B and Table S1. The proteins shown in the red letter were further analyzed. Single, double, and triple asterisk represented precipitated FLAG- UAP56 or - URH49, HA- CIP29, and IgG light chain, respectively. (D) FLAG- URH49 is specifically associated with each novel apo- AREX component in an ATP-depleted condition, but not in the presence of ATP. (E) FLAG- ILF2 and - ILF3 bind with apo URH49 in an ATP-deficient condition. Single and double asterisks represent FLAG and endogenous- ILF2 or ILF3, respectively. (F) FLAG- HNRNPM associates with URH49 in the absence of ATP. (G) FLAG- RUVBL1 and - RUVBL2 associate with URH49 in an ATP-depleted condition. Single and double asterisks represent FLAG and endogenous- endogenous RUVBL1 or RUVBL2, respectively.
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+  
+
+![](images/Figure_3.jpg)
+
+<center>Figure 2 </center>  
+
+The novel apo- AREX components are specifically associated with URH49- mediated mRNA processing and export (A) Depletion of apo- AREX components caused nuclear poly(A)+ RNA accumulation in U2OS cells. DAPI was used to visualize the nuclei. Scale bar, \(40 \mu \mathrm{m}\) . (B) Quantification of the nuclear poly(A)+ RNA accumulation caused by the knockdown of apo- AREX components. The graph indicates the fold changes in the ratio of nuclear per cytoplasmic distribution of mRNA. These data were normalized to the
+
+<--- Page Split --->
+
+score of the control knockdown condition. The signal intensities of bulk poly(A)+ RNA in the nucleus and the cytoplasm were quantified using ImageJ (n = 40 of each, respectively). Boxes show the median (centerline) and upper and lower quartiles. Whiskers show the lowest and highest values. Statistical analysis was performed using the Kruskal- Wallis test followed by the Steel test. \(***p < 0.001\) . (C), Localization of poly(A)+ RNA in U2OS cells. Poly(A)+ RNA localization (red) was observed under the knockdown of each apo- AREX component. Anti- SRRM2 antibody was used to stain the nuclear speckle (green). DAPI was used to visualize the nuclei (blue). Scale bar, 10 μm. In the right panels, signal intensities of poly(A)+ RNA and SRRM2 (same colors) were plotted between the A and B lines in the left panels. (D) Depletion of apo- AREX components resulted in the decreased expression of URH49- target mRNAs in the cytoplasm. RT qPCR was performed using the cytoplasmic RNA to compare the mRNA expression level. Values represent the relative expression of indicated mRNA normalized to PGK and the mean ± SEM of three independent experiments. Statistical analysis was performed using one- way ANOVA followed by Dunnett's test. \(**p < 0.01\) , \(***p < 0.001\) . n.s.: not significant
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 3 </center>  
+
+A single amino acid alteration between UAP56 and URH49 impacts their apo- complex formation and specific functions (A) Diagram of amino acids homology between UAP56 and URH49 and a list of chimeric mutants analyzed in this study. (B) FLAG- URH49 C223V mutant forms the apo- TREX- like complex. Immunoprecipitation was performed using anti- DYKDDDDK tag antibody beads and Flp- In T- REx 293 cells stably expressing FLAG- tagged proteins. Precipitated sample was separated and detected
+
+<--- Page Split --->
+
+by silver staining (left) or immunoblotting with the indicated antibodies (right). Single and double asterisks represented precipitated FLAG- UAP56 or - URH49, and IgG light chain, respectively. (C) Overexpression of chimeric mutants of FLAG- UAP56 or URH49 rescued the nuclear poly(A)+ RNA accumulation due to UAP56 or URH49 depletion. Poly(A)+ RNA (red), exogenously expressed FLAG- UAP56 or - URH49 (green), and chromosomal DNA (blue) were visualized in U2OS cells. Scale bar, \(40 \mu \mathrm{m}\) . (D) Quantification of the nuclear poly(A)+ RNA accumulation caused by each condition in (C). The fold changes in the ratio of nuclear to cytoplasmic distribution of poly(A)+ RNA are shown. These data were normalized to the score of control plasmid overexpression under the control knockdown condition. The signal intensities of bulk poly(A)+ RNA in the nucleus and the cytoplasm were quantified from at least 28 cells for each condition using ImageJ. Boxes show the median (centerline) and upper and lower quartiles. Whiskers show the lowest and highest values. Statistical analysis was performed using the Kruskal- Wallis test followed by the Steel test. \(***p < 0.001\) . n.s.: not significant
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 4 </center>  
+
+UAP56 and URH49 have different apo- structural features but similar ATP binding structural features, which were correlated with their complex formation Each purified protein was treated with trypsin. Aliquots were taken at each time point, separated by SDS- PAGE, and detected by Coomassie staining. (A, B) Full- length UAP56 and full- length URH49 had different partial digestion patterns in the absence of ADP, but similar limited proteolysis patterns upon ADP addition. (C) UAP56Δ42 and URH49Δ41 showed
+
+<--- Page Split --->
+
+different limited proteolysis patterns, and URH49 C223VΔ41 showed a pattern similar to apo-UAP56Δ42 but not apo-URH49Δ41 in the absence of ADP. (D) UAP56Δ42 and URH49Δ41 showed similar limited proteolysis patterns upon ADP addition. (E, F) Top: Analysis of cleavage sites by limited proteolysis. The "A1-4", "R1-4", and "Total" products obtained by limited proteolysis of UAP56Δ42 and URH49Δ41 in (C, D) were analyzed by LC-MS/MS. Relative peptide scores were obtained by dividing the detected prot-score of each peptide fragment derived from "A1-4" and "R1-4" by "Total". The start-site, the end-site, and the relative score of each peptide were described in Table S2. Bottom: Limited digestion models of UAP56 (blue) or URH49 (gray) predicted from the peptide containing within "A1-4" or "R1-4"  
+
+![](images/Figure_6.jpg)
+
+<center>Figure 5 </center>
+
+<--- Page Split --->
+
+Structural comparison between UAP56, URH49, and Sub2 (A) Comparison of the structure of apo- UAP56Δ42 (1XTI) and the structural model of apo- URH49Δ41 (Fr48) which is generated by molecular dynamics analysis of the structure of URH49Δ41 (8IJU). Detail of the generation of the Fr48 model were described in Extended Data Fig.8C, D. By aligning the N-domain of both structural models using pyMOL, the difference in the angle of the C-domains was calculated. (B) Comparison of N-domain and C-domain between apo- UAP56Δ42 crystal (1XTI) and apo- URH49Δ41 model structure (Fr48). (C) Top: the C-loop of the apo- URH49Δ41 structural model (Fr48) was located as covering the ATP binding pocket of apo URH49Δ41. Bottom: the structure of ADP- UAP56Δ42 (1XTI) was overlaid to the structure of apo- UAP56Δ42 (1XTI) or the structural model of apo- URH49Δ41 (Fr48). (D) Loop structure in the C-domain of the apo- Sub2Δ59 (5SUQ) was overlaid to apo UAP56Δ42 (1XTI) or URH49Δ41 structural model (Fr48)  
+
+![PLACEHOLDER_40_0]
+
+<center>Figure 6 </center>  
+
+Diversified Structurers and mRNA export machineries The model of a selective mRNA transcription and export machinery driven by the structural diversification from yeast Sub2 to human UAP56 and URH49.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- ExtendFigTableLegend.pdf- Table.pdf
+
+<--- Page Split --->

preprint/preprint__98621a0a9a75bfd4307471acfa62e2ac69cd7da3d48c9482423c2bb4b8db561a/preprint__98621a0a9a75bfd4307471acfa62e2ac69cd7da3d48c9482423c2bb4b8db561a_det.mmd

@@ -0,0 +1,542 @@
+<|ref|>title<|/ref|><|det|>[[44, 108, 950, 208]]<|/det|>
+# Structural differences between the closely related RNA helicases, UAP56 and URH49 fashions distinct functional apo-complexes  
+
+<|ref|>text<|/ref|><|det|>[[44, 228, 460, 248]]<|/det|>
+Seiji Masuda ( smasuda@nara.kindai.ac.jp )  
+
+<|ref|>text<|/ref|><|det|>[[50, 252, 205, 270]]<|/det|>
+Kindai University  
+
+<|ref|>text<|/ref|><|det|>[[44, 277, 260, 316]]<|/det|>
+Ken- ichi Fujita Fujita Health University  
+
+<|ref|>text<|/ref|><|det|>[[44, 323, 198, 362]]<|/det|>
+Misa Ito Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 368, 198, 407]]<|/det|>
+Midori Irie Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 414, 198, 453]]<|/det|>
+Kotaro Harada Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 460, 198, 499]]<|/det|>
+Naoko Fujiwara Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 506, 198, 545]]<|/det|>
+Yuya Ikeda Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 551, 198, 590]]<|/det|>
+Hanae Yoshioka Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 597, 221, 636]]<|/det|>
+Tomohiro Yamazaki Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 643, 198, 682]]<|/det|>
+Masaki Kojima Tokyo University of Pharmacy and Life Sciences  
+
+<|ref|>text<|/ref|><|det|>[[44, 689, 198, 728]]<|/det|>
+Bunzo Mikami Kyoto University  
+
+<|ref|>text<|/ref|><|det|>[[44, 735, 166, 754]]<|/det|>
+Akila Mayeda  
+
+<|ref|>text<|/ref|><|det|>[[44, 757, 916, 777]]<|/det|>
+Institute for Comprehensive Medical Science, Fujita Health University https://orcid.org/0000- 0002-  
+
+<|ref|>text<|/ref|><|det|>[[44, 781, 140, 799]]<|/det|>
+9562- 550X  
+
+<|ref|>text<|/ref|><|det|>[[44, 844, 102, 861]]<|/det|>
+Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 882, 136, 900]]<|/det|>
+Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 919, 297, 938]]<|/det|>
+Posted Date: May 12th, 2023
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[43, 45, 473, 64]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 2819840/v1  
+
+<|ref|>text<|/ref|><|det|>[[43, 82, 910, 125]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 144, 253, 162]]<|/det|>
+## Additional Declarations:  
+
+<|ref|>text<|/ref|><|det|>[[44, 167, 319, 186]]<|/det|>
+There is NO Competing Interest.  
+
+<|ref|>text<|/ref|><|det|>[[43, 205, 515, 224]]<|/det|>
+Table 1 is available in the Supplementary Files section  
+
+<|ref|>text<|/ref|><|det|>[[42, 275, 940, 317]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on January 15th, 2024. See the published version at https://doi.org/10.1038/s41467- 023- 44217- 8.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[88, 118, 844, 870]]<|/det|>
+1 Structural differences between the closely related RNA helicases, UAP56 and 2 URH49 fashions distinct functional apo-complexes 3 4 Ken-ichi Fujita1,2\\*, Misa Ito1, Midori Irie1, Kotaro Harada1, Naoko Fujiwara1,3, Yuya 5 Ikeda1, Hanae Yoshioka1, Tomohiro Yamazaki1,3, Masaki Kojima4, Bunzo Mikami5,6, 6 Akila Mayeda2, Seiji Masuda1,7,8,9,10\\* 7 8 1 Division of Integrated Life Sciences, Graduate School of Biostudies, Kyoto 9 University, Kyoto, Kyoto, 606-8502, Japan. 10 2 Division of Gene Expression Mechanism, Center for Medical Science, Fujita Health 11 University, Toyoake, Aichi 470-1192, Japan 12 3 Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka 565-0871, 13 Japan. 14 4 School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, Hachioji, 15 Tokyo, 192-0392, Japan. 16 5 Research Institute for Sustainable Humano sphere, Kyoto University 17 6 Institute of Advanced Energy, Kyoto University 18 7 Department of Food Science and Nutrition, Faculty of Agriculture Kindai University, 19 Nara, Nara 631-8505, Japan 20 8 Agricultural Technology and Innovation Research Institute, Kindai University, Nara, 21 Nara, 631-8505, Japan 22 9 Antiaging center, Kindai University, Higashiosaka, Osaka 577-8502, Japan 23 10 Lead Contact 24 \* To whom correspondence should be addressed. Tel: +81-742-43-1713; Email: 25 smasuda@nara.kindai.ac.jp 26 27 ORCID 28 Ken-ichi Fujita; 0000-0002-3104-5274, Naoko Fujiwara; 0000-0003-2366-9076, 29 Tomohiro Yamazaki; 0000-0003-0866-5173, Masaki Kojima; 0009-0001-9190-4923, 30 Bunzo Mikami; 0000-0003-0638-8619, Akila Mayeda; 0000-0002-9562-550X, Seiji 31 Masuda;. 0000-0003-0295-6789 32
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[139, 122, 228, 138]]<|/det|>
+## Summary  
+
+<|ref|>text<|/ref|><|det|>[[137, 145, 856, 472]]<|/det|>
+Messenger RNA export is a regulated pathway that control gene expression and specific physiological events. In humans, closely related RNA helicases, UAP56 and URH49 shape selective mRNA export pathways through distinct apo- complex formation and remodeling to similar ATP- complex to achieve precise gene regulation. The difference in apo- complex is the key to functional divergence. However, the profile of the apo- complex formed by URH49 (named apo- AREX complex) and why UAP56 and URH49 exhibit distinct complex formation remain unknown. Here, we investigated unidentified apo- AREX components. RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM were physically and functionally associated with URH49, indicating the key components of the apo- AREX complex. Integrating analysis of crystal structure and complex formation of UAP56/URH49 chimera mutants demonstrated that their structural features contribute to respective complex formation. This study provides insights into the specific function of two helicases and into how two helicases diverged from a single ancestral gene, Sub2.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[139, 121, 254, 138]]<|/det|>
+## Introduction  
+
+<|ref|>text<|/ref|><|det|>[[139, 144, 857, 281]]<|/det|>
+During the expression of protein- coding genes, pre- mRNAs are transcribed in the nucleus and undergo several RNA processing steps, including capping, splicing, and polyadenylation. Subsequently, the mature mRNA is exported to the cytoplasm for translation. These processes are coupled with one another through appropriate assembly and remodeling of mRNA- protein (mRNP) complexes to achieve accurate gene expression<sup>1</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[139, 287, 852, 400]]<|/det|>
+A key player integrating transcription and mRNA export is the evolutionarily conserved ATP- dependent multi- subunit Transcription- EXport (TREX) complex. The human ATP- bound TREX complex consists of the THO subcomplex, comprising THOC1, THOC2, THOC3, THOC5, THOC6, and THOC7, and several affiliated proteins: ALYREF, CIP29, CHTOP, PDIP3, ZC3H11A, and DEAD- box RNA helicase  
+
+<|ref|>text<|/ref|><|det|>[[140, 405, 755, 470]]<|/det|>
+UAP56<sup>2,3</sup> (and refs therein<sup>4,5,6</sup>). The TREX components are recruited onto transcribing RNA polymerase II (Pol II) and loaded onto spliced- mRNA in a splicing- dependent manner, which is crucial for subsequent export<sup>5,7</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[139, 475, 856, 732]]<|/det|>
+Perhaps the most crucial factor in the assembly of the TREX complex is UAP56 (Sub2 in yeast). During splicing, UAP56 is loaded onto pre- mRNA through the interaction with U2AF65<sup>8</sup> and in turn it regulates spliceosome assembly<sup>9,10</sup>. UAP56 interacts with the THO subcomplex in an ATP- independent manner<sup>11</sup>. When ATP binds UAP56, it recruits CIP29, ALYREF, CHTOP, PDIP3, and ZC11A into the TREX complex<sup>2,3,6</sup>. Thus, the TREX complex exists in two states that remodel depending on whether ATP binds to UAP56. We term the ATP- unbound form as the apo- TREX complex and the ATP- bound one as the ATP- TREX complex to distinguish both complexes<sup>11</sup> (Fig.1A). The formation of the ATP- TREX complex drives the export of bound mRNA because ALYREF, CHTOP, and THOC5 act as adaptors of the NXF1- NXT1 heterodimer which functions in the final step of the mRNA export<sup>2,12,13</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[139, 737, 852, 898]]<|/det|>
+In mammals, UAP56 has a paralogue that is \(90\%\) identical, URH49<sup>14</sup>. Furthermore, we have previously shown that UAP56 and URH49 form distinct apoproteins. UAP56 forms the apo- TREX complex, and URH49 forms the apo- Alternative- mRNA- EXport (AREX) complex. Unlike the apo- TREX complex, the apo- AREX complex contains CIP29 and it does not contain the THO subcomplex<sup>15</sup>. Like the apo- TREX complex, the apo- AREX complex is remodeled to ATP- complex when ATP is loaded onto URH49, and accesses NXF1- NXT1 heterodimer for mRNA
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[135, 120, 852, 163]]<|/det|>
+export<sup>11</sup>. Irrespective of whether the precursor is an apo- TREX or an apo- AREX complex, ATP- complexes resemble each other and are called the ATP- TREX complex.  
+
+<|ref|>text<|/ref|><|det|>[[135, 168, 855, 400]]<|/det|>
+In addition, each helicase selectively exports a specific subset of mRNAs. UAP56 and URH49 selectively regulate distinct subsets of key mitotic regulators<sup>15</sup>. URH49 is also required for the gene expression involved in cytokinesis<sup>11</sup>. Besides, both helicases and the components of their respective complexes are required for a variety of physiologically important roles in lifelong cell differentiation<sup>16,17,18</sup>. Consequently, abnormalities of their mRNA export pathways including disruption of their expression have been associated with serious diseases such as cancer and neurodegenerative disorders<sup>7,19</sup>. Thus, the evolutionarily diversified mRNA export pathways formed by UAP56 and URH49 contribute to fine- tuned gene expression and are required for various physiological events.  
+
+<|ref|>text<|/ref|><|det|>[[137, 404, 856, 684]]<|/det|>
+Therefore, elucidation of the functional machinery of UAP56 and URH49 and their differences is important, not only for a better understanding of gene regulation in higher organisms but also for an understanding of a variety of diseases caused by disruption of these two helicases. DEAD- box family helicases generally bind RNA in a sequence- independent manner, and target recognition is primarily provided via partner proteins<sup>20</sup>. Thus, identifying the compositions of the apo- TREX and the apo- AREX complexes, and elucidating the molecular basis of the involvement of UAP56 and URH49 in complex formation, may be the key to understanding their function. However, the factor(s) of the apo- AREX complex are unknown except for CIP29, which is also in the ATP- TREX complex (Fig.1A). Importantly, the mechanisms by which UAP56 and URH49 form distinct complexes, despite their extensive homology, remains unknown.  
+
+<|ref|>text<|/ref|><|det|>[[139, 690, 856, 757]]<|/det|>
+In this study, we first used tandem- immunoprecipitation and mass spectrometry to investigate the factors of the apo- AREX complex. Then, we determined the reason why UAP56 and URH49 form different apo- complexes.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[140, 121, 200, 137]]<|/det|>
+## Result  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 144, 574, 162]]<|/det|>
+## Identification of the novel apo-AREX components  
+
+<|ref|>text<|/ref|><|det|>[[137, 168, 852, 615]]<|/det|>
+To analyze the composition of the apo- AREX complex, we performed immunoprecipitation using nuclear extract prepared from Flp- In T- REx 293 cells expressing either FLAG- UAP56 or FLAG- URH49 in the ATP- depleted condition. FLAG- UAP56 or FLAG- URH49 coprecipitated different components: FLAG- UAP56 was associated with the apo- TREX components (THOC1, THOC2, and THOC5) and FLAG- URH49 precipitated the apo- AREX component CIP29 (Fig.1B). These interactions are consistent with previously reported different interactions of endogenous UAP56 and URH49<sup>11,15</sup>. We then added ATP and found that the ATP- TREX components (THOC1, THOC2, THOC5, ALYREF, and CIP29) interacted with both FLAG- UAP56 and FLAG- URH49. We refer to the ATP- TREX complex containing UAP56 as the ATP- TREX (UAP56) complex and the ATP- TREX complex containing URH49 as the ATP- TREX (URH49) complex. This remodeling was also observed in the presence of ADP or AMP- PNP, indicating that the ATP binding, but not the ATP hydrolysis is sufficient to exert the complex remodeling (Fig.1B, Extended Data Fig.1A). In the FLAG- URH49 precipitate, we detected many candidates for the novel apo- AREX components. To identify these factors as authentic apo- AREX components, we performed tandem purification of the apo- AREX complex with the nuclear extract expressing both known apo- AREX components, FLAG- URH49 and HA- CIP29, and identified isolated factors by LC- MS/MS (Fig.1C, Extended Data Fig.1B- D).  
+
+<|ref|>text<|/ref|><|det|>[[137, 619, 857, 876]]<|/det|>
+Among the coimmunoprecipitated factors of FLAG- URH49 and FLAG- UAP56, we observed enrichment of splicing- associated factors, indicating that involvements of both helicases with the splicing process (Extended Data Fig.2A- C see also Table S1). Moreover, various RNA- binding proteins are found to bind to URH49, but not UAP56. RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM were reliably detected from the tandem immune- precipitate as well as from the FLAG- URH49 precipitate, but not in the FLAG- UAP56 precipitate (Fig.1C, Extended Data Fig.2B). RUVBL1 and RUVBL2 form heterodimers, as do ILF2 and ILF3<sup>21,22</sup>. HNRNPM interacts with ILF2, ILF3, and other factors<sup>23</sup>. Thus, we focused on these factors, and interactions between these factors and URH49 were confirmed by immunoblotting of FLAG- URH49 precipitate (Fig.1D).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 120, 856, 377]]<|/det|>
+To confirm the physical association of the apo- AREX candidates with URH49, we generated cell lines stably expressing FLAG- RUVBL1, RUVBL2, ILF2, ILF3, or HNRNPM, respectively. Anti- FLAG- precipitates from nuclear extracts expressing each apo- AREX candidate efficiently captured URH49 whereas there was no obvious binding to THOC1 and ALYREF, components of the apo- and the ATP- TREX complex, was observed (Fig.1E- G). Please see Extended Data Fig.2D in which complex each factor is present. The interactions between RUVBL1 and RUVBL2, as well as between ILF2 and ILF3 were sustained in the presence of ATP, as previously reported<sup>21,22</sup>. In contrast, these factors dissociated from URH49 upon ATP addition. These results indicate that these factors interact with URH49 as the apo- AREX complex but do not interact with the ATP- TREX complex (Extended Data Fig.2D).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 404, 799, 446]]<|/det|>
+## The novel apo-AREX components are associated with URH49-target mRNA processing and export  
+
+<|ref|>text<|/ref|><|det|>[[137, 452, 857, 805]]<|/det|>
+Next, we evaluated the functional significance of novel apo- AREX candidates in mRNA processing and export. In addition to ILF3 (their alternative name is NF110), NF90, a truncated isoform of ILF3, was produced from ILF3 gene. This factor also interacted with ILF2 as in the case of ILF3<sup>21</sup>. A previous study reported that the use of an ILF3- specific siRNA can deplete ILF3 expression without affecting NF90 expression, while the knockdown of ILF2 downregulated the expression of both ILF3 and NF90<sup>21</sup>. The depletion of either RUVBL1 or RUVBL2 causes a co- depletion of the other<sup>24</sup>. Thus, we depleted RUVBL1, ILF3, HNRNPM, and CIP29, a known apo- AREX component by siRNA- mediated knockdown (Extended Data Fig.3A). Depletion of either factor induced the nuclear accumulation of poly(A)<sup>+</sup> RNAs, which co- localized with nuclear speckles (Fig.2A- C). This observation is probably reflecting the perturbed mRNA splicing and export by knocking down either factor as shown by previous reports<sup>25,26,27</sup>. Similar results were observed with other cell lines and with other siRNAs of any of the factors (Extended Data Fig.3B- D). These observations indicated that our apo- AREX candidates function in mRNA processing and export.  
+
+<|ref|>text<|/ref|><|det|>[[138, 809, 857, 899]]<|/det|>
+To further clarify the function of these factors in the apo- AREX complex, we depleted each apo- AREX candidate and assessed its effect on the cytoplasmic mRNA expression of UAP56 or URH49 targets. Depletion of each factor specifically reduced the expression of the URH49 targets but did not cause a reduction of the UAP56 targets,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[139, 119, 855, 234]]<|/det|>
+indeed in some cases it upregulated them (Fig.2D). Such upregulation of UAP56 targets was also observed when URH49 and CIP29 were depleted<sup>11</sup>, suggesting that depletion of the apo- AREX component probably enhances the UAP56 export pathway as a compensatory mechanism. These results indicated that each factor functions as the apo- AREX complex, and specifically regulates URH49- target mRNA export.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 262, 842, 304]]<|/det|>
+## A single amino acid difference between UAP56 and URH49 impacts apo-complex formation and function  
+
+<|ref|>text<|/ref|><|det|>[[139, 309, 858, 637]]<|/det|>
+Subsequently, we investigated how UAP56 and URH49 form distinct apo- complexes despite their high homology. DEAD- box RNA helicase contains a conserved core region with two domains (N- domain and C- domain, respectively), a linker region between them, and terminal regions<sup>20</sup> (Fig.3A, Extended Data Fig.4). We hypothesized that differences in a specific region(s) between UAP56 and URH49 are important to form their distinct apo- complexes. To identify the region(s) determining the distinct apo- complexes formed, we generated plasmids expressing mutants in which various regions of UAP56 and URH49 were swapped and examined apo- complex formation (Fig.3A, Extended Data Fig.5). The N- and C- terminal regions are relatively different compared to core regions of UAP56 and URH49. However, swapping of either terminal region did not affect the apo- complex formation (Extended Data Fig.5A). In contrast, mutants swapped of each N- domain (described as “UAP56 N- core” and “URH49 N- core”) dramatically altered apo- complex formation (Extended Data Fig.5B- C), suggesting the N- domain determines which apo- complex forms.  
+
+<|ref|>text<|/ref|><|det|>[[139, 642, 854, 900]]<|/det|>
+In humans, UAP56 and URH49 have twelve amino acid differences in the N- domain (Fig.3A, Extended Data Fig.4). Subsequently, we analyzed twelve point mutants in which different individual amino acids are swapped. Strikingly we found that the URH49 C223V mutant specifically switched the complex formation from the apo- AREX complex to the apo- TREX complex (Fig.3B, Extended Data Fig.5D). UAP56 V224C, the mutant corresponding to URH49 C223V, did not alter the apo- complex formation (Extended Data Fig.5D). To further examine the possibility that other amino acid differences besides UAP56- V224 and URH49- C223 also contribute to their distinct complex formation, we generated the UAP56 mutant described as “UAP56 N- core C224V”, in which the N- domains of UAP56 other than UAP56- V224 were replaced with the N- domains of URH49. This mutant lost the ability to form the apo
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[139, 119, 846, 187]]<|/det|>
+AREX complex but did form the apo- TREX complex (Extended Data Fig.5E). These results indicate that the difference between UAP56- V224 and URH49- C223 is the crucial determinant of apo- complex formation.  
+
+<|ref|>text<|/ref|><|det|>[[139, 191, 853, 377]]<|/det|>
+It has been reported that UAP56 and URH49 export not only mRNAs but also circular RNAs which are generated via "back- splicing"28. In that report, UAP56 plays a role in long circular RNAs export while URH49 exports short circular RNAs. The four different amino acids that are located in their N- domains between UAP56 and URH49 determine their specificity for circular RNAs. However, we did not find a difference in complex formation between the apo- TREX and the apo- AREX in the mutants with these four amino acid substitutions (Extended Data Fig.5F). Thus, the mechanism of mRNA export appears to be different from that of circular RNA export.  
+
+<|ref|>text<|/ref|><|det|>[[138, 381, 853, 760]]<|/det|>
+We next examined whether alteration of the apo- complex formation affects mRNA export activities of the two helicases. The depletion of UAP56 or URH49 induced bulk nuclear poly(A) \(^+\) RNA accumulation, respectively15,29. The forced expression of siRNA- resistant UAP56 rescued the nuclear poly(A) \(^+\) RNA accumulation induced by endogenous UAP56 knockdown, but did not rescue the nuclear poly(A) \(^+\) RNA accumulation provoked by the disruption of URH49, and vice versa (Fig.3C- D, Extended Data Fig.6). This result reflects that UAP56 and URH49 export distinct subsets of bulk mRNA substrates and do not the other15. URH49 C223V and URH49 chimera mutant, URH49 N- core which form the apo- TREX complex, could specifically rescue the nuclear poly(A) \(^+\) RNA accumulation caused by the knockdown of endogenous UAP56. In addition, UAP56 chimera mutant UAP56 N- core, which form the apo- AREX complex could rescue the nuclear poly(A) \(^+\) RNA accumulation induced by the disruption of endogenous URH49. These data clearly demonstrate that mRNA export selectivity was controlled at the apo- complex formation step. Taken together, the formation of distinct apo- complex due to the difference in a single amino acid between UAP56 and URH49 has a key role in the selective mRNA export by the two helicases.  
+
+<|ref|>sub_title<|/ref|><|det|>[[139, 785, 839, 826]]<|/det|>
+## UAP56 and URH49 form different apo-structures, but with similar ADP binding structures  
+
+<|ref|>text<|/ref|><|det|>[[139, 832, 847, 899]]<|/det|>
+DEAD- box helicases have similar structural features20. In the apo- state, DEAD- box family proteins adopt a variety of open structures with the configuration of N- domain and C- domain different for each member. The ATP- binding triggers the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 119, 858, 401]]<|/det|>
+rearrangement of the N- domain and the C- domain into similar closed- structure via interactions of ATP with both domains. These structural features determine what kind of complex forms according to their apo- and ATP- binding state. In fact, the remodeling of the apo- AREX complex to the ATP- TREX complex (URH49) dramatically altered the protein composition between the two complexes (Fig. 1D- G). These results led us to hypothesize that the structures of UAP56 and URH49 in the apo- and ATP- binding states dictate their apo- and ATP- complex formation. Supplying ADP caused the complex remodeling of UAP56 and URH49 as well as the addition of AMP- PNP, a non- hydrolysable analog of ATP (Extended Data Fig.1A), indicating that the ADP- bound structures of two helicases resemble the ATP- bound structures. Thus, we compared the structural features of UAP56 and URH49 in the apo- and the ADP- bound states by limited proteolysis.  
+
+<|ref|>text<|/ref|><|det|>[[137, 405, 855, 732]]<|/det|>
+The sites of the primary amino acid sequence predicted to be digested by trypsin were the same in both helicases. However, the digested fragments and thus the structures in the apo- state differed between UAP56 and URH49 (Fig.4A, B). As for UAP56, the digested products in the presence of ADP were similar to those in the absence of ADP. Previously, crystal structures of both apo- and ADP- bound form UAP56ΔN42, which lacked N terminal 42 residues of UAP56, were reported<sup>30</sup>. In that study, UAP56ΔN42 exhibits a relatively closed conformation in the apo- state compared to other DEAD- box proteins<sup>30</sup>. And ADP binding induces a slight structural rearrangement only around the ATP- binding pocket without the configurational change of their N- and C- domains. Our observations seem to reflect these findings. On the contrary, the digestion pattern of URH49 in the presence of ADP differed from that in the absence of ADP, and changed to that of UAP56 upon the addition of ADP. This indicates that unlike UAP56, URH49 undergoes the significant conformational change upon binding of ADP.  
+
+<|ref|>text<|/ref|><|det|>[[137, 737, 855, 900]]<|/det|>
+We also generated the UAP56ΔN42 and URH49ΔN41 which is the URH49 mutant corresponding to UAP56ΔN42. Their digested fragments were different under the apo- condition and became similar in the presence of ADP as well as in the case of UAP56 and URH49 (Fig.4C, D). These results indicate that UAP56ΔN42 has a similar structure to UAP56 and URH49ΔN41 to URH49. We, next, estimated the cleavage sites of the UAP56ΔN42 and URH49ΔN41 by detecting peptides using LC- MS/MS analysis (described A1- 4 and R1- 4 in Fig.4C, D). The A1 and A2 fragments generated in the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 118, 857, 352]]<|/det|>
+absence of ADP covered peptides from the N- domain and linker region of UAP56ΔN42. The peptide composition of R1 fragment was similar to that of A1 fragment while the peptide composition of R2 fragment contained the C- domain and the linker region, and was completely different from that of A2 fragment (Fig.4C, E). These results indicated that the sensitive site of UAP56 digestion by trypsin was different from that of URH49, probably based on their distinct structures in the absence of ATP (Fig.4E bottom, Extended Data Fig.7A). The A3 and 4 fragments generated in the presence of ADP had the same digestion pattern with the R3 and 4 fragments (Fig.4D, F), indicating that URH49 underwent a significant structural change by the loading of ADP and UAP56 did not.  
+
+<|ref|>text<|/ref|><|det|>[[137, 358, 850, 590]]<|/det|>
+To further confirm that URH49 underwent the structural rearrangement upon ADP- binding, we employed mutants lacking the ATP binding activities: UAP56ΔN42 K95N and URH49ΔN41 K94N<sup>11</sup>. These mutants had the same digestion pattern as one another in the presence of ADP (Extended Data Fig.7B, C). From these results, we concluded that the two helicases form different apo- structures, but were remodeled to similar structures on ADP binding. Importantly, the digested pattern of URH49 C223VΔN41 was similar to that of UAP56ΔN42 (Fig.4C). This indicates that V224 of UAP56 and C223 of URH49 play important roles in forming their different apo- structures, and raises the possibility that the structural feature of UAP56 and URH49 were associated with their apo- and ADP- complex formation.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 620, 629, 638]]<|/det|>
+## The Different Structures of apo-UAP56 and apo-URH49  
+
+<|ref|>text<|/ref|><|det|>[[137, 644, 855, 875]]<|/det|>
+To analyze the difference between both apo- structures, we solved the crystal structure of URH49ΔN41 by x- ray diffraction (8IJU) and compared it with the published apo- UAP56ΔN42 structure (1XTI)<sup>30</sup>. The folds of two N- and C- domains in URH49ΔN41 are essentially the same as the apo- UAP56ΔN42 structure<sup>30</sup>. However, the N- and C- domains of URH49ΔN41 were located in distinct positions from the respective domains of UAP56ΔN42 (Extended Data Fig.8A, Table 1). The crystal of URH49ΔN41 contained SO<sub>4</sub><sup>2-</sup> and polyethylene glycol (PEG) around the interspace between N- and C- domains and the ATP- binding pocket (Extended Data Fig.8B). This raised the possibility that the structure of apo- URH49ΔN41 containing SO<sub>4</sub><sup>2-</sup> and PEG differs from that of UAP56ΔN42 because of its interaction with these compounds.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 120, 855, 305]]<|/det|>
+To exclude this possibility, we generated URH49ΔN41 apo-structure models lacking \(\mathrm{SO_4^{2 - }}\) and PGE by molecular dynamics analysis<sup>31</sup>. Among these models, the Fr48 model is the representative conformation without thermodynamical destabilization (Extended Data Fig.8C, D). This structural model showed essentially the same structure to that of URH49ΔN41 containing \(\mathrm{SO_4^{2 - }}\) and PEG (Extended Data Fig.8A, E)). Thus, we concluded that \(\mathrm{SO_4^{2 - }}\) and PGE did not significantly affect the overall structure of URH49ΔN41 and continued our analysis of this model structure as the authentic apo- URH49ΔN41.  
+
+<|ref|>text<|/ref|><|det|>[[137, 310, 857, 876]]<|/det|>
+The apo- URH49ΔN41 model exhibited three different structural features compared to the apo- UAP56ΔN42 structure (Fig.5A). Firstly, although the amino acid sequence of the linker is the same in both helicases, the linker of UAP56 did not have any secondary structure, while this region of URH49 showed a clearly oriented \(\alpha\) - helical structure. This structural difference in the linker part was also implicated by the finding that the R2 fragment containing the linker and C domain was derived exclusively from URH49ΔN41 but not from UAP56ΔN42 (Fig.4E). Secondly, the relative orientation of the N- and C- domains differed between the apo- UAP56ΔN42 and the apo- URH49ΔN41. The overall folds of N- or C- domains were similar to each other, which characters are conserved in the DEAD- box families<sup>30</sup> (Fig.5B, Extended Data Fig.8E). The distinct spatial arrangement of the N- and C- domains in DEAD- box family proteins affects their complex formation<sup>20</sup>, implying that these differences may contribute to their unique apo- complex formation. We did not find any clear differences in the spatial arrangement of the residue V224 of UAP56 and the residue C223 of URH49 within their apo- conformation (Extended Data Fig.8F). Therefore, we described details about this point in the Discussion section. Finally, the loop structures in the C- domain formed by residues 344- 354 of UAP56 and residues 343- 353 of URH49 (hereafter referred to as C- domain loop) are positioned differently from each other (Fig.5C). In addition, the C- domain loop of URH49 covers its own ATP binding pocket. Consistent with this observation, URH49 had a lower ATP- binding and ATP- dependent helicase activity than UAP56 (Extended Data Fig.8G, H), implying that the C- domain loop prevented ATP from binding to the apo- URH49. Therefore, the URH49 C- domain loop likely gives URH49 less affinity for ATP in the apo- state, allowing it to maintain a different conformation from the apo- state of UAP56.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[137, 120, 777, 139]]<|/det|>
+## Evolutionary diversified apo-structures from Sub2 to UAP56 and URH49  
+
+<|ref|>text<|/ref|><|det|>[[137, 144, 849, 380]]<|/det|>
+Finally, we investigated whether the structural features of either UAP56 or URH49 observed in their apo states were conserved in yeast Sub2, the ancestor gene of UAP56. Firstly, we performed the limited proteolysis of Sub2ΔN59, a mutant corresponding to UAP56ΔN42, in ATP depleted conditions. The digestion pattern of Sub2ΔN59 was similar to that of UAP56ΔN42 (Extended Data Fig.8I). Secondly, we compared the structural difference of C- domains in UAP56, URH49, and Sub2 extracted from the co- crystal structure of Sub2- THO \(^{32}\) . The location of the C- domain loop of Sub2 was similar to that of UAP56 (Fig.5D). These data suggested that the structure of UAP56 is evolutionarily conserved with Sub2, while URH49 has diversified from UAP56 during evolution to form a different apo- structure.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[140, 121, 234, 138]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[140, 144, 857, 211]]<|/det|>
+In this study, we uncover unknown apo- AREX components and the molecular basis for their distinct complex formation, which is crucial for the functional divergence of both helicases playing distinct roles in mRNA processing and export.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 240, 796, 258]]<|/det|>
+## The apo-AREX complexes regulate gene expression of URH49 targets genes  
+
+<|ref|>text<|/ref|><|det|>[[139, 263, 858, 470]]<|/det|>
+With the exception of CIP29, the detail of the apo- AREX composition were not determined. Here, we identified RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM as factors which interact with URH49 by two- step affinity purification based on the apo- AREX complex under the ATP depletion condition. Depletion of each apo- AREX component induced the accumulation of poly(A) \(^+\) RNA in nuclear speckles. mRNAs with retained introns are tethered in nuclear speckles and thus inefficiently exported to the cytoplasm \(^{25,26}\) . This observation has led to the idea that the apo- AREX complex may have a link to upstream mRNA processing such as splicing as well as the apo- TREX complex does \(^{5,33}\) .  
+
+<|ref|>text<|/ref|><|det|>[[138, 476, 856, 853]]<|/det|>
+In addition, all of the newly identified apo- AREX components have other previously describe roles in nuclear RNA dynamics. RUVBL1 and RUVBL2 form heterodimers and function in chromatin remodeling as INO80 and SRCAP complexes \(^{22}\) . RUVBL1 and RUVBL2, as TIP160 complex components, are also involved in the regulation of transcription via histone acetylation at promoters \(^{22}\) . HNRNPM, which belongs to the hnRNPs family, together with various interacting factors, contributes to many aspects of RNA metabolism \(^{23,34,35}\) . ILF2 and ILF3 also form heterodimers and function in RNA splicing as the LASR complex with numerous proteins including HNRNPM \(^{23,35}\) . Moreover, CIP29 contains an evolutionarily conserved DNA- binding motif, SAF domain, and binds to DNA, which led to the speculation that CIP29 functions in transcription \(^{36,37}\) . These findings suggested that the novel factors identified in this study may form multiple complexes including the apo- AREX complex and are widely involved in RNA metabolism from chromatin regulation to splicing. Further studies will uncover how two closely related complexes recognize their target transcripts, which will reveal the individual function of these two complexes in coupling the processes from mRNA transcription to export.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[140, 120, 755, 161]]<|/det|>
+## The apo-AREX complexes is implicated in cell proliferation and cancer progression  
+
+<|ref|>text<|/ref|><|det|>[[139, 166, 852, 328]]<|/det|>
+UAP56 is continuously expressed during the cell cycle, while URH49 is expressed specifically during the proliferation phase and not during the quiescent phase<sup>14</sup>. URH49 are required for gene expression of subsets of key regulators of mitosis<sup>15</sup> and cytokinesis<sup>11</sup>. The apo- AREX components were also required for the expression of representative targets regulated by URH49. These results indicate that the apo- AREX complex is implicated in cell proliferation via regulation of its target gene expression.  
+
+<|ref|>text<|/ref|><|det|>[[137, 333, 848, 780]]<|/det|>
+Aberrant expression of UAP56 and URH49 is involved in tumorigenesis and cancer progression. UAP56 is upregulated in colorectal and ovarian cancers, and is associated with their progressions<sup>38,39</sup>. The association between URH49 and cancer has been reported more often observed than that of UAP56. In Cancer Genome Atlas (TCGA), pan- cancer cohort analysis showed that URH49 is upregulated in 18 cancer types than normal tissue<sup>40</sup>. Actually, aberrant up- regulation of URH49 is observed in various cancer tissues and cancer cell lines, and is positively correlated with advanced clinical stage and poor prognosis<sup>40,41,42</sup>. URH49 is important for the genes expression involved in cell proliferation and promotes malignancy of these cancer<sup>40,42</sup>. CIP29 is highly expressed in various cancer and is associated with cancer malignancy<sup>7,43</sup>. Overexpression of RUVBL1, RUVBL2, ILF2, ILF3, and HNRNPM are observed in various cancers with their progression (RUVBL1 and RUVBL2:22,44, ILF2 and ILF3:45,46,47, HNRNPM:48,49). Thus, it is possible that the apo- AREX components are important regulators of gene expression in cancer. Therefore, agents which impair the apo- AREX complex expressions and/or activity could be potential targets for cancer therapy. Indeed, YM155, an inhibitor of ILF3, and CB- 6644, an inhibitor of RUVBL1 and RUVBL2 sub- complex, exhibit anticancer activities<sup>50,51</sup>. The research focusing on the regulation of the apo- AREX complex may contribute to a therapeutic benefit in cancer.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 810, 630, 828]]<|/det|>
+## Complex formation and structure of UAP56 and URH49  
+
+<|ref|>text<|/ref|><|det|>[[140, 834, 851, 899]]<|/det|>
+We observed that UAP56 and URH49 have different structural features in the apo- state and remodeled to similar structures upon the ADP binding. The structure of UAP56 and URH49 in apo- state showed the distinct configuration of the N- and C-
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 119, 856, 283]]<|/det|>
+domain probably due to different linker structures. UAP56 interacts with the THO subcomplex via their N- and C- domains within the reconstituted UAP56- THO subcomplex<sup>52</sup>. These interactions were also conserved in the crystal structure of yeast Sub2- Tho<sup>32,53</sup>. These findings suggest that the spatial arrangement of N- and C- domains of UAP56 is important for the formation of the apo- and ATP- TREX (UAP56) complex. Therefore, the difference in the configurations between UAP56 and URH49 structures may play a pivotal role in their unique complex formation.  
+
+<|ref|>text<|/ref|><|det|>[[137, 287, 850, 496]]<|/det|>
+The loop structure within the C- domain of apo- URH49ΔN41 covers the ATP- binding pocket in their N- domain and prevents its own ATP binding. Although several DEAD- box proteins inhibit their ATP binding by intramolecular interactions<sup>20</sup>, the arrangement of the loop in the C- domain is not observed among other DEAD- box proteins, suggesting that URH49 has evolutionarily acquired a unique structure to repress ATP binding. Consistent with our findings, Sub2 exhibited sufficient helicase activity as well as UAP56<sup>54</sup>. In addition, CIP29 stimulates ATPase activity of URH49, followed by ATP binding in URH49<sup>55</sup>, implying that URH49 forms the apo- AREX complex as a steady state, then, remodels to the ATP- TREX complex in the cell.  
+
+<|ref|>text<|/ref|><|det|>[[137, 500, 857, 805]]<|/det|>
+Orthologs of UAP56 and URH49 are present in vertebrates, while only the ortholog of UAP56 is present in insects, implying that the ortholog of URH49 diversified during the evolution between vertebrates and invertebrates. Furthermore, amino acids corresponding to human UAP56- V224 and URH49- C223 are already present in orthologs of UAP56 and URH49. Thus, the amino acid substitution between UAP56- V224 and URH49- C223 probably occurred after diversification. But the precise effect of the differences between UAP56- V224 and URH49- C223 on their apo- conformation is not yet fully understood this time. Because UAP56- V224 and URH49- C223 are present inside N- domains of the apo- UAP56ΔN42 and URH49ΔN41, substitutions of these amino acids do not appear to affect the surface structures. One attractive possibility that needs to be verified later is that these substitutions affected the linker structure of UAP56 and URH49, resulting in the different orientations of the N- and C- domains between the two helicases.  
+
+<|ref|>text<|/ref|><|det|>[[138, 810, 850, 900]]<|/det|>
+Integrating our findings into previous observations, we proposed the following model that UAP56 and URH49 form the apo- TREX and - AREX complex based on their apo- conformation (Fig.6). The binding of ATP into UAP56 and URH49 promotes the conformational change to a highly similar closed conformation, triggering
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 120, 857, 280]]<|/det|>
+the remodeling of the respective apo-complex to the ATP- TREX complex. It has been implicated that UAP56 and URH49 function in selective mRNA export by forming respective complex formation. Therefore, we provided the possibility that diversified apo- structures of UAP56 and URH49 derived from Sub2 have contributed to the organization of gene regulation in humans. Further progress in genome analysis of other species is expected to advance our understanding of the diversification of UAP56 and URH49.  
+
+<|ref|>text<|/ref|><|det|>[[137, 286, 857, 472]]<|/det|>
+Interestingly, amino acids in UAP56 and URH49 required for selective circular RNA export are not linked to the apo- complex formations. ATPase activities of UAP56 and URH49 are not required for circular RNA export<sup>28</sup>. While ATP loading and ATPase activity are known to be essential for their function of mRNA export<sup>11,56</sup>. In addition, there is no significant difference in length between the mRNAs selectively exported by the two helicases and these pre- mRNAs<sup>11</sup>. Thus, the underlying mechanism of circular RNA export needs to be investigated separately from that of mRNA export including whether the complex formation is required for circular RNA export.  
+
+<|ref|>text<|/ref|><|det|>[[137, 477, 852, 732]]<|/det|>
+During evolution, many RNA- binding proteins have functionally diversified to execute well- tuned gene expression contributing to the complexity of living organisms. These include UAP56 and URH49 which have diversified to form different complexes and function in selective mRNA processing and export<sup>15</sup>. In addition to both helicases, several key mRNA processing and export factors such as NXFs, NXTs, DDX19s, and SR proteins have evolutionarily diversified from yeast to human. Some factors have gained different target specificities from their originated paralogs, but the molecular mechanism behind these differences is mostly unknown<sup>57</sup>. Further elucidation of the diversification of mRNA export- related proteins will uncover the mechanistic insight into the accurate gene expression through mRNA processing and export in humans and how it has developed during evolution.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 763, 368, 780]]<|/det|>
+## ACKNOWLEDGMENTS  
+
+<|ref|>text<|/ref|><|det|>[[137, 786, 857, 900]]<|/det|>
+We thank Mr. Yuzo Watanabe for help with the LC- MS/MS analysis and our lab members for their constructive discussions. Diffraction data were collected at the BL26B1 and BL44XU stations of SPring- 8 (Hyogo, Japan) with the approval of JASRI (proposal nos. 2015A1063, 2015B2063 and 2017B6750). This work was supported in part by "Grants- in- Aid" from JSPS KAKENHI (Grant Numbers 26292053, 17K19232,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[139, 119, 855, 186]]<|/det|>
+19K22280, 19H02884, 21K19078 and 22H02264 to S.M, 19K15807 to K.F). This work was also supported in part by "Grants- in- Aid" from The Sasakawa Scientific Research Grant from The Japan Science Society to K.F.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 216, 410, 234]]<|/det|>
+## AUTHOR CONTRIBUTIONS  
+
+<|ref|>text<|/ref|><|det|>[[139, 240, 845, 354]]<|/det|>
+K.F. and S.M. conceived and designed this study; K.F. performed the experiments and analyses, organized the data and drafted the manuscript; M.I, M.I, K.H, Y.I, H.Y, and T.Y performed biological experiments; M.K. performed MD analysis; B.M. performed crystal structure analysis; K.F., N.F, A.M, and S.M. analyzed the results and wrote the paper. All authors reviewed the final manuscript.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 381, 446, 399]]<|/det|>
+## DECLARATION OF INTERESTS  
+
+<|ref|>text<|/ref|><|det|>[[140, 406, 494, 423]]<|/det|>
+The authors declare no competing interests.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 429, 660, 447]]<|/det|>
+## CONTACT FOR REAGENT AND RESOURCE SHARING  
+
+<|ref|>text<|/ref|><|det|>[[140, 453, 848, 496]]<|/det|>
+For more information and requests regarding resources and reagents, please contact the Lead Contact, Seiji Masuda (smasuda@nara.kindai.ac.jp).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 525, 329, 542]]<|/det|>
+## METHOD DETAILS  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 549, 553, 566]]<|/det|>
+## Cell culture and establishment of stable cell line  
+
+<|ref|>text<|/ref|><|det|>[[139, 572, 848, 686]]<|/det|>
+Flp- In T- REx 293, U2OS, MCF7, A549 cells were maintained in Dulbecco's Modified Eagle's Medium (Fujifile Wako, Tokyo, Japan) supplemented with \(10\%\) heat- inactivated fetal bovine serum at \(37^{\circ}\mathrm{C}\) . Flp- In T- REx 293 cells stably expressing 3x FLAG- tagged protein were obtained by the transfection of pcDNA5 3x FLAG- tagged protein expression vector with pOG44, respectively.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 715, 546, 732]]<|/det|>
+## Reagents and antibodies, preparation of serum  
+
+<|ref|>text<|/ref|><|det|>[[139, 738, 850, 899]]<|/det|>
+4', 6- diamidino- 2- phenylindole (DAPI) was purchased from Fujifilm Wako. Antibodies were obtained as follows: FLAG M2 mouse monoclonal antibody, mouse anti- \(\beta\) - actin antibody, rabbit anti- HNRNPM antibody and mouse anti- SRRM2 antibody (Sigma- Aldrich Japan, Tokyo, Japan), HA (12CA5) mouse monoclonal antibody (GeneTex, Irvine, CA), mouse anti- GAPDH antibody (Fujifilm Wako), antibodies against THOC1, THOC2, THOC5, ALYREF, CIP29, UAP56 and URH49 have been described previously<sup>11</sup>. Anti- RUVBL1, anti- RUVBL2, anti- ILF2, and anti- ILF3 sera
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 118, 828, 258]]<|/det|>
+were prepared from immunized rats as described previously<sup>58</sup> in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Animal Committee in Kyoto University (Animal experiments were approved by the Committee on the Ethics of Animal Experiments of Kyoto University, Experiment permission number: Lif- K17002). The antibodies used in this study are listed in supplemental Tables S3.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 286, 408, 304]]<|/det|>
+## Plasmids, primers and siRNAs  
+
+<|ref|>text<|/ref|><|det|>[[137, 310, 857, 754]]<|/det|>
+To construct the following plasmids, the fragments were obtained by PCR amplification with the addition of restriction enzyme sites or infusion enzyme sites at both ends. pcDNA5- 3xFLAG and pcDNA5- HA vectors were generated as described previously<sup>11,59</sup>. 3xFLAG- UAP56, 3xFLAG- URH49, 3x FLAG- CIP29, 3xFLAG- RUVBL1, 3xFLAG- RUVBL2, 3xFLAG- ILF2, 3xFLAG- ILF3, and 3xFLAG- HNRNPM expression vectors were generated by the insertion of the respective open reading frame into pcDNA5- 3xFLAG, respectively. The HA- CIP29 expression vector was generated by the insertion of the CIP29 open reading frame into pcDNA5- HA. To construct GST- UAP56 and GST- URH49, and their derivative mutant expression plasmids, respective open reading frames of UAP56, URH49 and their mutants were inserted into pGEX6p2. The GST- Sub2Δ59 (60- 446 amino acids) expression plasmid was constructed by inserting the respective region into pGEX6p2. MBP- RUVBL1 (250- 456 amino acids), MBP- RUVBL2 (1- 225 amino acids), MBP- ILF2 (240- 390 amino acids), and MBP- ILF3 (280- 355 amino acids) expression plasmids were constructed by inserting their respective region into pMALc2X. To construct mutants of FLAG- UAP56, or - URH49 expression plasmids, overlap extension PCR was performed to induce the mutation. The construction of the plasmids was confirmed by sequencing. The primers and siRNAs used in this study are listed in supplemental Tables S4 and Tables S5.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 786, 407, 803]]<|/det|>
+## Plasmids or siRNA transfection  
+
+<|ref|>text<|/ref|><|det|>[[140, 809, 857, 850]]<|/det|>
+Transient transfection of siRNA and plasmids was performed using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's instructions.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 880, 535, 898]]<|/det|>
+## Total, cytoplasmic and nuclear RNA isolation
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 119, 860, 234]]<|/det|>
+Total RNA was isolated by Sepasol- RNA I super G (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions. For cytoplasmic RNA preparation, the cells were treated with lysis buffer (20 mM Tris- HCl (pH 8.0), 200 mM NaCl, 1 mM MgCl2, 1% NP- 40) on ice for 5 min. The cytoplasmic fraction was isolated by brief spin while the nuclear fraction was prepared from the precipitate.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 263, 789, 304]]<|/det|>
+## Quantitative and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR)  
+
+<|ref|>text<|/ref|><|det|>[[139, 308, 836, 444]]<|/det|>
+Quantitative RT- PCR (RT- qPCR) was performed with TB Green Premix Ex Taq II (TakaraBio, Tokyo, Japan) and analyzed by Thermal Cycler Dice real time system II (TakaraBio). PGK1 was used for standardization. The quantity of each mRNA was calculated by threshold cycle (Ct) values. The relative expression of each mRNA was evaluated by the values of \(2^{\circ}[\mathrm{Ct(TBP)} - \mathrm{Ct(eachmRNA)}]\) . Primer sets are listed in Table S6.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 477, 830, 497]]<|/det|>
+## Immunoprecipitation, immunoblotting, LC-MS/MS analysis and Silver staining  
+
+<|ref|>text<|/ref|><|det|>[[137, 501, 856, 875]]<|/det|>
+Immunoprecipitation, immunoblotting, LC- MS/MS analysis and Silver stainingPreparation of nuclear extract and immunoprecipitation were performed as described previously<sup>11</sup>. Briefly, nuclear extract was incubated for 30 min at \(20^{\circ}\mathrm{C}\) for the depletion of ATP and centrifuged to recover the supernatant. Then, RNaseA (100 ng/μL), ATP (500 μM), MgCl<sub>2</sub> (3.2 mM) and creatine phosphate (20 mM) were added to the nuclear extract, and the reaction mixture was incubated for 30 min at \(30^{\circ}\mathrm{C}\) . In the ATP (- ) condition, ATP, MgCl<sub>2</sub>, and creatine phosphate were omitted. After a brief spin, the clear supernatant was mixed with anti DYKDDDDK tag antibody beads (Fujifilm Wako) or anti HA antibody beads (Fujifilm Wako) and rotated overnight at \(4^{\circ}\mathrm{C}\) . The beads were extensively washed with PBS containing 0.1% TritonX100, 0.2 mM PMSF and 0.5 mM DTT to remove nonspecifically bound proteins. The proteins attached to the beads were dissolved in SDS sample buffer (250 mM Tris- HCl, 1% sodium lauryl sulfate), 0.002% bromophenol blue and 40% Glycerol for 10 min at \(37^{\circ}\mathrm{C}\) . The eluate was recovered to a new tube and DTT was added to 10 mM, and boiled for 2 min. For tandem immunoprecipitation, the proteins attached to the beads were eluted with FLAG peptide (M&S TechnoSystems Inc, Osaka, Japan) or HA peptide (MBL, Nagoya, Japan).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 120, 833, 354]]<|/det|>
+Samples were separated by SDS- polyacrylamide gel electrophoresis (SDS- PAGE) and blotted onto polyvinylidene difluoride membrane (Pall, Ann Arbor, MI). The blotted membrane was blocked with PBS containing \(0.1\%\) polyoxyethylene sorbitan monolaurate (Tween20) and \(5\%\) skim milk for \(1\mathrm{h}\) and reacted with primary antibodies at \(4^{\circ}\mathrm{C}\) overnight with gentle rotation. The membrane was extensively washed with PBS containing \(0.1\%\) Tween20. Secondary antibody conjugated with horseradish peroxidase was reacted with the membrane by rotating for \(2\mathrm{h}\) . After the extensive washing, the membrane was reacted with a chemiluminescence reagent (Millipore, Darmstadt, Germany). Signals were detected with LAS 4000 mini (GE Healthcare Japan, Tokyo, Japan).  
+
+<|ref|>text<|/ref|><|det|>[[139, 358, 848, 518]]<|/det|>
+The LC- MS/MS analysis was performed by Q Exactive Plus (Thermo Fisher Scientific, Waltham, MA). As outputs of the LC- MS/MS analysis, prot score was calculated using Mascot software (Matrix Science, London, UK). For factors with different prot_acc but the same GeneName, only the largest prot_score is listed. For comparison, protein scores were calculated by subtracting the prot_score of the control from each data. Gene ontology (GO) was analyzed using Database for Annotation, Visualization and Integrated Discovery (DAVID: version \(6.7)^{60}\) .  
+
+<|ref|>text<|/ref|><|det|>[[140, 524, 842, 568]]<|/det|>
+For silver staining, proteins were separated with SuperSep™ Ace, \(5\% - 20\%\) , 17- well (Fujifile Wako). Silver staining was performed as described previously.  
+
+<|ref|>sub_title<|/ref|><|det|>[[141, 597, 397, 614]]<|/det|>
+## Immunofluorescence staining  
+
+<|ref|>text<|/ref|><|det|>[[139, 619, 852, 852]]<|/det|>
+Cells ( \(5 \times 10^{4}\) cells/mL) on glass coverslips in a 12- well plate were cultured for \(24\mathrm{h}\) and transfected with siRNA or plasmid. After a \(48\mathrm{h}\) incubation, cells were fixed in \(4\%\) formaldehyde in PBS, permeabilized with \(0.1\%\) Triton X- 100 in PBS, and blocked with \(6\%\) bovine serum albumin (BSA) in PBS. The coverslips were reacted with primary antibodies in \(2\%\) BSA in PBS, secondary antibody conjugated with Alexa- 488 or Alexa- 594 (Molecular Probes, Eugene, OR) and DAPI to counterstain the nuclei. Fluorescence images were obtained with a fluorescent microscopy, Axioplan 2 (Carl Zeiss, Germany) or FV10i (Olympus, Tokyo, Japan), a laser scanning confocal microscopy, using the \(x60\) objective lens. Line Plot analysis was performed using FV10- ASW v4.1 software (Olympus)  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 880, 545, 898]]<|/det|>
+## RNA-fluorescence in situ hybridization (FISH)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[135, 118, 857, 401]]<|/det|>
+RNA- FISH was performed as described previously<sup>11</sup>. Briefly, cells \((5 \times 10^{4}\) cells/mL) were inoculated on glass coverslips in a 12- well plate, cultured for 24 h and transfected with siRNA or plasmid. After 24 to 48 h incubation, cells were fixed with \(10\%\) formaldehyde in PBS for 20 min and permeabilized in \(0.1\%\) Triton X- 100 in PBS for 10 min. The coverslip was washed three times with PBS for 10 min, and once with \(2\times\) Standard Saline Citrate (SSC) for 5 min. Cells were prehybridized with ULTRAhyb- Oligo Hybridization Buffer (Ambion, Austin, TX) for 1 h at \(42^{\circ}\mathrm{C}\) in a humidified chamber. Then, they were treated with 10 pmol Alexa Fluor 594- labeled oligo- dT<sub>45</sub> probe (Molecular Probes) overnight. Cells were washed for 20 min at \(42^{\circ}\mathrm{C}\) with \(2\times\) SSC, \(0.5\times \mathrm{SSC}\) , and \(0.1\times \mathrm{SSC}\) . Nuclei were counterstained with DAPI. Fluorescent images were obtained with Axioplan 2. Poly (A) \(^+\) RNA signal intensities in the nucleus and the cell were calculated with ImageJ software (https://imagej.nih.gov/ij/).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 428, 448, 446]]<|/det|>
+## Protein expression and purification  
+
+<|ref|>text<|/ref|><|det|>[[135, 451, 852, 757]]<|/det|>
+GST- fusion proteins were produced in \(E\) . coli BL21 strain. The production of recombinant protein was induced by the addition of \(0.05\mathrm{mM}\) IPTG at \(18^{\circ}\mathrm{C}\) overnight. Cells were pelleted by centrifugation at \(6000\times \mathrm{g}\) for \(10\mathrm{min}\) . The pellet was resuspended in PBS containing \(0.2\mathrm{mM}\) phenyl methyl sulfonyl fluoride (PMSF) and \(1\mathrm{mM}\) dithiothreitol (DTT), and then, sonicated 30 seconds four times on ice. The clear lysate was obtained by centrifugation at \(8000\times \mathrm{g}\) for \(15\mathrm{min}\) and transferred to a new tube. Glutathione- fixed beads (GE Healthcare) were added to the clear lysate and rotated for \(3\mathrm{h}\) at \(4^{\circ}\mathrm{C}\) . After the extensive washing with PBS containing \(0.2\mathrm{mM}\) PMSF and \(1\mathrm{mM}\) DTT, precision protease (GE Healthcare) was added to remove the GST- tag and rotated overnight at \(4^{\circ}\mathrm{C}\) . The eluate containing the GST- tag removed protein was further purified on a gel filtration column, HiPrep 16/60 Sephacryl S- 100 HiResolution (GE Healthcare). The purity and concentration of recombinant protein were confirmed by SDS- PAGE followed by Coomassie Brilliant Blue R- 250 (Nacalai Tesque) staining.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 787, 309, 803]]<|/det|>
+## Limited proteolysis  
+
+<|ref|>text<|/ref|><|det|>[[135, 809, 826, 898]]<|/det|>
+Recombinant protein was incubated with 1/100 (weight ratio) of trypsin (Promega Japan) at \(25^{\circ}\mathrm{C}\) for 0, 30, 120, or 300 min. In the ADP condition, ADP (1 mM) and \(\mathrm{MgCl_2}\) (10 mM) were added to the reaction mixture. The digestion was stopped by adding an equal volume of SDS sample buffer. Samples were boiled for 2 min, then,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 118, 830, 161]]<|/det|>
+separated by SDS- PAGE and stained with Coomassie Brilliant Blue R- 250. The LC- MS/MS analysis for some of the separated bands was performed by Q Exactive Plus.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 193, 306, 209]]<|/det|>
+## ATP binding assay  
+
+<|ref|>text<|/ref|><|det|>[[139, 216, 851, 329]]<|/det|>
+Recombinant protein was incubated with ATP beads (Jena Bioscience, Jena, Germany) and PBS containing \(0.1\%\) Triton X- 100, \(0.2 \mathrm{mM}\) PMSF, and \(1 \mathrm{mM}\) DTT at \(4^{\circ} \mathrm{C}\) for \(1 \mathrm{~h}\) , and then washed with PBS buffer. Input sample and bound proteins with ATP beads were eluted with SDS sample buffer and separated by SDS- PAGE, then stained with Coomassie Brilliant Blue R- 250 (Nacalai Tesque).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 360, 265, 376]]<|/det|>
+## Helicase assay  
+
+<|ref|>text<|/ref|><|det|>[[139, 382, 850, 781]]<|/det|>
+Helicase assays were performed as described previously with a slight modification \(^{62}\) . Substrate sequences were as follows; \(5^{\prime}\) duplex RNA was UGCUUGCUUUACGGUGCUAGUUUUGUUUGUUUGAUUUCGCCC, and \(5^{\prime}\) duplex DNA was GTAAAGCAAGCTTGAGT. Underlining indicates the region of the duplex. DNA was labeled by T4 polynucleotide kinase (Toyobo, Osaka, Japan) in reaction buffer with \([\gamma - ^{32}\mathrm{P}]\) ATP (GE healthcare) and purified on a Sephadex G- 25 column (GE healthcare). To make duplex, RNA and \(^{32}\mathrm{P}\) labeled DNA were annealed in annealing buffer (20 mM Tris- Cl, pH 8.0, 200 mM potassium acetate, 0.1 mM EDTA). Unwinding reaction buffer (50 nM duplex, \(1 \mu \mathrm{M}\) cold trap oligo DNA, 20mM HEPES- KOH pH 7.9, 50 mM potassium acetate, \(0.1 \mathrm{mM}\) MgCl₂, \(2 \mathrm{mM}\) DTT, \(0.01\%\) BSA, \(4 \mathrm{U} / \mu \mathrm{l}\) RNase inhibitor (Toyobo), \(1 \mathrm{mM}\) ATP, and \(50 \mathrm{ng}\) of GST fusion protein) containing duplex was preincubated without ATP at \(37^{\circ} \mathrm{C}\) for 5 minutes and incubated with ATP at \(37^{\circ} \mathrm{C}\) for 30 minutes. The reaction was stopped by the addition of proteinase K buffer (1 mg/ml proteinase K, \(10 \mathrm{mM}\) Tris- Cl, pH 8.0, \(12.5 \mathrm{mM}\) EDTA, \(150 \mathrm{mM}\) sodium chloride, \(1\%\) SDS) and incubated at \(37^{\circ} \mathrm{C}\) for 25 minutes. Sample was loaded to a \(15 \%\) polyacrylamide gel. Gels were exposed to an imaging plate (Fujifilm Wako and image data were obtained using a BAS2000 (Fujifilm Wako).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 811, 543, 828]]<|/det|>
+## Crystallization and Crystal structural analysis  
+
+<|ref|>text<|/ref|><|det|>[[139, 834, 838, 899]]<|/det|>
+URH49ΔN41 was concentrated at \(5 \mathrm{mg} / \mathrm{ml}\) and crystallized by the sitting- drop vapor diffusion. Briefly, \(1 \mu \mathrm{l}\) of a protein solution was mixed with \(1 \mu \mathrm{l}\) of a mother liquid containing \(0.2 \mathrm{mM}\) NaPO₄, pH 8.5, \(30 \%\) (w/v) PEG3350 at \(20^{\circ} \mathrm{C}\) . The diffraction of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[135, 119, 856, 401]]<|/det|>
+the crystals was confirmed by an in- house Bruker Hi- star detector after flash- cooling in a cold nitrogen gas stream (100 K) with \(25\%\) (v/v) ethylene glycol. The diffraction images were collected at \(100\mathrm{K}\) (in a cold nitrogen gas stream) on a Rayonix MX225 CCD detector (Rayonix, Evanston, IL) with a wavelength of \(1.0\mathrm{\AA}\) at BL26B2 in SPring- 8 (Hyogo, Japan). The resulting data sets were processed, merged, and scaled using XDS \(^{63}\) . The structure was solved by molecular replacement with UAP56ΔN42 (Protein Data Bank entry 1XTI) using a search mode by Molrep implemented in CCP4i software \(^{64}\) . The model was refined using PHENIX 1.20.1 software \(^{65}\) , rebuilt using COOT \(0.8.9^{66}\) and further modified based on sigma- weighted (2|Fo|-|Fc|) and (|Fo|-|Fc|) electron density maps. Protein structure images were depicted using PyMOL software (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC).  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 429, 448, 446]]<|/det|>
+## Molecular Dynamics (MD) analysis  
+
+<|ref|>text<|/ref|><|det|>[[135, 451, 856, 877]]<|/det|>
+MD simulation was performed for the apo- form of URH49ΔN41 using Desmond Molecular Dynamics System, version 5.2 (D. E. Shaw Research, New York, NY) \(^{31}\) . The atomic coordinates of \(\mathrm{SO_4^{2 - }}\) and PGE were removed from the crystal structure of their complex with URH49ΔN41 to generate the initial structure for the simulation. First, the structure was preprocessed with Protein Preparation Wizard of Maestro (version 11.4), the GUI for Desmond, to assign bond orders, add hydrogens, and create disulfide bonds. Then, it was solvated in a box with a buffer distance of \(10\mathrm{\AA}\) to the boundary. Afterwards, solvation was performed in a box with a buffer distance to the boundary of \(10\mathrm{\AA}\) . Sodium and chloride ions were added to neutralize the entire solvated system. OPLS_2005 force field \(^{67}\) and SPC model \(^{68}\) were used for the protein and water molecules, respectively. After relaxing the system according to the Maestro's default relaxation protocol, an MD run was performed in the constant- NPT ensemble at \(300\mathrm{K}\) and \(1.013\mathrm{bar}\) for \(1\mu \mathrm{s}\) . The coordinates were recorded every \(1\mathrm{ns}\) to yield 1,001 snapshots. Otherwise, the default setting in Desmond was adopted. The resulting MD trajectory was equidistantly divided into 101 frames so that each frame could contain ten consecutive snapshots. Then the Root Mean Square Deviation (RMSD) values between main chains of arbitrary two frames were calculated to generate an RMSD matrix. In the matrix, frames with an RMSD less than \(2\mathrm{\AA}\) were assigned to belong to
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[137, 119, 800, 161]]<|/det|>
+the same cluster. In each cluster, the frame with the minimal RMSDs to the other members was considered a representative structure of the cluster.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 192, 466, 209]]<|/det|>
+## Quantification and statistical analysis  
+
+<|ref|>text<|/ref|><|det|>[[137, 215, 853, 448]]<|/det|>
+RT- qPCR results were quantified using Thermal Cycler Dice real time system II (TakaraBio). FISH data was quantified using ImageJ software. Immunofluorescence staining data were quantified using FV10- ASW v4.1 software (Olympus). LC- MS/MS data were analyzed using Mascot software (Matrix Science, London, UK). The statistical significance for two- group and multiple comparisons was tested using R software<sup>69</sup>, as indicated in the legend of each Fig.. Non- adjusted (two- group comparison) and adjusted (multiple comparisons) P- values are indicated in each Fig.. In box plots, the first and third quartiles are indicated by both ends of the box, the median is indicated by a vertical line in the box, and the minimum and maximum excluding outliers are the ends of the whiskers. The outliers are indicated with open circles.  
+
+<|ref|>sub_title<|/ref|><|det|>[[140, 477, 368, 494]]<|/det|>
+## Data and code availability  
+
+<|ref|>text<|/ref|><|det|>[[140, 500, 831, 542]]<|/det|>
+Atomic coordinates and structure factors for the reported crystal structures have been deposited with the Protein Data bank under accession number 8IJU (URH49A41).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[75, 120, 849, 900]]<|/det|>
+Reference1. Singh, G., Pratt, G., Yeo, G. W. & Moore, M. J. The clothes make the mRNA: Past and present trends in mRNAFashion. Annu. Rev. Biochem. 84, 325–354 (2015).2. Chang, C. Te et al. Chtop is a component of the dynamic TREX mRNA export complex. EMBO J. 32, 473–486 (2013).3. Folco, E. G., Lee, C. S., Dufu, K., Yamazaki, T. & Reed, R. The proteins PDIP3 and ZC11A associate with the human TREX complex in an ATP-dependent manner and function in mRNA export. PLoS One 7, 1–7 (2012).4. Stäßer, K. et al. TREX is a conserved complex coupling transcription with messenger RNA export. Nature 417, 304–308 (2002).5. Masuda, S. et al. Recruitment of the human TREX complex to mRNA during splicing. Genes Dev. 19, 1512–1517 (2005).6. Dufu, K. et al. ATP is required for interactions between UAP56 and two conserved mRNA export proteins, Aly and CIP29, to assemble the TREX complex. Genes Dev. 24, 2043–2053 (2010).7. Heath, C. G., Viphakone, N. & Wilson, S. A. The role of TREX in gene expression and disease. Biochem. J. 473, 2911–2935 (2016).8. Fleckner, J., Zhang, M., Valcarcel, J. & Green, M. R. U2AF65 recruits a novel human DEAD box protein required for the U2 snRNP-branchpoint interaction. Genes Dev. 11, 1864–1872 (1997).9. Shen, J., Zhang, L. & Zhao, R. Biochemical characterization of the ATPase and helicase activity of UAP56, an essential pre-mRNA splicing and mRNA export factor. J. Biol. Chem. 282, 22544–22550 (2007).10. Shen, H. et al. Distinct activities of the DExD/H-box splicing factor hUAP56 facilitate stepwise assembly of the spliceosome. Genes Dev. 22, 1796–1803 (2008).11. Fujita, K. ichi et al. URH49 exports mRNA by remodeling complex formation and mediating the NXF1-dependent pathway. Biochim. Biophys. Acta - Gene Regul. Mech. 1863, 1–14 (2020).12. Katahira, J., Inoue, H., Hurt, E. & Yoneda, Y. Adaptor Aly and co-adaptor Thoc5 function in the Tap-p15-mediated nuclear export of HSP70 mRNA. EMBO J. 28, 556–567 (2009).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[75, 120, 857, 857]]<|/det|>
+763 13. Viphakone, N. et al. TREX exposes the RNA-binding domain of Nxf1 to enable mRNA export. Nat. Commun. 3, (2012).765 14. Pryor, A. et al. Growth-regulated expression and G0-specific turnover of the mRNA that encodes URH49, a mammalian DExH/D box protein that is highly related to the mRNA export protein UAP56. Nucleic Acids Res. 32, 1857–1865 (2004).769 15. Yamazaki, T. et al. The Closely Related RNA helicases, UAP56 and URH49, Preferentially Form Distinct mRNA Export Machineries and Coordinately Regulate Mitotic Progression. Mol. Biol. Cell 21, 2953–2965, (2010).772 16. Mancini, A., Koch, A., Whetton, A. D. & Tamura, T. The M-CSF receptor substrate and interacting protein FMIP is governed in its subcellular localization by protein kinase C-mediated phosphorylation, and thereby potentiates M-CSF-mediated differentiation. Oncogene 23, 6581–6589 (2004).776 17. Wang, L. et al. The THO complex regulates pluripotency gene mRNA export and controls embryonic stem cell self-renewal and somatic cell reprogramming. Cell Stem Cell 13, 676–690 (2013).779 18. Maeder, C. I. et al. The THO Complex Coordinates Transcripts for Synapse Development and Dopamine Neuron Survival. Cell 174, 1436–1449 (2018).781 19. Galarza-Muñoz, G. et al. Human Epistatic Interaction Controls IL7R Splicing and Increases Multiple Sclerosis Risk. Cell 169, 72–84 (2017).783 20. Ozgur, S. et al. The conformational plasticity of eukaryotic RNA-dependent ATPases. FEBS J. 282, 850–863 (2015).785 21. Guan, D. et al. Nuclear Factor 45 (NF45) Is a Regulatory Subunit of Complexes with NF90/110 Involved in Mitotic Control. Mol. Cell. Biol. 28, 4629–4641 (2008).788 22. Jha, S. & Dutta, A. RVB1/RVB2: Running Rings around Molecular Biology. Mol. Cell 34, 521–533 (2009).790 23. Damianov, A. et al. Rbfox Proteins Regulate Splicing as Part of a Large Multiprotein Complex LASR. Cell 165, 606–619 (2016).792 24. Gentili, C. et al. Chromosome missegregation associated with RUVBL1 deficiency. PLoS One 10, 1–19 (2015).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[75, 120, 856, 877]]<|/det|>
+794 25. Johnson, C. et al. Tracking COL1A1 RNA in osteogenesis imperfecta: Splice- 795 defective transcripts initiate transport from the gene but are retained within the 796 SC35 domain. J. Cell Biol. 150, 417-431 (2000). 797 26. Kaida, D. et al. Spliceostatin A targets SF3b and inhibits both splicing and 798 nuclear retention of pre-mRNA. Nat. Chem. Biol. 3, 576-83 (2007). 799 27. Ilik, I. A. et al. Son and srrr2 are essential for nuclear speckle formation. Elife 9, 800 1-48 (2020). 801 28. Huang, C., Liang, D., Tatomer, D. C. & Wilusz, J. E. A length-dependent 802 evolutionarily conserved pathway controls nuclear export of circular RNAs. 803 Genes Dev. 32, 639-644 (2018). 804 29. Kapadia, F., Pryor, A., Chang, T. H. & Johnson, L. F. Nuclear localization of 805 poly(A)+ mRNA following siRNA reduction of expression of the mammalian 806 RNA helicases UAP56 and URH49. Gene 384, 37-44 (2006). 807 30. Shi, H., Cordin, O., Minder, C. M., Linder, P. & Xu, R. M. Crystal structure of 808 the human ATP-dependent splicing and export factor UAP56. Proc. Natl. Acad. 809 Sci. U. S. A. 101, 17628-17633 (2004). 810 31. Bowers, K. J. et al. Scalable Algorithms for Molecular Dynamics Simulations on 811 Commodity Clusters. in Proceedings of the 2006 ACM/IEEE Conference on 812 Supercomputing 43-43 (2006). doi:10.1109/sc.2006.54. 813 32. Ren, Y., Schmiege, P. & Blobel, G. Structural and biochemical analyses of the 814 DEAD-box ATPase Sub2 in association with THO or Yra1. Elife 6, 1-17 (2017). 815 33. Luo, M. L. et al. Pre-mRNA splicing and mRNA export linked by direct 816 interactions between UAP56 and Aly. Nature 413, 644-647 (2001). 817 34. Geuens, T., Bouhy, D. & Timmerman, V. The hnRNP family: insights into their 818 role in health and disease. Hum. Genet. 135, 851-867 (2016). 819 35. Ying, Y. et al. Splicing Activation by Rbfox Requires Self-Aggregation through 820 Its Tyrosine-Rich Domain. Cell 170, 312-323 (2017). 821 36. Aravind, L. & Koonin, E. V. SAP - A putative DNA-binding motif involved in 822 chromosomal organization. Trends Biochem. Sci. 25, 112-114 (2000). 823 37. Matsuda-Hashii, Y. et al. A novel partner gene CIP29 containing a SAP domain 824 with MLL identified in infantile myelomonocytic leukemia [4]. Leukemia 18, 825 1546-1548 (2004).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[78, 120, 857, 900]]<|/det|>
+38. He, C. et al. The DDX39B/FUT3/TGFβR-I axis promotes tumor metastasis and EMT in colorectal cancer. Cell Death Dis. 12, (2021).
+39. Zhang, H. et al. DDX39B contributes to the proliferation of colorectal cancer through direct binding to CDK6/CCND1. Cell Death Discov. 8, (2022).
+40. Bao, Y. et al. DDX39 as a predictor of clinical prognosis and immune checkpoint therapy efficacy in patients with clear cell renal cell carcinoma. Int. J. Biol. Sci. 17, 3158–3172 (2021).
+41. Kuramitsu, Y. et al. Up-regulation of DDX39 in human malignant pleural mesothelioma cell lines compared to normal pleural mesothelial cells. Anticancer Res. 33, 2557–2560 (2013).
+42. Zhang, T. et al. DDX39 promotes hepatocellular carcinoma growth and metastasis through activating Wnt/β-catenin pathway article. Cell Death Dis. 9, (2018).
+43. Kang, G. J. et al. SARNP, a participant in mRNA splicing and export, negatively regulates E-cadherin expression via interaction with pinin. J. Cell. Physiol. 235, 1–13 (2019).
+44. Grigoletto, A., Lestienne, P. & Rosenbaum, J. The multifaceted proteins Reptin and Pontin as major players in cancer. Biochim. Biophys. Acta - Rev. Cancer 1815, 147–157 (2011).
+45. Castella, S., Bernard, R., Corno, M., Fradin, A. & Larcher, J. C. Ilf3 and NF90 functions in RNA biology. Wiley Interdiscip. Rev. RNA 6, 243–256 (2015).
+46. Cheng, S. et al. Expression and critical role of interleukin enhancer binding factor 2 in Hepatocellular Carcinoma. Int. J. Mol. Sci. 17, 1373 (2016).
+47. Marchesini, M. et al. ILF2 Is a Regulator of RNA Splicing and DNA Damage Response in 1q21-Amplified Multiple Myeloma. Cancer Cell 32, 88–100 (2017).
+48. Xu, Y. et al. Cell type-restricted activity of hnRNPM promotes breast cancer metastasis via regulating alternative splicing. Genes Dev. 28, 1191–1203 (2014).
+49. Yang, W. H., Ding, M. J., Cui, G. Z., Yang, M. & Dai, D. L. Heterogeneous nuclear ribonucleoprotein M promotes the progression of breast cancer by regulating the axin/β-catenin signaling pathway. Biomed. Pharmacother. 105, 848–855 (2018).
+50. Cheng, C. C. et al. EGFR-mediated interleukin enhancer-binding factor 3 contributes to formation and survival of cancer stem-like tumorspheres as a
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[78, 120, 856, 900]]<|/det|>
+859 therapeutic target against EGFR- positive non- small cell lung cancer. Lung 860 Cancer 116, 80- 89 (2018). 861 51. Assimon, V. A. et al. CB- 6644 Is a Selective Inhibitor of the RUVBL1/2 862 Complex with Anticancer Activity. ACS Chem. Biol. 14, 236- 244 (2019). 863 52. Pühringer, T. et al. Structure of the human core transcription- export complex 864 reveals a hub for multivalent interactions. Elife 9, 1- 21 (2020). 865 53. Schuller, S. K. et al. Structural insights into the nucleic acid remodeling 866 mechanisms of the yeast tho- SUB2 complex. Elife 9, 1- 51 (2020). 867 54. Saguez, C. et al. Mutational analysis of the yeast RNA helicase Sub2p reveals 868 conserved domains required for growth, mRNA export, and genomic stability. 869 Rna 19, 1363- 1371 (2013). 870 55. Sugiura, T., Sakurai, K. & Nagano, Y. Intracellular characterization of DDX39, a 871 novel growth- associated RNA helicase. Exp. Cell Res. 313, 782- 790 (2007). 872 56. Taniguchi, I. & Ohno, M. ATP- Dependent Recruitment of Export Factor 873 Aly/REF onto Intronless mRNAs by RNA Helicase UAP56. Mol. Cell. Biol. 28, 601- 608 (2008). 875 57. Siddiqui, N. & Borden, K. L. B. B. mRNA export and cancer. Wiley Interdiscip. 876 Rev. RNA 3, 13- 25 (2012). 877 58. Okamura, M. et al. Depletion of mRNA export regulator DBP5/ DDX19, GLE1 878 or IPPK that is a key enzyme for the production of IP6resulting in differentially 879 altered cytoplasmic mRNA expression and specific cell defect. PLoS One 13, 1- 880 24 (2018). 881 59. Fujiwara, N. et al. MPP6 stimulates both RRP6 and DIS3 to degrade a specified 882 subset of MTR4- sensitive substrates in the human nucleus. Nucleic Acids Res. 883 50, 8779- 8806 (2022). 884 60. Huang, D. W., Sherman, B. T. & Lempicki, R. A. Bioinformatics enrichment 885 tools: Paths toward the comprehensive functional analysis of large gene lists. 886 Nucleic Acids Res. 37, 1- 13 (2009). 887 61. Kurata, M. et al. Food- Derived Compounds Apigenin and Luteolin Modulate 888 mRNA Splicing of Introns with Weak Splice Sites. iScience 22, 336- 352 (2019). 889 62. Rocak, S., Emery, B., Tanner, N. K. & Linder, P. Characterization of the ATPase 890 and unwinding activities of the yeast DEAD- box protein Has1p and the analysis 891 of the roles of the conserved motifs. Nucleic Acids Res. 33, 999- 1009 (2005).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[75, 118, 852, 485]]<|/det|>
+892 63. Kabsch, W. XDS. Acta Crystallogr. D. Biol. Crystallogr. 66, 125- 32 (2010).893 64. Winn, M. D. et al. Overview of the CCP4 suite and current developments. Acta Crystallogr. Sect. D Biol. Crystallogr. 67, 235- 242 (2011).895 65. Liebschner, D. et al. Macromolecular structure determination using X- rays, neutrons and electrons: Recent developments in Phenix. Acta Crystallogr. Sect. D Struct. Biol. 75, 861- 877 (2019).896 66. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. Sect. D Biol. Crystallogr. 66, 486- 501 (2010).897 67. Banks, J. L. et al. Integrated Modeling Program, Applied Chemical Theory (IMPACT). J. Comput. Chem. 26, 1752- 1780 (2005).898 68. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F. & Hermans, J. Interaction Models for Water in Relation to Protein Hydration. in Intermolecular Forces 331- 342 (1981). doi:10.1007/978- 94- 015- 7658- 1_21.899 69. Andy Bunn, M. K. A language and environment for statistical computing. R Found. Stat. Comput. 10, 11- 18 (2017).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 44, 144, 69]]<|/det|>
+## Figures  
+
+<|ref|>image<|/ref|><|det|>[[44, 99, 720, 808]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 850, 115, 869]]<|/det|>
+<center>Figure 1 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 891, 945, 956]]<|/det|>
+Identification of novel apo- AREX components Immunoprecipitation using anti- DYKDDDDK (equivalent to FLAG) tag antibody beads to pull down nuclear extract of Flp- In T- REx 293 cells stably expressing FLAG tagged each protein. Each precipitated sample was separated and detected by silver staining (left) or
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 44, 960, 430]]<|/det|>
+immunoblotting (right) with the indicated antibodies. (A) The model of the apo- TREX, - AREX and the ATP- TREX complex formations of UAP56 and URH49. The dotted circles refer to apo- AREX complex components that were unidentified prior to this study. (B) FLAG- UAP56 and - URH49 differ in apo- complex formation but are similar in ATP- complex formation. In the right panel, known apo TREX, - AREX, and ATP- TREX components were detected by immunoblotting. Single asterisk represents precipitated FLAG- UAP56 or - URH49, double asterisk represents IgG light chain. (C) Identification of novel apo- AREX components by tandemimmunoprecipitation. Precipitated proteins by FLAG immunoprecipitation or tandemimmunoprecipitation (first: HA, second: FLAG) are detected respectively. Identified proteins are shown on the right side. Details of proteins identified are indicated in Extended Data Fig.2B and Table S1. The proteins shown in the red letter were further analyzed. Single, double, and triple asterisk represented precipitated FLAG- UAP56 or - URH49, HA- CIP29, and IgG light chain, respectively. (D) FLAG- URH49 is specifically associated with each novel apo- AREX component in an ATP-depleted condition, but not in the presence of ATP. (E) FLAG- ILF2 and - ILF3 bind with apo URH49 in an ATP-deficient condition. Single and double asterisks represent FLAG and endogenous- ILF2 or ILF3, respectively. (F) FLAG- HNRNPM associates with URH49 in the absence of ATP. (G) FLAG- RUVBL1 and - RUVBL2 associate with URH49 in an ATP-depleted condition. Single and double asterisks represent FLAG and endogenous- endogenous RUVBL1 or RUVBL2, respectively.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[66, 55, 787, 450]]<|/det|>  
+
+<|ref|>image<|/ref|><|det|>[[150, 512, 688, 740]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 802, 118, 820]]<|/det|>
+<center>Figure 2 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 841, 950, 955]]<|/det|>
+The novel apo- AREX components are specifically associated with URH49- mediated mRNA processing and export (A) Depletion of apo- AREX components caused nuclear poly(A)+ RNA accumulation in U2OS cells. DAPI was used to visualize the nuclei. Scale bar, \(40 \mu \mathrm{m}\) . (B) Quantification of the nuclear poly(A)+ RNA accumulation caused by the knockdown of apo- AREX components. The graph indicates the fold changes in the ratio of nuclear per cytoplasmic distribution of mRNA. These data were normalized to the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[41, 44, 952, 339]]<|/det|>
+score of the control knockdown condition. The signal intensities of bulk poly(A)+ RNA in the nucleus and the cytoplasm were quantified using ImageJ (n = 40 of each, respectively). Boxes show the median (centerline) and upper and lower quartiles. Whiskers show the lowest and highest values. Statistical analysis was performed using the Kruskal- Wallis test followed by the Steel test. \(***p < 0.001\) . (C), Localization of poly(A)+ RNA in U2OS cells. Poly(A)+ RNA localization (red) was observed under the knockdown of each apo- AREX component. Anti- SRRM2 antibody was used to stain the nuclear speckle (green). DAPI was used to visualize the nuclei (blue). Scale bar, 10 μm. In the right panels, signal intensities of poly(A)+ RNA and SRRM2 (same colors) were plotted between the A and B lines in the left panels. (D) Depletion of apo- AREX components resulted in the decreased expression of URH49- target mRNAs in the cytoplasm. RT qPCR was performed using the cytoplasmic RNA to compare the mRNA expression level. Values represent the relative expression of indicated mRNA normalized to PGK and the mean ± SEM of three independent experiments. Statistical analysis was performed using one- way ANOVA followed by Dunnett's test. \(**p < 0.01\) , \(***p < 0.001\) . n.s.: not significant
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[55, 55, 682, 760]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 800, 117, 819]]<|/det|>
+<center>Figure 3 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 841, 944, 954]]<|/det|>
+A single amino acid alteration between UAP56 and URH49 impacts their apo- complex formation and specific functions (A) Diagram of amino acids homology between UAP56 and URH49 and a list of chimeric mutants analyzed in this study. (B) FLAG- URH49 C223V mutant forms the apo- TREX- like complex. Immunoprecipitation was performed using anti- DYKDDDDK tag antibody beads and Flp- In T- REx 293 cells stably expressing FLAG- tagged proteins. Precipitated sample was separated and detected
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[40, 45, 951, 316]]<|/det|>
+by silver staining (left) or immunoblotting with the indicated antibodies (right). Single and double asterisks represented precipitated FLAG- UAP56 or - URH49, and IgG light chain, respectively. (C) Overexpression of chimeric mutants of FLAG- UAP56 or URH49 rescued the nuclear poly(A)+ RNA accumulation due to UAP56 or URH49 depletion. Poly(A)+ RNA (red), exogenously expressed FLAG- UAP56 or - URH49 (green), and chromosomal DNA (blue) were visualized in U2OS cells. Scale bar, \(40 \mu \mathrm{m}\) . (D) Quantification of the nuclear poly(A)+ RNA accumulation caused by each condition in (C). The fold changes in the ratio of nuclear to cytoplasmic distribution of poly(A)+ RNA are shown. These data were normalized to the score of control plasmid overexpression under the control knockdown condition. The signal intensities of bulk poly(A)+ RNA in the nucleus and the cytoplasm were quantified from at least 28 cells for each condition using ImageJ. Boxes show the median (centerline) and upper and lower quartiles. Whiskers show the lowest and highest values. Statistical analysis was performed using the Kruskal- Wallis test followed by the Steel test. \(***p < 0.001\) . n.s.: not significant
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[50, 55, 648, 747]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 801, 118, 820]]<|/det|>
+<center>Figure 4 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 840, 951, 955]]<|/det|>
+UAP56 and URH49 have different apo- structural features but similar ATP binding structural features, which were correlated with their complex formation Each purified protein was treated with trypsin. Aliquots were taken at each time point, separated by SDS- PAGE, and detected by Coomassie staining. (A, B) Full- length UAP56 and full- length URH49 had different partial digestion patterns in the absence of ADP, but similar limited proteolysis patterns upon ADP addition. (C) UAP56Δ42 and URH49Δ41 showed
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[41, 45, 954, 224]]<|/det|>
+different limited proteolysis patterns, and URH49 C223VΔ41 showed a pattern similar to apo-UAP56Δ42 but not apo-URH49Δ41 in the absence of ADP. (D) UAP56Δ42 and URH49Δ41 showed similar limited proteolysis patterns upon ADP addition. (E, F) Top: Analysis of cleavage sites by limited proteolysis. The "A1-4", "R1-4", and "Total" products obtained by limited proteolysis of UAP56Δ42 and URH49Δ41 in (C, D) were analyzed by LC-MS/MS. Relative peptide scores were obtained by dividing the detected prot-score of each peptide fragment derived from "A1-4" and "R1-4" by "Total". The start-site, the end-site, and the relative score of each peptide were described in Table S2. Bottom: Limited digestion models of UAP56 (blue) or URH49 (gray) predicted from the peptide containing within "A1-4" or "R1-4"  
+
+<|ref|>image<|/ref|><|det|>[[55, 260, 933, 836]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[42, 900, 117, 919]]<|/det|>
+<center>Figure 5 </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 44, 955, 272]]<|/det|>
+Structural comparison between UAP56, URH49, and Sub2 (A) Comparison of the structure of apo- UAP56Δ42 (1XTI) and the structural model of apo- URH49Δ41 (Fr48) which is generated by molecular dynamics analysis of the structure of URH49Δ41 (8IJU). Detail of the generation of the Fr48 model were described in Extended Data Fig.8C, D. By aligning the N-domain of both structural models using pyMOL, the difference in the angle of the C-domains was calculated. (B) Comparison of N-domain and C-domain between apo- UAP56Δ42 crystal (1XTI) and apo- URH49Δ41 model structure (Fr48). (C) Top: the C-loop of the apo- URH49Δ41 structural model (Fr48) was located as covering the ATP binding pocket of apo URH49Δ41. Bottom: the structure of ADP- UAP56Δ42 (1XTI) was overlaid to the structure of apo- UAP56Δ42 (1XTI) or the structural model of apo- URH49Δ41 (Fr48). (D) Loop structure in the C-domain of the apo- Sub2Δ59 (5SUQ) was overlaid to apo UAP56Δ42 (1XTI) or URH49Δ41 structural model (Fr48)  
+
+<|ref|>image<|/ref|><|det|>[[163, 312, 835, 805]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[42, 865, 117, 884]]<|/det|>
+<center>Figure 6 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 907, 936, 950]]<|/det|>
+Diversified Structurers and mRNA export machineries The model of a selective mRNA transcription and export machinery driven by the structural diversification from yeast Sub2 to human UAP56 and URH49.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 93, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 130, 318, 176]]<|/det|>
+- ExtendFigTableLegend.pdf- Table.pdf
+
+<--- Page Split --->

preprint/preprint__9866b1504884b2fa1c880a124ef92ec8435930a77685e3af1575260c12e23d80/images_list.json

@@ -0,0 +1,85 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1| Average spectral estimates of local land temperature over the Northern Hemisphere. a, Average spectral estimates of land air temperature from pollen-based reconstructions, instrumental data and model simulations extracted at the pollen record locations. Also shown are the spectra estimated after applying a 23-ka sinusoidal detrending (dashed, see Methods). 90% confidence intervals are given (shaded). The number of pollen records contributing to each timescale is indicated below (brown axis). b, Average spectral estimates from reconstructed annual sea-surface temperature derived from marine archives<sup>5</sup>, and from instrumental data at the corresponding locations. The pollen-based and instrumental average spectra from a are reproduced. Also shown are linear combinations of power-laws with slope \\(\\beta = 1.2\\) and white noise series (dashed green for land, dashed blue for sea and dashed grey for the white noise levels) corresponding to the energy-balance model approximation.",
+    "footnote": [],
+    "bbox": [
+      [
+        290,
+        149,
+        694,
+        647
+      ]
+    ],
+    "page_idx": 7
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2| Spatial patterns of temperature variability in instrumental data and pollen-based reconstructions. a, Map of millennial variability estimated from the temperature spectra of pollen-based reconstructions as the mean PSD over the 1000-3000 year timescale band. The PSD shown in the background are smoothed using Gaussian weights with a \\(300\\mathrm{km}\\) scale. Overlaid circles are the average of all pollen records closest to the corresponding instrumental grid point. b, Map of the multi-decadal scaling exponent \\(\\beta\\) from the spectra of instrumental temperature records fitted over the 10-60 year timescale band. c, Map of sub-decadal variability, mean PSD over the 2-10 year timescale band, estimated from the spectra of instrumental temperature records. In a-c, the circles indicate the instrumental grid points which were the closest to pollen records.",
+    "footnote": [],
+    "bbox": [
+      [
+        95,
+        150,
+        828,
+        631
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Extended_Data_Figure_7.jpg",
+    "caption": "Figure 3 | Spectral estimates of land temperature as a function of sub-decadal variability. a, Spectral estimates of the instrumental temperature ( \\(\\Delta t< 60\\) years) and of the pollen-based reconstructions ( \\(\\Delta t > 100\\) years) binned according to sub-decadal instrumental variability (See Methods and Extended Data Fig.7). Dashed lines indicate estimated instrumental multi-decadal scaling exponents \\(\\beta_{10 - 60 \\text{years}}\\) . \\(90\\%\\) confidence intervals are given (shaded). b, Relation of the instrumental sub-decadal temperature variability \\(\\mathrm{PSD}_{2 - 10 \\text{years}}\\) and the pollen-based millennial temperature variability \\(\\mathrm{PSD}_{1000 - 3000 \\text{years}}\\) . The size of the points is proportional to the errors on the millennial variability estimates. c, As in b, but between \\(\\beta_{10 - 60 \\text{years}}\\) and \\(\\mathrm{PSD}_{1000 - 3000 \\text{years}}\\) .",
+    "footnote": [],
+    "bbox": [
+      [
+        92,
+        241,
+        888,
+        650
+      ]
+    ],
+    "page_idx": 9
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 17
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2",
+    "footnote": [],
+    "bbox": [
+      [
+        44,
+        45,
+        875,
+        644
+      ]
+    ],
+    "page_idx": 18
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3",
+    "footnote": [],
+    "bbox": [
+      [
+        45,
+        92,
+        888,
+        565
+      ]
+    ],
+    "page_idx": 19
+  }
+]

preprint/preprint__9866b1504884b2fa1c880a124ef92ec8435930a77685e3af1575260c12e23d80/preprint__9866b1504884b2fa1c880a124ef92ec8435930a77685e3af1575260c12e23d80.mmd

@@ -0,0 +1,362 @@
+
+# Land temperature variability driven by oceans at millennial timescales  
+
+Raphael Hebert ( Raphael.hebert@awi.de )  
+
+Alfred Wegener Institut Helmholtz Centre for Polar and Marine Research https://orcid.org/0000- 0002- 9869- 4658  
+
+Ulrike Herzschuh  
+
+Alfred Wegener Institute for Polar and Marine Research https://orcid.org/0000- 0003- 0999- 1261  
+
+Thomas Laepple  
+
+Alfred- Wegener- Institut Helmholtz- Zentrum für Polar- und Meeresforschung  
+
+## Article  
+
+Keywords: land temperature variability, timescales  
+
+Posted Date: March 8th, 2021  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 147890/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Version of Record: A version of this preprint was published at Nature Geoscience on October 31st, 2022. See the published version at https://doi.org/10.1038/s41561- 022- 01056- 4.
+
+<--- Page Split --->
+
+## 1 Land temperature variability driven by oceans at millennial timescales  
+
+3 \*R. Hebert \(^{1,2}\) , U. Herzschuh \(^{1,2,3}\) & T. Laepple \(^{1,4}\)  
+
+4 Affiliations: 5 1 Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Potsdam, Germany. 6 2 Institute of Environmental Sciences and Geography, University of Potsdam, Germany 7 3 Institute of Biochemistry and Biology, University of Potsdam, Germany 8 4 University of Bremen, MARUM - Center for Marine Environmental Sciences and Faculty of Geosciences, 28334 Bremen, Germany 10 \*Correspondence: raphael.hebert@awi.de  
+
+11 Variations in regional temperature have widespread implications for society, but our 12 understanding of the amplitude and origin of long- term natural variability is insufficient. This 13 is especially the case for terrestrial temperature variability which is currently thought to be 14 weak over long timescales. Here, we provide the first comprehensive estimate of regional 15 temperature variability from annual to millennial timescales based on sedimentary pollen 16 records and instrumental data from the Northern Hemisphere. We show that the short- term 17 random variations are overprinted by strong ocean- driven climate variability on multi- 18 decadal and longer timescales. This may cause substantial climate change at the regional scale 19 in the coming century, contrasting the rather monotonous warming projected by climate 20 models. Due to the marine influence, regions characterized by stable oceanic climate at sub- 21 decadal timescales experience stronger long- term variability while continental regions with 22 higher sub- decadal variability show weaker long- term variability. This fundamental 23 relationship between the timescales provides unique insight into the mechanisms governing 24 slow terrestrial climate variability and sets the basis to project its amplitude.  
+
+25 Improving our ability to characterize internal and natural climate variability is pivotal to improve 26 our long- term climate projections \(^{1}\) , particularly at the regional scale \(^{2}\) . The statistics of the natural 27 variability can be quantified up to decadal timescales using instrumental observations, but indirect 28 climate proxies are needed for the longer timescales. 29 Proxy- model comparisons aiming to validate climate models generally focus on whether long term 30 (forced) trends found in proxies can be reproduced by Global Climate Models (GCMs) \(^{3,4}\) . In this 31 regard, land pollen- based climate reconstructions in northern mid- latitudes were shown to be 32 consistent with the forcing and simulated warming \(^{5}\) . However, the long- term multi- millennial trend 33 is a rather limited measure which does not evaluate the ability of climate models to simulate 34 realistic climate variability, both forced and internal, at all timescales. To evaluate climate 35 variability across timescales, we consider the power spectral density (PSD, or simply spectrum) 36 S(Δt), which provides an estimate of how variance is distributed with frequency, or equivalently 37 with timescale Δt. Mitchell (1976) \(^{6}\) provided the classical view of climate variability being 38 dominated by specific quasi- periodic processes such as solar variations linked to orbital cycles over 39 a mostly random and uncorrelated (i.e. white) noise background. Since then, an increasing number 40 of climate proxies have shown a contrasting view of a variance that continuously increases with 41 increasing timescale. Such a behaviour is often well approximated by a power- law scaling 42 relationship \(^{7,8}\) and can be summarized by the scaling exponent β such that \(S(\Delta t) \sim \Delta t^{\beta}\) .
+
+<--- Page Split --->
+
+Several studies have suggested that GCMs can simulate realistic climate variability for global mean temperature across timescales \(^{9,10}\) which is dominated by the response to external forcing. However, at the regional scale, where only a small fraction of the variability is forced \(^{11}\) , the skill of GCMs has been called into question for slow timescales \(^{12,13}\) . Marine proxies suggest a strong scaling of regional sea- surface temperature (SST) variance from annual to millennial timescales ( \(\beta \approx 1\) ) \(^{7}\) which contrasts with the weak scaling found in GCMs ( \(\beta \approx 0.1\) ) \(^{14,15}\) . This leads to an increasing discrepancy between reconstructed SST and model simulations at longer timescales, reaching two orders of magnitude in variance at the millennial timescale \(^{7}\) .  
+
+On land, instrumental temperatures show a fundamentally different behaviour compared to the oceans, with only a weak timescale dependency up to multi- decadal scales ( \(\beta \approx 0.1\) ) \(^{14 - 16}\) ; this low- frequency weather has been termed “macroweather regime” and starts at sub- monthly timescales \(^{17}\) . This difference can be explained by the much smaller heat capacity of land surfaces compared to the oceanic mixed layer \(^{18}\) . If the macroweather- type behaviour over land would hold on to longer timescales, internal variability would only play a minor role on the uncertainty of regional climate projections \(^{19}\) . However, a different scaling behaviour of ocean and land at longer timescales would imply an increasingly large variability discrepancy between terrestrial and marine regions which seems physically implausible and in contradiction with both diffusive energy balance models \(^{20}\) and GCMs \(^{15}\) . This leaves us with the conundrum that we must either reject altogether the marine proxies or see a fundamental change of variability in the terrestrial domain on longer timescales.  
+
+Given the limited instrumental record, this fundamental question on the scaling of variability between land and ocean cannot be answered from observations without terrestrial proxy data. Holocene Greenland ice- core \(\delta^{180}\) timeseries suggest that the weak scaling behaviour found in the instrumental data extends to millennial timescales (akin to white noise with \(\beta \approx 0\) ) \(^{21}\) . However, it is unclear whether the climate variability derived from the Greenland ice- cores is representative for other terrestrial regions \(^{22}\) and to what extent the proxy variability reflects temperature variability \(^{23}\) . In order to elucidate the behaviour of climate variability over land at timescales longer than centennial, we compiled and analysed an extensive collection of recent Holocene Northern Hemisphere pollen data covering the past 8000 years, providing the largest spatial coverage of any land- based proxy at millennial timescales. This compilation of pollen data comprises 985 unique records from the northern hemisphere, 363 of which from North America, 487 from Europe, and 135 from Asia including the most complete dataset for China \(^{24}\) and Siberia \(^{25}\) . We produced summer (June- July- August) temperature reconstructions (hereafter simply termed temperature) as summer temperatures are usually well correlated with other variables driving vegetation growth \(^{26}\) . Our results are not sensitive on the choice of summer temperature (Extended Data Fig. 1) or a potential influence of precipitation (Extended Data Fig. 2). Pollen- based reconstructions rely on the assumption of dynamical equilibrium between climate and vegetation. This assumption is timescale dependent and is generally more valid at longer timescales \(^{26}\) . In this respect, millennial scale estimates of temperature variability should thus be particularly robust. In addition, the comparison with instrumental and tree- ring data (Extended Data Figure 3) suggests that temperature variability derived from pollen assemblages also reliably captures faster timescales. Finally, we verified that the estimates of millennial variability were not systematically biased by human influences (see Extended Data Fig. 4).
+
+<--- Page Split --->
+
+Pollen- based reconstructions, once coupled with instrumental data \(^{27}\) , enable us to establish for the first time a comprehensive picture of regional land temperature variability from inter- annual to millennial timescales (Fig. 1). The spectrum of instrumental air temperature shows a rather flat scaling behaviour that is characteristic of the macroweather regime. At longer timescales, however, the pollen- based reconstructions show a strong increase in variability with increasing timescale (Fig. 1a). This suggests that even in the relatively stable recent Holocene, there exists significant centennial to millennial scale temperature variability. This behaviour clearly differs from the rather flat macroweather regime and resembles power- law scaling with an exponent \(\beta\) near unity for timescales longer than centennial. Traces of this increased scaling behaviour at multi- decadal timescales already appear in the instrumental record and is corroborated by dendrochronological timeseries (Extended Data Fig. 3).  
+
+We benchmark our result against three transient climate models simulations of the recent Holocene: IPSL \(^{28}\) , ECHAM5 \(^{29}\) and the last 8000 years of the TraCE- 21ka deglaciation experiment \(^{30}\) (i.e. after the last freshwater forcing events). The transient simulations of these GCMs show the weak scaling behaviour characteristic of the macroweather regime, although with higher annual to decadal variability relative to the instrumental data as previously recognized \(^{15}\) . However, they fail to capture the increase in variability observed in the reconstructions at multi- decadal timescales. Instead, they show a relatively weak increase at multi- centennial timescales and a sharp increase at millennial timescales due to spectral leaking from the orbital 23- ka precession cycle (Fig. 1a, Extended Data Fig. 5). As a result of this divergence in variability scaling there is an increasing deficit in temperature variability observed in the simulations compared to the reconstructions. The range of variance ratios between the reconstructions and the different model simulations increases from 7- 10 over 100- 300 years to 17- 56 (19- 102 with orbital detrending) over 1000- 3000 years, resembling the discrepancy between models and proxy data for regional SST variability \(^{7}\) .  
+
+Comparing the reconstructions over land with estimates of marine variability \(^{7}\) shows they have a very similar low- frequency behaviour (Fig. 1b). This supports the view that both components will vary more coherently as climatic variability becomes a global phenomenon over longer timescales \(^{20}\) also indicated by coherent land and ocean average temperatures \(^{5}\) . Energy- balance models suggest that this parallel behaviour of land and oceans on long timescales is due to heat exchange between the land and ocean compartments. In such models, land air temperature can be described as a linear combination of the SST and a time- dependent forcing over land \(^{31}\) ; the resulting variability spectrum over land is then a linear combination of the spectra of each term (Fig. 1b) when the two are uncorrelated (see Supplementary Information). In this framework, the change in scaling behaviour can be regarded as a transition from the macroweather regime at shorter timescales, dominated by a weakly scaling forcing component akin to white noise over land \(^{18}\) , to an oceanic regime dominated by the SST component at timescales longer than decadal. Interestingly, the parallel behaviour between land and ocean temperature spectra on multi- decadal to millennial timescales provides no evidence for additional terrestrial slow climate feedbacks. The oceanic component present in land temperature variability appears amplified by a factor of \(\sim 4\) in PSD or \(\sim 2\) in amplitude. This factor is similar to the land- sea warming contrast \(^{32}\) observed during the last century \(^{33}\) and within the range of land- sea warming ratios measured in GCMs \(^{34}\) . This is thought to be the result of local feedbacks, for example the evaporation feedback, when moisture availability over land limits evaporative cooling in comparison with marine regions \(^{35}\) , and also because of the asymmetry in the land- ocean heat exchange which favours land due to its lower specific humidity \(^{34}\) .
+
+<--- Page Split --->
+
+The extensive spatial coverage of the pollen- based reconstructions allows us to perform a spatial analysis of the millennial scale temperature variability \(\mathrm{PSD}_{1000 - 3000\mathrm{years}}\) (the mean PSD over 1000- 3000 year) (Fig. 2a) and investigate the potential link to oceanic influence. The spatial coherency (Moran's \(\mathrm{I} = 0.19\) , \(\mathrm{p}< 0.001\) , see Methods) demonstrates that the variability estimates are not drowned out by local noise. Over Europe, with its large number of records, millennial scale variability decreases inland along the path of prevailing winds blowing from the Atlantic Ocean, and is lowest over Fennoscandia where blocking events are most frequent<sup>36</sup>. Similarly, China's high millennial variability would be linked to the persistent oceanic influence carried by the dominant easterlies at that latitude, while further north in eastern Siberia the dominant westerlies bring little oceanic influence. This further suggests that higher millennial variability relies on higher connectivity to oceans, as implied by energy- balance models, although compounded with local sensitivity. The high variability in central Asia remains an outlier given the strong continentality there, but the significance is lower because of the sparseness of records. It is also possible that the lower connectivity to oceans is compensated by the stronger local climate sensitivity<sup>33</sup> which may be linked to hydrological feedback due to the arid conditions<sup>35</sup>. Meanwhile, in North America the lowest millennial variability is found in the prairies, near the centre of the continent, where the westerlies predominantly blow from the north- west and the oceanic influence is lowest.  
+
+We use the instrumental data to study the mechanisms governing the spatial distribution of the millennial variability and the continuum of variability. The scaling of variability in instrumental data has already been shown to be related to the strength of the annual cycle and of the sub- decadal variability<sup>8</sup>. If we aggregate the instrumental data and reconstructions based on the sub- decadal variability \(\mathrm{PSD}_{2 - 10\mathrm{years}}\) (the mean PSD over 2- 10 years; Fig. 3, Extended Data Fig. 6, Extended Data Fig. 7), a clear relationship appears with the emergence of the low- frequency regime, quantified by the multi- decadal scaling exponent \(\beta_{10 - 60\mathrm{years}}\) ( \(\beta\) regressed over 10- 60 years): locations with lower sub- decadal variability thus show a stronger increase of variability towards longer timescales, as indicated by higher multi- decadal scaling (Fig. 2b,c, Fig. 3, Extended Data Fig. 7). We should thus expect an inversion where regions of low (high) sub- decadal variability, typically characterized by more maritime (continental) influences, would become regions of high (low) variability at long timescales.  
+
+Indeed, this hypothesized relationship is confirmed by the pollen- based reconstructions. Their estimates of millennial temperature variability \(\mathrm{PSD}_{1000 - 3000\mathrm{years}}\) show a strong anti- correlation with the sub- decadal variability \(\mathrm{PSD}_{2 - 10\mathrm{years}}\) (r=- 0.95, \(\mathrm{p}< 0.01\) ) and a strong correlation with the multi- decadal scaling \(\beta_{10 - 60\mathrm{years}}\) (r=0.91, \(\mathrm{p}< 0.01\) ). These significant strong relationships between the pollen- based reconstructions and independent instrumental temperature data demonstrate a fundamental link of temperature variability from sub- decadal to millennial timescales. The spatial pattern of the variability (Fig. 2 and Extended Data Fig. 5) further suggests that this relationship is caused by varying marine influence. In addition, this can explain the relation of the amplitude of the annual cycle (an indicator for continentality) with the inter- annual variability scaling in the instrumental relationship proposed by Huybers and Curry (2006)<sup>8</sup>. Therefore, our findings complete the linkage between seasonal and millennial land temperature variability.
+
+<--- Page Split --->
+
+Our results indicate that current GCMs underestimate regional temperature variability over land at timescales longer than multi-decadal (Fig. 1a, Extended Data Fig. 5). In combination with the spatial pattern of variability (Fig. 2), this suggests that the deficit in low- frequency variability is related to an underestimation of marine variability<sup>7</sup>. The interpretation of climate sensitive proxies remains an area of active research, and in principle, it remains possible that the observed model- data mismatches stem from non- climatic variability. However, several lines of evidence argue against this interpretation. Firstly, there are no known archival processes yet which could artificially create such power- law scaling in sedimentary archives<sup>37</sup>. Specifically, the known processes such as counting errors, spatial or temporal aliasing, and bioturbation in the sediment cannot explain the power spectra of variability found here. Secondly, the consistency between independent marine and terrestrial archives (Fig. 1b) provides further support for the temperature variability reconstruction. Finally, the spatial relationship with the independent instrumental temperature data (Fig. 3) also indicates that this is no artefact from the proxy data.  
+
+Thus, pollen- based reconstructions support the paradigm of an increasing continuum of climate variability with increasing timescales<sup>8</sup>, in contradiction with the local temperature variability in current GCM simulations and the classical picture of Mitchell<sup>6</sup>. More importantly, our results extend previous findings from instrumental data<sup>8</sup> and demonstrate a fundamental link between interannual, multi- decadal and millennial timescales driven by the interaction of marine and terrestrial temperature variability modulated by continentality.  
+
+This fundamental behaviour of temperature variability has implications for the relative impact of natural and anthropogenically forced variability. High latitude regions characterised by high interannual variability show a weaker oceanic regime and, ultimately, less natural variability on long timescales. As these regions are also highly sensitive to anthropogenic forcing, the impact of anthropogenic warming, relative to natural variability, will be greater. However, regions of strong maritime influence, where most of the world's population is located, could see large natural variability, that is not covered by current GCM projections which tend to display monotonous warming<sup>33</sup>. It is thus possible that until now, the stronger natural variability at multi- decadal timescales in maritime regions has partly overshadowed the anthropogenic warming in those regions which could explain their lower observational transient climate sensitivity<sup>33</sup>. Integrative archives such as glaciers should be particularly sensitive to this increased memory<sup>16</sup> and could be used to verify our findings. Large compilations of climate archives have the potential to inform us on the spatial patterns of slow variability and their underlying causes, and further studies combining multiple proxies over land and ocean show great promise to improve our understanding of the spatio- temporal correlation structure of climate variability.  
+
+## Author Contributions  
+
+R.H., T.L. and U.H. designed the research. U.H. performed the pollen- based reconstruction. T.L. contributed the interpretation of the marine archives. R.H. and T.L. developed the methodology. R.H. performed the data analysis and wrote the first draft of the manuscript. R.H., T.L., and U.H. contributed to the interpretation and to the preparation of the final manuscript.
+
+<--- Page Split --->
+
+## Acknowledgements  
+
+AcknowledgementsThis is a contribution to the SPACE ERC and GLACIAL LEGACY ERC projects; these projects have received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement no. 716092 and no. 772852). This work is further a contribution to the PALMOD project. The work profited from discussions at the CVAS working group of the Past Global Changes (PAGES) programme. We thank A. Dallmeyer, A. Dolman, I. Kröner, T. Kunz, S. Lovejoy and K. Rehfeld for useful discussion. We acknowledge P. Huybers for comments on the manuscript. We thank P. Braconnot and J. Cretat, and J. Jungclaus for, respectively, providing the IPSL and ECHAM5 simulations, and T. Böhmer and X. Cao for their support compiling the database. We thank all original data contributors who made their proxy data available and acknowledge the Neotoma Palaeoecology Database.  
+
+## Competing interests  
+
+The authors R.H., U.H. and T.L. declare no competing interests.  
+
+## Materials & Correspondence  
+
+Correspondence and materials request should be addressed to R.H.
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1| Average spectral estimates of local land temperature over the Northern Hemisphere. a, Average spectral estimates of land air temperature from pollen-based reconstructions, instrumental data and model simulations extracted at the pollen record locations. Also shown are the spectra estimated after applying a 23-ka sinusoidal detrending (dashed, see Methods). 90% confidence intervals are given (shaded). The number of pollen records contributing to each timescale is indicated below (brown axis). b, Average spectral estimates from reconstructed annual sea-surface temperature derived from marine archives<sup>5</sup>, and from instrumental data at the corresponding locations. The pollen-based and instrumental average spectra from a are reproduced. Also shown are linear combinations of power-laws with slope \(\beta = 1.2\) and white noise series (dashed green for land, dashed blue for sea and dashed grey for the white noise levels) corresponding to the energy-balance model approximation. </center>
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2| Spatial patterns of temperature variability in instrumental data and pollen-based reconstructions. a, Map of millennial variability estimated from the temperature spectra of pollen-based reconstructions as the mean PSD over the 1000-3000 year timescale band. The PSD shown in the background are smoothed using Gaussian weights with a \(300\mathrm{km}\) scale. Overlaid circles are the average of all pollen records closest to the corresponding instrumental grid point. b, Map of the multi-decadal scaling exponent \(\beta\) from the spectra of instrumental temperature records fitted over the 10-60 year timescale band. c, Map of sub-decadal variability, mean PSD over the 2-10 year timescale band, estimated from the spectra of instrumental temperature records. In a-c, the circles indicate the instrumental grid points which were the closest to pollen records. </center>
+
+<--- Page Split --->
+![](images/Extended_Data_Figure_7.jpg)
+
+<center>Figure 3 | Spectral estimates of land temperature as a function of sub-decadal variability. a, Spectral estimates of the instrumental temperature ( \(\Delta t< 60\) years) and of the pollen-based reconstructions ( \(\Delta t > 100\) years) binned according to sub-decadal instrumental variability (See Methods and Extended Data Fig.7). Dashed lines indicate estimated instrumental multi-decadal scaling exponents \(\beta_{10 - 60 \text{years}}\) . \(90\%\) confidence intervals are given (shaded). b, Relation of the instrumental sub-decadal temperature variability \(\mathrm{PSD}_{2 - 10 \text{years}}\) and the pollen-based millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \text{years}}\) . The size of the points is proportional to the errors on the millennial variability estimates. c, As in b, but between \(\beta_{10 - 60 \text{years}}\) and \(\mathrm{PSD}_{1000 - 3000 \text{years}}\) . </center>
+
+<--- Page Split --->
+
+## Methods  
+
+## Reconstructions  
+
+The modern pollen dataset used for calibration consists of 15,532 sampling sites. For each fossil record location, we selected a unique subset of modern sites within a \(2000\mathrm{km}\) radius in order to increase the reliability of the reconstructions and avoid a bi- modal climate optimum<sup>38</sup>. The fossil dataset includes taxonomically harmonized fossil pollen data from North America and Europe obtained via the Neotoma Palaeoecological Database<sup>39</sup>, and from Asia combining Cao et al. (2020)<sup>25</sup> and Cao et al. (2014)<sup>24</sup>. A fossil database comprising 985 records was compiled based on the requirement that the resulting spectral estimates covered timescales at least one fifth of an order of magnitude, i.e. 0.2 in a base 10 logarithm, below one third the length (to avoid the well- known multitaper low- bias at long timescales)<sup>40</sup>. The European and Asian datasets were combined keeping the 70 most common taxa according to Hill's second number.  
+
+The modern climate data used for calibration was the average of the June, July and August climatologies, i.e. the summer temperature, for the years from 1970- 2000 as obtained from WorldClim 2.1<sup>41</sup>. The Weighted Averaging Partial Least Square (WAPLS)<sup>42</sup> method was used to calibrate transfer functions relating the pollen- assemblages to the summer temperature, with leave- one- out cross validation. The pollen percentages were square- root transformed to decrease the dominance of abundant taxa with high productivity. The number of retained WAPLS components was selected using a randomization t- test. The same method was also applied to reconstruct annual mean temperature (Extended Data Fig. 1) and annual precipitation (Extended Data Fig. 2), also using the climatologies from WorldClim 2.1<sup>41</sup> for the calibration of each variable.  
+
+## Significance Testing  
+
+Following Telford and Birks (2011)<sup>43</sup>, we tested the summer temperature reconstructions for significance using the R package "palaeoSig" and found that 228 out of 985 reconstructions were significant (p<0.1). However, this significance test is rather conservative and several reasons can create type II errors (false negatives), including for example a low diversity of taxa, a small number of sub- fossil observations, an input climate signal that is less variable, or an inadequate training set<sup>43</sup>. A higher p- values therefore does not necessarily mean that the summer temperature has not been recorded, but rather that the information is insufficient to confirm it. Thus, instead of unduly discarding most records, we decided to include all records in the main analysis. We show that our conclusions are robust and continue to hold even if we restrict our analysis to the 'significant locations' only (p<0.1) which yielded similar results to the 'not significant locations' (p>0.1) (Extended Data Fig. 8).  
+
+## Testing for Anthropogenic Impacts  
+
+We considered all series covering entirely the last 8- ka showed and defined two temporal windows: the more recent (0- 4ka) and the more distant (4ka- 8ka) past. The 1000- 2000 year timescale band was taken to calculate the variance ratio between the two 4ka windows. We only included those series in our analyses whose spectral estimates covered at least one tenth of an order of magnitude (on a base 10 logarithmic scale) for both time periods; a criterion that was met by 344 records. We
+
+<--- Page Split --->
+
+found no systematic variance increase in the more recent half of the series as the mean of the distribution of the logarithm of the variance ratios did not significantly differ from zero (p>0.1). In fact, the more recent period, where human impacts may have contributed to an increased variability, is about \(6\%\) less variable than the earlier one. Likewise, the spatial distribution (Extended Data Fig. 4) did not show any obvious spatial patterns that could be related to human occupation, displaying a non- significant Moran's I of 0.014 (p>0.1; see Methods). If human occupation was the dominant driver of millennial scale variability, we would have expected to observe an increase in variability over both Europe and China, where human occupation has been increasing the most over the last 4000 years compared to the preceding 4000 years. We thus conclude that human impacts on vegetation did not have a significant enough impact on the slow variability to systematically bias millennial scale variability estimates.  
+
+## Instrumental Data  
+
+For the instrumental dataset we used the Berkeley Earth Surface Temperature (BEST) land and ocean product covering the period from 1850- 2020. The equal area product is used for calculations, while the regular \(1^{\circ} \times 1^{\circ}\) product was used for visual display. The instrumental data was detrended from anthropogenic influences to a first- order component<sup>44</sup> proportional to historical timeseries of doublings in atmospheric carbon dioxide concentration<sup>45</sup>. There are 403 grid points which comprise pollen records (circles on Fig. 2). Using alternative instrumental datasets<sup>46</sup> leads to similar conclusions.  
+
+## Model Data  
+
+Three model simulations of the recent Holocene were considered, namely the IPSL<sup>28</sup>, the ECHAM5<sup>29</sup> and the CCSM3<sup>47</sup>. The first two are recent Holocene transient simulations of the past 6,000 years, and the latter is the TraCE- 21ka deglaciation experiment<sup>30</sup>. We only retained the last 8,000 years of TraCE- 21ka since it is comparable to the recent Holocene transient simulations as they contain no more freshwater forcing events which were the main drivers of the deglaciation. We selected the average 2- meter summer air temperature (June- July- August) and averaged it annually.  
+
+Since the long- term trends in summer temperature are linked to the precession in the Earth's orbit, we also analysed the timeseries after detrending for a 23- ka sinusoid rather than use the standard linear detrending performed before computing spectral estimates. This approach attempts to minimize power leakage from the orbital forcing frequencies onto the observed frequencies (Fig. 1a, dashed lines). The reduction in leaked power was not nearly as important in the case of the pollen- based reconstructions (Fig. 1a, Extended Data Fig. 5).  
+
+## Spectral Estimates  
+
+The power spectral density estimates were calculated using the multitaper method<sup>48</sup>, adapted for irregular sampling through linear interpolation<sup>49</sup>, with the number of tapers \(n_{\text{tapers}} = 3\) and the time- bandwidth parameter \(\omega = 2\) , which yield up to \(n_{\text{tapers}} * \omega = 6\) degrees of freedom for the individual spectral estimates. Only timescales greater than twice the maximal resolution were kept to minimize power loss due to the interpolation<sup>37</sup>.  
+
+The confidence intervals were derived from the chi- squared distribution \((\chi^2)\) of the multitaper estimates. The degrees of freedom of the \(\chi^2\) were limited to a maximum based on the expected
+
+<--- Page Split --->
+
+effective spatial degrees of freedom at a given timescale in order to avoid obtaining over- confident estimates. The approximate relationship assumed for the maximum degrees of freedom \(\mathrm{v}_{\mathrm{max}}\) as a function of timescale is: \(\mathrm{v}_{\mathrm{max}} = 40 \mathrm{At}^{- 0.3}\) . This corresponds to about 20, 10, and 5 degrees of freedom at the decadal, centennial and millennial timescales respectively<sup>37</sup>. They were multiplied by six, the number of degrees of freedom for the multitaper estimates with three independent tapers, and further modulated by a factor of spatial representativity \(\mathrm{f}_{\Delta \mathrm{t}}\) calculated as the fraction of the land area represented by the spatial distribution of the data.  
+
+For each point over land we calculated the effective number of records \(\mathrm{N}_{\mathrm{Eff}}\) on a regular \(1^{\circ}\) by \(1^{\circ}\) degree grid using a Gaussian kernel:  
+
+\[N_{\mathrm{Eff}} = \sum_{i = 1}^{n}\mathrm{e}^{\frac{-d_{i,j}}{2\sigma^{2}}}\]  
+
+where \(\mathrm{d}_{\mathrm{i,j}}\) is the geographical distance between the \(\mathrm{i}^{\mathrm{th}}\) data record (out of n records with a PSD estimate at the given timescale \(\Delta \mathrm{t}\) ) and the \(\mathrm{j}^{\mathrm{th}}\) grid point where \(\mathrm{N}_{\mathrm{Eff}}\) is calculated, and \(\sigma\) is a characteristic decorrelation scale which we took as \(\sigma = 300 \mathrm{km}\) . \(\mathrm{N}_{\mathrm{Eff}}\) was also used for Fig. 2a in order to modulate the opacity of the background as a function of nearby records. The factor of spatial representativity for a given timescale was then calculated as:  
+
+\[f_{\Delta t} = \frac{\sum_{j = 1}^{m}\min \left(N_{\mathrm{Eff},j},1\right)\cos(\theta_{j})}{\sum_{j = 1}^{m}\cos(\theta_{j})}\]  
+
+where m is the number of grid points covering the land area north of the southernmost pollen record.  
+
+## 319 Variance Ratios  
+
+Variance ratios were computed by taking the ratio between the mean PSD over the same timescale band between different series after interpolating in the spectral domain. Since the ratio of two \(\chi^2\) distributed variables follows an F- distribution, the ratios were multiplied by (d- 2) \(\mathrm{d}^{- 1}\) where d is the number of degrees of freedom of the denominator<sup>22</sup>.  
+
+## 324 Sub-Decadal Variability Binning  
+
+The data was aggregated based on the mean sub- decadal variability \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) , defined as the mean PSD over the 2- 10 year timescale band. We calculated \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) for each of the 403 instrumental grid point for which pollen records were present nearby, ordered the results, and split them into eight non- overlapping bins (Extended Data Fig. 7). Each pollen record was assigned to the nearest instrumental grid point and averaged in the spectral domain. Varying the number of bins, for example using twenty bins instead of eight, leads to similar correlations (Extended Data Fig. 7).  
+
+## 331 Correlation  
+
+The Pearson's correlation was calculated between the instrumental sub- decadal variability \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) , the instrumental multi- decadal scaling exponent \(\beta_{10 - 60 \mathrm{years}}\) , and the pollen- based millennial variability \(\mathrm{PSD}_{1,000 - 3,000 \mathrm{years}}\) after binning the instrumental grid points near pollen records into 8
+
+<--- Page Split --->
+
+335 subsets according to the PSD2- 10 years amplitudes (see previous section). The standard errors on the 336 millennial variability estimates, i.e. \(\sqrt{2DoF}\) , where DoF is the total degrees of freedom for the 337 PSD estimate, were used as weights for the correlation calculation and for visual representation in 338 Fig. 3b,c.  
+
+## 339 Moran's I  
+
+Moran's I Moran's I spatial autocorrelation index was calculated using the method from Gittleman and Kot (1990) as implemented in the R- package "ape"51. The weight matrix used corresponds to the inverse of the distance between sites.  
+
+## 343 Code and Data Availability  
+
+All relevant code and data is available to the editor and the reviewers of this article upon request, and will be made publicly available in a GitHub repository upon publication.  
+
+## 346 References  
+
+1. Thompson, D. W. J., Barnes, E. A., Deser, C., Foust, W. E. & Phillips, A. S. Quantifying the Role of Internal Climate Variability in Future Climate Trends. J. Clim. 28, 6443-6456 (2015).  
+2. McKinnon, K. A. & Deser, C. Internal Variability and Regional Climate Trends in an Observational Large Ensemble. J. Clim. 31, 6783-6802 (2018).  
+3. Marcott, S. A., Shakun, J. D., Clark, P. U. & Mix, A. C. A Reconstruction of Regional and Global Temperature for the Past 11,300 Years. Science 339, 1198-1201 (2013).  
+4. Roulson, C. C. et al. Mid-latitude net precipitation decreased with Arctic warming during the Holocene. Nature 568, 83-87 (2019).  
+5. Marsicek, J., Shuman, B. N., Bartlein, P. J., Shafer, S. L. & Brewer, S. Reconciling divergent trends and millennial variations in Holocene temperatures. Nature 554, 92-96 (2018).  
+6. Mitchell, J. M. & others. An overview of climatic variability and its causal mechanisms. Quart. Res. 6, 481-493 (1976).  
+7. Laepple, T. & Huybers, P. Ocean surface temperature variability: Large model-data differences at decadal and longer periods. Proc. Natl. Acad. Sci. 111, 16682-16687 (2014).  
+8. Huybers, P. & Curry, W. Links between annual, Milankovitch and continuum temperature variability. Nature 441, 329-332 (2006).  
+9. Zhu, F. et al. Climate models can correctly simulate the continuum of global-average temperature variability. Proc. Natl. Acad. Sci. 116, 8728-8733 (2019).  
+10. Crowley, T. J. Causes of Climate Change Over the Past 1000 Years. Science 289, 270-277 (2000).  
+11. Goosse, H., Renssen, H., Timmermann, A. & Bradley, R. S. Internal and forced climate variability during the last millennium: a model-data comparison using ensemble simulations. Quart. Sci. Rev. 24, 1345-1360 (2005).
+
+<--- Page Split --->
+
+12. Cheung, A. H. et al. Comparison of Low-Frequency Internal Climate Variability in CMIP5 Models and Observations. J. Clim. 30, 4763-4776 (2017).  
+
+13. Dee, S. G. et al. Improved spectral comparisons of paleoclimate models and observations via proxy system modeling: Implications for multi-decadal variability. Earth Planet. Sci. Lett. 476, 34-46 (2017).  
+
+14. Fraedrich, K. & Blender, R. Scaling of Atmosphere and Ocean Temperature Correlations in Observations and Climate Models. Phys. Rev. Lett. 90, (2003).  
+
+15. Fredriksen, H.-B. & Rypdal, K. Spectral Characteristics of Instrumental and Climate Model Surface Temperatures. J. Clim. 29, 1253-1268 (2015).  
+
+16. Roe, G. H. What do glaciers tell us about climate variability and climate change? J. Glaciol. 57, 567-578 (2011).  
+
+17. Lovejoy, S. & Schertzer, D. Low-Frequency Weather and the Emergence of the Climate. in Extreme Events and Natural Hazards: The Complexity Perspective 231-254 (American Geophysical Union (AGU), 2013). doi:10.1029/2011GM001087.  
+
+18. Hasselmann, K. Stochastic climate models Part I. Theory. Tellus 28, 473-485 (1976).  
+
+19. Hawkins, E., Sutton, R. & others. The potential to narrow uncertainty in regional climate predictions. Bull. Am. Meteorol. Soc. 90, 1095-1107 (2009).  
+
+20. North, G. R., Cahalan, R. F. & Coakley, J. A. Energy balance climate models. Rev. Geophys. 19, 91-121 (1981).  
+
+21. Nilsen, T., Rypdal, K. & Fredriksen, H.-B. Are there multiple scaling regimes in Holocene temperature records? Earth Syst. Dyn. 7, 419-439 (2016).  
+
+22. Rehfeld, K., Münch, T., Ho, S. L. & Laepple, T. Global patterns of declining temperature variability from the Last Glacial Maximum to the Holocene. Nature 554, 356-359 (2018).  
+
+23. Casado, M., Münch, T. & Laepple, T. Climatic information archived in ice cores: impact of intermittency and diffusion on the recorded isotopic signal in Antarctica. Clim. Past 16, 1581-1598 (2020).  
+
+24. Cao, X., Herzschuh, U., Telford, R. J. & Ni, J. A modern pollen-climate dataset from China and Mongolia: Assessing its potential for climate reconstruction. Rev. Palaeobot. Palynol. 211, 87-96 (2014).  
+
+25. Cao, X. et al. A taxonomically harmonized and temporally standardized fossil pollen dataset from Siberia covering the last 40&thinsp;kyr. Earth Syst. Sci. Data 12, 119-135 (2020).  
+
+26. Chevalier, M. et al. Pollen-based climate reconstruction techniques for late Quaternary studies. Earth-Sci. Rev. 210, 103384 (2020).  
+
+27. Rohde, R., Muller, R., Jacobsen, R., Perlmutter, S. & Mosher, S. Berkeley Earth Temperature Averaging Process. Geoinformatics Geostat. Overv. 01, (2013).  
+
+28. Braconnot, P., Zhu, D., Marti, O. & Servonnat, J. Strengths and challenges for transient Mid- to Late Holocene simulations with dynamical vegetation. Clim. Past 15, 997-1024 (2019).  
+
+29. Fischer, N. & Jungclaus, J. H. Holocene experiment with coupled atmosphere-ocean-model ECHAM5/MPI-OM. Supplement to: Fischer, N; Jungclaus, JH (2011): Evolution of the seasonal temperature cycle in a transient Holocene simulation: orbital forcing and sea-ice. Climate of the Past, 7, 1139-1148, https://doi.org/10.5194/cp-7-1139-2011 (2012) doi:https://doi.org/10.1594/PANGAEA.773607.  
+
+30. He, F. Simulating Transient Climate Evolution of the Last deglaciation with CCSM3. 161.
+
+<--- Page Split --->
+
+31. Geoffroy, O. et al. Transient Climate Response in a Two-Layer Energy-Balance Model. Part I: Analytical Solution and Parameter Calibration Using CMIP5 AOGCM Experiments. J. Clim. 26, 1841-1857 (2013).  
+
+32. Byrne, M. P. & O'Gorman, P. A. Link between land-ocean warming contrast and surface relative humidities in simulations with coupled climate models. Geophys. Res. Lett. 40, 5223-5227 (2013).  
+
+33. Hébert, R. & Lovejoy, S. Regional Climate Sensitivity- and Historical-Based Projections to 2100. Geophys. Res. Lett. 45, 4248-4254 (2018).  
+
+34. Geoffroy, O., Saint-Martin, D. & Voldoire, A. Land-sea warming contrast: the role of the horizontal energy transport. Clim. Dyn. 45, 3493-3511 (2015).  
+
+35. Sejas, S. A., Albert, O. S., Cai, M. & Deng, Y. Feedback attribution of the land-sea warming contrast in a global warming simulation of the NCAR CCSM4. Environ. Res. Lett. 9, 124005 (2014).  
+
+36. Schwander, M., Rohrer, M., Brönnimann, S. & Malik, A. Influence of solar variability on the occurrence of central European weather types from 1763 to 2009. Clim. Past 13, 1199-1212 (2017).  
+
+37. Kunz, T., Dolman, A. M. & Laepple, T. A spectral approach to estimating the timescale-dependent uncertainty of paleoclimate records - Part 1: Theoretical concept. Clim. Past 16, 1469-1492 (2020).  
+
+38. Cao, X., Ni, J., Herzschuh, U., Wang, Y. & Zhao, Y. A late Quaternary pollen dataset from eastern continental Asia for vegetation and climate reconstructions: Set up and evaluation. Rev. Palaeobot. Palynol. 194, 21-37 (2013).  
+
+39. Goring, S. et al. neotoma: A Programmatic Interface to the Neotoma Paleoecological Database. Open Quat. 1, (2015).  
+
+40. Prieto, G., Parker, R., Thomson, D., Vernon, F. & Graham, R. Reducing the bias of multitaper spectrum estimates. Geophys. J. Int. 171, 1269-1281 (2007).  
+
+41. Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302-4315 (2017).  
+
+42. Juggins, S. & Birks, H. J. B. Quantitative Environmental Reconstructions from Biological Data. in Tracking Environmental Change Using Lake Sediments (eds. Birks, H. J. B., Lotter, A. F., Juggins, S. & Smol, J. P.) vol. 5 431-494 (Springer Netherlands, 2012).  
+
+43. Telford, R. J. & Birks, H. J. B. A novel method for assessing the statistical significance of quantitative reconstructions inferred from biotic assemblages. Quat. Sci. Rev. 30, 1272-1278 (2011).  
+
+44. Wu, T., Hu, A., Gao, F., Zhang, J. & Meehl, G. A. New insights into natural variability and anthropogenic forcing of global/regional climate evolution. Npj Clim. Atmospheric Sci. 2, 1-13 (2019).  
+
+45. Meinshausen, M. et al. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Clim. Change 109, 213 (2011).  
+
+46. Cowtan, K. & Way, R. G. Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends. Q. J. R. Meteorol. Soc. n/a-n/a (2013) doi:10.1002/qj.2297.  
+
+47. Yeager, S. G., Shields, C. A., Large, W. G. & Hack, J. J. The Low-Resolution CCSM3. J. Clim. 19, 2545-2566 (2006).  
+
+48. Thomson, D. J. Spectrum estimation and harmonic analysis. Proc. IEEE 70, 1055-1096 (1982).
+
+<--- Page Split --->
+
+49. Laepple, T. & Huybers, P. Reconciling discrepancies between Uk37 and Mg/Ca reconstructions of Holocene marine temperature variability. Earth Planet. Sci. Lett. 375, 418–429 (2013).50. Gittleman, J. L. & Kot, M. Adaptation: Statistics and a Null Model for Estimating Phylogenetic Effects. Syst. Biol. 39, 227–241 (1990).51. Paradis, E., Claude, J. & Strimmer, K. APE: Analyses of Phylogenetics and Evolution in R language. Bioinformatics 20, 289–290 (2004).
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+  
+
+Extended Data Figure 1 | Summer and Annual Temperatures Spectra - Comparing the average spectra at the location of pollen records for the mean summer temperature (dashed) and the mean annual temperatures (solid). \(90\%\) confidence intervals are given (shaded). The IPSL and ECHAM5 model results exhibit a slightly lower variability in their annual temperature than in their summer temperature over all timescales, except for the longest timescale since it is dominated by leaked power from the Earth's orbital precession which mainly affects summer temperature in the Northern Hemisphere during the Holocene. On the other hand, TraCE- 21ka generally shows a slightly higher variability in its annual compared to its summer temperature. Although the pollen- based reconstructions calibrated for annual temperature are thought to be less reliable than the summer temperature reconstructions, they give a very similar result. This shows that our conclusions are robust against uncertainties in the seasonal attribution of pollen variability.
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+  
+
+Extended Data Figure 2 | Comparison of pollen- based precipitation reconstructions and climate models. Same as Extended Data Fig. 1, but for precipitation instead of summer temperature. While most locations should reflect temperature, here we also tested the boundary case of assuming that all sites reflect precipitation. Even in this extreme case, the main results hold, namely increasing climate variability over land as a function of timescale and a corresponding deficit of variability in the climate models. In all cases, there was little difference between the estimates with (dashed) and without (solid) sinusoidal detrending. The three climate models vastly disagree in terms of the amplitude of precipitation variability, but they all show a large deficit of variability at long timescales compared to the pollen- based reconstructions. Thus, even if precipitation would affect parts of our records, this cannot reconcile the model- proxy discrepancy.
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+  
+
+Extended Data Figure 3 |Spectral estimates from tree ring series. Spectral estimates of land temperature variability obtained from the pollen- based reconstructions and dendrochronological results from tree ring width (TRW) and maximum latewood density (MXD) measurements are shown (solid). The average instrumental spectral estimates at the corresponding locations for each dataset are shown alongside for reference (dashed). See Supplementary Information for details about the data, TRW and MXD differences, and interpretation. \(90\%\) confidence intervals are given (shaded)
+
+<--- Page Split --->
+![PLACEHOLDER_20_0]
+  
+
+Extended Data Figure 4 | Variance ratio between late and mid- Holocene temperature variability. Shown are the \(\log_{2}\) of the variance ratios \(\alpha\) (i.e. the number of doublings) for the 1000- 2000 years timescale band of the late Holocene (4ka- 0ka BP) and for the mid- Holocene (8ka- 4ka). A positive \(\log_{2}\alpha\) implies that the later period, where human impacts should be more important, was more variable.
+
+<--- Page Split --->
+![PLACEHOLDER_21_0]
+  
+
+Extended Data Figure 5| Comparison of millennial scale variability in the reconstructions and models. a- c Millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \mathrm{year}}\) (mean PSD for the timescale band 1000- 3000 years) for the pollen- based reconstructions after smoothing with a Gaussian kernel with a characteristic scale of \(300 \mathrm{km}\) (see Methods). d- I Millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \mathrm{year}}\) for the three climate models (without smoothing). The results with different detrending methods before computing the power spectra are compared: d,g,j without detrending, e,h,k with linear detrending and f,i,l with a 23- ka sinusoidal detrending. The same colour scale is used for all maps. While the typical pollen- based variability would be generally between 10 and \(1000 \mathrm{K}^2 \mathrm{year}^{- 1}\) , the models yields a variability 1- 2 orders of magnitude smaller, between 0.1 and \(10 \mathrm{K}^2 \mathrm{year}^{- 1}\) , with some regions reaching up to \(100 \mathrm{K}^2 \mathrm{year}^{- 1}\) .
+
+<--- Page Split --->
+![PLACEHOLDER_22_0]
+  
+
+Extended Data Figure 6 | Location of the records contributing to each spectral bin. a. Shown are the location of individual pollen records which were considered in the analysis. The colours correspond to those in Fig. 3, indicating which records are included in the binning of each spectrum. b. Shown are the grid points of the instrumental dataset which are near pollen records. The colours also indicate the corresponding bins as in a and Fig. 3. Each of the 8 non- overlapping bins contain 50 grid points based on the sub- decadal variability (see Methods).
+
+<--- Page Split --->
+![PLACEHOLDER_23_0]
+  
+
+Extended Data Figure 7 | Spectral estimates of land temperature as a function of sub- decadal variability using twenty bins. As in Fig.3, but using twenty bins instead of eight. While the result is noisier than in Fig. 3, the correlation is still highly significant, which shows that our result is not sensitive to the number of bins.
+
+<--- Page Split --->
+![PLACEHOLDER_24_0]
+  
+
+Extended Data Figure 8 | Comparison of pollen- based reconstructions based on their statistical significance. A total of 985 pollen records which provided spectral estimates were tested for statistical significance and separated into ‘significant locations’ (p<0.1) and ‘not significant locations’ (p>0.1). The resulting average spectra overlap and are fairly similar over a wide range of timescales. The average spectra of the instrumental data at the corresponding locations are shown alongside for reference (dashed). 90% confidence intervals are given (shaded).
+
+<--- Page Split --->
+
+## Figures  
+
+![PLACEHOLDER_25_0]
+
+<center>Figure 1 </center>  
+
+Average spectral estimates of local land temperature over the Northern Hemisphere. a, Average spectral estimates of land air temperature from pollen- based reconstructions, instrumental data and model simulations extracted at the pollen record locations. Also shown are the spectra estimated after applying a 23- ka sinusoidal detrending (dashed, see Methods). \(90\%\) confidence intervals are given (shaded). The number of pollen records contributing to each timescale is indicated below (brown axis). b, Average spectral estimates from reconstructed annual sea- surface temperature derived from marine archives5, and from instrumental data at the corresponding locations. The pollen- based and instrumental average spectra from a are reproduced. Also shown are linear combinations of power- laws with slope \(\beta = 1.2\) and white noise series (dashed green for land, dashed blue for sea and dashed grey for the white noise levels) corresponding to the energy- balance model approximation.
+
+<--- Page Split --->
+![PLACEHOLDER_26_0]
+
+<center>Figure 2 </center>  
+
+Spatial patterns of temperature variability in instrumental data and pollen- based reconstructions. a, Map of millennial variability estimated from the temperature spectra of pollen- based reconstructions as the mean PSD over the 1000- 3000 year timescale band. The PSD shown in the background are smoothed using Gaussian weights with a 300 km scale. Overlaid circles are the average of all pollen records closest to the corresponding instrumental grid point. b, Map of the multi- decadal scaling exponent \(\beta\) from the spectra of instrumental temperature records fitted over the 10- 60 year timescale band. c, Map of sub- decadal variability, mean PSD over the 2- 10 year timescale band, estimated from the spectra of instrumental temperature records. In a- c, the circles indicate the instrumental grid points which were the closest to pollen records. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning
+
+<--- Page Split --->
+
+the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  
+
+![PLACEHOLDER_27_0]
+
+<center>Figure 3 </center>  
+
+Spectral estimates of land temperature as a function of sub- decadal variability. a, Spectral estimates of the instrumental temperature \((\Delta t< 60\) years) and of the pollen- based reconstructions \((\Delta t > 100\) years) binned according to sub- decadal instrumental variability (See Methods and Extended Data Fig.7). Dashed lines indicate estimated instrumental multi- decadal scaling exponents \(\beta 10 - 60\) years. \(90\%\) confidence intervals are given (shaded). b, Relation of the instrumental sub- decadal temperature variability PSD2- 10 years and the pollen- based millennial temperature variability PSD1000- 3000 years. The size of the points is proportional to the errors on the millennial variability estimates. c, As in b, but between \(\beta 10 - 60\) years and PSD1000- 3000 years.  
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- ExtDataFig2.pdf
+
+<--- Page Split --->
+
+- HebertHerzschuhLaepple2021.odt- HebertHerzschuhLaepple2021.odt- SupplementaryInformationHebertHerzschuhLaepple2021.odt- SupplementaryInformationHebertHerzschnhLaepple2021.pdf- ExtDataFig8.pdf
+
+<--- Page Split --->

preprint/preprint__9866b1504884b2fa1c880a124ef92ec8435930a77685e3af1575260c12e23d80/preprint__9866b1504884b2fa1c880a124ef92ec8435930a77685e3af1575260c12e23d80_det.mmd

@@ -0,0 +1,466 @@
+<|ref|>title<|/ref|><|det|>[[44, 108, 880, 174]]<|/det|>
+# Land temperature variability driven by oceans at millennial timescales  
+
+<|ref|>text<|/ref|><|det|>[[44, 195, 435, 215]]<|/det|>
+Raphael Hebert ( Raphael.hebert@awi.de )  
+
+<|ref|>text<|/ref|><|det|>[[44, 218, 940, 259]]<|/det|>
+Alfred Wegener Institut Helmholtz Centre for Polar and Marine Research https://orcid.org/0000- 0002- 9869- 4658  
+
+<|ref|>text<|/ref|><|det|>[[44, 265, 193, 284]]<|/det|>
+Ulrike Herzschuh  
+
+<|ref|>text<|/ref|><|det|>[[50, 287, 888, 307]]<|/det|>
+Alfred Wegener Institute for Polar and Marine Research https://orcid.org/0000- 0003- 0999- 1261  
+
+<|ref|>text<|/ref|><|det|>[[44, 313, 195, 331]]<|/det|>
+Thomas Laepple  
+
+<|ref|>text<|/ref|><|det|>[[55, 335, 704, 353]]<|/det|>
+Alfred- Wegener- Institut Helmholtz- Zentrum für Polar- und Meeresforschung  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 393, 102, 411]]<|/det|>
+## Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 431, 479, 451]]<|/det|>
+Keywords: land temperature variability, timescales  
+
+<|ref|>text<|/ref|><|det|>[[44, 469, 301, 488]]<|/det|>
+Posted Date: March 8th, 2021  
+
+<|ref|>text<|/ref|><|det|>[[44, 507, 463, 526]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 147890/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 544, 910, 586]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[44, 623, 937, 666]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Geoscience on October 31st, 2022. See the published version at https://doi.org/10.1038/s41561- 022- 01056- 4.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[58, 81, 868, 129]]<|/det|>
+## 1 Land temperature variability driven by oceans at millennial timescales  
+
+<|ref|>text<|/ref|><|det|>[[58, 135, 503, 153]]<|/det|>
+3 \*R. Hebert \(^{1,2}\) , U. Herzschuh \(^{1,2,3}\) & T. Laepple \(^{1,4}\)  
+
+<|ref|>text<|/ref|><|det|>[[58, 153, 866, 252]]<|/det|>
+4 Affiliations: 5 1 Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Potsdam, Germany. 6 2 Institute of Environmental Sciences and Geography, University of Potsdam, Germany 7 3 Institute of Biochemistry and Biology, University of Potsdam, Germany 8 4 University of Bremen, MARUM - Center for Marine Environmental Sciences and Faculty of Geosciences, 28334 Bremen, Germany 10 \*Correspondence: raphael.hebert@awi.de  
+
+<|ref|>text<|/ref|><|det|>[[50, 265, 900, 530]]<|/det|>
+11 Variations in regional temperature have widespread implications for society, but our 12 understanding of the amplitude and origin of long- term natural variability is insufficient. This 13 is especially the case for terrestrial temperature variability which is currently thought to be 14 weak over long timescales. Here, we provide the first comprehensive estimate of regional 15 temperature variability from annual to millennial timescales based on sedimentary pollen 16 records and instrumental data from the Northern Hemisphere. We show that the short- term 17 random variations are overprinted by strong ocean- driven climate variability on multi- 18 decadal and longer timescales. This may cause substantial climate change at the regional scale 19 in the coming century, contrasting the rather monotonous warming projected by climate 20 models. Due to the marine influence, regions characterized by stable oceanic climate at sub- 21 decadal timescales experience stronger long- term variability while continental regions with 22 higher sub- decadal variability show weaker long- term variability. This fundamental 23 relationship between the timescales provides unique insight into the mechanisms governing 24 slow terrestrial climate variability and sets the basis to project its amplitude.  
+
+<|ref|>text<|/ref|><|det|>[[50, 582, 896, 920]]<|/det|>
+25 Improving our ability to characterize internal and natural climate variability is pivotal to improve 26 our long- term climate projections \(^{1}\) , particularly at the regional scale \(^{2}\) . The statistics of the natural 27 variability can be quantified up to decadal timescales using instrumental observations, but indirect 28 climate proxies are needed for the longer timescales. 29 Proxy- model comparisons aiming to validate climate models generally focus on whether long term 30 (forced) trends found in proxies can be reproduced by Global Climate Models (GCMs) \(^{3,4}\) . In this 31 regard, land pollen- based climate reconstructions in northern mid- latitudes were shown to be 32 consistent with the forcing and simulated warming \(^{5}\) . However, the long- term multi- millennial trend 33 is a rather limited measure which does not evaluate the ability of climate models to simulate 34 realistic climate variability, both forced and internal, at all timescales. To evaluate climate 35 variability across timescales, we consider the power spectral density (PSD, or simply spectrum) 36 S(Δt), which provides an estimate of how variance is distributed with frequency, or equivalently 37 with timescale Δt. Mitchell (1976) \(^{6}\) provided the classical view of climate variability being 38 dominated by specific quasi- periodic processes such as solar variations linked to orbital cycles over 39 a mostly random and uncorrelated (i.e. white) noise background. Since then, an increasing number 40 of climate proxies have shown a contrasting view of a variance that continuously increases with 41 increasing timescale. Such a behaviour is often well approximated by a power- law scaling 42 relationship \(^{7,8}\) and can be summarized by the scaling exponent β such that \(S(\Delta t) \sim \Delta t^{\beta}\) .
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[46, 84, 900, 234]]<|/det|>
+Several studies have suggested that GCMs can simulate realistic climate variability for global mean temperature across timescales \(^{9,10}\) which is dominated by the response to external forcing. However, at the regional scale, where only a small fraction of the variability is forced \(^{11}\) , the skill of GCMs has been called into question for slow timescales \(^{12,13}\) . Marine proxies suggest a strong scaling of regional sea- surface temperature (SST) variance from annual to millennial timescales ( \(\beta \approx 1\) ) \(^{7}\) which contrasts with the weak scaling found in GCMs ( \(\beta \approx 0.1\) ) \(^{14,15}\) . This leads to an increasing discrepancy between reconstructed SST and model simulations at longer timescales, reaching two orders of magnitude in variance at the millennial timescale \(^{7}\) .  
+
+<|ref|>text<|/ref|><|det|>[[45, 253, 900, 460]]<|/det|>
+On land, instrumental temperatures show a fundamentally different behaviour compared to the oceans, with only a weak timescale dependency up to multi- decadal scales ( \(\beta \approx 0.1\) ) \(^{14 - 16}\) ; this low- frequency weather has been termed “macroweather regime” and starts at sub- monthly timescales \(^{17}\) . This difference can be explained by the much smaller heat capacity of land surfaces compared to the oceanic mixed layer \(^{18}\) . If the macroweather- type behaviour over land would hold on to longer timescales, internal variability would only play a minor role on the uncertainty of regional climate projections \(^{19}\) . However, a different scaling behaviour of ocean and land at longer timescales would imply an increasingly large variability discrepancy between terrestrial and marine regions which seems physically implausible and in contradiction with both diffusive energy balance models \(^{20}\) and GCMs \(^{15}\) . This leaves us with the conundrum that we must either reject altogether the marine proxies or see a fundamental change of variability in the terrestrial domain on longer timescales.  
+
+<|ref|>text<|/ref|><|det|>[[45, 479, 895, 912]]<|/det|>
+Given the limited instrumental record, this fundamental question on the scaling of variability between land and ocean cannot be answered from observations without terrestrial proxy data. Holocene Greenland ice- core \(\delta^{180}\) timeseries suggest that the weak scaling behaviour found in the instrumental data extends to millennial timescales (akin to white noise with \(\beta \approx 0\) ) \(^{21}\) . However, it is unclear whether the climate variability derived from the Greenland ice- cores is representative for other terrestrial regions \(^{22}\) and to what extent the proxy variability reflects temperature variability \(^{23}\) . In order to elucidate the behaviour of climate variability over land at timescales longer than centennial, we compiled and analysed an extensive collection of recent Holocene Northern Hemisphere pollen data covering the past 8000 years, providing the largest spatial coverage of any land- based proxy at millennial timescales. This compilation of pollen data comprises 985 unique records from the northern hemisphere, 363 of which from North America, 487 from Europe, and 135 from Asia including the most complete dataset for China \(^{24}\) and Siberia \(^{25}\) . We produced summer (June- July- August) temperature reconstructions (hereafter simply termed temperature) as summer temperatures are usually well correlated with other variables driving vegetation growth \(^{26}\) . Our results are not sensitive on the choice of summer temperature (Extended Data Fig. 1) or a potential influence of precipitation (Extended Data Fig. 2). Pollen- based reconstructions rely on the assumption of dynamical equilibrium between climate and vegetation. This assumption is timescale dependent and is generally more valid at longer timescales \(^{26}\) . In this respect, millennial scale estimates of temperature variability should thus be particularly robust. In addition, the comparison with instrumental and tree- ring data (Extended Data Figure 3) suggests that temperature variability derived from pollen assemblages also reliably captures faster timescales. Finally, we verified that the estimates of millennial variability were not systematically biased by human influences (see Extended Data Fig. 4).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 65, 900, 273]]<|/det|>
+Pollen- based reconstructions, once coupled with instrumental data \(^{27}\) , enable us to establish for the first time a comprehensive picture of regional land temperature variability from inter- annual to millennial timescales (Fig. 1). The spectrum of instrumental air temperature shows a rather flat scaling behaviour that is characteristic of the macroweather regime. At longer timescales, however, the pollen- based reconstructions show a strong increase in variability with increasing timescale (Fig. 1a). This suggests that even in the relatively stable recent Holocene, there exists significant centennial to millennial scale temperature variability. This behaviour clearly differs from the rather flat macroweather regime and resembles power- law scaling with an exponent \(\beta\) near unity for timescales longer than centennial. Traces of this increased scaling behaviour at multi- decadal timescales already appear in the instrumental record and is corroborated by dendrochronological timeseries (Extended Data Fig. 3).  
+
+<|ref|>text<|/ref|><|det|>[[90, 291, 902, 536]]<|/det|>
+We benchmark our result against three transient climate models simulations of the recent Holocene: IPSL \(^{28}\) , ECHAM5 \(^{29}\) and the last 8000 years of the TraCE- 21ka deglaciation experiment \(^{30}\) (i.e. after the last freshwater forcing events). The transient simulations of these GCMs show the weak scaling behaviour characteristic of the macroweather regime, although with higher annual to decadal variability relative to the instrumental data as previously recognized \(^{15}\) . However, they fail to capture the increase in variability observed in the reconstructions at multi- decadal timescales. Instead, they show a relatively weak increase at multi- centennial timescales and a sharp increase at millennial timescales due to spectral leaking from the orbital 23- ka precession cycle (Fig. 1a, Extended Data Fig. 5). As a result of this divergence in variability scaling there is an increasing deficit in temperature variability observed in the simulations compared to the reconstructions. The range of variance ratios between the reconstructions and the different model simulations increases from 7- 10 over 100- 300 years to 17- 56 (19- 102 with orbital detrending) over 1000- 3000 years, resembling the discrepancy between models and proxy data for regional SST variability \(^{7}\) .  
+
+<|ref|>text<|/ref|><|det|>[[90, 555, 902, 932]]<|/det|>
+Comparing the reconstructions over land with estimates of marine variability \(^{7}\) shows they have a very similar low- frequency behaviour (Fig. 1b). This supports the view that both components will vary more coherently as climatic variability becomes a global phenomenon over longer timescales \(^{20}\) also indicated by coherent land and ocean average temperatures \(^{5}\) . Energy- balance models suggest that this parallel behaviour of land and oceans on long timescales is due to heat exchange between the land and ocean compartments. In such models, land air temperature can be described as a linear combination of the SST and a time- dependent forcing over land \(^{31}\) ; the resulting variability spectrum over land is then a linear combination of the spectra of each term (Fig. 1b) when the two are uncorrelated (see Supplementary Information). In this framework, the change in scaling behaviour can be regarded as a transition from the macroweather regime at shorter timescales, dominated by a weakly scaling forcing component akin to white noise over land \(^{18}\) , to an oceanic regime dominated by the SST component at timescales longer than decadal. Interestingly, the parallel behaviour between land and ocean temperature spectra on multi- decadal to millennial timescales provides no evidence for additional terrestrial slow climate feedbacks. The oceanic component present in land temperature variability appears amplified by a factor of \(\sim 4\) in PSD or \(\sim 2\) in amplitude. This factor is similar to the land- sea warming contrast \(^{32}\) observed during the last century \(^{33}\) and within the range of land- sea warming ratios measured in GCMs \(^{34}\) . This is thought to be the result of local feedbacks, for example the evaporation feedback, when moisture availability over land limits evaporative cooling in comparison with marine regions \(^{35}\) , and also because of the asymmetry in the land- ocean heat exchange which favours land due to its lower specific humidity \(^{34}\) .
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 84, 904, 404]]<|/det|>
+The extensive spatial coverage of the pollen- based reconstructions allows us to perform a spatial analysis of the millennial scale temperature variability \(\mathrm{PSD}_{1000 - 3000\mathrm{years}}\) (the mean PSD over 1000- 3000 year) (Fig. 2a) and investigate the potential link to oceanic influence. The spatial coherency (Moran's \(\mathrm{I} = 0.19\) , \(\mathrm{p}< 0.001\) , see Methods) demonstrates that the variability estimates are not drowned out by local noise. Over Europe, with its large number of records, millennial scale variability decreases inland along the path of prevailing winds blowing from the Atlantic Ocean, and is lowest over Fennoscandia where blocking events are most frequent<sup>36</sup>. Similarly, China's high millennial variability would be linked to the persistent oceanic influence carried by the dominant easterlies at that latitude, while further north in eastern Siberia the dominant westerlies bring little oceanic influence. This further suggests that higher millennial variability relies on higher connectivity to oceans, as implied by energy- balance models, although compounded with local sensitivity. The high variability in central Asia remains an outlier given the strong continentality there, but the significance is lower because of the sparseness of records. It is also possible that the lower connectivity to oceans is compensated by the stronger local climate sensitivity<sup>33</sup> which may be linked to hydrological feedback due to the arid conditions<sup>35</sup>. Meanwhile, in North America the lowest millennial variability is found in the prairies, near the centre of the continent, where the westerlies predominantly blow from the north- west and the oceanic influence is lowest.  
+
+<|ref|>text<|/ref|><|det|>[[90, 422, 898, 648]]<|/det|>
+We use the instrumental data to study the mechanisms governing the spatial distribution of the millennial variability and the continuum of variability. The scaling of variability in instrumental data has already been shown to be related to the strength of the annual cycle and of the sub- decadal variability<sup>8</sup>. If we aggregate the instrumental data and reconstructions based on the sub- decadal variability \(\mathrm{PSD}_{2 - 10\mathrm{years}}\) (the mean PSD over 2- 10 years; Fig. 3, Extended Data Fig. 6, Extended Data Fig. 7), a clear relationship appears with the emergence of the low- frequency regime, quantified by the multi- decadal scaling exponent \(\beta_{10 - 60\mathrm{years}}\) ( \(\beta\) regressed over 10- 60 years): locations with lower sub- decadal variability thus show a stronger increase of variability towards longer timescales, as indicated by higher multi- decadal scaling (Fig. 2b,c, Fig. 3, Extended Data Fig. 7). We should thus expect an inversion where regions of low (high) sub- decadal variability, typically characterized by more maritime (continental) influences, would become regions of high (low) variability at long timescales.  
+
+<|ref|>text<|/ref|><|det|>[[90, 667, 901, 874]]<|/det|>
+Indeed, this hypothesized relationship is confirmed by the pollen- based reconstructions. Their estimates of millennial temperature variability \(\mathrm{PSD}_{1000 - 3000\mathrm{years}}\) show a strong anti- correlation with the sub- decadal variability \(\mathrm{PSD}_{2 - 10\mathrm{years}}\) (r=- 0.95, \(\mathrm{p}< 0.01\) ) and a strong correlation with the multi- decadal scaling \(\beta_{10 - 60\mathrm{years}}\) (r=0.91, \(\mathrm{p}< 0.01\) ). These significant strong relationships between the pollen- based reconstructions and independent instrumental temperature data demonstrate a fundamental link of temperature variability from sub- decadal to millennial timescales. The spatial pattern of the variability (Fig. 2 and Extended Data Fig. 5) further suggests that this relationship is caused by varying marine influence. In addition, this can explain the relation of the amplitude of the annual cycle (an indicator for continentality) with the inter- annual variability scaling in the instrumental relationship proposed by Huybers and Curry (2006)<sup>8</sup>. Therefore, our findings complete the linkage between seasonal and millennial land temperature variability.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[90, 65, 901, 310]]<|/det|>
+Our results indicate that current GCMs underestimate regional temperature variability over land at timescales longer than multi-decadal (Fig. 1a, Extended Data Fig. 5). In combination with the spatial pattern of variability (Fig. 2), this suggests that the deficit in low- frequency variability is related to an underestimation of marine variability<sup>7</sup>. The interpretation of climate sensitive proxies remains an area of active research, and in principle, it remains possible that the observed model- data mismatches stem from non- climatic variability. However, several lines of evidence argue against this interpretation. Firstly, there are no known archival processes yet which could artificially create such power- law scaling in sedimentary archives<sup>37</sup>. Specifically, the known processes such as counting errors, spatial or temporal aliasing, and bioturbation in the sediment cannot explain the power spectra of variability found here. Secondly, the consistency between independent marine and terrestrial archives (Fig. 1b) provides further support for the temperature variability reconstruction. Finally, the spatial relationship with the independent instrumental temperature data (Fig. 3) also indicates that this is no artefact from the proxy data.  
+
+<|ref|>text<|/ref|><|det|>[[92, 329, 904, 441]]<|/det|>
+Thus, pollen- based reconstructions support the paradigm of an increasing continuum of climate variability with increasing timescales<sup>8</sup>, in contradiction with the local temperature variability in current GCM simulations and the classical picture of Mitchell<sup>6</sup>. More importantly, our results extend previous findings from instrumental data<sup>8</sup> and demonstrate a fundamental link between interannual, multi- decadal and millennial timescales driven by the interaction of marine and terrestrial temperature variability modulated by continentality.  
+
+<|ref|>text<|/ref|><|det|>[[90, 460, 890, 742]]<|/det|>
+This fundamental behaviour of temperature variability has implications for the relative impact of natural and anthropogenically forced variability. High latitude regions characterised by high interannual variability show a weaker oceanic regime and, ultimately, less natural variability on long timescales. As these regions are also highly sensitive to anthropogenic forcing, the impact of anthropogenic warming, relative to natural variability, will be greater. However, regions of strong maritime influence, where most of the world's population is located, could see large natural variability, that is not covered by current GCM projections which tend to display monotonous warming<sup>33</sup>. It is thus possible that until now, the stronger natural variability at multi- decadal timescales in maritime regions has partly overshadowed the anthropogenic warming in those regions which could explain their lower observational transient climate sensitivity<sup>33</sup>. Integrative archives such as glaciers should be particularly sensitive to this increased memory<sup>16</sup> and could be used to verify our findings. Large compilations of climate archives have the potential to inform us on the spatial patterns of slow variability and their underlying causes, and further studies combining multiple proxies over land and ocean show great promise to improve our understanding of the spatio- temporal correlation structure of climate variability.  
+
+<|ref|>sub_title<|/ref|><|det|>[[93, 771, 335, 790]]<|/det|>
+## Author Contributions  
+
+<|ref|>text<|/ref|><|det|>[[92, 799, 876, 873]]<|/det|>
+R.H., T.L. and U.H. designed the research. U.H. performed the pollen- based reconstruction. T.L. contributed the interpretation of the marine archives. R.H. and T.L. developed the methodology. R.H. performed the data analysis and wrote the first draft of the manuscript. R.H., T.L., and U.H. contributed to the interpretation and to the preparation of the final manuscript.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 66, 317, 85]]<|/det|>
+## Acknowledgements  
+
+<|ref|>text<|/ref|><|det|>[[90, 95, 900, 282]]<|/det|>
+AcknowledgementsThis is a contribution to the SPACE ERC and GLACIAL LEGACY ERC projects; these projects have received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement no. 716092 and no. 772852). This work is further a contribution to the PALMOD project. The work profited from discussions at the CVAS working group of the Past Global Changes (PAGES) programme. We thank A. Dallmeyer, A. Dolman, I. Kröner, T. Kunz, S. Lovejoy and K. Rehfeld for useful discussion. We acknowledge P. Huybers for comments on the manuscript. We thank P. Braconnot and J. Cretat, and J. Jungclaus for, respectively, providing the IPSL and ECHAM5 simulations, and T. Böhmer and X. Cao for their support compiling the database. We thank all original data contributors who made their proxy data available and acknowledge the Neotoma Palaeoecology Database.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 291, 324, 311]]<|/det|>
+## Competing interests  
+
+<|ref|>text<|/ref|><|det|>[[90, 320, 610, 338]]<|/det|>
+The authors R.H., U.H. and T.L. declare no competing interests.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 366, 416, 386]]<|/det|>
+## Materials & Correspondence  
+
+<|ref|>text<|/ref|><|det|>[[90, 395, 633, 412]]<|/det|>
+Correspondence and materials request should be addressed to R.H.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[290, 149, 694, 647]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[90, 661, 904, 868]]<|/det|>
+<center>Figure 1| Average spectral estimates of local land temperature over the Northern Hemisphere. a, Average spectral estimates of land air temperature from pollen-based reconstructions, instrumental data and model simulations extracted at the pollen record locations. Also shown are the spectra estimated after applying a 23-ka sinusoidal detrending (dashed, see Methods). 90% confidence intervals are given (shaded). The number of pollen records contributing to each timescale is indicated below (brown axis). b, Average spectral estimates from reconstructed annual sea-surface temperature derived from marine archives<sup>5</sup>, and from instrumental data at the corresponding locations. The pollen-based and instrumental average spectra from a are reproduced. Also shown are linear combinations of power-laws with slope \(\beta = 1.2\) and white noise series (dashed green for land, dashed blue for sea and dashed grey for the white noise levels) corresponding to the energy-balance model approximation. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[95, 150, 828, 631]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[91, 640, 902, 808]]<|/det|>
+<center>Figure 2| Spatial patterns of temperature variability in instrumental data and pollen-based reconstructions. a, Map of millennial variability estimated from the temperature spectra of pollen-based reconstructions as the mean PSD over the 1000-3000 year timescale band. The PSD shown in the background are smoothed using Gaussian weights with a \(300\mathrm{km}\) scale. Overlaid circles are the average of all pollen records closest to the corresponding instrumental grid point. b, Map of the multi-decadal scaling exponent \(\beta\) from the spectra of instrumental temperature records fitted over the 10-60 year timescale band. c, Map of sub-decadal variability, mean PSD over the 2-10 year timescale band, estimated from the spectra of instrumental temperature records. In a-c, the circles indicate the instrumental grid points which were the closest to pollen records. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[92, 241, 888, 650]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[90, 653, 880, 804]]<|/det|>
+<center>Figure 3 | Spectral estimates of land temperature as a function of sub-decadal variability. a, Spectral estimates of the instrumental temperature ( \(\Delta t< 60\) years) and of the pollen-based reconstructions ( \(\Delta t > 100\) years) binned according to sub-decadal instrumental variability (See Methods and Extended Data Fig.7). Dashed lines indicate estimated instrumental multi-decadal scaling exponents \(\beta_{10 - 60 \text{years}}\) . \(90\%\) confidence intervals are given (shaded). b, Relation of the instrumental sub-decadal temperature variability \(\mathrm{PSD}_{2 - 10 \text{years}}\) and the pollen-based millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \text{years}}\) . The size of the points is proportional to the errors on the millennial variability estimates. c, As in b, but between \(\beta_{10 - 60 \text{years}}\) and \(\mathrm{PSD}_{1000 - 3000 \text{years}}\) . </center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[92, 81, 209, 104]]<|/det|>
+## Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 123, 259, 141]]<|/det|>
+## Reconstructions  
+
+<|ref|>text<|/ref|><|det|>[[92, 150, 904, 337]]<|/det|>
+The modern pollen dataset used for calibration consists of 15,532 sampling sites. For each fossil record location, we selected a unique subset of modern sites within a \(2000\mathrm{km}\) radius in order to increase the reliability of the reconstructions and avoid a bi- modal climate optimum<sup>38</sup>. The fossil dataset includes taxonomically harmonized fossil pollen data from North America and Europe obtained via the Neotoma Palaeoecological Database<sup>39</sup>, and from Asia combining Cao et al. (2020)<sup>25</sup> and Cao et al. (2014)<sup>24</sup>. A fossil database comprising 985 records was compiled based on the requirement that the resulting spectral estimates covered timescales at least one fifth of an order of magnitude, i.e. 0.2 in a base 10 logarithm, below one third the length (to avoid the well- known multitaper low- bias at long timescales)<sup>40</sup>. The European and Asian datasets were combined keeping the 70 most common taxa according to Hill's second number.  
+
+<|ref|>text<|/ref|><|det|>[[92, 357, 895, 526]]<|/det|>
+The modern climate data used for calibration was the average of the June, July and August climatologies, i.e. the summer temperature, for the years from 1970- 2000 as obtained from WorldClim 2.1<sup>41</sup>. The Weighted Averaging Partial Least Square (WAPLS)<sup>42</sup> method was used to calibrate transfer functions relating the pollen- assemblages to the summer temperature, with leave- one- out cross validation. The pollen percentages were square- root transformed to decrease the dominance of abundant taxa with high productivity. The number of retained WAPLS components was selected using a randomization t- test. The same method was also applied to reconstruct annual mean temperature (Extended Data Fig. 1) and annual precipitation (Extended Data Fig. 2), also using the climatologies from WorldClim 2.1<sup>41</sup> for the calibration of each variable.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 536, 297, 555]]<|/det|>
+## Significance Testing  
+
+<|ref|>text<|/ref|><|det|>[[92, 563, 900, 770]]<|/det|>
+Following Telford and Birks (2011)<sup>43</sup>, we tested the summer temperature reconstructions for significance using the R package "palaeoSig" and found that 228 out of 985 reconstructions were significant (p<0.1). However, this significance test is rather conservative and several reasons can create type II errors (false negatives), including for example a low diversity of taxa, a small number of sub- fossil observations, an input climate signal that is less variable, or an inadequate training set<sup>43</sup>. A higher p- values therefore does not necessarily mean that the summer temperature has not been recorded, but rather that the information is insufficient to confirm it. Thus, instead of unduly discarding most records, we decided to include all records in the main analysis. We show that our conclusions are robust and continue to hold even if we restrict our analysis to the 'significant locations' only (p<0.1) which yielded similar results to the 'not significant locations' (p>0.1) (Extended Data Fig. 8).  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 794, 446, 814]]<|/det|>
+## Testing for Anthropogenic Impacts  
+
+<|ref|>text<|/ref|><|det|>[[92, 826, 897, 920]]<|/det|>
+We considered all series covering entirely the last 8- ka showed and defined two temporal windows: the more recent (0- 4ka) and the more distant (4ka- 8ka) past. The 1000- 2000 year timescale band was taken to calculate the variance ratio between the two 4ka windows. We only included those series in our analyses whose spectral estimates covered at least one tenth of an order of magnitude (on a base 10 logarithmic scale) for both time periods; a criterion that was met by 344 records. We
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[91, 66, 904, 272]]<|/det|>
+found no systematic variance increase in the more recent half of the series as the mean of the distribution of the logarithm of the variance ratios did not significantly differ from zero (p>0.1). In fact, the more recent period, where human impacts may have contributed to an increased variability, is about \(6\%\) less variable than the earlier one. Likewise, the spatial distribution (Extended Data Fig. 4) did not show any obvious spatial patterns that could be related to human occupation, displaying a non- significant Moran's I of 0.014 (p>0.1; see Methods). If human occupation was the dominant driver of millennial scale variability, we would have expected to observe an increase in variability over both Europe and China, where human occupation has been increasing the most over the last 4000 years compared to the preceding 4000 years. We thus conclude that human impacts on vegetation did not have a significant enough impact on the slow variability to systematically bias millennial scale variability estimates.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 291, 282, 309]]<|/det|>
+## Instrumental Data  
+
+<|ref|>text<|/ref|><|det|>[[91, 319, 900, 450]]<|/det|>
+For the instrumental dataset we used the Berkeley Earth Surface Temperature (BEST) land and ocean product covering the period from 1850- 2020. The equal area product is used for calculations, while the regular \(1^{\circ} \times 1^{\circ}\) product was used for visual display. The instrumental data was detrended from anthropogenic influences to a first- order component<sup>44</sup> proportional to historical timeseries of doublings in atmospheric carbon dioxide concentration<sup>45</sup>. There are 403 grid points which comprise pollen records (circles on Fig. 2). Using alternative instrumental datasets<sup>46</sup> leads to similar conclusions.  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 460, 214, 478]]<|/det|>
+## Model Data  
+
+<|ref|>text<|/ref|><|det|>[[91, 487, 901, 599]]<|/det|>
+Three model simulations of the recent Holocene were considered, namely the IPSL<sup>28</sup>, the ECHAM5<sup>29</sup> and the CCSM3<sup>47</sup>. The first two are recent Holocene transient simulations of the past 6,000 years, and the latter is the TraCE- 21ka deglaciation experiment<sup>30</sup>. We only retained the last 8,000 years of TraCE- 21ka since it is comparable to the recent Holocene transient simulations as they contain no more freshwater forcing events which were the main drivers of the deglaciation. We selected the average 2- meter summer air temperature (June- July- August) and averaged it annually.  
+
+<|ref|>text<|/ref|><|det|>[[91, 621, 888, 735]]<|/det|>
+Since the long- term trends in summer temperature are linked to the precession in the Earth's orbit, we also analysed the timeseries after detrending for a 23- ka sinusoid rather than use the standard linear detrending performed before computing spectral estimates. This approach attempts to minimize power leakage from the orbital forcing frequencies onto the observed frequencies (Fig. 1a, dashed lines). The reduction in leaked power was not nearly as important in the case of the pollen- based reconstructions (Fig. 1a, Extended Data Fig. 5).  
+
+<|ref|>sub_title<|/ref|><|det|>[[92, 754, 283, 773]]<|/det|>
+## Spectral Estimates  
+
+<|ref|>text<|/ref|><|det|>[[91, 781, 904, 874]]<|/det|>
+The power spectral density estimates were calculated using the multitaper method<sup>48</sup>, adapted for irregular sampling through linear interpolation<sup>49</sup>, with the number of tapers \(n_{\text{tapers}} = 3\) and the time- bandwidth parameter \(\omega = 2\) , which yield up to \(n_{\text{tapers}} * \omega = 6\) degrees of freedom for the individual spectral estimates. Only timescales greater than twice the maximal resolution were kept to minimize power loss due to the interpolation<sup>37</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[91, 894, 856, 931]]<|/det|>
+The confidence intervals were derived from the chi- squared distribution \((\chi^2)\) of the multitaper estimates. The degrees of freedom of the \(\chi^2\) were limited to a maximum based on the expected
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 65, 900, 198]]<|/det|>
+effective spatial degrees of freedom at a given timescale in order to avoid obtaining over- confident estimates. The approximate relationship assumed for the maximum degrees of freedom \(\mathrm{v}_{\mathrm{max}}\) as a function of timescale is: \(\mathrm{v}_{\mathrm{max}} = 40 \mathrm{At}^{- 0.3}\) . This corresponds to about 20, 10, and 5 degrees of freedom at the decadal, centennial and millennial timescales respectively<sup>37</sup>. They were multiplied by six, the number of degrees of freedom for the multitaper estimates with three independent tapers, and further modulated by a factor of spatial representativity \(\mathrm{f}_{\Delta \mathrm{t}}\) calculated as the fraction of the land area represented by the spatial distribution of the data.  
+
+<|ref|>text<|/ref|><|det|>[[39, 216, 870, 253]]<|/det|>
+For each point over land we calculated the effective number of records \(\mathrm{N}_{\mathrm{Eff}}\) on a regular \(1^{\circ}\) by \(1^{\circ}\) degree grid using a Gaussian kernel:  
+
+<|ref|>equation<|/ref|><|det|>[[105, 253, 222, 298]]<|/det|>
+\[N_{\mathrm{Eff}} = \sum_{i = 1}^{n}\mathrm{e}^{\frac{-d_{i,j}}{2\sigma^{2}}}\]  
+
+<|ref|>text<|/ref|><|det|>[[39, 299, 860, 392]]<|/det|>
+where \(\mathrm{d}_{\mathrm{i,j}}\) is the geographical distance between the \(\mathrm{i}^{\mathrm{th}}\) data record (out of n records with a PSD estimate at the given timescale \(\Delta \mathrm{t}\) ) and the \(\mathrm{j}^{\mathrm{th}}\) grid point where \(\mathrm{N}_{\mathrm{Eff}}\) is calculated, and \(\sigma\) is a characteristic decorrelation scale which we took as \(\sigma = 300 \mathrm{km}\) . \(\mathrm{N}_{\mathrm{Eff}}\) was also used for Fig. 2a in order to modulate the opacity of the background as a function of nearby records. The factor of spatial representativity for a given timescale was then calculated as:  
+
+<|ref|>equation<|/ref|><|det|>[[100, 393, 353, 473]]<|/det|>
+\[f_{\Delta t} = \frac{\sum_{j = 1}^{m}\min \left(N_{\mathrm{Eff},j},1\right)\cos(\theta_{j})}{\sum_{j = 1}^{m}\cos(\theta_{j})}\]  
+
+<|ref|>text<|/ref|><|det|>[[39, 475, 847, 510]]<|/det|>
+where m is the number of grid points covering the land area north of the southernmost pollen record.  
+
+<|ref|>sub_title<|/ref|><|det|>[[41, 540, 256, 558]]<|/det|>
+## 319 Variance Ratios  
+
+<|ref|>text<|/ref|><|det|>[[39, 567, 897, 641]]<|/det|>
+Variance ratios were computed by taking the ratio between the mean PSD over the same timescale band between different series after interpolating in the spectral domain. Since the ratio of two \(\chi^2\) distributed variables follows an F- distribution, the ratios were multiplied by (d- 2) \(\mathrm{d}^{- 1}\) where d is the number of degrees of freedom of the denominator<sup>22</sup>.  
+
+<|ref|>sub_title<|/ref|><|det|>[[41, 670, 425, 690]]<|/det|>
+## 324 Sub-Decadal Variability Binning  
+
+<|ref|>text<|/ref|><|det|>[[39, 698, 900, 811]]<|/det|>
+The data was aggregated based on the mean sub- decadal variability \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) , defined as the mean PSD over the 2- 10 year timescale band. We calculated \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) for each of the 403 instrumental grid point for which pollen records were present nearby, ordered the results, and split them into eight non- overlapping bins (Extended Data Fig. 7). Each pollen record was assigned to the nearest instrumental grid point and averaged in the spectral domain. Varying the number of bins, for example using twenty bins instead of eight, leads to similar correlations (Extended Data Fig. 7).  
+
+<|ref|>sub_title<|/ref|><|det|>[[41, 840, 213, 858]]<|/det|>
+## 331 Correlation  
+
+<|ref|>text<|/ref|><|det|>[[39, 868, 886, 924]]<|/det|>
+The Pearson's correlation was calculated between the instrumental sub- decadal variability \(\mathrm{PSD}_{2 - 10 \mathrm{years}}\) , the instrumental multi- decadal scaling exponent \(\beta_{10 - 60 \mathrm{years}}\) , and the pollen- based millennial variability \(\mathrm{PSD}_{1,000 - 3,000 \mathrm{years}}\) after binning the instrumental grid points near pollen records into 8
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[36, 65, 886, 141]]<|/det|>
+335 subsets according to the PSD2- 10 years amplitudes (see previous section). The standard errors on the 336 millennial variability estimates, i.e. \(\sqrt{2DoF}\) , where DoF is the total degrees of freedom for the 337 PSD estimate, were used as weights for the correlation calculation and for visual representation in 338 Fig. 3b,c.  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 170, 197, 189]]<|/det|>
+## 339 Moran's I  
+
+<|ref|>text<|/ref|><|det|>[[42, 198, 878, 253]]<|/det|>
+Moran's I Moran's I spatial autocorrelation index was calculated using the method from Gittleman and Kot (1990) as implemented in the R- package "ape"51. The weight matrix used corresponds to the inverse of the distance between sites.  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 282, 368, 302]]<|/det|>
+## 343 Code and Data Availability  
+
+<|ref|>text<|/ref|><|det|>[[42, 311, 880, 348]]<|/det|>
+All relevant code and data is available to the editor and the reviewers of this article upon request, and will be made publicly available in a GitHub repository upon publication.  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 432, 237, 454]]<|/det|>
+## 346 References  
+
+<|ref|>text<|/ref|><|det|>[[120, 465, 900, 920]]<|/det|>
+1. Thompson, D. W. J., Barnes, E. A., Deser, C., Foust, W. E. & Phillips, A. S. Quantifying the Role of Internal Climate Variability in Future Climate Trends. J. Clim. 28, 6443-6456 (2015).  
+2. McKinnon, K. A. & Deser, C. Internal Variability and Regional Climate Trends in an Observational Large Ensemble. J. Clim. 31, 6783-6802 (2018).  
+3. Marcott, S. A., Shakun, J. D., Clark, P. U. & Mix, A. C. A Reconstruction of Regional and Global Temperature for the Past 11,300 Years. Science 339, 1198-1201 (2013).  
+4. Roulson, C. C. et al. Mid-latitude net precipitation decreased with Arctic warming during the Holocene. Nature 568, 83-87 (2019).  
+5. Marsicek, J., Shuman, B. N., Bartlein, P. J., Shafer, S. L. & Brewer, S. Reconciling divergent trends and millennial variations in Holocene temperatures. Nature 554, 92-96 (2018).  
+6. Mitchell, J. M. & others. An overview of climatic variability and its causal mechanisms. Quart. Res. 6, 481-493 (1976).  
+7. Laepple, T. & Huybers, P. Ocean surface temperature variability: Large model-data differences at decadal and longer periods. Proc. Natl. Acad. Sci. 111, 16682-16687 (2014).  
+8. Huybers, P. & Curry, W. Links between annual, Milankovitch and continuum temperature variability. Nature 441, 329-332 (2006).  
+9. Zhu, F. et al. Climate models can correctly simulate the continuum of global-average temperature variability. Proc. Natl. Acad. Sci. 116, 8728-8733 (2019).  
+10. Crowley, T. J. Causes of Climate Change Over the Past 1000 Years. Science 289, 270-277 (2000).  
+11. Goosse, H., Renssen, H., Timmermann, A. & Bradley, R. S. Internal and forced climate variability during the last millennium: a model-data comparison using ensemble simulations. Quart. Sci. Rev. 24, 1345-1360 (2005).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[120, 67, 896, 103]]<|/det|>
+12. Cheung, A. H. et al. Comparison of Low-Frequency Internal Climate Variability in CMIP5 Models and Observations. J. Clim. 30, 4763-4776 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[123, 103, 880, 158]]<|/det|>
+13. Dee, S. G. et al. Improved spectral comparisons of paleoclimate models and observations via proxy system modeling: Implications for multi-decadal variability. Earth Planet. Sci. Lett. 476, 34-46 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[123, 160, 895, 196]]<|/det|>
+14. Fraedrich, K. & Blender, R. Scaling of Atmosphere and Ocean Temperature Correlations in Observations and Climate Models. Phys. Rev. Lett. 90, (2003).  
+
+<|ref|>text<|/ref|><|det|>[[123, 197, 896, 233]]<|/det|>
+15. Fredriksen, H.-B. & Rypdal, K. Spectral Characteristics of Instrumental and Climate Model Surface Temperatures. J. Clim. 29, 1253-1268 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[123, 234, 897, 270]]<|/det|>
+16. Roe, G. H. What do glaciers tell us about climate variability and climate change? J. Glaciol. 57, 567-578 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[123, 272, 894, 327]]<|/det|>
+17. Lovejoy, S. & Schertzer, D. Low-Frequency Weather and the Emergence of the Climate. in Extreme Events and Natural Hazards: The Complexity Perspective 231-254 (American Geophysical Union (AGU), 2013). doi:10.1029/2011GM001087.  
+
+<|ref|>text<|/ref|><|det|>[[123, 329, 840, 347]]<|/det|>
+18. Hasselmann, K. Stochastic climate models Part I. Theory. Tellus 28, 473-485 (1976).  
+
+<|ref|>text<|/ref|><|det|>[[123, 348, 875, 384]]<|/det|>
+19. Hawkins, E., Sutton, R. & others. The potential to narrow uncertainty in regional climate predictions. Bull. Am. Meteorol. Soc. 90, 1095-1107 (2009).  
+
+<|ref|>text<|/ref|><|det|>[[123, 385, 900, 421]]<|/det|>
+20. North, G. R., Cahalan, R. F. & Coakley, J. A. Energy balance climate models. Rev. Geophys. 19, 91-121 (1981).  
+
+<|ref|>text<|/ref|><|det|>[[123, 422, 886, 459]]<|/det|>
+21. Nilsen, T., Rypdal, K. & Fredriksen, H.-B. Are there multiple scaling regimes in Holocene temperature records? Earth Syst. Dyn. 7, 419-439 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[123, 460, 888, 497]]<|/det|>
+22. Rehfeld, K., Münch, T., Ho, S. L. & Laepple, T. Global patterns of declining temperature variability from the Last Glacial Maximum to the Holocene. Nature 554, 356-359 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[123, 498, 895, 552]]<|/det|>
+23. Casado, M., Münch, T. & Laepple, T. Climatic information archived in ice cores: impact of intermittency and diffusion on the recorded isotopic signal in Antarctica. Clim. Past 16, 1581-1598 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[123, 554, 897, 608]]<|/det|>
+24. Cao, X., Herzschuh, U., Telford, R. J. & Ni, J. A modern pollen-climate dataset from China and Mongolia: Assessing its potential for climate reconstruction. Rev. Palaeobot. Palynol. 211, 87-96 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[123, 610, 900, 647]]<|/det|>
+25. Cao, X. et al. A taxonomically harmonized and temporally standardized fossil pollen dataset from Siberia covering the last 40&thinsp;kyr. Earth Syst. Sci. Data 12, 119-135 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[123, 648, 860, 684]]<|/det|>
+26. Chevalier, M. et al. Pollen-based climate reconstruction techniques for late Quaternary studies. Earth-Sci. Rev. 210, 103384 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[123, 686, 803, 722]]<|/det|>
+27. Rohde, R., Muller, R., Jacobsen, R., Perlmutter, S. & Mosher, S. Berkeley Earth Temperature Averaging Process. Geoinformatics Geostat. Overv. 01, (2013).  
+
+<|ref|>text<|/ref|><|det|>[[123, 723, 870, 778]]<|/det|>
+28. Braconnot, P., Zhu, D., Marti, O. & Servonnat, J. Strengths and challenges for transient Mid- to Late Holocene simulations with dynamical vegetation. Clim. Past 15, 997-1024 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[123, 780, 895, 872]]<|/det|>
+29. Fischer, N. & Jungclaus, J. H. Holocene experiment with coupled atmosphere-ocean-model ECHAM5/MPI-OM. Supplement to: Fischer, N; Jungclaus, JH (2011): Evolution of the seasonal temperature cycle in a transient Holocene simulation: orbital forcing and sea-ice. Climate of the Past, 7, 1139-1148, https://doi.org/10.5194/cp-7-1139-2011 (2012) doi:https://doi.org/10.1594/PANGAEA.773607.  
+
+<|ref|>text<|/ref|><|det|>[[120, 874, 881, 893]]<|/det|>
+30. He, F. Simulating Transient Climate Evolution of the Last deglaciation with CCSM3. 161.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[120, 66, 900, 120]]<|/det|>
+31. Geoffroy, O. et al. Transient Climate Response in a Two-Layer Energy-Balance Model. Part I: Analytical Solution and Parameter Calibration Using CMIP5 AOGCM Experiments. J. Clim. 26, 1841-1857 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[120, 121, 875, 175]]<|/det|>
+32. Byrne, M. P. & O'Gorman, P. A. Link between land-ocean warming contrast and surface relative humidities in simulations with coupled climate models. Geophys. Res. Lett. 40, 5223-5227 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[120, 178, 895, 214]]<|/det|>
+33. Hébert, R. & Lovejoy, S. Regional Climate Sensitivity- and Historical-Based Projections to 2100. Geophys. Res. Lett. 45, 4248-4254 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[120, 216, 875, 252]]<|/det|>
+34. Geoffroy, O., Saint-Martin, D. & Voldoire, A. Land-sea warming contrast: the role of the horizontal energy transport. Clim. Dyn. 45, 3493-3511 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[120, 253, 900, 307]]<|/det|>
+35. Sejas, S. A., Albert, O. S., Cai, M. & Deng, Y. Feedback attribution of the land-sea warming contrast in a global warming simulation of the NCAR CCSM4. Environ. Res. Lett. 9, 124005 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[120, 309, 890, 363]]<|/det|>
+36. Schwander, M., Rohrer, M., Brönnimann, S. & Malik, A. Influence of solar variability on the occurrence of central European weather types from 1763 to 2009. Clim. Past 13, 1199-1212 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[120, 365, 892, 419]]<|/det|>
+37. Kunz, T., Dolman, A. M. & Laepple, T. A spectral approach to estimating the timescale-dependent uncertainty of paleoclimate records - Part 1: Theoretical concept. Clim. Past 16, 1469-1492 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[120, 421, 885, 476]]<|/det|>
+38. Cao, X., Ni, J., Herzschuh, U., Wang, Y. & Zhao, Y. A late Quaternary pollen dataset from eastern continental Asia for vegetation and climate reconstructions: Set up and evaluation. Rev. Palaeobot. Palynol. 194, 21-37 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[120, 478, 840, 514]]<|/det|>
+39. Goring, S. et al. neotoma: A Programmatic Interface to the Neotoma Paleoecological Database. Open Quat. 1, (2015).  
+
+<|ref|>text<|/ref|><|det|>[[120, 516, 825, 552]]<|/det|>
+40. Prieto, G., Parker, R., Thomson, D., Vernon, F. & Graham, R. Reducing the bias of multitaper spectrum estimates. Geophys. J. Int. 171, 1269-1281 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[120, 553, 890, 590]]<|/det|>
+41. Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302-4315 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[120, 591, 900, 646]]<|/det|>
+42. Juggins, S. & Birks, H. J. B. Quantitative Environmental Reconstructions from Biological Data. in Tracking Environmental Change Using Lake Sediments (eds. Birks, H. J. B., Lotter, A. F., Juggins, S. & Smol, J. P.) vol. 5 431-494 (Springer Netherlands, 2012).  
+
+<|ref|>text<|/ref|><|det|>[[120, 648, 904, 702]]<|/det|>
+43. Telford, R. J. & Birks, H. J. B. A novel method for assessing the statistical significance of quantitative reconstructions inferred from biotic assemblages. Quat. Sci. Rev. 30, 1272-1278 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[120, 704, 904, 758]]<|/det|>
+44. Wu, T., Hu, A., Gao, F., Zhang, J. & Meehl, G. A. New insights into natural variability and anthropogenic forcing of global/regional climate evolution. Npj Clim. Atmospheric Sci. 2, 1-13 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[120, 760, 884, 796]]<|/det|>
+45. Meinshausen, M. et al. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Clim. Change 109, 213 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[120, 798, 904, 834]]<|/det|>
+46. Cowtan, K. & Way, R. G. Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends. Q. J. R. Meteorol. Soc. n/a-n/a (2013) doi:10.1002/qj.2297.  
+
+<|ref|>text<|/ref|><|det|>[[120, 836, 874, 871]]<|/det|>
+47. Yeager, S. G., Shields, C. A., Large, W. G. & Hack, J. J. The Low-Resolution CCSM3. J. Clim. 19, 2545-2566 (2006).  
+
+<|ref|>text<|/ref|><|det|>[[120, 873, 866, 910]]<|/det|>
+48. Thomson, D. J. Spectrum estimation and harmonic analysis. Proc. IEEE 70, 1055-1096 (1982).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[120, 66, 884, 199]]<|/det|>
+49. Laepple, T. & Huybers, P. Reconciling discrepancies between Uk37 and Mg/Ca reconstructions of Holocene marine temperature variability. Earth Planet. Sci. Lett. 375, 418–429 (2013).50. Gittleman, J. L. & Kot, M. Adaptation: Statistics and a Null Model for Estimating Phylogenetic Effects. Syst. Biol. 39, 227–241 (1990).51. Paradis, E., Claude, J. & Strimmer, K. APE: Analyses of Phylogenetics and Evolution in R language. Bioinformatics 20, 289–290 (2004).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[108, 135, 802, 615]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 664, 904, 871]]<|/det|>
+Extended Data Figure 1 | Summer and Annual Temperatures Spectra - Comparing the average spectra at the location of pollen records for the mean summer temperature (dashed) and the mean annual temperatures (solid). \(90\%\) confidence intervals are given (shaded). The IPSL and ECHAM5 model results exhibit a slightly lower variability in their annual temperature than in their summer temperature over all timescales, except for the longest timescale since it is dominated by leaked power from the Earth's orbital precession which mainly affects summer temperature in the Northern Hemisphere during the Holocene. On the other hand, TraCE- 21ka generally shows a slightly higher variability in its annual compared to its summer temperature. Although the pollen- based reconstructions calibrated for annual temperature are thought to be less reliable than the summer temperature reconstructions, they give a very similar result. This shows that our conclusions are robust against uncertainties in the seasonal attribution of pollen variability.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[108, 108, 802, 545]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 586, 904, 775]]<|/det|>
+Extended Data Figure 2 | Comparison of pollen- based precipitation reconstructions and climate models. Same as Extended Data Fig. 1, but for precipitation instead of summer temperature. While most locations should reflect temperature, here we also tested the boundary case of assuming that all sites reflect precipitation. Even in this extreme case, the main results hold, namely increasing climate variability over land as a function of timescale and a corresponding deficit of variability in the climate models. In all cases, there was little difference between the estimates with (dashed) and without (solid) sinusoidal detrending. The three climate models vastly disagree in terms of the amplitude of precipitation variability, but they all show a large deficit of variability at long timescales compared to the pollen- based reconstructions. Thus, even if precipitation would affect parts of our records, this cannot reconcile the model- proxy discrepancy.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[105, 60, 808, 545]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 590, 888, 724]]<|/det|>
+Extended Data Figure 3 |Spectral estimates from tree ring series. Spectral estimates of land temperature variability obtained from the pollen- based reconstructions and dendrochronological results from tree ring width (TRW) and maximum latewood density (MXD) measurements are shown (solid). The average instrumental spectral estimates at the corresponding locations for each dataset are shown alongside for reference (dashed). See Supplementary Information for details about the data, TRW and MXD differences, and interpretation. \(90\%\) confidence intervals are given (shaded)
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[122, 108, 884, 277]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 348, 894, 441]]<|/det|>
+Extended Data Figure 4 | Variance ratio between late and mid- Holocene temperature variability. Shown are the \(\log_{2}\) of the variance ratios \(\alpha\) (i.e. the number of doublings) for the 1000- 2000 years timescale band of the late Holocene (4ka- 0ka BP) and for the mid- Holocene (8ka- 4ka). A positive \(\log_{2}\alpha\) implies that the later period, where human impacts should be more important, was more variable.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[90, 65, 890, 540]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[90, 572, 900, 760]]<|/det|>
+Extended Data Figure 5| Comparison of millennial scale variability in the reconstructions and models. a- c Millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \mathrm{year}}\) (mean PSD for the timescale band 1000- 3000 years) for the pollen- based reconstructions after smoothing with a Gaussian kernel with a characteristic scale of \(300 \mathrm{km}\) (see Methods). d- I Millennial temperature variability \(\mathrm{PSD}_{1000 - 3000 \mathrm{year}}\) for the three climate models (without smoothing). The results with different detrending methods before computing the power spectra are compared: d,g,j without detrending, e,h,k with linear detrending and f,i,l with a 23- ka sinusoidal detrending. The same colour scale is used for all maps. While the typical pollen- based variability would be generally between 10 and \(1000 \mathrm{K}^2 \mathrm{year}^{- 1}\) , the models yields a variability 1- 2 orders of magnitude smaller, between 0.1 and \(10 \mathrm{K}^2 \mathrm{year}^{- 1}\) , with some regions reaching up to \(100 \mathrm{K}^2 \mathrm{year}^{- 1}\) .
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[95, 70, 900, 435]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 470, 904, 585]]<|/det|>
+Extended Data Figure 6 | Location of the records contributing to each spectral bin. a. Shown are the location of individual pollen records which were considered in the analysis. The colours correspond to those in Fig. 3, indicating which records are included in the binning of each spectrum. b. Shown are the grid points of the instrumental dataset which are near pollen records. The colours also indicate the corresponding bins as in a and Fig. 3. Each of the 8 non- overlapping bins contain 50 grid points based on the sub- decadal variability (see Methods).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[95, 67, 888, 472]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 510, 893, 585]]<|/det|>
+Extended Data Figure 7 | Spectral estimates of land temperature as a function of sub- decadal variability using twenty bins. As in Fig.3, but using twenty bins instead of eight. While the result is noisier than in Fig. 3, the correlation is still highly significant, which shows that our result is not sensitive to the number of bins.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[106, 60, 808, 545]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[91, 592, 900, 705]]<|/det|>
+Extended Data Figure 8 | Comparison of pollen- based reconstructions based on their statistical significance. A total of 985 pollen records which provided spectral estimates were tested for statistical significance and separated into ‘significant locations’ (p<0.1) and ‘not significant locations’ (p>0.1). The resulting average spectra overlap and are fairly similar over a wide range of timescales. The average spectra of the instrumental data at the corresponding locations are shown alongside for reference (dashed). 90% confidence intervals are given (shaded).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 44, 143, 70]]<|/det|>
+## Figures  
+
+<|ref|>image<|/ref|><|det|>[[45, 95, 456, 650]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 684, 115, 703]]<|/det|>
+<center>Figure 1 </center>  
+
+<|ref|>text<|/ref|><|det|>[[41, 724, 950, 950]]<|/det|>
+Average spectral estimates of local land temperature over the Northern Hemisphere. a, Average spectral estimates of land air temperature from pollen- based reconstructions, instrumental data and model simulations extracted at the pollen record locations. Also shown are the spectra estimated after applying a 23- ka sinusoidal detrending (dashed, see Methods). \(90\%\) confidence intervals are given (shaded). The number of pollen records contributing to each timescale is indicated below (brown axis). b, Average spectral estimates from reconstructed annual sea- surface temperature derived from marine archives5, and from instrumental data at the corresponding locations. The pollen- based and instrumental average spectra from a are reproduced. Also shown are linear combinations of power- laws with slope \(\beta = 1.2\) and white noise series (dashed green for land, dashed blue for sea and dashed grey for the white noise levels) corresponding to the energy- balance model approximation.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[44, 45, 875, 644]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 670, 118, 688]]<|/det|>
+<center>Figure 2 </center>  
+
+<|ref|>text<|/ref|><|det|>[[41, 710, 951, 937]]<|/det|>
+Spatial patterns of temperature variability in instrumental data and pollen- based reconstructions. a, Map of millennial variability estimated from the temperature spectra of pollen- based reconstructions as the mean PSD over the 1000- 3000 year timescale band. The PSD shown in the background are smoothed using Gaussian weights with a 300 km scale. Overlaid circles are the average of all pollen records closest to the corresponding instrumental grid point. b, Map of the multi- decadal scaling exponent \(\beta\) from the spectra of instrumental temperature records fitted over the 10- 60 year timescale band. c, Map of sub- decadal variability, mean PSD over the 2- 10 year timescale band, estimated from the spectra of instrumental temperature records. In a- c, the circles indicate the instrumental grid points which were the closest to pollen records. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 943, 87]]<|/det|>
+the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  
+
+<|ref|>image<|/ref|><|det|>[[45, 92, 888, 565]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[42, 584, 117, 603]]<|/det|>
+<center>Figure 3 </center>  
+
+<|ref|>text<|/ref|><|det|>[[40, 626, 955, 806]]<|/det|>
+Spectral estimates of land temperature as a function of sub- decadal variability. a, Spectral estimates of the instrumental temperature \((\Delta t< 60\) years) and of the pollen- based reconstructions \((\Delta t > 100\) years) binned according to sub- decadal instrumental variability (See Methods and Extended Data Fig.7). Dashed lines indicate estimated instrumental multi- decadal scaling exponents \(\beta 10 - 60\) years. \(90\%\) confidence intervals are given (shaded). b, Relation of the instrumental sub- decadal temperature variability PSD2- 10 years and the pollen- based millennial temperature variability PSD1000- 3000 years. The size of the points is proportional to the errors on the millennial variability estimates. c, As in b, but between \(\beta 10 - 60\) years and PSD1000- 3000 years.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 829, 311, 857]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 880, 764, 900]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[61, 919, 226, 937]]<|/det|>
+- ExtDataFig2.pdf
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[59, 45, 616, 172]]<|/det|>
+- HebertHerzschuhLaepple2021.odt- HebertHerzschuhLaepple2021.odt- SupplementaryInformationHebertHerzschuhLaepple2021.odt- SupplementaryInformationHebertHerzschnhLaepple2021.pdf- ExtDataFig8.pdf
+
+<--- Page Split --->

preprint/preprint__9888a9f622f768075271e80afca4b2931de0af2d78287779b00526c00cd54162/images_list.json

@@ -0,0 +1,55 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1",
+    "footnote": [],
+    "bbox": [
+      [
+        52,
+        50,
+        940,
+        479
+      ]
+    ],
+    "page_idx": 19
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3",
+    "footnote": [],
+    "bbox": [
+      [
+        55,
+        150,
+        741,
+        884
+      ]
+    ],
+    "page_idx": 20
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4",
+    "footnote": [],
+    "bbox": [
+      [
+        50,
+        45,
+        480,
+        500
+      ]
+    ],
+    "page_idx": 22
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 22
+  }
+]

preprint/preprint__9888a9f622f768075271e80afca4b2931de0af2d78287779b00526c00cd54162/preprint__9888a9f622f768075271e80afca4b2931de0af2d78287779b00526c00cd54162.mmd

@@ -0,0 +1,442 @@
+
+# Quantifying how single dose Ad26.COV2.S vaccine efficacy depends on Spike sequence features  
+
+Craig Magaret Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 5056- 2664  
+
+Li Li Fred Hutchinson Cancer Center  
+
+Allan deCamp Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 1404- 4322  
+
+Morgane Rolland MHRP- HJF https://orcid.org/0000- 0003- 3650- 8490  
+
+Michal Juraska Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 0920- 2915  
+
+Brian Williamson Kaiser Permanente Washington Health Research Institute https://orcid.org/0000- 0002- 7024- 548X  
+
+James Ludwig Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 3500- 3490  
+
+Cindy Molitor Fred Hutchinson Cancer Center  
+
+David Benkeser Emory https://orcid.org/0000- 0002- 1019- 8343  
+
+Alex Luedtke University of Washington  
+
+Brian Simpkins Pitzer College https://orcid.org/0009- 0002- 1044- 0053  
+
+Lindsay Carpp Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 0333- 5925  
+
+Hongjun Bai WRAIR https://orcid.org/0000- 0002- 3501- 3974  
+
+Bethany Dearlove Walter Reed Army Institute of Research https://orcid.org/0000- 0003- 3653- 4592  
+
+Alexander Greninger University of Washington  
+
+Pavitra Roychoudhury University of Washington  
+
+Jerald Sadoff
+
+<--- Page Split --->
+
+Janssen Research & Development, LLC https://orcid.org/0000- 0002- 4839- 3013  
+
+Glenda Gray South African Medical Research Council  
+
+Sanne Roels Janssen R&D  
+
+An Vandebosch Janssen R&D  
+
+Daniel Stieh Janssen Vaccines & Prevention BV  
+
+Mathieu Le Gars Janssen Vaccines and Prevention B.V.  
+
+Johan Vingerhoets Janssen Pharmaceutica N.V., Beerse, Belgium https://orcid.org/0000- 0001- 5939- 9501  
+
+Beatriz Grinsztejn Evandro Chagas National Institute of Infectious Diseases- Fundacao Oswaldo Cruz  
+
+Paul Goepfert Department of Medicine, Division of Infectious Diseases, University of Alabama at Birmingham  
+
+Carla Truyers Janssen Pharmaceutica N.V., Beerse, Belgium  
+
+Ilse Van Dromme Janssen R&D, a division of Janssen Pharmaceutica NV  
+
+Edith Swann NIAID/NIH  
+
+Mary Marovich National Institute of Allergy and Infectious Diseases  
+
+Dean Follmann National Institutes of Health https://orcid.org/0000- 0003- 4073- 0393  
+
+Kathleen Neuzil University of Maryland School of Medicine https://orcid.org/0000- 0001- 9480- 2714  
+
+Lawrence Corey Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 2179- 2436  
+
+Ollivier Hyrien Fred Hutchinson Cancer Research Center  
+
+Leonardo Paiva de Sousa Evandro Chagas National Institute of Infectious Diseases- Fundacao Oswaldo Cruz https://orcid.org/0000- 0001- 9004- 5154  
+
+Martin Casapia Asociación Civil Selva Amazónica https://orcid.org/0000- 0002- 5972- 0948  
+
+Marcelo Losso
+
+<--- Page Split --->
+
+Hospital General de Agudos José María Ramos Mejia https://orcid.org/0000- 0002- 4273- 4833 Susan Little Department of Medicine, University of California, San Diego, CA 92903 https://orcid.org/0000- 0002- 7645- 9737  
+
+Aditya Gaur St. Jude Children's Research Hospital  
+
+Linda- Gail Bekker Desmond Tutu HIV centre https://orcid.org/0000- 0002- 0755- 4386  
+
+Nigel Garrett Centre for the AIDS Program of Research in South Africa (CAPRISA), University of KwaZulu- Natal, Durban, South Africa 4041 https://orcid.org/0000- 0002- 4530- 234X  
+
+Fei Heng University of North Florida https://orcid.org/0000- 0002- 6701- 7873  
+
+Yanqing Sun University of North Carolina at Charlotte  
+
+Peter Gilbert  
+
+pgi1bert@fredhutch.org  
+
+Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 2662- 9427  
+
+## Article  
+
+Keywords: Antibody- epitope escape score, COVID- 19 vaccine, ENSEMBLE trial, genetic distance, Hamming distance, neutralization resistance, SARS- CoV- 2, sieve analysis, vaccine efficacy, viral variants  
+
+Posted Date: May 31st, 2023  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 2743022/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: Yes there is potential Competing Interest. ALG reports contract testing from Abbott, Cepheid, Novavax, Pfizer, Janssen, and Hologic and research support from Gilead and Merck. JS declares support for the submitted work from the Janssen Pharmaceutical Companies of Johnson & Johnson and partial support (in the form of funding to his institution) from BARDA for the submitted work, declares support within the past 36 months from the Janssen Pharmaceutical Companies of Johnson & Johnson and BARDA funding for part of this work, has patents on invention of the Janssen COVID- 19 vaccine, and has Johnson & Johnson stock and stock options. SR had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution)
+
+<--- Page Split --->
+
+for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. AV had partial support from BARDA (in the form of contract payments to her institution) for the submitted work, had all patent rights transferred to Johnson & Johnson, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. DJS had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen. MLG had partial support from BARDA (in the form of contract payments to his institution) for the submitted work, has patents on invention of the Janssen COVID- 19 vaccine, has shares in Johnson & Johnson, and is an employee of Johnson & Johnson. JV has stock and stock options in Johnson and Johnson and is an employee of Janssen Pharmaceutica NV. CT and IVD both had partial support from BARDA (in the form of contract payments to their institution) for the submitted work, hold stock in Janssen Pharmaceuticals, and are employees of Janssen Pharmaceutica NV.  
+
+Version of Record: A version of this preprint was published at Nature Communications on March 11th, 2024. See the published version at https://doi.org/10.1038/s41467-024-46536-w.
+
+<--- Page Split --->
+
+## Abstract  
+
+It is of interest to pinpoint SARS- CoV- 2 sequence features defining vaccine resistance. In the ENSEMBLE randomized, placebo- controlled phase 3 trial, estimated single- dose Ad26.COV2.S vaccine efficacy (VE) was \(56\%\) against moderate to severe- critical COVID- 19. SARS- CoV- 2 Spike sequences were measured from 484 vaccine and 1,067 placebo recipients who acquired COVID- 19 during the trial. In Latin America, where Spike diversity was greatest, VE was significantly lower against Lambda than against Reference and against all non- Lambda variants [family- wise error rate (FWER) \(\mathsf{p}< 0.05]\) . VE also differed by residue match vs. mismatch to the vaccine- strain residue at 16 amino acid positions (4 FWER \(\mathsf{p}< 0.05\) ; 12 q- value \(\leq 0.20\) ). VE significantly decreased with physicochemical- weighted Hamming distance to the vaccine- strain sequence for Spike, receptor- binding domain, N- terminal domain, and S1 (FWER \(\mathsf{p}< 0.001\) ); differed (FWER \(\leq 0.05\) ) by distance to the vaccine strain measured by 9 different antibody- epitope escape scores and by 4 NTD neutralization- impacting features; and decreased ( \(\mathsf{p} = 0.011\) ) with neutralization resistance level to vaccine recipient sera. VE against severe- critical COVID- 19 was stable across most sequence features but lower against viruses with greatest distances. These results help map antigenic specificity of in vivo vaccine protection.  
+
+## Main Text  
+
+Initial SARS- CoV- 2 vaccine candidates were based on the virus's original lineage, as represented by the Wuhan- Hu- 1 index strain with Spike D614 (NC_045512). As the virus has evolved, \(^{1 - 4}\) efficacy of these vaccines against symptomatic infection has waned, \(^{5,6}\) and new vaccine inserts have been developed.  
+
+Based on data from a randomized, placebo- controlled vaccine efficacy (VE) trial on clinical outcomes and pathogen sequences isolated from participants experiencing clinical outcomes, sieve analysis assesses how VE depends on pathogen sequence features. \(^{7,8}\) Pajon et al. \(^{9}\) and Sadoff et al. \(^{10}\) showed how the VE against symptomatic COVID- 19 was lower against certain variants than against the Reference strain in the phase 3 COVE trial of two doses of Moderna's mRNA- 1273 vaccine and the phase 3 ENSEMBLE trial of a single dose of Janssen's Ad26.COV2.S vaccine, respectively. [As in ref. \(^{10}\) , Reference is defined as the basal outbreak lineage B.1, which bears the D614G mutation.] Cao et al. showed that VE was higher in COVID- 19 VE trials where circulating viruses had shorter Spike sequence Hamming distances to the vaccine strain. \(^{11}\) These sieve analyses only considered Spike viral variation defined by the WHO- defined variant category or the unweighted Spike protein distance. They did not assess how VE depends on other Spike sequence features, such as at the level of individual mutations or features that impact immunological functions such as anti- SARS- CoV- 2 neutralization, \(^{12 - 17}\) relevant given the strong evidence of neutralizing antibodies as a cross- platform correlate of protection. \(^{18 - 20}\)  
+
+We report here the results of a sieve analysis of the ENSEMBLE trial, which enrolled over 40,000 participants and was conducted in the US, South Africa, and six countries in Latin America. The sieve analysis considers baseline SARS- CoV- 2 seronegative per- protocol participants and the primary endpoint (moderate to severe- critical COVID- 19), as well as the severe- critical COVID- 19 endpoint, during the
+
+<--- Page Split --->
+
+double- blinded period of follow- up. We focus the main text on the Latin America results given the greatest information for sieve analysis as noted below.  
+
+## Results  
+
+## SARS-CoV-2 sequence data  
+
+A total of 1,345 SARS- CoV- 2 Spike amino acid sequences were obtained from 1,224 participants experiencing the moderate to severe- critical primary endpoint. All sequences were variant- typed to either the Reference lineage or to one of nine different WHO- defined variants (Fig. 1A) (Table S5). Lineages that circulated at the beginning of the study period, e.g., Reference, were closer to the sequence from the vaccine insert than later emerging lineages, with Lambda the most distant (Fig. 1B- C).  
+
+## Greater SARS-CoV-2 Spike diversity in Latin America than in South Africa and the US  
+
+Most sequences were obtained from participants in Latin America \((n = 776)\) with additional sequences from the US \((n = 323)\) and South Africa \((n = 125)\) (Table S6). Five main variants circulated in Latin America (Reference, Zeta, Gamma, Lambda, Mu), while the South African sequences were \(76\%\) Beta and \(17\%\) Delta, and the US sequences were \(85\%\) Reference (Fig. 1A). There was greater Spike AA sequence diversity in Latin America compared to South Africa and the US (Rao's \(Q = 10.1\) vs. 7.7 vs. 3.3, respectively; Fig. S1).  
+
+The succession of distinct co- circulating variants in Latin America and the resulting broadest dynamic range of inter- individual sequence diversity, and the greatest number of COVID- 19 endpoints, implies that sieve analyses of the Latin America region have the greatest statistical power. In contrast, the domination of the Reference lineage in the US and the Beta and Delta lineages in South Africa constrained the sequence diversity's dynamic range and limited the power of these sieve analyses. Therefore, we focus on the results from Latin America, with the US and South Africa results reported in the Supplementary Materials.  
+
+## Differential vaccine efficacy against COVID-19 by SARS-CoV-2 lineage  
+
+All reported results on VE by SARS- CoV- 2 features are based on feature- specific proportional- hazards models \(^{21,22}\) (see the SAP). Figure 2A shows VE against the primary COVID- 19 endpoint caused by the Reference, Gamma, Zeta, Lambda, and Mu lineages, and Fig. 2B shows VE against the primary COVID- 19 endpoint caused by the groupings of all other lineages excluding each individual lineage ("not- lineage"). Figure 2C shows differential VE against pairs of lineages or against pairs of lineage vs. not- lineage. VE was significantly higher against Reference than against Lambda and against not- Reference lineages [family- wise error rate (FWER) \(p < 0.05\) ]. It was also significantly higher against not- Lambda vs. Lambda
+
+<--- Page Split --->
+
+and against Zeta vs. Lambda (FWER \(\mathsf{p}\leq 0.05\) ), and higher against Reference vs. Gamma, Reference vs. Mu, Zeta vs. Gamma, and Zeta vs. Mu (q- value \(\leq 0.20\) ).  
+
+## Vaccine efficacy greater against COVID-19 caused by SARS-CoV-2 genotypes defined by individual Spike AA position residues matching the vaccine strain  
+
+We scanned across all Spike AA positions with sufficient residue variability (at least 20 endpoints with a vaccine- mismatched residue: \(\mathsf{n} = 37\) positions). VE significantly differed (q- value \(\leq 0.20\) ) by residue match vs. mismatch to the vaccine strain residue at 16 positions (Fig. 2D; 4 positions with FWER \(\mathsf{p}\leq 0.05\) : 75, 76, 253, 490). Similarly, when assessing the presence or absence of specific residues at each AA position, VE significantly differed (q- value \(\leq 0.20\) ) for 38 residues (75V vs. not- 75V and 76l vs. not- 76l with FWER \(\mathsf{p}\leq 0.05\) ) at the same 16 positions. Figure S4 shows the distributions of residues at these 16 positions. Thirteen of these 16 AA sites (Fig. 2D) were sites harboring characteristic mutations of the Lambda variant and not for any other variants, and very highly covaried with Lambda vs. not- Lambda (Fig. S5, Mstar \(^{23} > 0.85\) ), thereby providing nearly equivalent signatures of differential VE captured by Lambda vs. not- Lambda. The full results of the covariability analysis are in the Supplementary Materials.  
+
+Four of the 1277 analyzed Spike positions (417, 452, 484, 490) were pre- specified as being hypothesized to impact neutralization based on an association with a reduced neutralizing antibody response in mRNA vaccine recipients, \(^{24 - 26}\) or evidence for increased transmissibility (452) \(^{24}\) or increased infectivity in vitro (452, 490). \(^{24,26,27}\) Of these sites, positions 452 and 490 were found to significantly impact VE (FWER \(\mathsf{p}\leq 0.05\) ).  
+
+Figures S2B, S3B, and S6 provide complete results including by geographic region.  
+
+## Vaccine efficacy against COVID-19 decreases with increasing protein distance to the vaccine-strain in Spike, receptor-binding domain (RBD), N-terminal domain (NTD), and S1  
+
+VE significantly decreased with physicochemical- weighted Hamming distance (between the observed vs. vaccine insert sequence) for Spike, RBD, NTD, and S1 (Fig. 3, FWER \(\mathsf{p}< 0.001\) ) but not for S2 ( \(\mathsf{p} = 0.78\) ). Against viruses with shortest Spike distances (average 6 residue mismatches), VE was \(69\%\) (95% CI: 60- 76%), and against viruses with 25th, 50th, 75th, and 95th percentile Spike distances (average 8.1, 12.9, 17.8, 18.6 residue mismatches), VE was \(64\%\) (56%, 71%), 52% (44%, 58%), 34% (19%, 46%), and 30% (13%, 44%), respectively. The median distances of sequences for vaccine: placebo were 15.0:9.5 for Spike, 2.6:1.0 for RBD, 4.0:1.6 for NTD, 11.7:6.2 for S1, and 3.1:3.2 for S2. Tables S7 and S8 show inferences about differences in mean distances of vaccine vs. placebo sequences. Figs. S7- S11 and Table S9 provide complete results including by geographic region, where Table S9 shows that VE decreased with weighted Hamming distance for RBD, NTD, and S1 in the US (q- value \(\leq 0.20\) ).  
+
+By lineage, ordered by placebo arm COVID- 19 endpoint Spike distance to the vaccine strain, Reference viruses had 6.0- 17.7 residue mismatches, Zeta 8.1- 22.1 mismatches, Epsilon 10.7 mismatches, Mu 12.2- 16.8 mismatches, Alpha 14.5- 16.8 mismatches, Gamma 16.7- 20.2 mismatches, and Lambda
+
+<--- Page Split --->
+
+17.2–27.7 mismatches. This ordering of lineages by protein distance matches the ordering of the VE estimates by lineage category, suggesting that overall Spike evolution is a reasonable metric capturing VE decline with variant. The results are generally similarly ordered for the RBD, NTD, and S1 distances (Fig. S12).  
+
+## Vaccine efficacy against COVID-19 decreases with increasing spike antibody-escape score to the vaccine-strain  
+
+Neutralization- relevant RBD features were defined where mutations impact binding in deep mutational scanning (DMS) experiments<sup>28</sup> (see Supplementary Materials. Escape scores were defined for whole- RBD and for each of 10 epitope- specific clusters of AA sites (see Methods), labeled DMS (whole- RBD) and DMS1 through DMS10. Vaccine efficacy significantly decreased (q- value \(\leq 0.20\) ) with each of the DMS, DMS2, DMS6, DMS7, and DMS8 escape scores (FWER \(\mathsf{p}\leq 0.05\) ) as well as for DMS1, DMS5, DMS9 (q- value \(\leq 0.20\) and FWER \(>0.05\) ) (Table S12). Tables S10 and S11 show inferences about differences in mean escape scores of vaccine vs. placebo sequences.  
+
+Alternatively, we defined putative antibody footprint site sets (including whole Spike) based on structures of SARS- CoV- 2 in complex with antibodies available from the PDB. Each sequence was assigned an escape score based on a class of epitopes (see Supplementary Materials). These features are referred to as PDB1 through PDB14, with the first 12 clusters in the RBD and PDB13 and PDB14 in the NTD. Vaccine efficacy significantly decreased (q- value \(\leq 0.20\) ) with the escape scores for PDB4, PDB7, PDB8, and PDB13 (FWER \(\mathsf{p}\leq 0.05\) ) as well as for PDB1 and PDB3 (q- value \(\leq 0.20\) and FWER \(>0.05\) ) (Table S15). Tables S13 and S14 show inferences about differences in mean escape scores of vaccine vs. placebo sequences.  
+
+To interpret the DMS and PDB results, we focus on the epitope- specific features with FWER \(\mathsf{p}\leq 0.05\) that carry the greatest amount of independent information based on inter- correlation and hierarchical clustering analysis (Supplementary Text, Figs. S13 and S14): DMS2, PDB7, PDB8, and PDB13. The sieve analysis results are similar across these four features, with estimated VE at \(60 - 70\%\) against viruses with escape score zero and decreasing to \(0\% - 20\%\) against viruses with maximum escape score. PDB8 and PDB13 rank highest for discriminating VE with slightly greater span of VE point estimates over the range of escape scores (spans \(20 - 60\%\) , \(16 - 60\%\) , \(21 - 69\%\) , and \(1 - 57\%\) for DMS2, PDB7, PDB8, and PDB13, respectively) (Fig. 4A- D). Figure 4E lists the Spike AA residues in each epitope footprint and the visualizations in Fig. 4F- I show the positions comprising the four antibody epitope footprints on a Spike monomer structure. Figures S15- S23 and S24- S30 provide complete results for DMS and PDB features, respectively. Another reason PDB8 was highlighted is its balanced contacts across the whole receptor- binding motif (RBM) whereas the other RBM- specific clusters (PDB1- PDB6) are more tightly grouped within a region of the RBM. Among the non- RBM focusing antibodies (PDB7, PDB9- PDB14), PDB7 and PDB13 correspond to the most accessible sites on Spike in a closed prefusion trimer (Fig. S31) and these sites are relatively variable among SARS- CoV- 2 sequences.
+
+<--- Page Split --->
+
+## Lower vaccine efficacy against COVID-19 with NTD features hypothesized to abrogate neutralization  
+
+Lower vaccine efficacy against COVID- 19 with NTD features hypothesized to abrogate neutralizationSeven dichotomous NTD features (see Supplementary Materials) were assessed for a sieve effect as for vaccine- match vs. vaccine- mismatch binary features. Six of the 7 NTD features significantly impacted VE (q- value \(\leq 0.20\) ): NTD4, NTD6, NTD1, NTD3, NTD5, and NTD7 (where the last four also had FWER \(p \leq 0.05\) ) (Fig. 5). Figure S32 shows the spatial locations in the NTD of the features that impacted VE (FWER \(p \leq 0.05\) ).  
+
+## Vaccine efficacy greater against lineages with lower variant-neutralization resistance to Ad26.COV2.S vaccine recipient sera  
+
+All of the sieve analyses study how VE depends on Spike AA features except one: a neutralization sieve analysis that scores each virus's lineage by its experimentally measured sensitivity to neutralization by Ad26. COV2. S vaccine sera. \(^{29,30}\) VE decreased with this variant- neutralization resistance score ( \(p = 0.011\) ) (Fig. 5B). Under one model for the neutralization assay being a perfect correlate of protection, the estimates of VE for each of the five lineages would fall on the curve of VE by variant- neutralization resistance score. Lambda had evidence of deviating from the curve, with VE \(55\%\) (48, \(62\%\) ) based on its measured neutralization sensitivity compared to VE \(11\%\) (- 35, \(41\%\) ) based on direct analysis of Lambda ignoring neutralization data. In contrast, the weighted Hamming distance analyses yielded VE estimates at Lambda- variant distance values that are closer to the VE \(11\%\) figure.  
+
+Figure S33 provides complete results by geographic region.  
+
+## Multivariable virus features as predictors of treatment arm  
+
+A variable importance measure analysis by ensemble machine learning \(^{31}\) of COVID- 19 endpoint cases compared how well AA sequence features predicted treatment arm (results in Fig. S34 and the Supplementary Text).  
+
+## Assessing the severe-critical COVID 19 endpoint  
+
+Differential VE against severe- critical COVID- 19 by lineage could only be assessed for Latin America, with VE of \(83\%\) (64, \(92\%\) ) against Reference, \(64\%\) (26, \(83\%\) ) against Gamma, \(94\%\) (- 27, \(100\%\) ) against Zeta, \(62\%\) (- 31, \(89\%\) ) against Lambda, and \(84\%\) (42, \(96\%\) ) against Mu (Table S16). There was no evidence of variation in VE across the lineages ( \(p = 0.50\) ) (Table S16, S17). The estimates of VE were similar/stable across AA positions with vaccine- matched vs. vaccine- mismatched residue, with all unadjusted p- values for differential VE above 0.05 (Fig. S35). For the key positions 452 and 490 found to show sieve effects for the primary COVID- 19 endpoint, the results for the severe- critical COVID- 19 endpoint were VE \(79\%\) (68, \(87\%\) ) against 452- matched virus compared to VE \(70\%\) (3, \(91\%\) ) against 452- mismatched virus ( \(p = 0.58\)
+
+<--- Page Split --->
+
+for difference), and VE \(80\%\) (68, \(87\%\) ) against 490- matched virus compared to VE \(62\%\) (- 31, \(89\%\) ) against 490- mismatched virus ( \(\mathrm{p} = 0.34\) for differential VE). For the DMS antibody escape score distances, the data support stable VE across the distances (Table S18). Similarly, the data support stable VE across RBD and PDB Spike- antibody escape scores (Table S19). VE was stable by variant- neutralization resistance score, with \(\mathrm{VE} = 84\%\) (67%, \(92\%\) ) for the most sensitive lineage (ancestral) and \(\mathrm{VE} = 73\%\) (50, \(85\%\) ) for the least sensitive lineage (Mu) ( \(\mathrm{p} = 0.33\) , Fig. S36).  
+
+## Vaccine efficacy against severe–critical COVID-19 decreases with increasing protein distance to the vaccine-strain and by NTD features hypothesized to abrogate neutralization  
+
+There was a trend of VE against severe–critical COVID- 19 decreasing with the weighted Hamming distance for the Spike, NTD, and S1 regions (q- values \(= 0.20\) ) (Table S20, Figs. S37, S39, S40). The point estimates of VE suggested moderate declines of VE with distances. For example, the VE for Spike was \(87\%\) (71%, \(94\%\) ) against viruses with shortest distance of 6 and \(66\%\) (34%, \(83\%\) ) against viruses with long distance of 20 ( \(\mathrm{p} = 0.12\) ). Figs. S37- S41 and Table S20 provide complete information by geographic region. In addition, while VE was stable across levels of NTD1 through NTD4 ( \(\mathrm{p} > 0.20\) ), it differed by levels of NTD5, NTD6, and NTD7, with VE of \(61\%\) (31, \(78\%\) ) vs. \(88\%\) (76, \(94\%\) ) for the two NTD5 genotypes ( \(\mathrm{q} = 0.10\) for difference), VE of \(60\%\) (20, \(80\%\) ) vs. \(84\%\) (72, \(91\%\) ) for the two NTD6 genotypes ( \(\mathrm{q} = 0.12\) for difference), and VE of \(64\%\) (32, \(80\%\) ) vs. \(85\%\) (73, \(92\%\) ) for the two NTD7 genotypes ( \(\mathrm{q} = 0.12\) for difference) (Table S21).  
+
+## Discussion  
+
+Sieve analysis compares genotype- specific or immunophenotype- specific COVID- 19 incidence between randomized study groups, therefore directly assessing causal effects of vaccination and providing inferences for how vaccine efficacy depends on SARS- CoV- 2 features. In addition to the strength of a randomized, double- blinded placebo- controlled phase 3 trial, the present sieve analysis of ENSEMBLE had ample statistical precision due to the large number of SARS- CoV- 2 Spike sequences (measured from more than 1,200 participants) and the broad proteomic variability of the SARS- CoV- 2 Spike sequences causing these endpoints. Consequently, the sieve analysis could provide many insights into how the efficacy of the Ad26. COV2. S vaccine, evaluated in baseline SARS- CoV- 2 negative individuals, depended on virus features.  
+
+In the Latin American cohort, VE against the moderate to severe–critical COVID- 19 primary endpoint significantly declined with Spike sequence distance as measured in myriad ways, including lineage, weighted Hamming distances calculated for Spike, RBD, NTD, and S1, scores reflecting degree of escape from epitope- specific antibodies computed using deep mutational scanning or based on crystal structures in the Protein Data Bank (PDB), and NTD features previously shown to impact neutralization. Estimates of VE by lineage were consistently ordered by the distances of the different lineages to the vaccine strain. VE declined similarly with Spike, RBD, NTD, and S1 distances (VE about \(70\%\) against viruses closest to the vaccine and \(20\%\) against viruses beyond the 90- 95th percentile of distances) but
+
+<--- Page Split --->
+
+did not depend on S2 distances. This may be explained by S2's relative conservation when compared to S1. As such, almost all variant- characteristic mutations are not in S2, and none of the prescribed antibody epitope footprint clusters included S2 positions (only rare epitopes in PDB mapped to S2), reflecting S2's 'stalk' location and relative lack of exposure to the immune system.  
+
+VE significantly declined with 14 of the 20 evaluable antibody epitope escape scores. Six antibody- epitope clusters had no evidence of impacting VE: DMS3, PDB2, PDB5, PDB6, PDB9, PDB14. Of the 14 clusters with a sieve effect, 9 include at least one site that harbors a characteristic mutation of Lambda, whereas 3 include site 417 which is a characteristic mutation of Mu and Gamma, 1 includes site 501 that harbors a characteristic mutation of Gamma, Alpha, and Mu, and 1 includes both sites 417 and 501. Thus the 9 sieve- effect clusters appear to be driven by the differential VE by Lambda vs. not- Lambda, whereas the other 5 appear to be driven by mutations at the important sieve- effect sites 417 and 501 that impact neutralization. Of the 6 non- sieve- effect clusters, only one (PDB14) included a site harboring a characteristic mutation of Lambda, site 75, which was a sieve- effect site with FWER \(p \leq 0.05\) . The potential for sieve effects in different epitope sets depends on many factors including level of accessibility to neutralizing antibodies, conservation, and the narrowness of the footprints on the tridimensional structure they target (Fig. S31).  
+
+Neutralizing antibody assays have performed well at predicting vaccine efficacy against COVID- 19 and severe- critical COVID- 19 across SARS- CoV- 2 lineages. \(^{19,20,32}\) Importantly, one of the sieve analyses in the present work scored viruses by their lineage's directly measured resistance to neutralization by sera from ENSEMBLE Ad26. COV2. S vaccine recipients, providing a way to study a neutralization correlate of protection (CoP) in a complementary way to individual- level and population- level immune correlates analyses. \(^{33 - 35}\) VE significantly declined against lineages with greater neutralization resistance scores, providing validation of pseudovirus neutralization titer as a CoP. However, the lineage scores were estimated from only eight ENSEMBLE vaccine recipients, albeit the scores are supported by additional data from 17 Ad26. COV2. S vaccine recipients in the COV2001 phase 1/2a study. \(^{36}\)  
+
+The relative prevalence of SARS- CoV- 2 lineages changed over time (Fig. 1A and Fig. 1 of ref. \(^{10}\) ) where in Latin America the median (range) number of days from enrollment until the COVID- 19 endpoint among placebo recipients was 48 (15, 197) for Reference, 45 (15, 141) for Zeta, 114 (42, 220) for Gamma, 126 (57, 204) for Lambda, and 170 (109, 219) for Mu. If newer variants tended to expose participants later in follow- up than older variants it could cause spurious genotypic sieve effects that are instead due to waning vaccine efficacy. This potential bias was mitigated by controlling for calendar time of enrollment in the sieve analyses.  
+
+The Ad26. COV2. S vaccine sieve effects observed here, based on data collected prior to July 10, 2021, revealed broader vaccine adaptation features as several sieve signature sites showed mutations in subsequent variant waves. Hence, mutations at sites 452, 484 and 501 are dominant in currently circulating Omicron sub- lineages [global proportion between 2022- 12- 04 and 2022- 12- 10: L452R = 87.2%, E484A = 98.5%, N501Y = 99.2% \(^{37}\) ]. While the sieve signature F490S had been rare until the end of 2022,
+
+<--- Page Split --->
+
+this mutation became dominant in early 2023 with the global rapid spread of XBB.1.5 variants. The fact that sieve analysis predicted currently relevant mutations could be expected since SARS- CoV- 2 has shown remarkable patterns of convergent evolution since the initial appearance of variants, with numerous recurrent mutations, especially in the RBD, shared across lineages over time. \(^{38}\)  
+
+A strength of this study was it was conducted in three separate geographic regions with different circulating lineages, which contribute insights based on these lineages and their characteristic signature mutations, and different distributions of genetic distances of circulating sequences to the vaccine strain. The analyses of Latin American study sites provided the greatest insights given that \(63\%\) of primary COVID- 19 endpoints with sequence data were in Latin America where the circulating SARS- CoV- 2 sequences were the most diversified. All features showing sieve effects in the US also showed sieve effects in Latin America, constituting independent replication of results. The result of no sieve effects in South African study sites can likely be explained by the vast majority of circulating sequences being Beta or Delta variants with limited dynamic range of genetic distances within each variant and a lack of Reference viruses that are close to the vaccine strain.  
+
+Another strength of this study was that VE against severe- critical COVID- 19 could be assessed. The results support that VE against this endpoint also declines with Spike sequence distance as measured in multiple ways, yet with VE starting higher against viruses closest to the vaccine strain and diminishing less rapidly with increasing degrees of sequence mismatch. Overall, the finding that protection against severe- critical COVID- 19 is more invariant to sequence changes than against less- symptomatic COVID- 19 may have clinical implications for planning updates of vaccines with new variants. The severe- critical classification covers a broad spectrum of clinical phenotypes ranging from individuals with only repeated low partial pressure of oxygen to severe pneumonia requiring respiratory support. Protection against hospitalization with severe consequences is clinically most important but sieve analysis specific to this outcome could not be performed given small numbers of cases. Yet, ENSEMBLE and post- approval trials have shown high Ad26. COV2. S efficacy against this outcome especially in South Africa after a 6- month boost, suggesting that neutralization resistance and sequence variation may be playing a less dominant role in vaccine- induced protection against the most serious disease, perhaps due to CD8 + T cells. \(^{39}\)  
+
+## Methods  
+
+## Trial design, study cohort, and COVID-19 endpoints  
+
+Trial enrollment began on September 21, 2020. The end of the double- blind period varied by country; the data cutoff for this analysis was July 9, 2021. The main endpoint for sieve analysis is the same COVID- 19 primary endpoint (moderate to severe- critical) as in the primary analyses, \(^{10,40}\) restricting to endpoints starting 14 days post vaccination. Sieve analyses were also conducted for severe- critical COVID- 19, again using the same definition as used in the primary papers. \(^{10,40}\) Analyses were conducted in the per
+
+<--- Page Split --->
+
+protocol baseline seronegative cohort. \(^{40}\) See Section 1 of the Statistical Analysis Plan (SAP, provided in ref. \(^{41}\) and as supplementary material) and the Supplementary Materials for further details.  
+
+## SARS-CoV-2 sequencing and sequence data  
+
+SARS- CoV- 2 Spike sequences were generated and variant- typed as described. \(^{40}\) Sequences were selected for analysis if they were obtained within 36 days following the first RNA- positive timepoint associated with the first moderate to severe- critical COVID- 19 primary endpoint. See the Supplementary Materials for further details.  
+
+## Neutralizing antibody titers  
+
+Neutralizing antibody titers were measured to a panel of Spike antigens representing the Reference strain B.1.D614G and several variants. \(^{29,30}\) Each variant was assigned a score defined as the log10- transformed ratio of geometric mean titer of vaccinee sera against the variant and the geometric mean titer of vaccinee sera against the Reference strain.  
+
+## Sieve analysis  
+
+This analysis was specified in advance and documented in the SAP. The sieve analyses were conducted for each of the four geographic regions: Latin America, South Africa, the US, and the three geographic regions pooled (hereafter, 'geographic- region analyses'). Details on specification of spike amino acid (AA) sequence features for sieve analysis, prospective vaccine efficacy sieve analysis, neutralization hypothesis- driven sieve analysis, and multiple hypothesis testing adjustment for AA sequence sieve analysis are provided in the Supplementary Materials.  
+
+Additional details on covariability analysis, quantification of viral diversity, antibody escape scores [deep mutational scanning (DMS) and Protein Data Bank (PDB)], variant- neutralization sensitivity score assigned to variants, handling of missing sequences, and structural modeling is also in the Supplementary Materials.  
+
+## Declarations  
+
+Acknowledgments: The authors thank Jesse Bloom for input in defining deep mutational scanning Spike sequence features for sieve analysis. We gratefully acknowledge all data contributors, i.e., the Authors and their Originating laboratories responsible for obtaining the specimens, and their Submitting laboratories for generating the genetic sequence and metadata and sharing via the GISAID Initiative, on which this research is based.  
+
+## Funding:
+
+<--- Page Split --->
+
+Administration for Strategic Preparedness and Response, Biomedical Advanced Research and Development Authority, Government Contract Nos. HHSO100201700018C with Janssen.  
+
+National Institute of Allergy and Infectious Diseases (NIAID) grant UM1 AI068635 (HVTN SDMC) (PBG), UM1 AI068614 (HVTN LOC) (LC), and R37AI054165 (PBG).  
+
+Intramural Research Program of the NIAID Scientific Computing Infrastructure at Fred Hutch, ORIP grant S100D028685.  
+
+Janssen Research and Development, an affiliate of Janssen Vaccines and Prevention and part of the Janssen pharmaceutical companies of Johnson & Johnson.  
+
+The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the Department of Health and Human Services or its components.  
+
+## Author contributions:  
+
+Conceptualization: CAM, AdC, MR, SR, AV, DJS, MLG, DF, PBG  Methodology: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, BS, AG, PR, OH, PBG  Software: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, BLD, AG, PR, OH, PBG  Validation: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, AG, PR, OH, PBG  Formal Analysis: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, AL, HB, AG, PR, OH, PBG  Investigation: JS, GEG, BG, SDR, DJS, MLG, JV, PAG, CT, IVD, ES, MAM, KMN, LC  Resources: JS, GEG, SDR, AV, DJS, MLG, JV, BG, PAG, LPdS, MC, MHL, SJL, AG, LGB, NG, CT, IVD, ES, MAM, KMN, LC  Data Curation: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, AG, PR, CT, OH, PBG  Visualization: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, LNC, AG, PR, OH, PBG  Funding acquisition: LC, PBG  Project administration: PBG  
+
+Writing - original draft: CAM, AdC, MR, LNC, PBG
+
+<--- Page Split --->
+
+Writing - review & editing: All  
+
+Competing interests: ALG reports contract testing from Abbott, Cepheid, Novavax, Pfizer, Janssen, and Hologic and research support from Gilead and Merck. JS declares support for the submitted work from the Janssen Pharmaceutical Companies of Johnson & Johnson and partial support (in the form of funding to his institution) from BARDA for the submitted work, declares support within the past 36 months from the Janssen Pharmaceutical Companies of Johnson & Johnson and BARDA funding for part of this work, has patents on invention of the Janssen COVID- 19 vaccine, and has Johnson & Johnson stock and stock options. SR had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. AV had partial support from BARDA (in the form of contract payments to her institution) for the submitted work, had all patent rights transferred to Johnson & Johnson, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. DJS had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. CT and IVD both had partial support from BARDA (in the form of contract payments to their institution) for the submitted work, hold stock in Janssen Pharmaceuticals, and are employees of Janssen Pharmaceutica NV.  
+
+The views expressed by MR, HB and BLD are those of the authors and should not be construed to represent the positions of the U.S. Army, the Department of Defense, or the Department of Health and Human Services.  
+
+Data and materials availability: All sequences involved with this study are available on GISAID, including their contributors' details, such as accession number, virus name, collection date, originating lab, submitting lab and the list of authors. The sequences are available in two groups: the sequences obtained from study participants (Supplementary Data 1) and the sequences curated by LANL to define the canonical variant sequences (GISAID Identifier: EPL_SET_221208yn; doi: 10.55876/gis8.221208yn). Custom code for the structural modeling has been deposited at Zenodo (doi: 10.5281/zenodo.7869358).  
+
+## References  
+
+1. Sievers BL, Chakraborty S, Xue Y, et al. Antibodies elicited by SARS-CoV-2 infection or mRNA vaccines have reduced neutralizing activity against Beta and Omicron pseudoviruses. Sci Transl Med 2022; 14(634): eabn7842.  
+2. Planas D, Veyer D, Baidaliuk A, et al. Reduced sensitivity of SARS-CoV-2 variant Delta to antibody neutralization. Nature 2021; 596(7871): 276-80.
+
+<--- Page Split --->
+
+3. Syed AM, Ciling A, Taha TY, et al. Omicron mutations enhance infectivity and reduce antibody neutralization of SARS-CoV-2 virus-like particles. Proc Natl Acad Sci U S A 2022; 119(31): e2200592119.  
+
+4. Harvey WT, Carabelli AM, Jackson B, et al. SARS-CoV-2 variants, spike mutations and immune escape. Nat Rev Microbiol 2021; 19(7): 409-24.  
+
+5. Feikin DR, Higdon MM, Abu-Raddad LJ, et al. Duration of effectiveness of vaccines against SARS-CoV-2 infection and COVID-19 disease: results of a systematic review and meta-regression. Lancet 2022; 399(10328): 924-44.  
+
+6. Ssentongo P, Ssentongo AE, Voleti N, et al. SARS-CoV-2 vaccine effectiveness against infection, symptomatic and severe COVID-19: a systematic review and meta-analysis. BMC Infect Dis 2022; 22(1): 439.  
+
+7. Gilbert P, Self S, Rao M, Naficy A, Clemens J. Sieve analysis: methods for assessing from vaccine trial data how vaccine efficacy varies with genotypic and phenotypic pathogen variation. J Clin Epidemiol 2001; 54(1): 68-85.  
+
+8. Rolland M, Gilbert PB. Sieve analysis to understand how SARS-CoV-2 diversity can impact vaccine protection. PLoS Pathog 2021; 17(3): e1009406.  
+
+9. Pajon R, Paila YD, Girard B, et al. Initial analysis of viral dynamics and circulating viral variants during the mRNA-1273 Phase 3 COVE trial. Nat Med 2022.  
+
+10. Sadoff J, Gray G, Vandebosch A, et al. Final Analysis of Efficacy and Safety of Single-Dose Ad26.COV2.S. N Engl J Med 2022; 386(9): 847-60.  
+
+11. Cao L, Lou J, Chan SY, et al. Rapid evaluation of COVID-19 vaccine effectiveness against symptomatic infection with SARS-CoV-2 variants by analysis of genetic distance. Nat Med 2022; 28(8): 1715-22.  
+
+12. Greaney AJ, Eguia RT, Starr TN, et al. The SARS-CoV-2 Delta variant induces an antibody response largely focused on class 1 and 2 antibody epitopes. PLoS Pathog 2022; 18(6): e1010592.  
+
+13. Greaney AJ, Starr TN, Bloom JD. An antibody-escape estimator for mutations to the SARS-CoV-2 receptor-binding domain. Virus Evol 2022; 8(1): veac021.  
+
+14. Greaney AJ, Starr TN, Eguia RT, et al. A SARS-CoV-2 variant elicits an antibody response with a shifted immunodominance hierarchy. PLoS Pathog 2022; 18(2): e1010248.  
+
+15. Alenquer M, Ferreira F, Lousa D, et al. Signatures in SARS-CoV-2 spike protein conferring escape to neutralizing antibodies. PLoS Pathog 2021; 17(8): e1009772.  
+
+16. Greaney AJ, Loes AN, Crawford KHD, et al. Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies. Cell Host Microbe 2021; 29(3): 463-76 e6.  
+
+17. McCallum M, De Marco A, Lempp FA, et al. N-terminal domain antigenic mapping reveals a site of vulnerability for SARS-CoV-2. Cell 2021; 184(9): 2332-47 e16.
+
+<--- Page Split --->
+
+18. Gilbert PB, Donis RO, Koup RA, Fong Y, Plotkin SA, Follmann D. A Covid-19 Milestone Attained – A Correlate of Protection for Vaccines. New England Journal of Medicine 2022; 387: 2203-6.  
+
+19. Cromer D, Steain M, Reynaldi A, et al. Neutralising antibody titres as predictors of protection against SARS-CoV-2 variants and the impact of boosting: a meta-analysis. Lancet Microbe 2022; 3(1): e52-e61.  
+
+20. Cromer D, Steain M, Reynaldi A, et al. Neutralising antibodies predict protection from severe COVID-19. medRxiv 2022: 2022.06.09.22275942.  
+
+21. Heng F, Sun Y, Gilbert PB. Estimation and Hypothesis Testing of Strain-Specific Vaccine Efficacy with Missing Strain Types, with Applications to a COVID-19 Vaccine Trial. [Preprint] arXiv:2201.08946 [stat.ME]. 22 Jan 2022. Cited 9 Feb 2023.  
+
+22. Juraska M, Gilbert PB. Mark-specific hazard ratio model with missing multivariate marks. Lifetime Data Anal 2016; 22(4): 606-25.  
+
+23. Gilbert PB, Novitsky V, Essex M. Covariability of selected amino acid positions for HIV type 1 subtypes C and B. AIDS Res Hum Retroviruses 2005; 21(12): 1016-30.  
+
+24. Deng X, Garcia-Knight MA, Khalid MM, et al. Transmission, infectivity, and neutralization of a spike L452R SARS-CoV-2 variant. Cell 2021; 184(13): 3426-37 e8.  
+
+25. Liu Y, Liu J, Xia H, et al. Neutralizing Activity of BNT162b2-Elicited Serum. N Engl J Med 2021; 384(15): 1466-8.  
+
+26. Acevedo ML, Gaete-Argel A, Alonso-Palomares L, et al. Differential neutralizing antibody responses elicited by CoronaVac and BNT162b2 against SARS-CoV-2 Lambda in Chile. Nat Microbiol 2022; 7(4): 524-9.  
+
+27. Motozono C, Toyoda M, Zahradnik J, et al. SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity. Cell Host Microbe 2021; 29(7): 1124-36 e11.  
+
+28. Starr TN, Greaney AJ, Hilton SK, et al. Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding. bioRxiv 2020.  
+
+29. Sadoff J, Le Gars M, Brandenburg B, et al. Durable antibody responses elicited by 1 dose of Ad26.COV2.S and substantial increase after boosting: 2 randomized clinical trials. Vaccine 2022; 40(32): 4403-11.  
+
+30. Jongeneelen M, Kaszas K, Veldman D, et al. Ad26.COV2.S elicited neutralizing activity against Delta and other SARS-CoV-2 variants of concern. bioRxiv 2021: 2021.07.01.450707.  
+
+31. Williamson B, Gilbert PB, Simon N, Carone M. A General Framework for Inference on Algorithm-Agnostic Variable Importance. doi: 10.1080/01621459.2021.2003200. Journal of the American Statistical Association 2022.  
+
+32. Khoury DS, Schlub TE, Cromer D, et al. Correlates of protection, thresholds of protection, and immunobridging in SARS-CoV-2 infection. [Preprint] Posted 6 June, 2022. Access date 8 December, 2022. medRxiv 2022.
+
+<--- Page Split --->
+
+33. Gilbert PB, Montefiori DC, McDermott AB, et al. Immune correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy clinical trial. \*Science\* 2022; 375(6576): 43-50.  
+34. Fong Y, McDermott AB, Benkeser D, et al. Immune Correlates Analysis of the ENSEMBLE Single Ad26.COV2.S Dose Vaccine Efficacy Clinical Trial. \*Nature Microbiology\* 2022; 7(12): 1996-2010.  
+35. Khoury DS, Cromer D, Reynaldi A, et al. Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. \*Nat Med\* 2021; 27(7): 1205-11.  
+36. VIPER Group COVID19 Vaccine Tracker Team. COVID19 Vaccine Tracker. "World Health Organization (WHO)." Last updated Oct 5, 2022. Available at https://covid19.trackvaccines.org/agency/who/ Access date Oct 5, 2022.  
+37. Korber B, Fischer WM, Gnanakaran S, et al. Tracking Changes in SARS-CoV-2 Spike: Evidence that D614G Increases Infectivity of the COVID-19 Virus. \*Cell\* 2020; 182(4): 812-27 e19.  
+38. Corey L, Beyrer C, Cohen MS, Michael NL, Bedford T, Rolland M. SARS-CoV-2 Variants in Patients with Immunosuppression. \*N Engl J Med\* 2021; 385(6): 562-6.  
+39. Alter G, Yu J, Liu J, et al. Immunogenicity of Ad26.COV2.S vaccine against SARS-CoV-2 variants in humans. \*Nature\* 2021; 596(7871): 268-72.  
+40. Sadoff J, Gray G, Vandebosch A, et al. Safety and Efficacy of Single-Dose Ad26.COV2.S Vaccine against Covid-19. \*N Engl J Med\* 2021; 384(23): 2187-201.  
+41. Gilbert PB, Rolland M, DeCamp AC, et al. ENSEMBLE Phase 3 Trial Sieve Analysis Statistical Analysis Plan. figshare. Online resource. Posted 18 Jan, 2023. Access date 24 Mar, 2023. https://doi.org/10.6084/m9.figshare.21920652.v1. 2023.  
+
+## Tables
+
+<--- Page Split --->
+
+Table 1
+
+Numbers of primary endpoint COVID-19 cases with Spike amino acid sequence data by treatment arm and geographic region. A primary endpoint case is defined as the moderate to severe-critical primary COVID-19 endpoint in the per-protocol baseline seronegative cohort, with disease onset starting 14 days post vaccination through to a participant's unblinding date.
+
+<table><tr><td colspan="9">Geographic Region</td></tr><tr><td></td><td colspan="2">Latin America</td><td colspan="2">South Africa</td><td colspan="2">United States</td><td colspan="2">Pooled</td></tr><tr><td>Primary<br>endpoint<br>case<br>lineage</td><td>Vaccine (329)1</td><td>Placebo (634)</td><td>Vaccine (62)</td><td>Placebo (110)</td><td>Vaccine (93)</td><td>Placebo (323)</td><td>Vaccine (484)</td><td>Placebo (1067)</td></tr><tr><td>Reference</td><td>72</td><td>196</td><td>1</td><td>4</td><td>52</td><td>221</td><td>125</td><td>421</td></tr><tr><td>Alpha</td><td>4</td><td>10</td><td>1</td><td>2</td><td>4</td><td>16</td><td>9</td><td>28</td></tr><tr><td>Beta</td><td>-</td><td>-</td><td>36</td><td>59</td><td>-</td><td>-</td><td>36</td><td>59</td></tr><tr><td>Delta</td><td>-</td><td>-</td><td>11</td><td>10</td><td>-</td><td>-</td><td>11</td><td>10</td></tr><tr><td>Epsilon</td><td>-</td><td>2</td><td>-</td><td>-</td><td>8</td><td>15</td><td>8</td><td>17</td></tr><tr><td>Gamma</td><td>73</td><td>111</td><td>-</td><td>-</td><td>1</td><td>-</td><td>74</td><td>111</td></tr><tr><td>lota</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>4</td><td>0</td><td>4</td></tr><tr><td>Lambda</td><td>43</td><td>45</td><td>-</td><td>1</td><td>-</td><td>-</td><td>43</td><td>46</td></tr><tr><td>Mu</td><td>38</td><td>57</td><td>-</td><td>-</td><td>-</td><td>-</td><td>38</td><td>57</td></tr><tr><td>Zeta</td><td>33</td><td>92</td><td>-</td><td>-</td><td>1</td><td>1</td><td>34</td><td>93</td></tr><tr><td>No<br>Sequence Obtained</td><td>66</td><td>121</td><td>13</td><td>34</td><td>27</td><td>66</td><td>106</td><td>221</td></tr></table>
+
+1Numbers in parentheses are numbers of moderate to severe-critical COVID-19 primary endpoints caused by the listed SARS-CoV-2 lineage, regardless of availability of SARS-CoV-2 sequence data
+
+# Figures
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1 </center>  
+
+Circulating SARS- CoV- 2 lineages in Latin America have greater diversity than in South Africa or the United States. (A) The distribution of SARS- CoV- 2 lineages of COVID- 19 primary endpoints. The number of lineage sequences identified each month is shown for vaccine and placebo participants. (B) A phylogenetic tree based on the amino acid sequences from Latin America for the Spike protein. Tips are colored to indicate vaccine (red) or placebo (blue). (C) The distribution of variant sequences identified in Latin America as a function of their Spike Hamming distance from the vaccine insert.
+
+<--- Page Split --->
+
+for mismatched residue genotypes in maroon. In (D), the two amino acid positions hypothesized to impact VE (452 and 490) \(^{24,26,27}\) are identified with an asterisk. For each geographic-region analysis, lineages with at least 20 COVID-19 endpoints were included, and amino acid positions with at least 20 vaccine-mismatched COVID-19 endpoints were included. CI, confidence interval.  
+
+![](images/Figure_3.jpg)
+
+<center>Figure 3 </center>
+
+<--- Page Split --->
+
+For the Latin America cohort, vaccine efficacy (VE) against the primary COVID- 19 endpoint by physicochemical- weighted Hamming distances in (A) Spike, (B) the RBD domain, (3) the NTD domain, or (4) the S1 region of the disease- causing SARS- CoV- 2 isolate to that of the vaccine- insert sequence. The top plot in each panel shows the distributions of distances by treatment arm, color- coded by lineage. The bottom plot in each panel shows the estimated VE by SARS- CoV- 2 sequence distance. The dotted lines are pointwise \(95\%\) confidence intervals. The dots are overall VE estimates for the given lineage placed at the lineage- specific median distance of placebo arm endpoints, with vertical bars indicating their pointwise \(95\%\) confidence intervals. Two Zeta sequences are visible outliers from other Zeta sequences; both sequences have two large deletions (9AA and 7AA in length) in the NTD. The plots reveal that Lambda has two sub- lineages, one (n = 79) with range of distances 17.2- 18.9 and a second (n = 9) with range of distances 25.8- 27.7, due to a 13- AA deletion between sites 64 and 76.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4 </center>  
+
+![](images/Figure_5.jpg)
+  
+
+In the Latin America cohort, vaccine efficacy (VE) against the primary COVID- 19 endpoint by the SARS- CoV- 2 antibody escape score. VE (point estimates as solid line, 95% confidence intervals as dashed lines) is shown by the antibody escape scores for: (A) DMS2, (B) PDB7, (C) PDB8, and (D) PDB13. The plot at the top of each panel shows the reverse cumulative distribution function (RCDF) of the relevant antibody- binding escape score across SARS- CoV- 2 viruses by treatment arm. (E) Spike amino acid (AA) residues constituting each antibody escape score- based putative epitope footprint. (F- I) For each set of residues constituting an antibody epitope footprint for DMS2, PDB7, PDB8, and PDB13, the image shows the set of AA positions comprising the footprint on a Spike monomer NTD or RBD structure. Cyan ribbons
+
+<--- Page Split --->
+
+highlight epitope footprint residues while red ribbons make up the rest of RBD [(F) DMS2, (G) PDB7, and (H) PDB8)] or NTD (I) (PDB13). Residue numbers and cyan dashed lines are used to label footprint residues. Each structure's orientation was chosen to best visualize all residues of a footprint. Residues are colored based on their cluster weights going from white to blue with increasing weight. 
+
+A 
+
+<table><tr><td rowspan="2">NTD Features</td><td rowspan="2">No. of Cases (V vs. P)<br/>(Incidence per 100 PYRs)</td><td rowspan="2">VE (%) (95% CI)</td><td colspan="2">Two-sided</td><td colspan="2">Two-sided Differential VE</td></tr><tr><td>Two-sided<br/>P-value</td><td>P-value</td><td>FWER<br/>P-value</td><td>Q-value</td></tr><tr><td>NTD1</td><td></td><td></td><td></td><td>0.0016</td><td>0.0065</td><td>0.0025</td></tr><tr><td>Mark Value = 1</td><td>93 (3.3) vs. 126 (4.6)</td><td>31.0 (10.3, 46.8)</td><td></td><td>0.0055</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>170 (6) vs. 387 (14)</td><td>58.7 (51.0, 65.3)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD3</td><td></td><td></td><td></td><td>0.0017</td><td>0.0065</td><td>0.0025</td></tr><tr><td>Mark Value = 1</td><td>46 (1.6) vs. 50 (1.8)</td><td>12.5 (-30.1, 41.2)</td><td></td><td>0.51</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>217 (7.6) vs. 463 (16.7)</td><td>56.1 (48.9, 62.2)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD4</td><td></td><td></td><td></td><td>0.097</td><td>0.19</td><td>0.12</td></tr><tr><td>Mark Value = 1</td><td>133 (4.7) vs. 223 (8.1)</td><td>45.3 (33.0, 55.4)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>130 (4.6) vs. 290 (10.5)</td><td>57.2 (47.9, 64.9)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD5</td><td></td><td></td><td></td><td>&amp;lt;0.001</td><td>0.0021</td><td>0.0012</td></tr><tr><td>Mark Value = 1</td><td>132 (4.6) vs. 188 (6.8)</td><td>35.2 (19.8, 47.6)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>131 (4.6) vs. 325 (11.7)</td><td>61.8 (53.7, 68.6)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD6</td><td></td><td></td><td></td><td>0.12</td><td>0.19</td><td>0.12</td></tr><tr><td>Mark Value = 1</td><td>77 (2.7) vs. 120 (4.3)</td><td>41.5 (22.7, 55.7)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>186 (6.5) vs. 393 (14.2)</td><td>55.2 (47.1, 62.0)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD7</td><td></td><td></td><td></td><td>&amp;lt;0.001</td><td>0.0013</td><td>0.0012</td></tr><tr><td>Mark Value = 1</td><td>117 (4.1) vs. 157 (5.7)</td><td>31.7 (14.2, 45.7)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>146 (5.1) vs. 356 (12.9)</td><td>61.0 (53.2, 67.5)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr></table>
+
+![PLACEHOLDER_23_0]
+
+ 
+
+<center>Figure 5</center>
+
+<--- Page Split --->
+
+In the Latin America cohort, NTD sequence feature sieve analysis and neutralization phenotype sieve analysis. (A) Vaccine efficacy (VE) estimates against the primary COVID- 19 endpoint caused by SARS- CoV- 2 with (vs. without) a NTD feature value, screened in as a specific hypothesis- driven neutralizing antibody (nAb) correlate of protection. VE estimates against SARS- CoV- 2 harboring the NTD feature value are shown in blue; those against SARS- CoV- 2 without the NTD feature value are shown in maroon. (B) VE against the primary COVID- 19 endpoint by geometric fold change in neutralizing antibody titer against the disease- causing SARS- CoV- 2 variant vs. against the D614G Reference strain. The top plot shows the numbers of cases by treatment arm and color- coded by lineage. The bottom plot shows the estimated vaccine efficacy by geometric fold change in nAb titer against the disease- causing SARS- CoV- 2 variant vs. against the D614G Reference strain. The dashed lines are pointwise 95% confidence intervals. The dots are VE point estimates against the given lineage, with the vertical bars showing 95% confidence intervals.  
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+SupplementaryData1. csv  ENSEMBLEsieveSuppNatCommun16May2023tosubmit.docx
+
+<--- Page Split --->

preprint/preprint__9888a9f622f768075271e80afca4b2931de0af2d78287779b00526c00cd54162/preprint__9888a9f622f768075271e80afca4b2931de0af2d78287779b00526c00cd54162_det.mmd

@@ -0,0 +1,601 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 931, 177]]<|/det|>
+# Quantifying how single dose Ad26.COV2.S vaccine efficacy depends on Spike sequence features  
+
+<|ref|>text<|/ref|><|det|>[[44, 195, 685, 238]]<|/det|>
+Craig Magaret Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 5056- 2664  
+
+<|ref|>text<|/ref|><|det|>[[44, 243, 328, 283]]<|/det|>
+Li Li Fred Hutchinson Cancer Center  
+
+<|ref|>text<|/ref|><|det|>[[44, 290, 685, 333]]<|/det|>
+Allan deCamp Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 1404- 4322  
+
+<|ref|>text<|/ref|><|det|>[[44, 337, 512, 377]]<|/det|>
+Morgane Rolland MHRP- HJF https://orcid.org/0000- 0003- 3650- 8490  
+
+<|ref|>text<|/ref|><|det|>[[44, 382, 685, 425]]<|/det|>
+Michal Juraska Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 0920- 2915  
+
+<|ref|>text<|/ref|><|det|>[[44, 429, 911, 472]]<|/det|>
+Brian Williamson Kaiser Permanente Washington Health Research Institute https://orcid.org/0000- 0002- 7024- 548X  
+
+<|ref|>text<|/ref|><|det|>[[44, 476, 685, 517]]<|/det|>
+James Ludwig Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 3500- 3490  
+
+<|ref|>text<|/ref|><|det|>[[44, 522, 328, 562]]<|/det|>
+Cindy Molitor Fred Hutchinson Cancer Center  
+
+<|ref|>text<|/ref|><|det|>[[44, 568, 472, 609]]<|/det|>
+David Benkeser Emory https://orcid.org/0000- 0002- 1019- 8343  
+
+<|ref|>text<|/ref|><|det|>[[44, 615, 277, 655]]<|/det|>
+Alex Luedtke University of Washington  
+
+<|ref|>text<|/ref|><|det|>[[44, 660, 533, 702]]<|/det|>
+Brian Simpkins Pitzer College https://orcid.org/0009- 0002- 1044- 0053  
+
+<|ref|>text<|/ref|><|det|>[[44, 707, 685, 749]]<|/det|>
+Lindsay Carpp Fred Hutchinson Cancer Center https://orcid.org/0000- 0003- 0333- 5925  
+
+<|ref|>text<|/ref|><|det|>[[44, 754, 475, 795]]<|/det|>
+Hongjun Bai WRAIR https://orcid.org/0000- 0002- 3501- 3974  
+
+<|ref|>text<|/ref|><|det|>[[44, 800, 752, 842]]<|/det|>
+Bethany Dearlove Walter Reed Army Institute of Research https://orcid.org/0000- 0003- 3653- 4592  
+
+<|ref|>text<|/ref|><|det|>[[44, 847, 277, 887]]<|/det|>
+Alexander Greninger University of Washington  
+
+<|ref|>text<|/ref|><|det|>[[44, 892, 277, 932]]<|/det|>
+Pavitra Roychoudhury University of Washington  
+
+<|ref|>text<|/ref|><|det|>[[44, 938, 166, 956]]<|/det|>
+Jerald Sadoff
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[55, 45, 752, 65]]<|/det|>
+Janssen Research & Development, LLC https://orcid.org/0000- 0002- 4839- 3013  
+
+<|ref|>text<|/ref|><|det|>[[44, 70, 405, 110]]<|/det|>
+Glenda Gray South African Medical Research Council  
+
+<|ref|>text<|/ref|><|det|>[[44, 117, 175, 155]]<|/det|>
+Sanne Roels Janssen R&D  
+
+<|ref|>text<|/ref|><|det|>[[44, 163, 183, 201]]<|/det|>
+An Vandebosch Janssen R&D  
+
+<|ref|>text<|/ref|><|det|>[[44, 208, 360, 247]]<|/det|>
+Daniel Stieh Janssen Vaccines & Prevention BV  
+
+<|ref|>text<|/ref|><|det|>[[44, 254, 387, 293]]<|/det|>
+Mathieu Le Gars Janssen Vaccines and Prevention B.V.  
+
+<|ref|>text<|/ref|><|det|>[[44, 300, 808, 343]]<|/det|>
+Johan Vingerhoets Janssen Pharmaceutica N.V., Beerse, Belgium https://orcid.org/0000- 0001- 5939- 9501  
+
+<|ref|>text<|/ref|><|det|>[[44, 348, 771, 389]]<|/det|>
+Beatriz Grinsztejn Evandro Chagas National Institute of Infectious Diseases- Fundacao Oswaldo Cruz  
+
+<|ref|>text<|/ref|><|det|>[[44, 394, 877, 437]]<|/det|>
+Paul Goepfert Department of Medicine, Division of Infectious Diseases, University of Alabama at Birmingham  
+
+<|ref|>text<|/ref|><|det|>[[44, 441, 454, 481]]<|/det|>
+Carla Truyers Janssen Pharmaceutica N.V., Beerse, Belgium  
+
+<|ref|>text<|/ref|><|det|>[[44, 486, 536, 527]]<|/det|>
+Ilse Van Dromme Janssen R&D, a division of Janssen Pharmaceutica NV  
+
+<|ref|>text<|/ref|><|det|>[[44, 532, 150, 571]]<|/det|>
+Edith Swann NIAID/NIH  
+
+<|ref|>text<|/ref|><|det|>[[44, 578, 508, 619]]<|/det|>
+Mary Marovich National Institute of Allergy and Infectious Diseases  
+
+<|ref|>text<|/ref|><|det|>[[44, 625, 660, 666]]<|/det|>
+Dean Follmann National Institutes of Health https://orcid.org/0000- 0003- 4073- 0393  
+
+<|ref|>text<|/ref|><|det|>[[44, 671, 784, 713]]<|/det|>
+Kathleen Neuzil University of Maryland School of Medicine https://orcid.org/0000- 0001- 9480- 2714  
+
+<|ref|>text<|/ref|><|det|>[[44, 718, 684, 759]]<|/det|>
+Lawrence Corey Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 2179- 2436  
+
+<|ref|>text<|/ref|><|det|>[[44, 764, 411, 805]]<|/det|>
+Ollivier Hyrien Fred Hutchinson Cancer Research Center  
+
+<|ref|>text<|/ref|><|det|>[[44, 811, 770, 874]]<|/det|>
+Leonardo Paiva de Sousa Evandro Chagas National Institute of Infectious Diseases- Fundacao Oswaldo Cruz https://orcid.org/0000- 0001- 9004- 5154  
+
+<|ref|>text<|/ref|><|det|>[[44, 880, 710, 921]]<|/det|>
+Martin Casapia Asociación Civil Selva Amazónica https://orcid.org/0000- 0002- 5972- 0948  
+
+<|ref|>text<|/ref|><|det|>[[44, 927, 175, 945]]<|/det|>
+Marcelo Losso
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[50, 45, 930, 134]]<|/det|>
+Hospital General de Agudos José María Ramos Mejia https://orcid.org/0000- 0002- 4273- 4833 Susan Little Department of Medicine, University of California, San Diego, CA 92903 https://orcid.org/0000- 0002- 7645- 9737  
+
+<|ref|>text<|/ref|><|det|>[[46, 140, 380, 181]]<|/det|>
+Aditya Gaur St. Jude Children's Research Hospital  
+
+<|ref|>text<|/ref|><|det|>[[46, 186, 636, 228]]<|/det|>
+Linda- Gail Bekker Desmond Tutu HIV centre https://orcid.org/0000- 0002- 0755- 4386  
+
+<|ref|>text<|/ref|><|det|>[[45, 233, 897, 297]]<|/det|>
+Nigel Garrett Centre for the AIDS Program of Research in South Africa (CAPRISA), University of KwaZulu- Natal, Durban, South Africa 4041 https://orcid.org/0000- 0002- 4530- 234X  
+
+<|ref|>text<|/ref|><|det|>[[45, 302, 642, 344]]<|/det|>
+Fei Heng University of North Florida https://orcid.org/0000- 0002- 6701- 7873  
+
+<|ref|>text<|/ref|><|det|>[[45, 349, 404, 391]]<|/det|>
+Yanqing Sun University of North Carolina at Charlotte  
+
+<|ref|>text<|/ref|><|det|>[[45, 395, 152, 413]]<|/det|>
+Peter Gilbert  
+
+<|ref|>text<|/ref|><|det|>[[52, 421, 301, 440]]<|/det|>
+pgi1bert@fredhutch.org  
+
+<|ref|>text<|/ref|><|det|>[[52, 466, 682, 487]]<|/det|>
+Fred Hutchinson Cancer Center https://orcid.org/0000- 0002- 2662- 9427  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 527, 102, 545]]<|/det|>
+## Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 565, 943, 609]]<|/det|>
+Keywords: Antibody- epitope escape score, COVID- 19 vaccine, ENSEMBLE trial, genetic distance, Hamming distance, neutralization resistance, SARS- CoV- 2, sieve analysis, vaccine efficacy, viral variants  
+
+<|ref|>text<|/ref|><|det|>[[44, 626, 295, 645]]<|/det|>
+Posted Date: May 31st, 2023  
+
+<|ref|>text<|/ref|><|det|>[[44, 664, 473, 684]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 2743022/v1  
+
+<|ref|>text<|/ref|><|det|>[[42, 701, 910, 745]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 761, 951, 943]]<|/det|>
+Additional Declarations: Yes there is potential Competing Interest. ALG reports contract testing from Abbott, Cepheid, Novavax, Pfizer, Janssen, and Hologic and research support from Gilead and Merck. JS declares support for the submitted work from the Janssen Pharmaceutical Companies of Johnson & Johnson and partial support (in the form of funding to his institution) from BARDA for the submitted work, declares support within the past 36 months from the Janssen Pharmaceutical Companies of Johnson & Johnson and BARDA funding for part of this work, has patents on invention of the Janssen COVID- 19 vaccine, and has Johnson & Johnson stock and stock options. SR had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 44, 955, 338]]<|/det|>
+for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. AV had partial support from BARDA (in the form of contract payments to her institution) for the submitted work, had all patent rights transferred to Johnson & Johnson, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. DJS had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen. MLG had partial support from BARDA (in the form of contract payments to his institution) for the submitted work, has patents on invention of the Janssen COVID- 19 vaccine, has shares in Johnson & Johnson, and is an employee of Johnson & Johnson. JV has stock and stock options in Johnson and Johnson and is an employee of Janssen Pharmaceutica NV. CT and IVD both had partial support from BARDA (in the form of contract payments to their institution) for the submitted work, hold stock in Janssen Pharmaceuticals, and are employees of Janssen Pharmaceutica NV.  
+
+<|ref|>text<|/ref|><|det|>[[42, 372, 925, 417]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on March 11th, 2024. See the published version at https://doi.org/10.1038/s41467-024-46536-w.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 159, 68]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[40, 82, 953, 400]]<|/det|>
+It is of interest to pinpoint SARS- CoV- 2 sequence features defining vaccine resistance. In the ENSEMBLE randomized, placebo- controlled phase 3 trial, estimated single- dose Ad26.COV2.S vaccine efficacy (VE) was \(56\%\) against moderate to severe- critical COVID- 19. SARS- CoV- 2 Spike sequences were measured from 484 vaccine and 1,067 placebo recipients who acquired COVID- 19 during the trial. In Latin America, where Spike diversity was greatest, VE was significantly lower against Lambda than against Reference and against all non- Lambda variants [family- wise error rate (FWER) \(\mathsf{p}< 0.05]\) . VE also differed by residue match vs. mismatch to the vaccine- strain residue at 16 amino acid positions (4 FWER \(\mathsf{p}< 0.05\) ; 12 q- value \(\leq 0.20\) ). VE significantly decreased with physicochemical- weighted Hamming distance to the vaccine- strain sequence for Spike, receptor- binding domain, N- terminal domain, and S1 (FWER \(\mathsf{p}< 0.001\) ); differed (FWER \(\leq 0.05\) ) by distance to the vaccine strain measured by 9 different antibody- epitope escape scores and by 4 NTD neutralization- impacting features; and decreased ( \(\mathsf{p} = 0.011\) ) with neutralization resistance level to vaccine recipient sera. VE against severe- critical COVID- 19 was stable across most sequence features but lower against viruses with greatest distances. These results help map antigenic specificity of in vivo vaccine protection.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 422, 176, 449]]<|/det|>
+## Main Text  
+
+<|ref|>text<|/ref|><|det|>[[42, 462, 931, 533]]<|/det|>
+Initial SARS- CoV- 2 vaccine candidates were based on the virus's original lineage, as represented by the Wuhan- Hu- 1 index strain with Spike D614 (NC_045512). As the virus has evolved, \(^{1 - 4}\) efficacy of these vaccines against symptomatic infection has waned, \(^{5,6}\) and new vaccine inserts have been developed.  
+
+<|ref|>text<|/ref|><|det|>[[40, 548, 953, 852]]<|/det|>
+Based on data from a randomized, placebo- controlled vaccine efficacy (VE) trial on clinical outcomes and pathogen sequences isolated from participants experiencing clinical outcomes, sieve analysis assesses how VE depends on pathogen sequence features. \(^{7,8}\) Pajon et al. \(^{9}\) and Sadoff et al. \(^{10}\) showed how the VE against symptomatic COVID- 19 was lower against certain variants than against the Reference strain in the phase 3 COVE trial of two doses of Moderna's mRNA- 1273 vaccine and the phase 3 ENSEMBLE trial of a single dose of Janssen's Ad26.COV2.S vaccine, respectively. [As in ref. \(^{10}\) , Reference is defined as the basal outbreak lineage B.1, which bears the D614G mutation.] Cao et al. showed that VE was higher in COVID- 19 VE trials where circulating viruses had shorter Spike sequence Hamming distances to the vaccine strain. \(^{11}\) These sieve analyses only considered Spike viral variation defined by the WHO- defined variant category or the unweighted Spike protein distance. They did not assess how VE depends on other Spike sequence features, such as at the level of individual mutations or features that impact immunological functions such as anti- SARS- CoV- 2 neutralization, \(^{12 - 17}\) relevant given the strong evidence of neutralizing antibodies as a cross- platform correlate of protection. \(^{18 - 20}\)  
+
+<|ref|>text<|/ref|><|det|>[[42, 869, 950, 959]]<|/det|>
+We report here the results of a sieve analysis of the ENSEMBLE trial, which enrolled over 40,000 participants and was conducted in the US, South Africa, and six countries in Latin America. The sieve analysis considers baseline SARS- CoV- 2 seronegative per- protocol participants and the primary endpoint (moderate to severe- critical COVID- 19), as well as the severe- critical COVID- 19 endpoint, during the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 955, 88]]<|/det|>
+double- blinded period of follow- up. We focus the main text on the Latin America results given the greatest information for sieve analysis as noted below.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 110, 145, 136]]<|/det|>
+## Results  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 150, 465, 181]]<|/det|>
+## SARS-CoV-2 sequence data  
+
+<|ref|>text<|/ref|><|det|>[[42, 196, 950, 308]]<|/det|>
+A total of 1,345 SARS- CoV- 2 Spike amino acid sequences were obtained from 1,224 participants experiencing the moderate to severe- critical primary endpoint. All sequences were variant- typed to either the Reference lineage or to one of nine different WHO- defined variants (Fig. 1A) (Table S5). Lineages that circulated at the beginning of the study period, e.g., Reference, were closer to the sequence from the vaccine insert than later emerging lineages, with Lambda the most distant (Fig. 1B- C).  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 336, 911, 390]]<|/det|>
+## Greater SARS-CoV-2 Spike diversity in Latin America than in South Africa and the US  
+
+<|ref|>text<|/ref|><|det|>[[42, 406, 940, 542]]<|/det|>
+Most sequences were obtained from participants in Latin America \((n = 776)\) with additional sequences from the US \((n = 323)\) and South Africa \((n = 125)\) (Table S6). Five main variants circulated in Latin America (Reference, Zeta, Gamma, Lambda, Mu), while the South African sequences were \(76\%\) Beta and \(17\%\) Delta, and the US sequences were \(85\%\) Reference (Fig. 1A). There was greater Spike AA sequence diversity in Latin America compared to South Africa and the US (Rao's \(Q = 10.1\) vs. 7.7 vs. 3.3, respectively; Fig. S1).  
+
+<|ref|>text<|/ref|><|det|>[[42, 557, 956, 715]]<|/det|>
+The succession of distinct co- circulating variants in Latin America and the resulting broadest dynamic range of inter- individual sequence diversity, and the greatest number of COVID- 19 endpoints, implies that sieve analyses of the Latin America region have the greatest statistical power. In contrast, the domination of the Reference lineage in the US and the Beta and Delta lineages in South Africa constrained the sequence diversity's dynamic range and limited the power of these sieve analyses. Therefore, we focus on the results from Latin America, with the US and South Africa results reported in the Supplementary Materials.  
+
+<|ref|>sub_title<|/ref|><|det|>[[45, 743, 955, 773]]<|/det|>
+## Differential vaccine efficacy against COVID-19 by SARS-CoV-2 lineage  
+
+<|ref|>text<|/ref|><|det|>[[42, 787, 949, 947]]<|/det|>
+All reported results on VE by SARS- CoV- 2 features are based on feature- specific proportional- hazards models \(^{21,22}\) (see the SAP). Figure 2A shows VE against the primary COVID- 19 endpoint caused by the Reference, Gamma, Zeta, Lambda, and Mu lineages, and Fig. 2B shows VE against the primary COVID- 19 endpoint caused by the groupings of all other lineages excluding each individual lineage ("not- lineage"). Figure 2C shows differential VE against pairs of lineages or against pairs of lineage vs. not- lineage. VE was significantly higher against Reference than against Lambda and against not- Reference lineages [family- wise error rate (FWER) \(p < 0.05\) ]. It was also significantly higher against not- Lambda vs. Lambda
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 937, 88]]<|/det|>
+and against Zeta vs. Lambda (FWER \(\mathsf{p}\leq 0.05\) ), and higher against Reference vs. Gamma, Reference vs. Mu, Zeta vs. Gamma, and Zeta vs. Mu (q- value \(\leq 0.20\) ).  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 105, 946, 149]]<|/det|>
+## Vaccine efficacy greater against COVID-19 caused by SARS-CoV-2 genotypes defined by individual Spike AA position residues matching the vaccine strain  
+
+<|ref|>text<|/ref|><|det|>[[41, 165, 952, 393]]<|/det|>
+We scanned across all Spike AA positions with sufficient residue variability (at least 20 endpoints with a vaccine- mismatched residue: \(\mathsf{n} = 37\) positions). VE significantly differed (q- value \(\leq 0.20\) ) by residue match vs. mismatch to the vaccine strain residue at 16 positions (Fig. 2D; 4 positions with FWER \(\mathsf{p}\leq 0.05\) : 75, 76, 253, 490). Similarly, when assessing the presence or absence of specific residues at each AA position, VE significantly differed (q- value \(\leq 0.20\) ) for 38 residues (75V vs. not- 75V and 76l vs. not- 76l with FWER \(\mathsf{p}\leq 0.05\) ) at the same 16 positions. Figure S4 shows the distributions of residues at these 16 positions. Thirteen of these 16 AA sites (Fig. 2D) were sites harboring characteristic mutations of the Lambda variant and not for any other variants, and very highly covaried with Lambda vs. not- Lambda (Fig. S5, Mstar \(^{23} > 0.85\) ), thereby providing nearly equivalent signatures of differential VE captured by Lambda vs. not- Lambda. The full results of the covariability analysis are in the Supplementary Materials.  
+
+<|ref|>text<|/ref|><|det|>[[41, 409, 952, 527]]<|/det|>
+Four of the 1277 analyzed Spike positions (417, 452, 484, 490) were pre- specified as being hypothesized to impact neutralization based on an association with a reduced neutralizing antibody response in mRNA vaccine recipients, \(^{24 - 26}\) or evidence for increased transmissibility (452) \(^{24}\) or increased infectivity in vitro (452, 490). \(^{24,26,27}\) Of these sites, positions 452 and 490 were found to significantly impact VE (FWER \(\mathsf{p}\leq 0.05\) ).  
+
+<|ref|>text<|/ref|><|det|>[[42, 543, 755, 564]]<|/det|>
+Figures S2B, S3B, and S6 provide complete results including by geographic region.  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 580, 920, 625]]<|/det|>
+## Vaccine efficacy against COVID-19 decreases with increasing protein distance to the vaccine-strain in Spike, receptor-binding domain (RBD), N-terminal domain (NTD), and S1  
+
+<|ref|>text<|/ref|><|det|>[[40, 640, 953, 867]]<|/det|>
+VE significantly decreased with physicochemical- weighted Hamming distance (between the observed vs. vaccine insert sequence) for Spike, RBD, NTD, and S1 (Fig. 3, FWER \(\mathsf{p}< 0.001\) ) but not for S2 ( \(\mathsf{p} = 0.78\) ). Against viruses with shortest Spike distances (average 6 residue mismatches), VE was \(69\%\) (95% CI: 60- 76%), and against viruses with 25th, 50th, 75th, and 95th percentile Spike distances (average 8.1, 12.9, 17.8, 18.6 residue mismatches), VE was \(64\%\) (56%, 71%), 52% (44%, 58%), 34% (19%, 46%), and 30% (13%, 44%), respectively. The median distances of sequences for vaccine: placebo were 15.0:9.5 for Spike, 2.6:1.0 for RBD, 4.0:1.6 for NTD, 11.7:6.2 for S1, and 3.1:3.2 for S2. Tables S7 and S8 show inferences about differences in mean distances of vaccine vs. placebo sequences. Figs. S7- S11 and Table S9 provide complete results including by geographic region, where Table S9 shows that VE decreased with weighted Hamming distance for RBD, NTD, and S1 in the US (q- value \(\leq 0.20\) ).  
+
+<|ref|>text<|/ref|><|det|>[[42, 883, 925, 950]]<|/det|>
+By lineage, ordered by placebo arm COVID- 19 endpoint Spike distance to the vaccine strain, Reference viruses had 6.0- 17.7 residue mismatches, Zeta 8.1- 22.1 mismatches, Epsilon 10.7 mismatches, Mu 12.2- 16.8 mismatches, Alpha 14.5- 16.8 mismatches, Gamma 16.7- 20.2 mismatches, and Lambda
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 958, 134]]<|/det|>
+17.2–27.7 mismatches. This ordering of lineages by protein distance matches the ordering of the VE estimates by lineage category, suggesting that overall Spike evolution is a reasonable metric capturing VE decline with variant. The results are generally similarly ordered for the RBD, NTD, and S1 distances (Fig. S12).  
+
+<|ref|>sub_title<|/ref|><|det|>[[42, 163, 924, 216]]<|/det|>
+## Vaccine efficacy against COVID-19 decreases with increasing spike antibody-escape score to the vaccine-strain  
+
+<|ref|>text<|/ref|><|det|>[[42, 232, 951, 391]]<|/det|>
+Neutralization- relevant RBD features were defined where mutations impact binding in deep mutational scanning (DMS) experiments<sup>28</sup> (see Supplementary Materials. Escape scores were defined for whole- RBD and for each of 10 epitope- specific clusters of AA sites (see Methods), labeled DMS (whole- RBD) and DMS1 through DMS10. Vaccine efficacy significantly decreased (q- value \(\leq 0.20\) ) with each of the DMS, DMS2, DMS6, DMS7, and DMS8 escape scores (FWER \(\mathsf{p}\leq 0.05\) ) as well as for DMS1, DMS5, DMS9 (q- value \(\leq 0.20\) and FWER \(>0.05\) ) (Table S12). Tables S10 and S11 show inferences about differences in mean escape scores of vaccine vs. placebo sequences.  
+
+<|ref|>text<|/ref|><|det|>[[42, 408, 950, 586]]<|/det|>
+Alternatively, we defined putative antibody footprint site sets (including whole Spike) based on structures of SARS- CoV- 2 in complex with antibodies available from the PDB. Each sequence was assigned an escape score based on a class of epitopes (see Supplementary Materials). These features are referred to as PDB1 through PDB14, with the first 12 clusters in the RBD and PDB13 and PDB14 in the NTD. Vaccine efficacy significantly decreased (q- value \(\leq 0.20\) ) with the escape scores for PDB4, PDB7, PDB8, and PDB13 (FWER \(\mathsf{p}\leq 0.05\) ) as well as for PDB1 and PDB3 (q- value \(\leq 0.20\) and FWER \(>0.05\) ) (Table S15). Tables S13 and S14 show inferences about differences in mean escape scores of vaccine vs. placebo sequences.  
+
+<|ref|>text<|/ref|><|det|>[[41, 604, 951, 944]]<|/det|>
+To interpret the DMS and PDB results, we focus on the epitope- specific features with FWER \(\mathsf{p}\leq 0.05\) that carry the greatest amount of independent information based on inter- correlation and hierarchical clustering analysis (Supplementary Text, Figs. S13 and S14): DMS2, PDB7, PDB8, and PDB13. The sieve analysis results are similar across these four features, with estimated VE at \(60 - 70\%\) against viruses with escape score zero and decreasing to \(0\% - 20\%\) against viruses with maximum escape score. PDB8 and PDB13 rank highest for discriminating VE with slightly greater span of VE point estimates over the range of escape scores (spans \(20 - 60\%\) , \(16 - 60\%\) , \(21 - 69\%\) , and \(1 - 57\%\) for DMS2, PDB7, PDB8, and PDB13, respectively) (Fig. 4A- D). Figure 4E lists the Spike AA residues in each epitope footprint and the visualizations in Fig. 4F- I show the positions comprising the four antibody epitope footprints on a Spike monomer structure. Figures S15- S23 and S24- S30 provide complete results for DMS and PDB features, respectively. Another reason PDB8 was highlighted is its balanced contacts across the whole receptor- binding motif (RBM) whereas the other RBM- specific clusters (PDB1- PDB6) are more tightly grouped within a region of the RBM. Among the non- RBM focusing antibodies (PDB7, PDB9- PDB14), PDB7 and PDB13 correspond to the most accessible sites on Spike in a closed prefusion trimer (Fig. S31) and these sites are relatively variable among SARS- CoV- 2 sequences.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 828, 96]]<|/det|>
+## Lower vaccine efficacy against COVID-19 with NTD features hypothesized to abrogate neutralization  
+
+<|ref|>text<|/ref|><|det|>[[42, 111, 951, 223]]<|/det|>
+Lower vaccine efficacy against COVID- 19 with NTD features hypothesized to abrogate neutralizationSeven dichotomous NTD features (see Supplementary Materials) were assessed for a sieve effect as for vaccine- match vs. vaccine- mismatch binary features. Six of the 7 NTD features significantly impacted VE (q- value \(\leq 0.20\) ): NTD4, NTD6, NTD1, NTD3, NTD5, and NTD7 (where the last four also had FWER \(p \leq 0.05\) ) (Fig. 5). Figure S32 shows the spatial locations in the NTD of the features that impacted VE (FWER \(p \leq 0.05\) ).  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 253, 882, 307]]<|/det|>
+## Vaccine efficacy greater against lineages with lower variant-neutralization resistance to Ad26.COV2.S vaccine recipient sera  
+
+<|ref|>text<|/ref|><|det|>[[42, 322, 944, 525]]<|/det|>
+All of the sieve analyses study how VE depends on Spike AA features except one: a neutralization sieve analysis that scores each virus's lineage by its experimentally measured sensitivity to neutralization by Ad26. COV2. S vaccine sera. \(^{29,30}\) VE decreased with this variant- neutralization resistance score ( \(p = 0.011\) ) (Fig. 5B). Under one model for the neutralization assay being a perfect correlate of protection, the estimates of VE for each of the five lineages would fall on the curve of VE by variant- neutralization resistance score. Lambda had evidence of deviating from the curve, with VE \(55\%\) (48, \(62\%\) ) based on its measured neutralization sensitivity compared to VE \(11\%\) (- 35, \(41\%\) ) based on direct analysis of Lambda ignoring neutralization data. In contrast, the weighted Hamming distance analyses yielded VE estimates at Lambda- variant distance values that are closer to the VE \(11\%\) figure.  
+
+<|ref|>text<|/ref|><|det|>[[44, 543, 555, 562]]<|/det|>
+Figure S33 provides complete results by geographic region.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 593, 809, 620]]<|/det|>
+## Multivariable virus features as predictors of treatment arm  
+
+<|ref|>text<|/ref|><|det|>[[44, 639, 928, 705]]<|/det|>
+A variable importance measure analysis by ensemble machine learning \(^{31}\) of COVID- 19 endpoint cases compared how well AA sequence features predicted treatment arm (results in Fig. S34 and the Supplementary Text).  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 735, 678, 763]]<|/det|>
+## Assessing the severe-critical COVID 19 endpoint  
+
+<|ref|>text<|/ref|><|det|>[[42, 778, 950, 958]]<|/det|>
+Differential VE against severe- critical COVID- 19 by lineage could only be assessed for Latin America, with VE of \(83\%\) (64, \(92\%\) ) against Reference, \(64\%\) (26, \(83\%\) ) against Gamma, \(94\%\) (- 27, \(100\%\) ) against Zeta, \(62\%\) (- 31, \(89\%\) ) against Lambda, and \(84\%\) (42, \(96\%\) ) against Mu (Table S16). There was no evidence of variation in VE across the lineages ( \(p = 0.50\) ) (Table S16, S17). The estimates of VE were similar/stable across AA positions with vaccine- matched vs. vaccine- mismatched residue, with all unadjusted p- values for differential VE above 0.05 (Fig. S35). For the key positions 452 and 490 found to show sieve effects for the primary COVID- 19 endpoint, the results for the severe- critical COVID- 19 endpoint were VE \(79\%\) (68, \(87\%\) ) against 452- matched virus compared to VE \(70\%\) (3, \(91\%\) ) against 452- mismatched virus ( \(p = 0.58\)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 44, 944, 180]]<|/det|>
+for difference), and VE \(80\%\) (68, \(87\%\) ) against 490- matched virus compared to VE \(62\%\) (- 31, \(89\%\) ) against 490- mismatched virus ( \(\mathrm{p} = 0.34\) for differential VE). For the DMS antibody escape score distances, the data support stable VE across the distances (Table S18). Similarly, the data support stable VE across RBD and PDB Spike- antibody escape scores (Table S19). VE was stable by variant- neutralization resistance score, with \(\mathrm{VE} = 84\%\) (67%, \(92\%\) ) for the most sensitive lineage (ancestral) and \(\mathrm{VE} = 73\%\) (50, \(85\%\) ) for the least sensitive lineage (Mu) ( \(\mathrm{p} = 0.33\) , Fig. S36).  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 196, 905, 240]]<|/det|>
+## Vaccine efficacy against severe–critical COVID-19 decreases with increasing protein distance to the vaccine-strain and by NTD features hypothesized to abrogate neutralization  
+
+<|ref|>text<|/ref|><|det|>[[42, 257, 955, 483]]<|/det|>
+There was a trend of VE against severe–critical COVID- 19 decreasing with the weighted Hamming distance for the Spike, NTD, and S1 regions (q- values \(= 0.20\) ) (Table S20, Figs. S37, S39, S40). The point estimates of VE suggested moderate declines of VE with distances. For example, the VE for Spike was \(87\%\) (71%, \(94\%\) ) against viruses with shortest distance of 6 and \(66\%\) (34%, \(83\%\) ) against viruses with long distance of 20 ( \(\mathrm{p} = 0.12\) ). Figs. S37- S41 and Table S20 provide complete information by geographic region. In addition, while VE was stable across levels of NTD1 through NTD4 ( \(\mathrm{p} > 0.20\) ), it differed by levels of NTD5, NTD6, and NTD7, with VE of \(61\%\) (31, \(78\%\) ) vs. \(88\%\) (76, \(94\%\) ) for the two NTD5 genotypes ( \(\mathrm{q} = 0.10\) for difference), VE of \(60\%\) (20, \(80\%\) ) vs. \(84\%\) (72, \(91\%\) ) for the two NTD6 genotypes ( \(\mathrm{q} = 0.12\) for difference), and VE of \(64\%\) (32, \(80\%\) ) vs. \(85\%\) (73, \(92\%\) ) for the two NTD7 genotypes ( \(\mathrm{q} = 0.12\) for difference) (Table S21).  
+
+<|ref|>sub_title<|/ref|><|det|>[[45, 505, 191, 530]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[42, 545, 949, 747]]<|/det|>
+Sieve analysis compares genotype- specific or immunophenotype- specific COVID- 19 incidence between randomized study groups, therefore directly assessing causal effects of vaccination and providing inferences for how vaccine efficacy depends on SARS- CoV- 2 features. In addition to the strength of a randomized, double- blinded placebo- controlled phase 3 trial, the present sieve analysis of ENSEMBLE had ample statistical precision due to the large number of SARS- CoV- 2 Spike sequences (measured from more than 1,200 participants) and the broad proteomic variability of the SARS- CoV- 2 Spike sequences causing these endpoints. Consequently, the sieve analysis could provide many insights into how the efficacy of the Ad26. COV2. S vaccine, evaluated in baseline SARS- CoV- 2 negative individuals, depended on virus features.  
+
+<|ref|>text<|/ref|><|det|>[[42, 765, 944, 944]]<|/det|>
+In the Latin American cohort, VE against the moderate to severe–critical COVID- 19 primary endpoint significantly declined with Spike sequence distance as measured in myriad ways, including lineage, weighted Hamming distances calculated for Spike, RBD, NTD, and S1, scores reflecting degree of escape from epitope- specific antibodies computed using deep mutational scanning or based on crystal structures in the Protein Data Bank (PDB), and NTD features previously shown to impact neutralization. Estimates of VE by lineage were consistently ordered by the distances of the different lineages to the vaccine strain. VE declined similarly with Spike, RBD, NTD, and S1 distances (VE about \(70\%\) against viruses closest to the vaccine and \(20\%\) against viruses beyond the 90- 95th percentile of distances) but
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 44, 955, 134]]<|/det|>
+did not depend on S2 distances. This may be explained by S2's relative conservation when compared to S1. As such, almost all variant- characteristic mutations are not in S2, and none of the prescribed antibody epitope footprint clusters included S2 positions (only rare epitopes in PDB mapped to S2), reflecting S2's 'stalk' location and relative lack of exposure to the immune system.  
+
+<|ref|>text<|/ref|><|det|>[[41, 151, 953, 423]]<|/det|>
+VE significantly declined with 14 of the 20 evaluable antibody epitope escape scores. Six antibody- epitope clusters had no evidence of impacting VE: DMS3, PDB2, PDB5, PDB6, PDB9, PDB14. Of the 14 clusters with a sieve effect, 9 include at least one site that harbors a characteristic mutation of Lambda, whereas 3 include site 417 which is a characteristic mutation of Mu and Gamma, 1 includes site 501 that harbors a characteristic mutation of Gamma, Alpha, and Mu, and 1 includes both sites 417 and 501. Thus the 9 sieve- effect clusters appear to be driven by the differential VE by Lambda vs. not- Lambda, whereas the other 5 appear to be driven by mutations at the important sieve- effect sites 417 and 501 that impact neutralization. Of the 6 non- sieve- effect clusters, only one (PDB14) included a site harboring a characteristic mutation of Lambda, site 75, which was a sieve- effect site with FWER \(p \leq 0.05\) . The potential for sieve effects in different epitope sets depends on many factors including level of accessibility to neutralizing antibodies, conservation, and the narrowness of the footprints on the tridimensional structure they target (Fig. S31).  
+
+<|ref|>text<|/ref|><|det|>[[41, 438, 945, 647]]<|/det|>
+Neutralizing antibody assays have performed well at predicting vaccine efficacy against COVID- 19 and severe- critical COVID- 19 across SARS- CoV- 2 lineages. \(^{19,20,32}\) Importantly, one of the sieve analyses in the present work scored viruses by their lineage's directly measured resistance to neutralization by sera from ENSEMBLE Ad26. COV2. S vaccine recipients, providing a way to study a neutralization correlate of protection (CoP) in a complementary way to individual- level and population- level immune correlates analyses. \(^{33 - 35}\) VE significantly declined against lineages with greater neutralization resistance scores, providing validation of pseudovirus neutralization titer as a CoP. However, the lineage scores were estimated from only eight ENSEMBLE vaccine recipients, albeit the scores are supported by additional data from 17 Ad26. COV2. S vaccine recipients in the COV2001 phase 1/2a study. \(^{36}\)  
+
+<|ref|>text<|/ref|><|det|>[[41, 664, 947, 824]]<|/det|>
+The relative prevalence of SARS- CoV- 2 lineages changed over time (Fig. 1A and Fig. 1 of ref. \(^{10}\) ) where in Latin America the median (range) number of days from enrollment until the COVID- 19 endpoint among placebo recipients was 48 (15, 197) for Reference, 45 (15, 141) for Zeta, 114 (42, 220) for Gamma, 126 (57, 204) for Lambda, and 170 (109, 219) for Mu. If newer variants tended to expose participants later in follow- up than older variants it could cause spurious genotypic sieve effects that are instead due to waning vaccine efficacy. This potential bias was mitigated by controlling for calendar time of enrollment in the sieve analyses.  
+
+<|ref|>text<|/ref|><|det|>[[41, 840, 951, 955]]<|/det|>
+The Ad26. COV2. S vaccine sieve effects observed here, based on data collected prior to July 10, 2021, revealed broader vaccine adaptation features as several sieve signature sites showed mutations in subsequent variant waves. Hence, mutations at sites 452, 484 and 501 are dominant in currently circulating Omicron sub- lineages [global proportion between 2022- 12- 04 and 2022- 12- 10: L452R = 87.2%, E484A = 98.5%, N501Y = 99.2% \(^{37}\) ]. While the sieve signature F490S had been rare until the end of 2022,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 44, 941, 136]]<|/det|>
+this mutation became dominant in early 2023 with the global rapid spread of XBB.1.5 variants. The fact that sieve analysis predicted currently relevant mutations could be expected since SARS- CoV- 2 has shown remarkable patterns of convergent evolution since the initial appearance of variants, with numerous recurrent mutations, especially in the RBD, shared across lineages over time. \(^{38}\)  
+
+<|ref|>text<|/ref|><|det|>[[41, 153, 951, 378]]<|/det|>
+A strength of this study was it was conducted in three separate geographic regions with different circulating lineages, which contribute insights based on these lineages and their characteristic signature mutations, and different distributions of genetic distances of circulating sequences to the vaccine strain. The analyses of Latin American study sites provided the greatest insights given that \(63\%\) of primary COVID- 19 endpoints with sequence data were in Latin America where the circulating SARS- CoV- 2 sequences were the most diversified. All features showing sieve effects in the US also showed sieve effects in Latin America, constituting independent replication of results. The result of no sieve effects in South African study sites can likely be explained by the vast majority of circulating sequences being Beta or Delta variants with limited dynamic range of genetic distances within each variant and a lack of Reference viruses that are close to the vaccine strain.  
+
+<|ref|>text<|/ref|><|det|>[[41, 394, 955, 690]]<|/det|>
+Another strength of this study was that VE against severe- critical COVID- 19 could be assessed. The results support that VE against this endpoint also declines with Spike sequence distance as measured in multiple ways, yet with VE starting higher against viruses closest to the vaccine strain and diminishing less rapidly with increasing degrees of sequence mismatch. Overall, the finding that protection against severe- critical COVID- 19 is more invariant to sequence changes than against less- symptomatic COVID- 19 may have clinical implications for planning updates of vaccines with new variants. The severe- critical classification covers a broad spectrum of clinical phenotypes ranging from individuals with only repeated low partial pressure of oxygen to severe pneumonia requiring respiratory support. Protection against hospitalization with severe consequences is clinically most important but sieve analysis specific to this outcome could not be performed given small numbers of cases. Yet, ENSEMBLE and post- approval trials have shown high Ad26. COV2. S efficacy against this outcome especially in South Africa after a 6- month boost, suggesting that neutralization resistance and sequence variation may be playing a less dominant role in vaccine- induced protection against the most serious disease, perhaps due to CD8 + T cells. \(^{39}\)  
+
+<|ref|>sub_title<|/ref|><|det|>[[45, 713, 163, 738]]<|/det|>
+## Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[45, 752, 820, 784]]<|/det|>
+## Trial design, study cohort, and COVID-19 endpoints  
+
+<|ref|>text<|/ref|><|det|>[[42, 799, 955, 914]]<|/det|>
+Trial enrollment began on September 21, 2020. The end of the double- blind period varied by country; the data cutoff for this analysis was July 9, 2021. The main endpoint for sieve analysis is the same COVID- 19 primary endpoint (moderate to severe- critical) as in the primary analyses, \(^{10,40}\) restricting to endpoints starting 14 days post vaccination. Sieve analyses were also conducted for severe- critical COVID- 19, again using the same definition as used in the primary papers. \(^{10,40}\) Analyses were conducted in the per
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 46, 935, 92]]<|/det|>
+protocol baseline seronegative cohort. \(^{40}\) See Section 1 of the Statistical Analysis Plan (SAP, provided in ref. \(^{41}\) and as supplementary material) and the Supplementary Materials for further details.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 121, 623, 150]]<|/det|>
+## SARS-CoV-2 sequencing and sequence data  
+
+<|ref|>text<|/ref|><|det|>[[42, 166, 951, 255]]<|/det|>
+SARS- CoV- 2 Spike sequences were generated and variant- typed as described. \(^{40}\) Sequences were selected for analysis if they were obtained within 36 days following the first RNA- positive timepoint associated with the first moderate to severe- critical COVID- 19 primary endpoint. See the Supplementary Materials for further details.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 286, 400, 313]]<|/det|>
+## Neutralizing antibody titers  
+
+<|ref|>text<|/ref|><|det|>[[42, 328, 956, 418]]<|/det|>
+Neutralizing antibody titers were measured to a panel of Spike antigens representing the Reference strain B.1.D614G and several variants. \(^{29,30}\) Each variant was assigned a score defined as the log10- transformed ratio of geometric mean titer of vaccinee sera against the variant and the geometric mean titer of vaccinee sera against the Reference strain.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 450, 231, 476]]<|/det|>
+## Sieve analysis  
+
+<|ref|>text<|/ref|><|det|>[[42, 492, 951, 626]]<|/det|>
+This analysis was specified in advance and documented in the SAP. The sieve analyses were conducted for each of the four geographic regions: Latin America, South Africa, the US, and the three geographic regions pooled (hereafter, 'geographic- region analyses'). Details on specification of spike amino acid (AA) sequence features for sieve analysis, prospective vaccine efficacy sieve analysis, neutralization hypothesis- driven sieve analysis, and multiple hypothesis testing adjustment for AA sequence sieve analysis are provided in the Supplementary Materials.  
+
+<|ref|>text<|/ref|><|det|>[[42, 644, 945, 734]]<|/det|>
+Additional details on covariability analysis, quantification of viral diversity, antibody escape scores [deep mutational scanning (DMS) and Protein Data Bank (PDB)], variant- neutralization sensitivity score assigned to variants, handling of missing sequences, and structural modeling is also in the Supplementary Materials.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 756, 213, 781]]<|/det|>
+## Declarations  
+
+<|ref|>text<|/ref|><|det|>[[42, 796, 947, 908]]<|/det|>
+Acknowledgments: The authors thank Jesse Bloom for input in defining deep mutational scanning Spike sequence features for sieve analysis. We gratefully acknowledge all data contributors, i.e., the Authors and their Originating laboratories responsible for obtaining the specimens, and their Submitting laboratories for generating the genetic sequence and metadata and sharing via the GISAID Initiative, on which this research is based.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 927, 121, 945]]<|/det|>
+## Funding:
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 857, 88]]<|/det|>
+Administration for Strategic Preparedness and Response, Biomedical Advanced Research and Development Authority, Government Contract Nos. HHSO100201700018C with Janssen.  
+
+<|ref|>text<|/ref|><|det|>[[42, 104, 937, 149]]<|/det|>
+National Institute of Allergy and Infectious Diseases (NIAID) grant UM1 AI068635 (HVTN SDMC) (PBG), UM1 AI068614 (HVTN LOC) (LC), and R37AI054165 (PBG).  
+
+<|ref|>text<|/ref|><|det|>[[42, 165, 940, 209]]<|/det|>
+Intramural Research Program of the NIAID Scientific Computing Infrastructure at Fred Hutch, ORIP grant S100D028685.  
+
+<|ref|>text<|/ref|><|det|>[[42, 226, 920, 270]]<|/det|>
+Janssen Research and Development, an affiliate of Janssen Vaccines and Prevention and part of the Janssen pharmaceutical companies of Johnson & Johnson.  
+
+<|ref|>text<|/ref|><|det|>[[42, 286, 955, 376]]<|/det|>
+The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the Department of Health and Human Services or its components.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 394, 228, 413]]<|/det|>
+## Author contributions:  
+
+<|ref|>text<|/ref|><|det|>[[42, 430, 790, 900]]<|/det|>
+Conceptualization: CAM, AdC, MR, SR, AV, DJS, MLG, DF, PBG  Methodology: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, BS, AG, PR, OH, PBG  Software: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, BLD, AG, PR, OH, PBG  Validation: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, AG, PR, OH, PBG  Formal Analysis: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, BS, AL, HB, AG, PR, OH, PBG  Investigation: JS, GEG, BG, SDR, DJS, MLG, JV, PAG, CT, IVD, ES, MAM, KMN, LC  Resources: JS, GEG, SDR, AV, DJS, MLG, JV, BG, PAG, LPdS, MC, MHL, SJL, AG, LGB, NG, CT, IVD, ES, MAM, KMN, LC  Data Curation: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, AG, PR, CT, OH, PBG  Visualization: CAM, LL, AdC, MR, MJ, BDW, JL, CM, DB, AL, LNC, AG, PR, OH, PBG  Funding acquisition: LC, PBG  Project administration: PBG  
+
+<|ref|>text<|/ref|><|det|>[[42, 909, 468, 929]]<|/det|>
+Writing - original draft: CAM, AdC, MR, LNC, PBG
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 299, 64]]<|/det|>
+Writing - review & editing: All  
+
+<|ref|>text<|/ref|><|det|>[[39, 80, 950, 536]]<|/det|>
+Competing interests: ALG reports contract testing from Abbott, Cepheid, Novavax, Pfizer, Janssen, and Hologic and research support from Gilead and Merck. JS declares support for the submitted work from the Janssen Pharmaceutical Companies of Johnson & Johnson and partial support (in the form of funding to his institution) from BARDA for the submitted work, declares support within the past 36 months from the Janssen Pharmaceutical Companies of Johnson & Johnson and BARDA funding for part of this work, has patents on invention of the Janssen COVID- 19 vaccine, and has Johnson & Johnson stock and stock options. SR had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. AV had partial support from BARDA (in the form of contract payments to her institution) for the submitted work, had all patent rights transferred to Johnson & Johnson, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. DJS had partial support from the Department of Health and Human Services BARDA (in the form of contract payments to his institution) for the submitted work, has stock and/or stock options in Johnson & Johnson, and is an employee of Janssen Pharmaceutica NV. CT and IVD both had partial support from BARDA (in the form of contract payments to their institution) for the submitted work, hold stock in Janssen Pharmaceuticals, and are employees of Janssen Pharmaceutica NV.  
+
+<|ref|>text<|/ref|><|det|>[[42, 551, 920, 618]]<|/det|>
+The views expressed by MR, HB and BLD are those of the authors and should not be construed to represent the positions of the U.S. Army, the Department of Defense, or the Department of Health and Human Services.  
+
+<|ref|>text<|/ref|><|det|>[[42, 635, 950, 771]]<|/det|>
+Data and materials availability: All sequences involved with this study are available on GISAID, including their contributors' details, such as accession number, virus name, collection date, originating lab, submitting lab and the list of authors. The sequences are available in two groups: the sequences obtained from study participants (Supplementary Data 1) and the sequences curated by LANL to define the canonical variant sequences (GISAID Identifier: EPL_SET_221208yn; doi: 10.55876/gis8.221208yn). Custom code for the structural modeling has been deposited at Zenodo (doi: 10.5281/zenodo.7869358).  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 794, 196, 819]]<|/det|>
+## References  
+
+<|ref|>text<|/ref|><|det|>[[58, 835, 955, 951]]<|/det|>
+1. Sievers BL, Chakraborty S, Xue Y, et al. Antibodies elicited by SARS-CoV-2 infection or mRNA vaccines have reduced neutralizing activity against Beta and Omicron pseudoviruses. Sci Transl Med 2022; 14(634): eabn7842.  
+2. Planas D, Veyer D, Baidaliuk A, et al. Reduced sensitivity of SARS-CoV-2 variant Delta to antibody neutralization. Nature 2021; 596(7871): 276-80.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[58, 44, 890, 110]]<|/det|>
+3. Syed AM, Ciling A, Taha TY, et al. Omicron mutations enhance infectivity and reduce antibody neutralization of SARS-CoV-2 virus-like particles. Proc Natl Acad Sci U S A 2022; 119(31): e2200592119.  
+
+<|ref|>text<|/ref|><|det|>[[58, 116, 900, 160]]<|/det|>
+4. Harvey WT, Carabelli AM, Jackson B, et al. SARS-CoV-2 variants, spike mutations and immune escape. Nat Rev Microbiol 2021; 19(7): 409-24.  
+
+<|ref|>text<|/ref|><|det|>[[58, 165, 933, 233]]<|/det|>
+5. Feikin DR, Higdon MM, Abu-Raddad LJ, et al. Duration of effectiveness of vaccines against SARS-CoV-2 infection and COVID-19 disease: results of a systematic review and meta-regression. Lancet 2022; 399(10328): 924-44.  
+
+<|ref|>text<|/ref|><|det|>[[58, 238, 925, 304]]<|/det|>
+6. Ssentongo P, Ssentongo AE, Voleti N, et al. SARS-CoV-2 vaccine effectiveness against infection, symptomatic and severe COVID-19: a systematic review and meta-analysis. BMC Infect Dis 2022; 22(1): 439.  
+
+<|ref|>text<|/ref|><|det|>[[58, 310, 923, 377]]<|/det|>
+7. Gilbert P, Self S, Rao M, Naficy A, Clemens J. Sieve analysis: methods for assessing from vaccine trial data how vaccine efficacy varies with genotypic and phenotypic pathogen variation. J Clin Epidemiol 2001; 54(1): 68-85.  
+
+<|ref|>text<|/ref|><|det|>[[58, 382, 930, 427]]<|/det|>
+8. Rolland M, Gilbert PB. Sieve analysis to understand how SARS-CoV-2 diversity can impact vaccine protection. PLoS Pathog 2021; 17(3): e1009406.  
+
+<|ref|>text<|/ref|><|det|>[[58, 431, 903, 476]]<|/det|>
+9. Pajon R, Paila YD, Girard B, et al. Initial analysis of viral dynamics and circulating viral variants during the mRNA-1273 Phase 3 COVE trial. Nat Med 2022.  
+
+<|ref|>text<|/ref|><|det|>[[55, 480, 870, 525]]<|/det|>
+10. Sadoff J, Gray G, Vandebosch A, et al. Final Analysis of Efficacy and Safety of Single-Dose Ad26.COV2.S. N Engl J Med 2022; 386(9): 847-60.  
+
+<|ref|>text<|/ref|><|det|>[[50, 529, 920, 596]]<|/det|>
+11. Cao L, Lou J, Chan SY, et al. Rapid evaluation of COVID-19 vaccine effectiveness against symptomatic infection with SARS-CoV-2 variants by analysis of genetic distance. Nat Med 2022; 28(8): 1715-22.  
+
+<|ref|>text<|/ref|><|det|>[[50, 600, 928, 646]]<|/det|>
+12. Greaney AJ, Eguia RT, Starr TN, et al. The SARS-CoV-2 Delta variant induces an antibody response largely focused on class 1 and 2 antibody epitopes. PLoS Pathog 2022; 18(6): e1010592.  
+
+<|ref|>text<|/ref|><|det|>[[50, 650, 916, 695]]<|/det|>
+13. Greaney AJ, Starr TN, Bloom JD. An antibody-escape estimator for mutations to the SARS-CoV-2 receptor-binding domain. Virus Evol 2022; 8(1): veac021.  
+
+<|ref|>text<|/ref|><|det|>[[50, 699, 900, 744]]<|/det|>
+14. Greaney AJ, Starr TN, Eguia RT, et al. A SARS-CoV-2 variant elicits an antibody response with a shifted immunodominance hierarchy. PLoS Pathog 2022; 18(2): e1010248.  
+
+<|ref|>text<|/ref|><|det|>[[50, 749, 930, 794]]<|/det|>
+15. Alenquer M, Ferreira F, Lousa D, et al. Signatures in SARS-CoV-2 spike protein conferring escape to neutralizing antibodies. PLoS Pathog 2021; 17(8): e1009772.  
+
+<|ref|>text<|/ref|><|det|>[[50, 799, 947, 866]]<|/det|>
+16. Greaney AJ, Loes AN, Crawford KHD, et al. Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies. Cell Host Microbe 2021; 29(3): 463-76 e6.  
+
+<|ref|>text<|/ref|><|det|>[[50, 870, 927, 916]]<|/det|>
+17. McCallum M, De Marco A, Lempp FA, et al. N-terminal domain antigenic mapping reveals a site of vulnerability for SARS-CoV-2. Cell 2021; 184(9): 2332-47 e16.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[47, 44, 933, 92]]<|/det|>
+18. Gilbert PB, Donis RO, Koup RA, Fong Y, Plotkin SA, Follmann D. A Covid-19 Milestone Attained – A Correlate of Protection for Vaccines. New England Journal of Medicine 2022; 387: 2203-6.  
+
+<|ref|>text<|/ref|><|det|>[[47, 95, 941, 161]]<|/det|>
+19. Cromer D, Steain M, Reynaldi A, et al. Neutralising antibody titres as predictors of protection against SARS-CoV-2 variants and the impact of boosting: a meta-analysis. Lancet Microbe 2022; 3(1): e52-e61.  
+
+<|ref|>text<|/ref|><|det|>[[47, 166, 940, 210]]<|/det|>
+20. Cromer D, Steain M, Reynaldi A, et al. Neutralising antibodies predict protection from severe COVID-19. medRxiv 2022: 2022.06.09.22275942.  
+
+<|ref|>text<|/ref|><|det|>[[47, 215, 949, 281]]<|/det|>
+21. Heng F, Sun Y, Gilbert PB. Estimation and Hypothesis Testing of Strain-Specific Vaccine Efficacy with Missing Strain Types, with Applications to a COVID-19 Vaccine Trial. [Preprint] arXiv:2201.08946 [stat.ME]. 22 Jan 2022. Cited 9 Feb 2023.  
+
+<|ref|>text<|/ref|><|det|>[[47, 286, 934, 331]]<|/det|>
+22. Juraska M, Gilbert PB. Mark-specific hazard ratio model with missing multivariate marks. Lifetime Data Anal 2016; 22(4): 606-25.  
+
+<|ref|>text<|/ref|><|det|>[[47, 336, 877, 381]]<|/det|>
+23. Gilbert PB, Novitsky V, Essex M. Covariability of selected amino acid positions for HIV type 1 subtypes C and B. AIDS Res Hum Retroviruses 2005; 21(12): 1016-30.  
+
+<|ref|>text<|/ref|><|det|>[[47, 385, 935, 430]]<|/det|>
+24. Deng X, Garcia-Knight MA, Khalid MM, et al. Transmission, infectivity, and neutralization of a spike L452R SARS-CoV-2 variant. Cell 2021; 184(13): 3426-37 e8.  
+
+<|ref|>text<|/ref|><|det|>[[47, 435, 896, 479]]<|/det|>
+25. Liu Y, Liu J, Xia H, et al. Neutralizing Activity of BNT162b2-Elicited Serum. N Engl J Med 2021; 384(15): 1466-8.  
+
+<|ref|>text<|/ref|><|det|>[[47, 484, 935, 551]]<|/det|>
+26. Acevedo ML, Gaete-Argel A, Alonso-Palomares L, et al. Differential neutralizing antibody responses elicited by CoronaVac and BNT162b2 against SARS-CoV-2 Lambda in Chile. Nat Microbiol 2022; 7(4): 524-9.  
+
+<|ref|>text<|/ref|><|det|>[[47, 556, 875, 600]]<|/det|>
+27. Motozono C, Toyoda M, Zahradnik J, et al. SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity. Cell Host Microbe 2021; 29(7): 1124-36 e11.  
+
+<|ref|>text<|/ref|><|det|>[[47, 605, 920, 650]]<|/det|>
+28. Starr TN, Greaney AJ, Hilton SK, et al. Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding. bioRxiv 2020.  
+
+<|ref|>text<|/ref|><|det|>[[47, 654, 923, 721]]<|/det|>
+29. Sadoff J, Le Gars M, Brandenburg B, et al. Durable antibody responses elicited by 1 dose of Ad26.COV2.S and substantial increase after boosting: 2 randomized clinical trials. Vaccine 2022; 40(32): 4403-11.  
+
+<|ref|>text<|/ref|><|det|>[[47, 726, 940, 771]]<|/det|>
+30. Jongeneelen M, Kaszas K, Veldman D, et al. Ad26.COV2.S elicited neutralizing activity against Delta and other SARS-CoV-2 variants of concern. bioRxiv 2021: 2021.07.01.450707.  
+
+<|ref|>text<|/ref|><|det|>[[47, 776, 910, 842]]<|/det|>
+31. Williamson B, Gilbert PB, Simon N, Carone M. A General Framework for Inference on Algorithm-Agnostic Variable Importance. doi: 10.1080/01621459.2021.2003200. Journal of the American Statistical Association 2022.  
+
+<|ref|>text<|/ref|><|det|>[[47, 848, 936, 914]]<|/det|>
+32. Khoury DS, Schlub TE, Cromer D, et al. Correlates of protection, thresholds of protection, and immunobridging in SARS-CoV-2 infection. [Preprint] Posted 6 June, 2022. Access date 8 December, 2022. medRxiv 2022.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[45, 44, 950, 530]]<|/det|>
+33. Gilbert PB, Montefiori DC, McDermott AB, et al. Immune correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy clinical trial. \*Science\* 2022; 375(6576): 43-50.  
+34. Fong Y, McDermott AB, Benkeser D, et al. Immune Correlates Analysis of the ENSEMBLE Single Ad26.COV2.S Dose Vaccine Efficacy Clinical Trial. \*Nature Microbiology\* 2022; 7(12): 1996-2010.  
+35. Khoury DS, Cromer D, Reynaldi A, et al. Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. \*Nat Med\* 2021; 27(7): 1205-11.  
+36. VIPER Group COVID19 Vaccine Tracker Team. COVID19 Vaccine Tracker. "World Health Organization (WHO)." Last updated Oct 5, 2022. Available at https://covid19.trackvaccines.org/agency/who/ Access date Oct 5, 2022.  
+37. Korber B, Fischer WM, Gnanakaran S, et al. Tracking Changes in SARS-CoV-2 Spike: Evidence that D614G Increases Infectivity of the COVID-19 Virus. \*Cell\* 2020; 182(4): 812-27 e19.  
+38. Corey L, Beyrer C, Cohen MS, Michael NL, Bedford T, Rolland M. SARS-CoV-2 Variants in Patients with Immunosuppression. \*N Engl J Med\* 2021; 385(6): 562-6.  
+39. Alter G, Yu J, Liu J, et al. Immunogenicity of Ad26.COV2.S vaccine against SARS-CoV-2 variants in humans. \*Nature\* 2021; 596(7871): 268-72.  
+40. Sadoff J, Gray G, Vandebosch A, et al. Safety and Efficacy of Single-Dose Ad26.COV2.S Vaccine against Covid-19. \*N Engl J Med\* 2021; 384(23): 2187-201.  
+41. Gilbert PB, Rolland M, DeCamp AC, et al. ENSEMBLE Phase 3 Trial Sieve Analysis Statistical Analysis Plan. figshare. Online resource. Posted 18 Jan, 2023. Access date 24 Mar, 2023. https://doi.org/10.6084/m9.figshare.21920652.v1. 2023.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 550, 133, 574]]<|/det|>
+## Tables
+
+<--- Page Split --->
+<|ref|>table_caption<|/ref|><|det|>[[465, 48, 532, 63]]<|/det|>
+Table 1
+
+<|ref|>text<|/ref|><|det|>[[57, 72, 945, 156]]<|/det|>
+Numbers of primary endpoint COVID-19 cases with Spike amino acid sequence data by treatment arm and geographic region. A primary endpoint case is defined as the moderate to severe-critical primary COVID-19 endpoint in the per-protocol baseline seronegative cohort, with disease onset starting 14 days post vaccination through to a participant's unblinding date.
+
+<|ref|>table<|/ref|><|det|>[[44, 161, 955, 700]]<|/det|>
+
+<table><tr><td colspan="9">Geographic Region</td></tr><tr><td></td><td colspan="2">Latin America</td><td colspan="2">South Africa</td><td colspan="2">United States</td><td colspan="2">Pooled</td></tr><tr><td>Primary<br>endpoint<br>case<br>lineage</td><td>Vaccine (329)1</td><td>Placebo (634)</td><td>Vaccine (62)</td><td>Placebo (110)</td><td>Vaccine (93)</td><td>Placebo (323)</td><td>Vaccine (484)</td><td>Placebo (1067)</td></tr><tr><td>Reference</td><td>72</td><td>196</td><td>1</td><td>4</td><td>52</td><td>221</td><td>125</td><td>421</td></tr><tr><td>Alpha</td><td>4</td><td>10</td><td>1</td><td>2</td><td>4</td><td>16</td><td>9</td><td>28</td></tr><tr><td>Beta</td><td>-</td><td>-</td><td>36</td><td>59</td><td>-</td><td>-</td><td>36</td><td>59</td></tr><tr><td>Delta</td><td>-</td><td>-</td><td>11</td><td>10</td><td>-</td><td>-</td><td>11</td><td>10</td></tr><tr><td>Epsilon</td><td>-</td><td>2</td><td>-</td><td>-</td><td>8</td><td>15</td><td>8</td><td>17</td></tr><tr><td>Gamma</td><td>73</td><td>111</td><td>-</td><td>-</td><td>1</td><td>-</td><td>74</td><td>111</td></tr><tr><td>lota</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>4</td><td>0</td><td>4</td></tr><tr><td>Lambda</td><td>43</td><td>45</td><td>-</td><td>1</td><td>-</td><td>-</td><td>43</td><td>46</td></tr><tr><td>Mu</td><td>38</td><td>57</td><td>-</td><td>-</td><td>-</td><td>-</td><td>38</td><td>57</td></tr><tr><td>Zeta</td><td>33</td><td>92</td><td>-</td><td>-</td><td>1</td><td>1</td><td>34</td><td>93</td></tr><tr><td>No<br>Sequence Obtained</td><td>66</td><td>121</td><td>13</td><td>34</td><td>27</td><td>66</td><td>106</td><td>221</td></tr></table>
+
+<|ref|>text<|/ref|><|det|>[[59, 693, 900, 725]]<|/det|>
+1Numbers in parentheses are numbers of moderate to severe-critical COVID-19 primary endpoints caused by the listed SARS-CoV-2 lineage, regardless of availability of SARS-CoV-2 sequence data
+
+<|ref|>title<|/ref|><|det|>[[43, 781, 145, 803]]<|/det|>
+# Figures
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[52, 50, 940, 479]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 519, 115, 538]]<|/det|>
+<center>Figure 1 </center>  
+
+<|ref|>text<|/ref|><|det|>[[42, 560, 951, 695]]<|/det|>
+Circulating SARS- CoV- 2 lineages in Latin America have greater diversity than in South Africa or the United States. (A) The distribution of SARS- CoV- 2 lineages of COVID- 19 primary endpoints. The number of lineage sequences identified each month is shown for vaccine and placebo participants. (B) A phylogenetic tree based on the amino acid sequences from Latin America for the Spike protein. Tips are colored to indicate vaccine (red) or placebo (blue). (C) The distribution of variant sequences identified in Latin America as a function of their Spike Hamming distance from the vaccine insert.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 925, 135]]<|/det|>
+for mismatched residue genotypes in maroon. In (D), the two amino acid positions hypothesized to impact VE (452 and 490) \(^{24,26,27}\) are identified with an asterisk. For each geographic-region analysis, lineages with at least 20 COVID-19 endpoints were included, and amino acid positions with at least 20 vaccine-mismatched COVID-19 endpoints were included. CI, confidence interval.  
+
+<|ref|>image<|/ref|><|det|>[[55, 150, 741, 884]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 909, 115, 927]]<|/det|>
+<center>Figure 3 </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 44, 950, 293]]<|/det|>
+For the Latin America cohort, vaccine efficacy (VE) against the primary COVID- 19 endpoint by physicochemical- weighted Hamming distances in (A) Spike, (B) the RBD domain, (3) the NTD domain, or (4) the S1 region of the disease- causing SARS- CoV- 2 isolate to that of the vaccine- insert sequence. The top plot in each panel shows the distributions of distances by treatment arm, color- coded by lineage. The bottom plot in each panel shows the estimated VE by SARS- CoV- 2 sequence distance. The dotted lines are pointwise \(95\%\) confidence intervals. The dots are overall VE estimates for the given lineage placed at the lineage- specific median distance of placebo arm endpoints, with vertical bars indicating their pointwise \(95\%\) confidence intervals. Two Zeta sequences are visible outliers from other Zeta sequences; both sequences have two large deletions (9AA and 7AA in length) in the NTD. The plots reveal that Lambda has two sub- lineages, one (n = 79) with range of distances 17.2- 18.9 and a second (n = 9) with range of distances 25.8- 27.7, due to a 13- AA deletion between sites 64 and 76.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[50, 45, 480, 500]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[44, 737, 116, 755]]<|/det|>
+<center>Figure 4 </center>  
+
+<|ref|>image<|/ref|><|det|>[[504, 45, 930, 500]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[42, 777, 952, 958]]<|/det|>
+In the Latin America cohort, vaccine efficacy (VE) against the primary COVID- 19 endpoint by the SARS- CoV- 2 antibody escape score. VE (point estimates as solid line, 95% confidence intervals as dashed lines) is shown by the antibody escape scores for: (A) DMS2, (B) PDB7, (C) PDB8, and (D) PDB13. The plot at the top of each panel shows the reverse cumulative distribution function (RCDF) of the relevant antibody- binding escape score across SARS- CoV- 2 viruses by treatment arm. (E) Spike amino acid (AA) residues constituting each antibody escape score- based putative epitope footprint. (F- I) For each set of residues constituting an antibody epitope footprint for DMS2, PDB7, PDB8, and PDB13, the image shows the set of AA positions comprising the footprint on a Spike monomer NTD or RBD structure. Cyan ribbons
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[41, 45, 937, 133]]<|/det|>
+highlight epitope footprint residues while red ribbons make up the rest of RBD [(F) DMS2, (G) PDB7, and (H) PDB8)] or NTD (I) (PDB13). Residue numbers and cyan dashed lines are used to label footprint residues. Each structure's orientation was chosen to best visualize all residues of a footprint. Residues are colored based on their cluster weights going from white to blue with increasing weight. 
+
+<|ref|>table_caption<|/ref|><|det|>[[62, 151, 80, 163]]<|/det|>
+A 
+
+<|ref|>table<|/ref|><|det|>[[105, 159, 897, 490]]<|/det|>
+<table><tr><td rowspan="2">NTD Features</td><td rowspan="2">No. of Cases (V vs. P)<br/>(Incidence per 100 PYRs)</td><td rowspan="2">VE (%) (95% CI)</td><td colspan="2">Two-sided</td><td colspan="2">Two-sided Differential VE</td></tr><tr><td>Two-sided<br/>P-value</td><td>P-value</td><td>FWER<br/>P-value</td><td>Q-value</td></tr><tr><td>NTD1</td><td></td><td></td><td></td><td>0.0016</td><td>0.0065</td><td>0.0025</td></tr><tr><td>Mark Value = 1</td><td>93 (3.3) vs. 126 (4.6)</td><td>31.0 (10.3, 46.8)</td><td></td><td>0.0055</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>170 (6) vs. 387 (14)</td><td>58.7 (51.0, 65.3)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD3</td><td></td><td></td><td></td><td>0.0017</td><td>0.0065</td><td>0.0025</td></tr><tr><td>Mark Value = 1</td><td>46 (1.6) vs. 50 (1.8)</td><td>12.5 (-30.1, 41.2)</td><td></td><td>0.51</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>217 (7.6) vs. 463 (16.7)</td><td>56.1 (48.9, 62.2)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD4</td><td></td><td></td><td></td><td>0.097</td><td>0.19</td><td>0.12</td></tr><tr><td>Mark Value = 1</td><td>133 (4.7) vs. 223 (8.1)</td><td>45.3 (33.0, 55.4)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>130 (4.6) vs. 290 (10.5)</td><td>57.2 (47.9, 64.9)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD5</td><td></td><td></td><td></td><td>&amp;lt;0.001</td><td>0.0021</td><td>0.0012</td></tr><tr><td>Mark Value = 1</td><td>132 (4.6) vs. 188 (6.8)</td><td>35.2 (19.8, 47.6)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>131 (4.6) vs. 325 (11.7)</td><td>61.8 (53.7, 68.6)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD6</td><td></td><td></td><td></td><td>0.12</td><td>0.19</td><td>0.12</td></tr><tr><td>Mark Value = 1</td><td>77 (2.7) vs. 120 (4.3)</td><td>41.5 (22.7, 55.7)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>186 (6.5) vs. 393 (14.2)</td><td>55.2 (47.1, 62.0)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>NTD7</td><td></td><td></td><td></td><td>&amp;lt;0.001</td><td>0.0013</td><td>0.0012</td></tr><tr><td>Mark Value = 1</td><td>117 (4.1) vs. 157 (5.7)</td><td>31.7 (14.2, 45.7)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr><tr><td>Mark Value = 0</td><td>146 (5.1) vs. 356 (12.9)</td><td>61.0 (53.2, 67.5)</td><td></td><td>&amp;lt;0.001</td><td></td><td></td></tr></table>
+
+<|ref|>image<|/ref|><|det|>[[60, 515, 415, 870]]<|/det|>
+ 
+
+<|ref|>image_caption<|/ref|><|det|>[[44, 909, 120, 927]]<|/det|>
+<center>Figure 5</center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[39, 44, 959, 316]]<|/det|>
+In the Latin America cohort, NTD sequence feature sieve analysis and neutralization phenotype sieve analysis. (A) Vaccine efficacy (VE) estimates against the primary COVID- 19 endpoint caused by SARS- CoV- 2 with (vs. without) a NTD feature value, screened in as a specific hypothesis- driven neutralizing antibody (nAb) correlate of protection. VE estimates against SARS- CoV- 2 harboring the NTD feature value are shown in blue; those against SARS- CoV- 2 without the NTD feature value are shown in maroon. (B) VE against the primary COVID- 19 endpoint by geometric fold change in neutralizing antibody titer against the disease- causing SARS- CoV- 2 variant vs. against the D614G Reference strain. The top plot shows the numbers of cases by treatment arm and color- coded by lineage. The bottom plot shows the estimated vaccine efficacy by geometric fold change in nAb titer against the disease- causing SARS- CoV- 2 variant vs. against the D614G Reference strain. The dashed lines are pointwise 95% confidence intervals. The dots are VE point estimates against the given lineage, with the vertical bars showing 95% confidence intervals.  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 339, 311, 367]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 389, 765, 410]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 426, 611, 473]]<|/det|>
+SupplementaryData1. csv  ENSEMBLEsieveSuppNatCommun16May2023tosubmit.docx
+
+<--- Page Split --->

preprint/preprint__988c6e6669851bc7703fb2127307b0846f2be30632bdbe99cbb1061f47c152b9/images_list.json

@@ -0,0 +1,40 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "A Population 4 bound to CD4 and FP-directed antibody VRC34.01",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 29
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_1.jpg",
+    "caption": "B Population 5 bound to CD4 and FP-directed antibody VRC34.01",
+    "footnote": [],
+    "bbox": [
+      [
+        102,
+        303,
+        900,
+        472
+      ]
+    ],
+    "page_idx": 29
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_2.jpg",
+    "caption": "C Step-wise CD4-induced Env opening.",
+    "footnote": [],
+    "bbox": [
+      [
+        102,
+        512,
+        902,
+        820
+      ]
+    ],
+    "page_idx": 29
+  }
+]

preprint/preprint__988c6e6669851bc7703fb2127307b0846f2be30632bdbe99cbb1061f47c152b9/preprint__988c6e6669851bc7703fb2127307b0846f2be30632bdbe99cbb1061f47c152b9.mmd

@@ -0,0 +1,470 @@
+
+# Conformational trajectory of the HIV-1 fusion peptide during CD4-induced envelope opening  
+
+Priyamvada Acharya  
+
+priyamvada.acharya@duke.edu  
+
+Duke University https://orcid.org/0000- 0002- 0089- 277X  
+
+Bhishem Thakur Duke University  
+
+Revansiddha Katte University of Texas at Tyler Health Science Center  
+
+Wang Xu University of Texas at Tyler Health Science Center https://orcid.org/0000- 0003- 4452- 9240  
+
+Katarzyna Janowska Duke Human Vaccine Institute  
+
+Salam Sammour Duke Human Vaccine Institute  
+
+Rory Henderson  
+
+Duke Human Vaccine Institute; Department of Medicine; Department of Immunology, Duke University https://orcid.org/0000- 0002- 4301- 6382  
+
+Maolin Lu Duke University  
+
+Peter Kwong Columbia University  
+
+Biological Sciences - Article  
+
+Keywords:  
+
+Posted Date: November 6th, 2024  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 5090208/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: There is NO Competing Interest.
+
+<--- Page Split --->
+
+Version of Record: A version of this preprint was published at Nature Communications on May 17th, 2025. See the published version at https://doi.org/10.1038/s41467-025-59721-2.
+
+<--- Page Split --->
+
+# Conformational trajectory of the HIV-1 fusion peptide during CD4-induced envelope opening  
+
+Bhishem Thakur \(^{1}\) , Revansiddha H. Katte \(^{2}\) , Wang Xu \(^{2}\) , Katarzyna Janowska \(^{1}\) , Salam Sammour \(^{1}\) , Rory  
+
+Henderson \(^{1,3}\) , Maolin Lu \(^{2}\) , Peter D. Kwong \(^{4,5}\) , Priyamvada Acharya \(^{1,6,7*}\)  
+
+## Affiliations:  
+
+\(^{1}\) Duke Human Vaccine Institute, Durham NC 27710, USA \(^{2}\) Department of Cellular and Molecular Biology, School of Medicine, University of Texas at Tyler Health Science Center, Tyler, Texas, 75708, USA \(^{3}\) Duke University, Department of Medicine, Durham NC 27710, USA \(^{4}\) Aaron Diamond AIDS Research Center, Columbia University Vagelos College of Physicians and Surgeons, New York, NY 10032, USA \(^{5}\) Vaccine Research Center, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892, USA \(^{6}\) Duke University, Department of Surgery, Durham NC 27710, USA \(^{7}\) Duke University, Department of Biochemistry, Durham NC 27710, USA  
+
+\*To whom correspondence should be addressed  
+
+Correspondence to: Priyamvada Acharya (priyamvada.acharya@duke.edu)
+
+<--- Page Split --->
+
+## Abstract  
+
+The hydrophobic fusion peptide (FP), a critical component of the HIV- 1 entry machinery, is located at the N terminal stretch of the envelope (Env) gp41 subunit \(^{1,3}\) . The receptor- binding gp120 subunit of Env forms a heterodimer with gp41 and assembles into a trimer, in which FP is accessible for antibody binding \(^{3}\) . Env conformational changes or “opening” that follow receptor binding result in FP relocating to a newly formed interprotomer pocket at the gp41- gp120 interface where it is sterically inaccessible to antibody \(^{4}\) . The mechanistic steps connecting the entry- related transition of antibody accessible- to- inaccessible FP configurations remain unresolved. Here, using SOSIP- stabilized Env ectodomains \(^{5}\) , we visualized atomic- level details of a functional entry intermediate, where partially open Env was bound to receptor CD4, co- receptor mimetic antibody 17b, and FP- targeting antibody VRC34.01, demonstrating that FP remains antibody accessible despite substantial receptor- induced Env opening. We determined a series of structures delineating stepwise opening of Env from its closed state to a newly resolved intermediate and defining downstream re- organizations of the gp120- gp41 interface that ultimately resulted in FP burial in an antibody- inaccessible configuration. Our studies improve our understanding of HIV- 1 entry and provide information on entry- related conformation reorganization of a key site of HIV vulnerability to neutralizing antibody.
+
+<--- Page Split --->
+
+The HIV- 1 envelope glycoprotein (Env), a homotrimer of gp120- gp41 heterodimers, mediates viral entry into host cells. The gp120 subunit engages host receptors, while the gp41 subunit contains a fusion peptide (FP) that is inserted into the host membrane to induce host and virus membrane fusion<sup>6</sup>. Prior to its binding to host receptors, the HIV- 1 Env is characterized by a closed configuration with gp120 protomers packed against each other and the gp41 subunit<sup>1</sup>, while the highly conserved and immunodominant coreceptor- binding region at the Env trimer apex remains occluded by packing of the first and second (V1V2) as well as the third (V3) variable loops (Figure 1A)<sup>2</sup>. At the trimer base, FP comprises a hydrophobic stretch of about 20 amino acids at the gp41 N terminus that is accessible for antibody binding in the closed configuration of Env. FP is a site of vulnerability to broadly neutralizing antibodies (bnAbs) and thus of vaccine focus<sup>3,7,8</sup>.  
+
+HIV- 1 Env uses its gp120 subunits to engage the CD4 receptor on the surface of human immune cells. CD4- induced conformational changes have been structurally characterized in virus- associated Env by cryo- electron tomography (cryo- ET)<sup>9- 11</sup>, while high- resolution structural definition of receptor- induced Env opening has been obtained by single- particle cryo- EM analysis of stabilized, soluble Env ectodomains<sup>12- 14</sup>. Both lines of evidence have synergized to facilitate our understanding of HIV- 1 entry- related and enabled visualization of functionally relevant Env structural changes across resolution scales.  
+
+CD4- induced Env conformational changes, collectively termed as “Env opening”, include rigidity body displacement and rotation of the gp120 subunits resulting in up to \(\sim 40\) - Å shift in the positioning of the V1V2 base (Figure 1B)<sup>4</sup>. Env opening is accompanied by internal rearrangements within gp120 that involve disruption of inter- protomer interactions formed by the gp120 V1V2 and V3 regions, release of the V3 loop, and formation and exposure of the bridging sheet. The V3 loop and bridging sheet are the structural elements that form the binding site for a GPCR coreceptor, either CCR5 or CXCR4<sup>15- 17</sup>. The structural signatures of CD4- induced Env opening include the bridging sheet and the \(\alpha 0\) helix in gp120 that were first defined in crystal structures of CD4- bound monomeric gp120<sup>18</sup>. CD4- induction of Env also
+
+<--- Page Split --->
+
+re- organizes the gp41 subunit<sup>4,19</sup> resulting in burial of FP within a gp41 cavity such that it is no longer accessible for antibody binding (Figure 1 A, B).  
+
+While high- resolution structural details have been elucidated for FP in an antibody- accessible conformation (the closed configuration of Env prior to receptor engagement) and in an antibody- inaccessible conformation after CD4 receptor- induced opening of Env, mechanistic details of this FP relocation remain unclear. Here, we use conformation- sensitive antibodies as molecular probes to simultaneously track the trajectories of Env opening and of FP accessibility. For FP accessibility, we used the prototype FP- directed antibody VRC34.01<sup>3</sup>, isolated from a chronically HIV- 1- infected individual, which binds at an epitope comprised primarily of the gp41 FP residues 512- 519 (contributing \(\sim 55\%\) of total interactive surface area) and gp120 glycan N88 ( \(\sim 26\%\) of the total interactive surface area). For Env opening, we used the CD4- induced (CD4i) antibody 17b to assess the formation and exposure of the bridging sheet upon CD4- triggering of Env. As the formation of the bridging sheet requires disruption of the V1V2 cap at the trimer apex and at least partial Env opening, binding to 17b was also an indicator of Env opening<sup>18,20</sup>.  
+
+Using Env ectodomains stabilized by an intraprotomer gp120/gp41 disulfide and an Ile to Pro change in gp41 (SOSIP)<sup>5</sup>, we performed cryogenic electron microscopy (cryo- EM) to define intermediates where FP remains accessible to antibody binding despite substantial Env opening. Among these conformations were populations with their gp120 promoters either partially rotated from the pre- receptor closed Env conformation or more substantially rotated to resemble the geometry observed in the CD4- induced fully open conformation described previously<sup>4,21</sup>. The partially rotated gp120 were associated with antibody- accessible FP, whereas further gp120 displacement along an axis orthogonal to the central trimer axis resulted in FP burial, suggesting an association of FP burial with the extent of gp120 displacement. Taken together, our data provide evidence that accessibility of FP to antibody binding persists post- receptor engagement despite substantial Env opening. Furthermore, we define the mechanistic steps that lead to FP burial and antibody inaccessibility upon further Env opening. Our
+
+<--- Page Split --->
+
+results resolve several gaps in our knowledge HIV- 1 entry and provide information relevant to the development of vaccines and therapeutics.  
+
+## Fusion peptide remains exposed even after substantial opening of the HIV-1 Env trimer  
+
+To assess CD4- induced changes in FP accessibility, we measured binding to the FP- targeted antibody VRC34.01 at different time points following incubation of BG505.SOSIP Env with either CD4 alone, or together with the fragment antigen binding (Fab) of the coreceptor- mimicking antibody 17b that recognizes an epitope presented upon CD4- induced Env opening (Figure 1C). VRC34.01 binding decreased after CD4- induction, and the decrease was more profound in the presence of 17b Fab. A control experiment without addition of CD4 or 17b showed no change in VRC34.01 binding to BG505.SOSIP Env.  
+
+We next assessed simultaneous changes in FP exposure measured by binding to VRC34.01, and Env opening measured by binding to 17b (Figure 1D). 17b binding increased post CD4 addition, indicating Env opening and exposure of the bridging sheet, while VRC34.01 binding decreased. Although these overall trends of increase in 17b binding and decrease in VRC34.01 binding upon CD4 induction were as expected, we noted the retention of substantial VRC34.01 binding at the time- points where near- saturation 17b binding had occurred, suggesting that the FP remained exposed and available for antibody binding despite substantial Env opening.  
+
+## FP remains accessible for antibody binding after CD4-induced Env opening  
+
+To visualize the impact of CD4- bound Env conformations on FP positioning, we incubated BG505 SOSIP Env with CD4 and 17b Fab, and performed single particle cryo- EM on the Env complexes at selected time- points, 1.3 hour (hr), 20 hr, and 3 days, post CD4/17b addition, with VRC34.01 Fab added 30 minutes before the samples were vitrified for cryo- EM analysis (Figures 1E, 2, S1- S4, Table S1). We identified three particle populations across the three cryo- EM datasets that differed in their stoichiometries of bound VRC34.01 Fab (Figure 1E and Supplemental Figure S4). Population 1 dominated at all three time points
+
+<--- Page Split --->
+
+and consisted of partially open Env in which each of the three gp120- gp41 protomers was bound to one each of CD4, 17b Fab and VRC34.01 Fab. Another population, named Population 2, was detected at all three time points, albeit in smaller proportions relative to Population 1 (Supplemental Figure S4). In Population 2, each of the three gp120 subunits were bound to one each of CD4 and 17b Fab, while only two protomers were bound to VRC34.01 Fab. The proportion of Population 2 relative to Population 1 increased with longer incubation times. At the 3- day time point, a third population, named Population 3, was detected that resembled Populations 1 and 2 in their bound CD4 and 17b stoichiometries but only had a single VRC34.01 Fab bound, leaving two protomers not bound to VRC34.01.  
+
+In summary, we identified three populations of CD4- induced Env in our cryo- EM datasets with differing stoichiometries of bound VRC34.01. These results confirmed that FP remained accessible to VRC34.01 binding despite substantial Env opening and suggested FP accessibility to antibody binding to be hindered at the sites in Populations 2 and 3 that were not bound by VRC34.01.  
+
+## FP remains antibody-accessible in a CD4- triggered partially open Env intermediate on the HIV-1 entry pathway  
+
+Two distinct structural configurations of the FP have thus far been defined in the literature, one that is antibody accessible in the pre- fusion closed \(\mathrm{Env}^{22}\) and a second that is sequestered within a gp41- gp120 pocket in a partially or fully open CD4- induced \(\mathrm{Env}^{4,19}\) . In this study, we have discovered new CD4- induced Env intermediates that are open enough to bind the bridging sheet- directed antibody 17b and yet retain the ability to bind an FP- targeting antibody. To understand Env- structural changes that enable CD4- induced opening, while the FP remains in an antibody- accessible configuration, we first examined Population 1, which was the dominant population in all the cryo- EM datasets (Figures 1D, 2, S1- S4 and Table S1). We selected the Population 1 reconstruction from the 1.3- hr time- point for our analysis as it contained the largest number of particles and the highest resolution among the Population 1 structures from the three datasets.
+
+<--- Page Split --->
+
+In Population 1, the gp120 subunits exhibited known structural markers of the CD4- induced conformation<sup>1,4,18</sup>, including the bridging sheet at the 17b- binding interface and residues 63- 73 assembled into the \(\alpha 0\) helix (Figure 2A). The gp41 subunit appeared conformationally less perturbed. FP bound VRC34.01, with a similar interaction interface as previously observed in the structure with closed BG505 SOSIP dominated by the FP and the glycan at gp120 residue position N88 (Figure S5)<sup>3</sup>. Although no symmetry had been applied during the cryo- EM data processing, the three protomers were highly similar in the symmetrically open Population 1 intermediate (Figure S5). Our Population 1 structure revealed a similar gp120 opening geometry as the cryo- ET structure of membrane- associated HIV- 1<sub>ADA,CM</sub> Env bound to three membrane- associated CD4 molecules (Figure 2B)<sup>11</sup>, suggesting that Population 1 represents a physiologically relevant entry intermediate.  
+
+We next studied the Population 1 structure using a previously defined set of vectors that report on structural rearrangements associated with rigid body movements in gp120 relative to gp41 (Figure 2C)<sup>23</sup>. These vectors describe the orientation of gp120 relative to the gp41 three- helix bundle, capturing rotation of gp120 away from the trimer central axis and rotation orthogonal to the trimer central axis. These measures effectively capture differences between closed, open, and intermediate state Envs. All three Population 1 protomers clustered together in all measures examined and were similar to previously published structures of BG505 (PDB: 6CM3) or B41 (PDB: 6EDU) SOSIP bound to CD4, 17b and 8ANC195 Fab<sup>19</sup>. The Population 1 structures were distinct from previously published open and open occluded state structures (PDBs: 5VN3 and 5VN8, respectively) in gp120 rotations described by a dihedral angle \((\phi)\) that defines orthogonal rotation and angles \((\theta_{1}\) and \(\theta_{2}\) ) describing rotations relative to the trimer central axis (Figure 2C). However, the distance between the gp120 core and W571 was similar between the open and intermediate structures. Contrasting each with the closed state structure clusters indicates the open and open occluded structures occupy distinct angles in the dihedral and the angle between the gp41 three- helix bundle and gp120 termini while the Population 1 and CD4, 17b, 8ANC195- bound Envs differ in the angle describing gp120 rotation away from the central trimer axis. In summary,
+
+<--- Page Split --->
+
+our vector analysis indicates that the partially open Population 1 intermediate described here shifts the gp120 domains away from the central axis, while the open and open- occluded shift the gp120 domains orthogonal to the trimer central axis.  
+
+We compared the configuration of the FP in the previously published partially open CD4, 17b, 8ANC195- bound structure (PDB: 6CM3) and the partially open CD4,17b, VRC34.01- bound Population 1 structure resolved in this study (PDB: 9D90) (Figure 2D). In the Population 1 structure (VRC34.01- bound), the FP was extended out of the Env core to bind the VRC34.01 antibody, whereas, in the 8ANC195- bound structures, the FP was buried in an intra- protomer gp120/gp41 hydrophobic pocket. The formation of the pocket for FP sequestration was facilitated by a shift in the position of the FP proximal region (FPPR) with the major contribution coming from straightening of the FPPR helix that created space for FP burial. The distance between the Cα atoms of FPPR residue Gln540 and the gp120 residue Phe233 measured at 13 Å for the Population 1 structure and at 19 Å for the CD4,17b, 8ANC195 bound BG505 SOSIP structure (Figure 2D). A similar FP configuration was observed in the B41- complex with CD4, 17b and 8ANC195, suggesting that this is an isolate- independent conformational state (Figure S6). The difference in FP accessibility while maintaining overall similar protomer geometry suggested that in this intermediate state the FP can either be antibody- accessible or it can be occluded. While VRC34.01 binds FP and stabilizes its accessible configuration, 8ANC195 stabilizes the FP occluded configuration.  
+
+In summary, we identified a CD4- triggered partially open Env intermediate on the HIV- 1 entry pathway, with a protomer geometry that accommodates an antibody- accessible or a buried FP, with a conformational change in the FPPR being the major facilitator for this conformational switch of FP.  
+
+## Mechanism of downstream FP sequestering and antibody inaccessibility  
+
+In addition to the near symmetric, partially open Population 1 state where VRC34.01 was bound at each of the three FP sites, we also identified populations that were bound to either one or two VRC34.01 Fabs, leaving two and one FP sites unbound, respectively, despite saturating amounts of
+
+<--- Page Split --->
+
+VRC34.01 Fab being used for sample preparation (Figures 1D, 3A-B, S4 and Table S1). As expected, based on binding to antibody 17b, the bridging sheet and the \(\alpha 0\) helix were formed in all gp120 protomers in the Population 2 and Population 3 structures (Figures 3A and B). Examination of these unbound sites revealed FP sequestered within a gp120/gp41 pocket in an antibody-inaccessible configuration (Figure 3C and D), thus providing a structural explanation for the lack of antibody binding to these FP sites.  
+
+Unlike the near- symmetric Population 1 structure, Populations 2 and 3 displayed marked asymmetry. To quantify Env opening, we measured interprotomer distances between the CD4- binding site gp120- residue Asp368 and gp120- residue Pro124 at the Env trimer apex (Figure 3D and F). As previously recognized, the closed and open Env structures showed substantial differences in these distances<sup>21</sup> (Figure 3D). In the closed Env trimer (PDB: 5ACO), the distance between the Asp368 residues and Pro124 residues measured 14.7 Å and 56 Å, respectively. These distances were much larger in the CD4,17b- bound open Env trimer (PDB: 5VN3) at 76.7 Å and 84.3 Å, respectively. By contrast, in the CD4,17b,8ANC195- bound partially open Env (PDB: 6CM3), these distances were intermediate between the open and closed, at 65 Å and 79.4 Å, respectively. Since all three structures were reconstructed by imposing C3 symmetry during cryo- EM map refinement, each of these interprotomer distances was identical within each structure. These distances measured in the Population 1 structure were similar to the distances in the CD4,17b,8ANC195- bound partially open structure (PDB: 6CM3) (Figure 3D and E). Since no symmetry was applied during the reconstruction of the Population 1 map, three distances were noted for each measure: 63 Å, 66 Å, and 66.9 Å for the interprotomer distances between residue Asp368, and 75 Å, 77.6 Å and 78.2 Å for the interprotomer distances between residue Pro124. In Population 2, the two protomers that were bound to VRC34.01 showed similar separation as observed in Population 1, 61.4 Å and 74.6 Å between residues Asp368 and Pro124, respectively. The protomer that was not bound to VRC34.01 showed greater gp120 displacement where these distances approached closer to those observed in the CD4,17b- bound open Env trimer (PDB: 5VN3). In Population 3, the two
+
+<--- Page Split --->
+
+protomers that were not bound to VRC34.01 had buried FPs and showed gp120 geometries closer to the fully open Env conformation.  
+
+Taken together, our results demonstrate that FP burial that renders it inaccessible to an FP- directed antibody requires further Env opening, involving transition of Env geometry past an entry intermediate that occurs earlier on CD4- induced Env opening trajectory. This early intermediate resembles the one previously captured in the cryo- ET structure of the CD4- bound HIV- 1 \(\mathrm{ADACM}\) Env, in the single particle cryo- EM structures of CD4,17b,8ANC195- bound BG505 SOSIP (PDB: 6CM3), CD4,17b,8ANC195- bound B41 SOSIP (PDB: 6EDU), and the CD4,17b, VRC34.01- bound BG505 SOSIP Population 1 resolved in this study (PDB: 9D90). In this early intermediate structure, FP can adopt both a buried (antibody- inaccessible) or an antibody- accessible conformation. This intermediate is characterized by gp120 protomers opening like the petals of a tulip where the Env trimer apex separates and gp120 is displaced from the trimer central axis, as a rigid body hinging about the gp120 N/C termini at the trimer base. For stable sequestration of FP that renders it unavailable for antibody binding, further displacement of the gp120 protomers is needed, in the form of a lateral rotation in a plane roughly parallel to the viral membrane and about an axis orthogonal to the central trimer axis.  
+
+## Conformational changes in gp41 required for stable FP sequestering  
+
+To elucidate the gp41 structural features involved in stable FP sequestration, we examined differences in the vicinity of the FP between the partially open intermediate Population 1 conformation (PDB: 9D90; this study) and the previously described fully open Env conformation (PDB 5VN3) \(^4\) . The fully open structure was characterized by a greater displacement of the gp120 subunits that could be visualized by the clear separation of signature \(\alpha 0\) helix from gp41, while this region in the Population 1 structure, although helical, remained associated with the gp41 subunit (Figure 4A).  
+
+At the FP site, the most striking difference was observed in the gp120/gp41 pocket where FP was buried in the fully open structure versus this region in the partially open intermediate (Figures 4B and C).
+
+<--- Page Split --->
+
+In the CD4,17b- bound fully open structure (PDB; 5VN3), this pocket was much larger and measured at \(\sim 26 \mathrm{\AA}\) between FPPR residue Gln540 and gp120 residue Phe233, with the buried FP adopting an extended loop conformation to fill the larger space of the pocket. By contrast, in the partially open intermediate, this distance measured \(13 \mathrm{\AA}\) in Population 1 (CD4,17b, VRC34.01- bound structure) where the FP was exposed and \(19 \mathrm{\AA}\) in the CD4,17b, 8ANC195- bound structure where the FP was buried and assumed a helical conformation. Progressive straightening of FPPR along with changes in both HR1 and HR2 regions of gp41 orchestrated the enlargement of this pocket, which in the fully open structure assumes an interprotomer character with one of its walls lined with the HR1 helix of the adjacent protomer. Thus, the concerted gp120/gp41 re- organizations that resulted in the formation of a larger FP- binding pocket may be responsible for the stable sequestration of the FP in the fully open structure.  
+
+We observed that the Population 2 gp41 conformation with buried FP was most similar to that of the fully open Env (PDB: 5VN3), with HR1 helices showing close overlap, and the FPPR straightened out further compared to the partially open Population 1 and the 6CM3 structures, albeit not to the extent of the fully open structure (Figure 4D). In both Population 2 and fully open 5VN3 structures, the movement of the HR2 region around residues 638- 662 (indicated by red arrow in Figure 4D) creates room for the FPPR unbending. The gp120/gp41 pocket in this Population 2 protomer measured \(21 \mathrm{\AA}\) , with the cavity size approaching that of the cavity measured in the fully open structure.  
+
+In summary, our data show that the FP undergoes stepwise changes in its configuration as a consequence of CD4- induced movements in gp120 and gp41. From a closed, pre- receptor state where FP is accessible to antibodies, Env proceeds to partially open states where FP remains available to the FP- directed antibody VRC34.01. Only upon more extensive rotation of gp120 and widespread changes in gp41 does FP become fully buried within a gp120/gp41 pocket and, as a result, no longer accessible to antibody binding.  
+
+smFRET analysis of Env opening on the virion surface
+
+<--- Page Split --->
+
+smFRET analysis of Env on the surface of intact virions has revealed conformational shifts of virus Env from a pre- triggered (PT) state through a pre- receptor closed (PC) state to a fully open CD4- bound conformational state (CO) in response to CD4 activation \(^{24,25}\) . The pre- fusion, pre- receptor closed Env on virions resembles the FP- accessible Env structure complexed with three VRC34.01 (PDB: 5I8H), while the fully open Env was associated with the symmetric Env structure bound with three CD4 and three 17b (PDB: 5VN3), and a pre- triggered state was suggested that is undefined in currently available structures \(^{24,25}\) . We asked whether the structural differences between partially open VRC34.01- bound Env structures characterized in this study and the fully open FP- sequestered Env would be reflected at the global population level of Env conformations presented on virions. We performed smFRET experiments of two different fluorescently click- labeled Env \(_{\mathrm{BG505}}\) on intact HIV- \(1_{Q23}\) virions \(^{26}\) , in which donor/acceptor fluorescent probes were placed between gp120 V1 and V4 or between gp120 V4 and gp41 \(\alpha 6\) , respectively (Figure S7A). Placing FRET probes at different paired structural elements of Env allowed us to visualize global conformational changes of Env from two different structural perspectives, gp120 V1- V4 and gp120- gp41 (Figure 4E, S7B and S7C, Table S2). Using these two imaging systems, we observed distinct FRET histograms of virus Env in the ligand- free and presence of ligands VRC34.01, VRC34.01+ sCD4+17b, and sCD4+17b (Figure 4E and S7B- E, Table S2). We applied the previously well- defined three- state (PT, PC, CO) Gaussian distributions \(^{26}\) to describe the FRET histograms, which reflect the overall conformational landscape of Env on virions. As expected, ligand- free Env exhibited predominance of the pre- triggered conformation, and Env, in the presence of sCD4 and 17b, prevailed in the fully open CD4- bound state. In the presence of VRC34.01, a decrease was observed in the PT population with an increase in the PC population, consistent with previously published results with the JR- FL Env \(^{3}\) . For the VRC34.01+ sCD4+17b sample, the smFRET histograms suggested that the Env conformational distributions resided between the PC and CO conformations (Figures 4E, S7B- F). Quantifying and comparing the propensity of each primary conformational state occupied by virus Env under ligand- free and different ligand- bound conditions, we observed distinct conformational effect on Env by VRC34.01 in the presence of sCD4+17b, positioned on the Env activation pathway between the
+
+<--- Page Split --->
+
+effect of VRC34.01 alone and the sCD4+17b CO state (Figures 4E, S7). These results were consistent between the observations from the gp120- gp41 (Figure 4E and S7E) and the gp120 V1- V4 structural perspectives (Figure S7B- D). Thus, smFRET analysis of the impact of VRC34.01 on the CD4,17b- bound Env was consistent with VRC34.01 stabilizing an intermediate state on the path of CD4- induced Env opening.  
+
+## Visualizing Preceding CD4-induced transitions of HIV-1 Env  
+
+We next sought to visualize CD4- induced Env conformational transitions that occur upstream to Population 1 by performing single particle cryo- EM analysis on a sample of BG505 SOSIP that was incubated with sCD4 for 2 hr, followed by the addition of VRC34.01 Fab 30 minutes before sample vitrification. As antibody 17b works synergistically with CD4 to open Env (Figure 1C), we rationalized that this strategy where we excluded 17b may allow us to capture earlier stages of CD4- induced Env conformational changes. Two distinct particle populations were revealed in the cryo- EM dataset, which yielded reconstructions of 4.08 Å (Population 4) and 4.14 Å (Population 5) global resolutions. For both populations, all three protomers were bound to one each of CD4 and VRC34.01 Fab (Figures 5A, 5B, S8 and Table S1). The two populations differed in the extent of rotation of their gp120 subunits. In Population 4, one of the three gp120 protomers was rotated roughly to the extent observed in the Population 1 structure, while the other two protomers were minimally rotated from their pre- receptor conformation. The distances between the CD4 binding site residue Asp 368 in the two minimally rotated protomers measured 60.2 Å and was thus closer to the distance observed in the closed, pre- receptor Env ( \(\sim 56\) Å) (Figure 3E) than to the distances measured in the partially open Population 1 intermediate ( \(\sim 75\) - 78.2 Å) (Figure 3F). The third protomer that had rotated was separated in this measure from the two other protomers by 61.7 Å and 66.9 Å. The bridging sheet and the α0 helix, which are the structural components associated with CD4- induced Env opening, were only seen in the rotated gp120 protomers. Although two protomers were minimally rotated, their V1V2 regions were unstructured, thus suggesting that CD4- induced disruption of the V1V2 cap of the closed, pre- receptor Env precedes the rotation of the
+
+<--- Page Split --->
+
+protomers. This is consistent with time- resolved, temperature- jump small- angle x- ray scattering studies that suggest an order- to- disorder transition in the trimer apex precedes Env transitions involving protomer rotation<sup>27</sup>. In Population 5, two gp120 protomers were rotated and showed formation of the bridging sheet and \(\alpha 0\) helix, while the third protomer remained minimally rotated. Both Populations 4 and 5 were bound to VRC34.01 Fab at all the three FP sites, as expected, since the gp120 subunits had not rotated far enough to allow the gp41 conformational changes required for FP burial and steric inaccessibility to antibodies.  
+
+In summary, Population 4 and 5 structures represent conformational states preceding the partially opened Population 1 conformation. Taken together, our results demonstrate sequential CD4- induced opening of the gp120 protomers and are consistent with previous studies that show initiation of CD4 binding to the closed Env begins with a single CD4, which induces opening of gp120 protomers needed for binding of additional CD4 molecules<sup>14,28</sup>.  
+
+## Discussion  
+
+HIV- 1 entry involves sequential receptor mediated conformational changes. Knowledge of how each part of the Env moves, in synergy with other parts, provides clues that have enabled the design of immunogens<sup>4,29,30</sup>. Despite years of intense research, critical gaps remain in our knowledge of the HIV- 1 entry mechanism. In this study, we have addressed one such gap related to the fate of the FP during receptor- mediated Env conformational changes. The HIV- 1 FP is accessible and a target for bnAbs in the closed Env and a target for bnAbs but becomes buried within a gp120/gp41 cavity in the receptor- bound fully open Env. Here, we add new knowledge by elucidating the structure and conformations of a receptor- bound, partially open, functional Env intermediate where the FP remains accessible to FP- directed antibodies (Step 5 of Figure 5C). We elucidate the stepwise mechanism that leads to the formation of this intermediate (Steps 1- 4, Figure 5C), and the stepwise process that leads from this intermediate to the fully open Env with a stably sequestered FP (Steps 6- 8, Figure 5C). We elucidate the gp120 subunit motions associated with an exposed, transiently buried or a stably sequestered FP. From
+
+<--- Page Split --->
+
+its receptor- bound, fully open conformation, further conformational changes are needed for Env to release its fusion peptide for insertion into the host membrane. Future studies will reveal the mechanistic details of these downstream steps in the HIV- 1 entry and virus- cell membrane fusion processes.  
+
+The FP site of vulnerability on HIV- 1 is one of the few sites of Env vulnerability against which vaccination has succeeded in elicited antibodies of over \(50\%\) neutralization breadth<sup>8</sup>. Like VRC34.01, these antibodies recognize the accessible conformation of FP in the prefusion- closed state, with the revealed mechanistic details indicating antibody recognition of FP extends into the early entry intermediates identified in this study. Thus, in both the prefusion- closed state as well as early entry intermediates, FP appears to have considerable conformational flexibility, which is crucial for vaccine priming with flexible peptides linked to carriers<sup>7,31</sup>  
+
+From being solvent exposed and accessible to antibodies to becoming transiently, then stably sequestered in an antibody inaccessible conformation, to finally being released from this sequestered state to insert into the host membrane and mediate fusion, the HIV- 1 FP follows a trajectory that is unique among Type 1 fusion proteins. The typical norm for other viruses is to hold their FP in a partially or wholly occluded, sometimes metastable conformation in the pre- receptor state, which receptor binding mediated conformational changes then release for insertion into the host membrane<sup>32</sup>. For SARS- CoV- 2, for example, FP- directed antibodies have been described but these antibodies cannot access their FP epitopes in the pre- receptor conformation of the spike protein<sup>33,36</sup>. Receptor binding mediated conformational changes reveal these cryptic epitopes to allow antibody binding<sup>34,37,38</sup>. In summary, the HIV- 1 FP tracks a unique trajectory among Type 1 fusion proteins with its multistep receptor- induced conformational transitions. Despite these differences, the central role of receptor- induced conformational changes in controlling and maneuvering the FP through its pre- receptor conformation to its fusion competent state is a common feature amongst all Type 1 fusion proteins.  
+
+One apparent conundrum that this study elucidates is the genesis of the FP site of in HIV- 1. Why in a virus that has evolved exemplary defenses to shield its vulnerabilities from the immune system,
+
+<--- Page Split --->
+
+would the highly conserved FP be exposed and a focus for targeting by broadly neutralizing antibodies? Why would it not be hidden from the immune system within the pre- receptor, closed Env? In this study, we have shown that at least partial opening of the HIV- 1 Env trimer is required for burial of FP within a gp41 cavity that forms because of this opening. While a partially open Env can accommodate FP burial, this study demonstrated that such burial is not stable and only upon more substantial Env opening does FP become stably sequestered. Since opening of Env, even partial, exposes epitopes that make the virus more susceptible to neutralization, on balance, it may have been advantageous for maintaining the neutralization resistant compact and closed HIV- 1 Env conformation to leave the FP exposed, and shield it with a few strategically placed glycans. The exposure of the FP, and thus the creation of the site of vulnerability to antibody, may be a mitigating step towards preventing greater vulnerability due to Env opening that may accompany burial of the flexible hydrophobic FP.  
+
+Another partially open Env conformation that has been described is the open- occluded conformation that is sampled in a receptor- independent manner by both the native Env as well as Env ectodomain constructs used for vaccine applications<sup>39</sup>. A recent preprint combining MD simulations and smFRET analysis has proposed the open- occluded conformation as neutralization- relevant intermediate of Env on the transition trajectory<sup>40</sup>. Structures of the open- occluded Env ectodomain show FP to be buried within a gp41 cavity<sup>39</sup>. Indeed, the hydrophobic solvent- exposed FP may be a source of Env metastability and its proclivity to shield itself, even if transiently, within a hydrophobic pocket may provide the underlying rationale for the accessibility of partially open Env conformations that allow such FP occlusion.  
+
+In summary, our study provides a stepwise mechanism for the receptor- induced opening of the HIV- 1 Env and elucidates the trajectory of the fusion peptide from its solvent exposed configuration in the pre- receptor, closed Env to its buried, antibody- inaccessible configuration in the receptor- bound, fully open Env. Collective evidence in our study suggests that the structure of Population 1 represents a general intermediate on the HIV- 1 entry pathway and that this intermediate is accessible for
+
+<--- Page Split --->
+
+binding to broadly neutralizing antibodies such as VRC34.01 and 8ANC195. The evidence includes the recurrence of similar geometry in structures obtained from different isolates (BG505 or B41) and in complex with different gp120/gp41 targeting antibodies (8ANC195 and VRC34.01), and their concurrence with the cryo- ET resolved structure of the CD4- bound HIV- 1ADA.CM Env in the membrane context. Our work thus reveals a functional intermediate with subunit geometry compatible with both a solvent exposed, antibody accessible and an occluded/buried, antibody inaccessible FP configurations. By elucidating the accessibility of FP, a major target for bnAbs, during receptor- induced Env conformational changes our study reveals key insights into this site of vaccine focus.  
+
+## Acknowledgements  
+
+Cryo- EM data were collected at the Duke Krios at the Duke University Shared Materials Instrumentation Facility (SMIF), a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), which is supported by the National Science Foundation (award number ECCS- 2025064) as part of the National Nanotechnology Coordinated Infrastructure (NNCI). This study utilized the computational resources offered by Duke Research Computing (http://rc.duke.edu; NIH 1S10OD018164- 01) at Duke University. This work was supported by NIH grants R01 AI145687 (P.A.), U54 AI170752 (P.A., R.H. and M.L.), R01 AI181600 from NIH/NIAID, an R35 GM151169 from NIH/NIGMS to M.L., and the Vaccine Research Center, an Intramural Division of NIAID, NIH.  
+
+## Author contributions  
+
+P.A. conceived the project and oversaw the study. B.T. and P.A. designed binding studies and cryo- EM experiments. B.T. expressed and purified proteins, performed SPR assays, optimized specimen, and prepared cryo- EM grids, collected cryo- EM data, performed map and coordinate refinement, and performed structural analysis. R.K., W.X., and M.L. performed smFRET analysis of virus Env, K.J.
+
+<--- Page Split --->
+
+419 assisted with cryo- EM data collection. S.S. assisted with protein purification. R.H. performed vector analysis. P.D.K. assist with initial SPR experiments. B.T. and P.A. wrote the first draft of the manuscript. 421 All authors reviewed and commented on the manuscript.
+
+<--- Page Split --->
+
+## 422 Methods  
+
+Protein expression and purification  
+
+HIV- 1 Env ectodomain constructs used in this study were purified form HEK293S GnTI cells (Thermo Fisher Scientific) diluted at the time of transfection to \(1.25 \times 10^{6}\) cells/mL. Before transfection, cells were diluted in Freestyle™ 293 Expression Medium (Cat No. 12338018) to \(1.25 \times 10^{6}\) cells/mL at a volume of \(950 \mathrm{mL}\) . Plasmid DNA expressing the Env ectodomain and furin were co- transfected at a 4:1 ratio ( \(650 \mu \mathrm{g}\) and \(150 \mu \mathrm{g}\) per transfection liter, respectively) and incubated with 293fectin™ transfection reagent (ThermoFisher Cat No. 12347019) in Opti- MEM I Reduced Serum Medium (ThermoFisher Cat No. 31985062). The diluted mixture was added to the cell culture which was incubated at \(37^{\circ}\mathrm{C}\) , \(9\%\) \(\mathrm{CO_2}\) on a shaker at \(120 \mathrm{rpm}\) for 6 days. On day 6 the cell supernatant was harvested by centrifuging the cell culture at \(4000 \mathrm{xg}\) for 30 minutes. The supernatant was filtered with a \(0.45 \mu \mathrm{m}\) PES filter and concentrated to approximately \(100 \mathrm{mL}\) using a Vivaflow® 200 cross- flow cassette (Sartorius Cat No. VF20P2).  
+
+The cell culture supernatant was passed through \(10 \mathrm{mL}\) PGT145 IgG- conjugated affinity column equilibrated in \(20 \mathrm{mM}\) PBS, pH 7.5. Following loading and washing, Env trimers were eluted using \(3 \mathrm{M}\) \(\mathrm{MgCl}_2\) , pH 7.2. The eluted Env were concentrated to \(\sim 1 \mathrm{mL}\) with a Centricon- 70 \(100 \mathrm{kDa}\) filter (Millipore Sigma). After concentrating, Env were filtered through \(0.22 \mu \mathrm{M}\) filter to remove any aggregates before loading on Superose 10/300 GL column (Cytiva) size exclusion column pre- equilibrated in PBS on an AKTA Pure (Cytiva) system. The fractions corresponding to the Env trimers were pooled, concentrated, flash frozen in liquid nitrogen frozen for long- term storage at \(- 80^{\circ}\mathrm{C}\) .  
+
+Antibodies were produced in Expi293 cells and purified using a Protein A affinity column followed by size exclusion chromatography using a HiLoad Superdex 200 column equilibrated in \(20 \mathrm{mM}\) PBS, pH 7.5, \(0.002\%\) w/v Azide.
+
+<--- Page Split --->
+
+4- domain CD4 was produced in Expj293 cells and purified by Q425- affinity chromatography, followed by size exclusion chromatography using a HiLoad Superdex 200 column equilibrated in 20mM PBS, pH 7.5, 0.002% w/v Azide.  
+
+## Surface Plasmon Resonance  
+
+Surface Plasmon ResonanceSurface Plasmon Resonance binding assays were performed on a T- 200 Biacore system (GE- Healthcare) operating at \(25^{\circ}\mathrm{C}\) . HBS- EP+ (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA and 0.05% surfactant P- 20) was used as running buffer. A 40 nM solution of BG505.SOSIP prepared in the running buffer was incubated at \(25^{\circ}\mathrm{C}\) with either 200 nM of CD4, or with 200 nM CD4 and 200 nM 17b Fab. The binding surface was prepared by flowing 100 nM each of, 17b IgG and VRC34.01 IgG over each flow cells 2 and 4, respectively at \(10~\mu \mathrm{l} / \mathrm{min}\) flow rate for 30 seconds with the \(1^{\mathrm{st}}\) and \(3^{\mathrm{rd}}\) flow cells serving as reference for \(2^{\mathrm{nd}}\) and \(4^{\mathrm{th}}\) flow cells, respectively. After surface preparation, the analyte (either BG505 SOSIP alone or BG505 SOSIP with CD4 or BG505 SOSIP with CD4 and 17b Fab) was flowed at \(30~\mu \mathrm{l} / \mathrm{min}\) flow rate for 60 seconds. The same injections were carried out using HBS- EP buffer in order to obtain a reference curve. The sensorgrams were blank corrected in Biacore T- 200 evaluation software.  
+
+## Cryo-EM  
+
+Purified BG505 SOSIP.664 trimer were diluted to a concentration of \(1.3\mathrm{mg / mL}\) and were incubated with five molar excess of 4D CD4 and 5- molar excess of 17b Fab. After mixing, the samples were incubated at \(25^{\circ}\mathrm{C}\) for different incubation times. VRC34.01 Fab in 5- fold molar excess concentration was added 30 minutes before freezing grids. To prevent interaction of the trimer complexes with the air- water interface during vitrification, the samples were incubated in \(0.085\mathrm{mM}\) n- dodecyl \(\beta\) - D- maltoside (DDM). A \(3.5\mathrm{- }\mu \mathrm{L}\) drop of protein was deposited on a Quantifoil- 1.2/1.3 grid (Electron Microscopy Sciences, PA) that had been glow discharged for 10 s using a PELCO easiGlow Cleaning System (Ted Pella). After a 30 seconds incubation in \(>95\%\) humidity, excess protein was blotted away for 2.5 s before being plunge frozen into liquid ethane
+
+<--- Page Split --->
+
+using a Leica EM GP2 plunge freezer (Leica Microsystems). Frozen grids were imaged using a Titan Krios (Thermo Fisher) equipped with a K3 detector (Gatan). The cryoSPARC (Punjani et al., 2017) software was used for data processing<sup>41</sup>. Raw movies were motion corrected using Patch Motion Correction and Contrast Transfer Function (CTF) were estimated. Micrographs with CTF estimates greater than 8 Å were discarded. Automated blob picker software was used to assign the particle position, and the particles were extracted with the 320- pixel extraction box size Fourier cropped to 80 pixels. Following particle extraction, multiple rounds of 2D classification was performed to remove junk particles and re- extraction of clean particles with 320 pixel box size. A reference free ab- initio 3D reconstruction was used to create 3D reconstructions representing diverse conformational states of the Env. Further, multiple rounds of heterogeneous refinement was performed to get rid of the noise. Finally, non- uniform refinement was used on the pure set of particles to get high resolution cryo- EM map.  
+
+Phenix, Coot, Pymol, Chimera, ChimeraX and Isolde were used for model building and refinement<sup>42- 47</sup>.  
+
+Virus packaging and fluorescent labeling  
+
+The methods of packaging and fluorescent labeling of replication- incompetent amber- free HIV- \(1_{Q23}\) viral particles with incorporated EnvBG505 have been described previously<sup>48</sup>. HIV- 1 virions that lack reverse transcriptase (ΔRT) were prepared and used for imaging. Amber- free HIV- \(1_{Q23}\) virions incorporated with two different double- tagged Env were used in this study, including dual- amber N136TAG S401TAG and hybrid click/peptide V4- A1 R542TAG. Amber- free V1V4 N136\* S401\* (\\*, unnatural amino acid - ncAA) viruses carrying click- chemistry- reactive ncAA at 136 in V1 and 401 in V4 were produced by co- transfecting HEK293T cells with a tag- free ΔRT plasmid, an Env- tagged variant N136TAG S401TAG (TAG, amber stop codon) ΔRT plasmid, and an amber suppressor plasmid tRNA<sup>Py1</sup>/NESPylRS<sup>AF</sup>. The amber suppressor can express tRNA and its cognate amino acid acyl- tRNA- synthetase in HEK293T cells. ncAA TCO\* (250 μM)
+
+<--- Page Split --->
+
+was added to the transfection system. Similarly, V4A1 R542\* viruses were prepared using the Env- tagged V4A1 (peptide A1 tag: DSLDMLEM in V4 loop) R542TAG \(\Delta\) RT plasmid. The ratio of tag- free vs. tagged Env plasmids used during transfection was adjusted based on previously characterized Env expression levels \(^{48}\) to ensure that, statistically, on average, one tagged protomer within an Env trimer on a virion was available for fluorescent labeling (enzymatically or click) \(^{48 - 51}\) . 40 hours post- transfection, the supernatant was harvested and filtered, then viruses were concentrated at 25000 rpm for 2 hours using an ultracentrifuge. Next, the virus pellet was resuspended using the labeling buffer containing 50 mM HEPES, 10 mM MgCl₂, and 10 mM CaCl₂. The fluorescent labeling of prepared virus Env was similar to the previously described \(^{48 - 51}\) . For the amber- free V1V4 N136\* S401\* viruses, two TCO\* were fluorescently labeled by 0.1 μM tetrazine- conjugated LD555- TTZ and LD655- TTZ by click chemistry. For the amber- free V4A1 R542\* viruses, the A1 peptide in V4 was labeled by LD655- CoA, 0.65 μM in the presence of enzyme AcpS (5 μM), and the TCO\* in gp41 R542 were click labeled by LD555- TTZ. Dyes were customized by Lumidyne Technologies. The above reaction mixture was incubated at room temperature overnight in the dark. PEG2000- biotin was then added at a final concentration of 0.1 mg/ml to the labeled viruses, followed by 30 min incubation at room temperature. Then, the labeled viruses were further purified using a 6%- 18% gradient of Opti- prep (Sigma- Aldrich) and centrifuged at 40,000 rpm for 1 hour at 4°C.  
+
+smFRET data acquisition and analysis of virus Env  
+
+All single- molecule fluorescence resonance energy transfer (smFRET) data of fluorescently labeled viruses were collected using a custom- made prism- based total internal reflection fluorescence (prism- TIRF) microscope equipped with a fluorescence signal detection system. The detailed operating manual has been described previously \(^{48}\) . Briefly, the sample loading module, a streptavidin- coated PEG passivated biotin quartz imaging chamber, was cleaned with the imaging buffer, and the background fluorescence signal was removed using the high- intensity laser. The imaging buffer contains 50 mM Tris pH 7.4, 50 mM NaCl, a cocktail of triplet- state quenchers, and oxygen scavenger: 2 mM protocatechuic acid and 8 nM protocatechuic- 3,4- dioxygenase. The labeled viruses were then loaded into the sample loading module. Un
+
+<--- Page Split --->
+
+immobilized viruses were removed using the imaging buffer, and the fluorescence signals were collected. Under the ligand- present experimental conditions, fluorescently labeled viruses were incubated with the indicated \(0.1\mathrm{mg / ml}\) antibody/ligand ( \(>5\mathrm{x}\) above IC95) for 30 mins at room temperature before imaging. All fluorescence signals were recorded simultaneously on two synchronized sCMOS cameras (Hamamatsu ORCA- Flash4.0 V3) at \(25\mathrm{Hz}\) for 80 seconds. The smFRET data were viewed, processed, and analyzed using the SPARTAN software package \(^{52}\) shared by the Scott Blanchard lab and custom MATLAB- based scripts.  
+
+Recorded 80- second movies (2000 frames/movie) were extracted as donor/acceptor fluorescence traces (time series) with background subtracted and crosstalk corrected. The energy transfer efficiency (FRET efficiency values or simplified as FRET in figures) from the donor fluorophore to the acceptor fluorophore was calculated using \(\mathrm{FRET} = \mathrm{I}_{\mathrm{A}} / (\gamma \mathrm{I}_{\mathrm{D}} + \mathrm{I}_{\mathrm{A}})\) , in which \(\mathrm{I}_{\mathrm{D}}\) and \(\mathrm{I}_{\mathrm{A}}\) represent the fluorescence of the donor and acceptor, respectively, and \(\gamma\) is the correlation coefficient compensating for variations in detection efficiencies. FRET traces (FRET efficiency traces) were further derived. FRET traces reflect real- time relative distance changes between donor and acceptor, resulting from the global conformational dynamics of Env. Under each experimental condition, approximately more than 200 individual traces were included in the final FRET histogram. These included traces meet the following filter settings: 1) a single photo bleaching point (ruling out cases of multiple labeled protomers in a trimer, multiple labeled Envs on one virion, no- labeled Env on one virion; 2) sufficient signal- to- noise ratio; 3) anti- correlated feature between donor and acceptor fluorescence (indicating active Env undergoing conformational changes, ruling out inactive Env as well as Env lacking either donor or acceptor or both). We used automatic filters in combination with manual visualization to ensure that traces of molecules with only one Cy3/Cy5- labeled protomer in a trimer on a viral particle were included for further data processing. FRET traces that meet all the above- mentioned criteria were included to compile FRET histograms/distributions. FRET histograms (conformational distributions) were presented as mean \(\pm\) s.e.m. and fitted into a sum of three distinct Gaussian/Normal distributions using the least- squares fitting algorithm in MATLAB. Parameters were
+
+<--- Page Split --->
+
+determined based on visual inspection of all traces that exhibit state- to- state transitions and the idealization of individual traces using three- state hidden Markov modeling. Each Gaussian represented one conformational state of virus Env. The area under each Gaussian curve was further calculated as an estimation of relative state occupancy, which is the probability of the corresponding state Env occupies.
+
+<--- Page Split --->
+
+549
+
+550
+
+551 **References**
+
+552 1 Pancera, M. et al. Structure and immune recognition of trimeric pre-fusion HIV-1 Env.
+
+553 Nature 514, 455-461 (2014).
+
+554 2 Lyumkis, D. et al. Cryo-EM structure of a fully glycosylated soluble cleaved HIV-1 envelope trimer. Science 342, 1484-1490 (2013).
+
+555 Kong, R. et al. Fusion peptide of HIV-1 as a site of vulnerability to neutralizing antibody. Science 352, 828-833 (2016).
+
+556 Ozorowski, G. et al. Open and closed structures reveal allostery and pliability in the HIV-1 envelope spike. Nature 547, 360-363 (2017).
+
+557 Sanders, R. W. et al. A next-generation cleaved, soluble HIV-1 Env trimer, BG505
+
+561 SOSIP.664 gp140, expresses multiple epitopes for broadly neutralizing but not non-neutralizing antibodies. PLoS Pathog 9, e1003618 (2013).
+
+562 https://doi.org:10.1371/journal.ppat.1003618
+
+564 Harrison, S. C. Viral membrane fusion. Virology 479, 498-507 (2015).
+
+565 Xu, K. et al. Epitope-based vaccine design yields fusion peptide-directed antibodies that neutralize diverse strains of HIV-1. Nature medicine 24, 857-867 (2018).
+
+567 Kong, R. et al. Antibody lineages with vaccine-induced antigen-binding hotspots develop broad HIV neutralization. Cell 178, 567-584. e519 (2019).
+
+569 Liu, J., Bartesaghi, A., Borgnia, M. J., Sapiro, G. & Subramaniam, S. Molecular architecture of native HIV-1 gp120 trimers. Nature 455, 109-113 (2008).
+
+571 Li, Z. et al. Subnanometer structures of HIV-1 envelope trimers on aldrithiol-2-inactivated virus particles. Nature Structural & Molecular Biology 27, 726-734 (2020).
+
+573 Li, W. et al. HIV-1 Env trimers asymmetrically engage CD4 receptors in membranes. Nature 623, 1026-1033 (2023).
+
+575 Harris, A. et al. Trimeric HIV-1 glycoprotein gp140 immunogens and native HIV-1 envelope glycoproteins display the same closed and open quaternary molecular architectures. Proceedings of the National Academy of Sciences 108, 11440-11445 (2011).
+
+579 13 Stadtmueller, B. M. et al. DEER spectroscopy measurements reveal multiple conformations of HIV-1 SOSIP envelopes that show similarities with envelopes on native virions. Immunity 49, 235-246. e234 (2018).
+
+582 Dam, K.-M. A., Fan, C., Yang, Z. & Bjorkman, P. J. Intermediate conformations of CD4-bound HIV-1 Env heterotrimers. Nature 623, 1017-1025 (2023).
+
+584 Huang, C.-c. et al. Structure of a V3-containing HIV-1 gp120 core. Science 310, 1025-1028 (2005).
+
+586 Shaik, M. M. et al. Structural basis of coreceptor recognition by HIV-1 envelope spike. Nature 565, 318-323 (2019).
+
+587 Hoffman, T. L. et al. Stable exposure of the coreceptor-binding site in a CD4-
+
+589 independent HIV-1 envelope protein. Proceedings of the National Academy of Sciences 96, 6359-6364 (1999).
+
+591 Kwong, P. D. et al. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393, 648-659 (1998).
+
+593 19 Wang, H., Barnes, C. O., Yang, Z., Nussenzweig, M. C. & Bjorkman, P. J. Partially open HIV-1 envelope structures exhibit conformational changes relevant for coreceptor binding and fusion. Cell Host & Microbe 24, 579-592. e574 (2018).
+
+<--- Page Split --->
+
+596 20 Thali, M. et al. Characterization of conserved human immunodeficiency virus type 1 597 gp120 neutralization epitopes exposed upon gp120- CD4 binding. Journal of virology 67, 598 3978- 3988 (1993). 599 21 Yang, Z., Wang, H., Liu, A. Z., Gristick, H. B. & Bjorkman, P. J. Asymmetric opening of 600 HIV- 1 Env bound to CD4 and a coreceptor- mimicking antibody. Nature structural & 601 molecular biology 26, 1167- 1175 (2019). 602 Lee, J. H., De Val, N., Lyumkis, D. & Ward, A. B. Model building and refinement of a 603 natively glycosylated HIV- 1 Env protein by high- resolution cryoelectron microscopy. 604 Structure 23, 1943- 1951 (2015). 605 Henderson, R. et al. Disruption of the HIV- 1 Envelope allosteric network blocks CD4- 606 induced rearrangements. Nature communications 11, 520 (2020). 607 24 Lu, M. et al. Associating HIV- 1 envelope glycoprotein structures with states on the virus 608 observed by smFRET. Nature 568, 415- 419 (2019). 609 25 Munro, J. B. et al. Conformational dynamics of single HIV- 1 envelope trimers on the 610 surface of native virions. Science 346, 759- 763 (2014). 611 26 Ao, Y. et al. Bioorthogonal click labeling of an amber- free HIV- 1 provirus for in- virus 612 single molecule imaging. Cell Chemical Biology 31, 487- 501. e487 (2024). 613 27 Bennett, A. L. et al. Microsecond dynamics control the HIV- 1 Envelope conformation. Sci 614 Adv 10, eadj0396 (2024). https://doi.org:10.1126/sciadv.adj0396 615 28 Liu, Q. et al. Quaternary contact in the initial interaction of CD4 with the HIV- 1 envelope 616 trimer. Nature structural & molecular biology 24, 370- 378 (2017). 617 29 Henderson, R. et al. (2020). 618 Joyce, M. G. et al. Soluble prefusion closed DS- SOSIP. 664- Env trimers of diverse HIV- 619 1 strains. Cell reports 21, 2992- 3002 (2017). 620 31 Ou, L. et al. Preclinical development of a fusion peptide conjugate as an HIV vaccine 621 immunogen. Scientific reports 10, 3032 (2020). 622 32 May, A. J., Pothula, K. R., Janowska, K. & Acharya, P. Structures of Langya Virus 623 Fusion Protein Ectodomain in Pre- and Postfusion Conformation. J Virol 97, e0043323 624 (2023). https://doi.org:10.1128/jvi.00433- 23 625 33 Jackson, C. B., Farzan, M., Chen, B. & Choe, H. Mechanisms of SARS- CoV- 2 entry into 626 cells. Nature reviews Molecular cell biology 23, 3- 20 (2022). 627 34 Low, J. S. et al. ACE2- binding exposes the SARS- CoV- 2 fusion peptide to broadly 628 neutralizing coronavirus antibodies. Science 377, 735- 742 (2022). 629 35 Gobeil, S. M.- C. et al. Effect of natural mutations of SARS- CoV- 2 on spike structure, 630 conformation, and antigenicity. Science 373, eabi6226 (2021). 631 36 Gobeil, S. M. et al. Structural diversity of the SARS- CoV- 2 Omicron spike. Mol Cell 82, 632 2050- 2068 e2056 (2022). https://doi.org:10.1016/j.molcel.2022.03.028 633 37 Aguilar, H. C., Henderson, B. A., Zamora, J. L. & Johnston, G. P. Paramyxovirus 634 glycoproteins and the membrane fusion process. Current clinical microbiology reports 3, 635 142- 154 (2016). 636 Baker, K. A., Dutch, R. E., Lamb, R. A. & Jardetzky, T. S. Structural basis for 637 paramyxovirus- mediated membrane fusion. Molecular cell 3, 309- 319 (1999). 638 39 Yang, Z. et al. Neutralizing antibodies induced in immunized macaques recognize the 639 CD4- binding site on an occluded- open HIV- 1 envelope trimer. Nature communications 640 13, 732 (2022). 641 40 Lee, M. et al. HIV- 1- envelope trimer transitions from prefusion- closed to CD4- bound- 642 open conformations through an occluded- intermediate state. bioRxiv, 2024.2007. 643 2015.603531 (2024). 644 41 Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for 645 rapid unsupervised cryo- EM structure determination. Nature methods 14, 290- 296 646 (2017).
+
+<--- Page Split --->
+
+647 42 Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and 648 electrons: recent developments in Phenix. Acta Crystallographica Section D: Structural 649 Biology 75, 861- 877 (2019). 650 43 Afonine, P. V. et al. Real- space refinement in PHENIX for cryo- EM and crystallography. 651 Acta Crystallographica Section D: Structural Biology 74, 531- 544 (2018). 652 Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. 653 Acta Crystallographica Section D: Biological Crystallography 66, 486- 501 (2010). 654 Schrödinger, L. The PyMOL Molecular Graphics System, Version 1.8. (No Title) (2015). 655 Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research 656 and analysis. Journal of computational chemistry 25, 1605- 1612 (2004). 657 Goddard, T. D. et al. UCSF ChimeraX: Meeting modern challenges in visualization and 658 analysis. Protein science 27, 14- 25 (2018). 659 Ao, Y. et al. Bioorthogonal click labeling of an amber- free HIV- 1 provirus for in- virus 660 single molecule imaging. Cell Chem Biol 31, 487- 501 e487 (2024). 661 https://doi.org:10.1016/j.chembiol.2023.12.017 662 Lu, M. et al. Associating HIV- 1 envelope glycoprotein structures with states on the virus 663 observed by smFRET. Nature 568, 415- 419 (2019). https://doi.org:10.1038/s41586- 019- 664 1101- y 665 50 Ma, X. et al. HIV- 1 Env trimer opens through an asymmetric intermediate in which 666 individual protomers adopt distinct conformations. Elife 7, e34271 (2018). 667 https://doi.org:10.7554/eLife.34271 668 Munro, J. B. et al. Conformational dynamics of single HIV- 1 envelope trimers on the 669 surface of native virions. Science 346, 759- 763 (2014). 670 https://doi.org:10.1126/science.1254426 671 52 Juette, M. F. et al. Single- Molecule imaging of non- equilibrium molecular ensembles on 672 the millisecond timescale. Nat Methods 13, 341- 344 (2016). 673 https://doi.org:10.1038/nmeth.3769
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+  
+
+![](images/Figure_unknown_1.jpg)
+  
+
+E Cryo- EM identified partially open populations with antibody accessible FP  
+
+![](images/Figure_unknown_2.jpg)
+
+
+<--- Page Split --->
+
+Figure 1. Time- dependent conformational changes in HIV- 1 BG505 SOSIP Env upon incubation with CD4. A. Structure of pre- fusion, pre- receptor, closed HIV- 1 Env (PDB: 518H) bound to broadly neutralizing, fusion peptide- directed antibody VRC34.01. The Env is shown in surface representation with the gp120 subunits colored light gray, and within the gp120 subunits, the V1V2 loop colored wheat, V3 loop olive and the residues contributing to the bridging sheet in the open Env colored in red. The gp41 subunits are colored black with the fusion peptide (FP) colored cyan. The antibody VRC34.01 is shown in ribbon representation bound to its FP- centered epitope. B. Structure of pre- fusion, CD4- bound open HIV- 1 Env bound to CD4- induced antibody 17b. The Env is colored similarly as in panel A. CD4 is shown as a yellow ribbon and 17b Fab is shown as an orange ribbon. C. Surface plasmon based binding (SPR) analysis monitoring FP burial. Env was incubated with either sCD4 alone or with CD4 and the coreceptor mimicking antibody 17b. At different time- points after incubation, binding was measured to the fusion peptide targeting antibody VRC34.01. D. Simultaneous Env opening and fusion peptide burial were measured by incubating Env with CD4 and at different time- points injecting over a VRC34.01 IgG or a 17b IgG surface. E. Cryo- EM reconstructions of three distinct populations of CD4/17b- bound, partially open Env bound to VRC34.01. Population 1 is bound to VRC34.01 at all three sites. The blue arrows indicate sites unoccupied by VRC34.01 in Populations 2 and 3.
+
+<--- Page Split --->
+
+A Population 1 bound to CD4, 17b, and FP- directed antibody VRC34.01  
+
+![PLACEHOLDER_31_0]
+  
+
+B Population 1 fit into in situ cryo- ET of CD4- bound Env intermediate  
+
+![PLACEHOLDER_31_1]
+  
+
+C Population 1 comparison to Env in diverse conformations  
+
+![PLACEHOLDER_31_2]
+  
+
+D Two distinct intermediate configurations of the HIV- 1 Env FP  
+
+![PLACEHOLDER_31_3]
+
+
+<--- Page Split --->
+
+Figure 2: A partially open intermediate on the HIV- 1 entry pathway retains FP accessibility to antibody binding. A. Three views of the Population 1 structure shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. Within gp120, the bridging sheet is colored red and the α0 helix green. B. Population 1 coordinates including Env (gp120 in light gray, gp41 in black) and CD4 (yellow) fitted into the in situ cryo- ET reconstruction of a partially open CD4- bound Env (EMD- 29294). C. (Left to right) Vectors describing the position of gp120 relative to gp41. The gp120 structure (blue), gp120 V1/V2 region (green), and gp41 (orange) in the closed state overlayed with the centroid locations depicting the dihedral, angles, and distances describing the position of gp120 relative to gp41. Dihedral, angle, and distance values for closed, intermediate, and open state structures. D. (Left) Population 1 protomer shown in surface representation zoomed- in at the location of the FP. FP is shown in cartoon representation. The gp120 subunit is colored light grey, gp41 black, FP cyan and FPPR pale green. (Middle) One protomer of the partially open Env bound to CD4, 17b Fab and 8ANC195 Fab (PDB ID: 6CM3) shown in surface representation zoomed- in at the location of the FP. FP is shown in cartoon representation. The gp120 subunit is colored dark grey, gp41 magenta, FP dark teal and FPPR light pink. (Right) Overlay of a Population 1 protomer with a protomer of a partially open Env bound to CD4, 17b Fab and 8ANC195 Fab (PDB ID: 6CM3). The gp120 subunits of each structure were used for the superposition. Inset zooms in on the FP and FPPR. Zoomed- in panel is slightly rotated compared to the zoomed- out view for better visualization. The solid lines (colored pale green for Population 1 and light pink for 6CM3) show the distance between FPPR residues Gln 540 and gp120 residue Phe 223.
+
+<--- Page Split --->
+
+A Population 2 bound to CD4, 17b, and 2X FP- directed antibody VRC34.01  
+
+![PLACEHOLDER_33_0]
+  
+
+![PLACEHOLDER_33_1]
+  
+
+## E Extent of Env opening  
+
+Env: BG505 SOSIP Ligands: CD4, 17b, 8ANC195 Conformation: Closed (PDB:5ACO)  
+
+B41 SOSIP CD4, 17b Open (PDB:5VN3)  
+
+BG505 SOSIP CD4, 17b, 8ANC195 Partially open (PDB:6CM3)  
+
+![PLACEHOLDER_33_2]
+  
+
+V1- V2 base: 14.7 Å CD4bs: 56 Å  
+
+![PLACEHOLDER_33_3]
+  
+
+76.7 Å 84.3 Å  
+
+![PLACEHOLDER_33_4]
+  
+
+65 Å 79.4 Å  
+
+## F  
+
+Env: BG505 SOSIP Ligands: CD4, 17b, (3X) VRC34.01 Conformation: Partially open (Population 1)  
+
+BG505 SOSIP CD4, 17b, (2X) VRC34.01 Partially open (Population 2)  
+
+BG505 SOSIP CD4, 17b, (1X) VRC34.01 Partially open (Population 3)  
+
+![PLACEHOLDER_33_5]
+  
+
+V1- V2 base: 63 Å, 66 Å, 66.9 CD4bs: 75 Å, 77.6 Å, 78.2 Å  
+
+![PLACEHOLDER_33_6]
+  
+
+61.4 Å, 71.8 Å, 70.1 Å, 74.6 Å, 82.8 Å, 80.2 Å  
+
+![PLACEHOLDER_33_7]
+  
+
+73.6 Å, 71.4 Å, 69.6 Å 61.6 Å, 79.5 Å, 82.8 Å
+
+<--- Page Split --->
+
+Figure 3: Burial of FP upon gp120 opening. A. Three views of the Population 2 structure shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. Within gp120, the bridging sheet is colored red and the α0 helix green. Blue circle headed arrows indicate the gp41 positions that are not bound to VRC4.01 Fab. B. View of Population 3 coordinates from the viral membrane shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. C. Population 2 structure zoomed-in view of gp41 subunit that was not bound to VRC34.01 showing the buried FP in cyan and the FPPR in light green. The cryo-EM reconstruction is shown as a transparent surface with fitted coordinates shown in cartoon representation. D. Population 3 structure zoomed-in view of its two gp41 subunits that were not bound to VRC34.01 showing the buried FP in cyan and the FPPR in light green. The cryo-EM reconstruction is shown as a transparent surface with fitted coordinates shown in cartoon representation. E and F. Extent of Env openness measured as the distance between residue 368 (blue spheres) and residue 124 (red spheres) in E. previously published Env conformational states and F. Env conformational states defined in this study.
+
+<--- Page Split --->
+
+## A State of the FP along CD4-induced Env opening pathway  
+
+![PLACEHOLDER_35_0]
+  
+
+## D gp41 reorganization required to stably sequester FP  
+
+![PLACEHOLDER_35_1]
+  
+
+## E smFRET analysis of impact of VRC34.01 binding on Env opening trajectory  
+
+![PLACEHOLDER_35_2]
+
+
+<--- Page Split --->
+
+Figure 4: FP trajectory upon CD4- induced Env opening. A. HIV- 1 Env structures organized based on extent of opening. Left to right: closed Env bound to VRC34.01 (PDB: 5I8H) with gp41 colored olive, FPPR orange and FP cyan; partially open, bound to CD4, 17b and VRC34.01 (PDB:9D90; This study) with gp41 colored black, FPPR light green and FP cyan; partially open, bound to CD4, 17b and 8ANC195 (PDB:6CM3) with gp41 colored magenta, FPPR light pink and FP teal; partially open, bound to CD4, 17b and VRC34.01 (EMD- 46671; This study) with gp41 colored black, FPPR light green and FP cyan; fully open, bound to CD4 and 17b (PDB:5VN3) with gp41 colored blue, FPPR light blue and FP cyan. The gp120 subunit is colored light grey. Inset shows a zoomed in view of the region around the \(\alpha 0\) helix (green). B. 180° rotated views of structures are shown in A. with a brown square indicating the FP region. The FP is colored cyan except in the partially open CD,17b,8ANC195 bound BG505 structure where the FP is colored teal. C. zoomed in views of the region around the FP. The red arrows indicate the direction of CD4- induced Env opening from the pre- CD4, closed Env to the fully CD4- induced fully open Env. D. Comparisons of gp41 organization between fully open Env with sequestered and inaccessible FP (PDB: 5VN3), and (left) P1 (transiently exposed FP), (middle) partially open Env with transiently buried FP (PDB: 6CM3), and (right) P2 (inaccessible FP). The red arrows in the left and middle panels indicate the HR2 movement that occurs between the partially open and the full open Envs allowing the FPPR to reorient creating space for FP burial. E- F. Three- dimensional presentation (E) and quantification (F) of conformational distribution- indicated FRET histograms shown in from the gp120- gp41 perspective in Figure S7. Virus EnvBG505 samples three primary conformational states (PT: Pre- triggered, PC: Prefusion Closed, and CO: CD4- bound open). PT predominates in the ligand- free condition, while VRC34 shifts the conformational landscape differently from that of the CD4- bound opening.
+
+<--- Page Split --->
+![PLACEHOLDER_37_0]
+
+<center>A Population 4 bound to CD4 and FP-directed antibody VRC34.01 </center>  
+
+![PLACEHOLDER_37_1]
+
+<center>B Population 5 bound to CD4 and FP-directed antibody VRC34.01 </center>  
+
+![PLACEHOLDER_37_2]
+
+<center>C Step-wise CD4-induced Env opening. </center>
+
+<--- Page Split --->
+
+Figure 5: Partially open early intermediates and a step- wise mechanism for CD4- induced Env opening. A, CD4, VRC34.01- bound BG505 SOSIP Env with a single gp120 rotated and showing the formation of the bridging sheet (red) and \(\alpha 0\) helix that are the markers for CD4- induced Env. B, CD4, VRC34.01- bound BG505 SOSIP Env with two gp120s rotated and showing the formation of the bridging sheet (red) and \(\alpha 0\) helix that are the markers for CD4- induced Env. C, A structure- guided mechanism for stepwise Env opening along the HIV- 1 entry pathway. Top panel structures were determined previously, bottom panel structures were determined in this study. Stepwise transitions are marked with numbers within a circle on top of each structure starting from (1) the binding of a single CD4 to a closed Env trimer (PDB: 5U1F, 8FYI). This is followed by (2) opening of the Env trimer (EMD- 29292) that allows additional CD4 molecules to bind. (5) A partially open Env conformation was described bound to CD4, a coreceptor mimicking antibody, and the gp120/gp41 interface targeting antibody 8ANC195 (PDB: 6CM3, 6EDU) where the FP was buried within a gp41 cavity. The CD4- induced opening of the HIV- 1 Env culminates in the complete rotation of all the gp120 subunits that are accompanied by gp41 conformational changes and resulting in the burial of FP. This state is numbered (8) in this schematic. This study showed that the geometry of the functional entry intermediate (5) that was also visualized on membrane- associated Env (EMD- 29294), was compatible with the FP being either buried or exposed, and thus, in this conformation the FP was accessible to antibodies. Further, this study filled in mechanistic gaps between (2) and (5) by showing stepwise gp120 rotation to reach this functional entry intermediate. Finally, this study visualized a stepwise mechanism for how the functional entry intermediate (5) may transition to the fully open Env (8), yet again by stepwise opening of the each gp120 subunit from its partially rotated to the fully rotated conformation, which was accompanied by burial of the FP in the corresponding protomer.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- TableS1.xlsx- FPpaperSupplementfigures14Sept2024.pdf
+
+<--- Page Split --->

preprint/preprint__988c6e6669851bc7703fb2127307b0846f2be30632bdbe99cbb1061f47c152b9/preprint__988c6e6669851bc7703fb2127307b0846f2be30632bdbe99cbb1061f47c152b9_det.mmd

@@ -0,0 +1,577 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 847, 178]]<|/det|>
+# Conformational trajectory of the HIV-1 fusion peptide during CD4-induced envelope opening  
+
+<|ref|>text<|/ref|><|det|>[[44, 196, 225, 215]]<|/det|>
+Priyamvada Acharya  
+
+<|ref|>text<|/ref|><|det|>[[52, 223, 355, 240]]<|/det|>
+priyamvada.acharya@duke.edu  
+
+<|ref|>text<|/ref|><|det|>[[50, 269, 553, 289]]<|/det|>
+Duke University https://orcid.org/0000- 0002- 0089- 277X  
+
+<|ref|>text<|/ref|><|det|>[[44, 294, 191, 333]]<|/det|>
+Bhishem Thakur Duke University  
+
+<|ref|>text<|/ref|><|det|>[[44, 339, 490, 381]]<|/det|>
+Revansiddha Katte University of Texas at Tyler Health Science Center  
+
+<|ref|>text<|/ref|><|det|>[[44, 386, 850, 430]]<|/det|>
+Wang Xu University of Texas at Tyler Health Science Center https://orcid.org/0000- 0003- 4452- 9240  
+
+<|ref|>text<|/ref|><|det|>[[44, 433, 320, 472]]<|/det|>
+Katarzyna Janowska Duke Human Vaccine Institute  
+
+<|ref|>text<|/ref|><|det|>[[44, 478, 320, 518]]<|/det|>
+Salam Sammour Duke Human Vaccine Institute  
+
+<|ref|>text<|/ref|><|det|>[[44, 525, 188, 543]]<|/det|>
+Rory Henderson  
+
+<|ref|>text<|/ref|><|det|>[[50, 547, 933, 590]]<|/det|>
+Duke Human Vaccine Institute; Department of Medicine; Department of Immunology, Duke University https://orcid.org/0000- 0002- 4301- 6382  
+
+<|ref|>text<|/ref|><|det|>[[44, 595, 191, 634]]<|/det|>
+Maolin Lu Duke University  
+
+<|ref|>text<|/ref|><|det|>[[44, 640, 230, 680]]<|/det|>
+Peter Kwong Columbia University  
+
+<|ref|>text<|/ref|><|det|>[[44, 722, 288, 741]]<|/det|>
+Biological Sciences - Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 760, 137, 779]]<|/det|>
+Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 798, 339, 817]]<|/det|>
+Posted Date: November 6th, 2024  
+
+<|ref|>text<|/ref|><|det|>[[44, 836, 475, 856]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 5090208/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 874, 914, 916]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 934, 535, 954]]<|/det|>
+Additional Declarations: There is NO Competing Interest.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 77, 912, 120]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on May 17th, 2025. See the published version at https://doi.org/10.1038/s41467-025-59721-2.
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[135, 297, 860, 315]]<|/det|>
+# Conformational trajectory of the HIV-1 fusion peptide during CD4-induced envelope opening  
+
+<|ref|>text<|/ref|><|det|>[[75, 380, 844, 398]]<|/det|>
+Bhishem Thakur \(^{1}\) , Revansiddha H. Katte \(^{2}\) , Wang Xu \(^{2}\) , Katarzyna Janowska \(^{1}\) , Salam Sammour \(^{1}\) , Rory  
+
+<|ref|>text<|/ref|><|det|>[[75, 411, 629, 429]]<|/det|>
+Henderson \(^{1,3}\) , Maolin Lu \(^{2}\) , Peter D. Kwong \(^{4,5}\) , Priyamvada Acharya \(^{1,6,7*}\)  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 480, 209, 494]]<|/det|>
+## Affiliations:  
+
+<|ref|>text<|/ref|><|det|>[[112, 495, 877, 655]]<|/det|>
+\(^{1}\) Duke Human Vaccine Institute, Durham NC 27710, USA \(^{2}\) Department of Cellular and Molecular Biology, School of Medicine, University of Texas at Tyler Health Science Center, Tyler, Texas, 75708, USA \(^{3}\) Duke University, Department of Medicine, Durham NC 27710, USA \(^{4}\) Aaron Diamond AIDS Research Center, Columbia University Vagelos College of Physicians and Surgeons, New York, NY 10032, USA \(^{5}\) Vaccine Research Center, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892, USA \(^{6}\) Duke University, Department of Surgery, Durham NC 27710, USA \(^{7}\) Duke University, Department of Biochemistry, Durham NC 27710, USA  
+
+<|ref|>text<|/ref|><|det|>[[115, 688, 464, 704]]<|/det|>
+\*To whom correspondence should be addressed  
+
+<|ref|>text<|/ref|><|det|>[[115, 718, 658, 735]]<|/det|>
+Correspondence to: Priyamvada Acharya (priyamvada.acharya@duke.edu)
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 102, 186, 118]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[111, 130, 880, 602]]<|/det|>
+The hydrophobic fusion peptide (FP), a critical component of the HIV- 1 entry machinery, is located at the N terminal stretch of the envelope (Env) gp41 subunit \(^{1,3}\) . The receptor- binding gp120 subunit of Env forms a heterodimer with gp41 and assembles into a trimer, in which FP is accessible for antibody binding \(^{3}\) . Env conformational changes or “opening” that follow receptor binding result in FP relocating to a newly formed interprotomer pocket at the gp41- gp120 interface where it is sterically inaccessible to antibody \(^{4}\) . The mechanistic steps connecting the entry- related transition of antibody accessible- to- inaccessible FP configurations remain unresolved. Here, using SOSIP- stabilized Env ectodomains \(^{5}\) , we visualized atomic- level details of a functional entry intermediate, where partially open Env was bound to receptor CD4, co- receptor mimetic antibody 17b, and FP- targeting antibody VRC34.01, demonstrating that FP remains antibody accessible despite substantial receptor- induced Env opening. We determined a series of structures delineating stepwise opening of Env from its closed state to a newly resolved intermediate and defining downstream re- organizations of the gp120- gp41 interface that ultimately resulted in FP burial in an antibody- inaccessible configuration. Our studies improve our understanding of HIV- 1 entry and provide information on entry- related conformation reorganization of a key site of HIV vulnerability to neutralizing antibody.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 130, 884, 440]]<|/det|>
+The HIV- 1 envelope glycoprotein (Env), a homotrimer of gp120- gp41 heterodimers, mediates viral entry into host cells. The gp120 subunit engages host receptors, while the gp41 subunit contains a fusion peptide (FP) that is inserted into the host membrane to induce host and virus membrane fusion<sup>6</sup>. Prior to its binding to host receptors, the HIV- 1 Env is characterized by a closed configuration with gp120 protomers packed against each other and the gp41 subunit<sup>1</sup>, while the highly conserved and immunodominant coreceptor- binding region at the Env trimer apex remains occluded by packing of the first and second (V1V2) as well as the third (V3) variable loops (Figure 1A)<sup>2</sup>. At the trimer base, FP comprises a hydrophobic stretch of about 20 amino acids at the gp41 N terminus that is accessible for antibody binding in the closed configuration of Env. FP is a site of vulnerability to broadly neutralizing antibodies (bnAbs) and thus of vaccine focus<sup>3,7,8</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[111, 459, 883, 640]]<|/det|>
+HIV- 1 Env uses its gp120 subunits to engage the CD4 receptor on the surface of human immune cells. CD4- induced conformational changes have been structurally characterized in virus- associated Env by cryo- electron tomography (cryo- ET)<sup>9- 11</sup>, while high- resolution structural definition of receptor- induced Env opening has been obtained by single- particle cryo- EM analysis of stabilized, soluble Env ectodomains<sup>12- 14</sup>. Both lines of evidence have synergized to facilitate our understanding of HIV- 1 entry- related and enabled visualization of functionally relevant Env structural changes across resolution scales.  
+
+<|ref|>text<|/ref|><|det|>[[111, 660, 875, 907]]<|/det|>
+CD4- induced Env conformational changes, collectively termed as “Env opening”, include rigidity body displacement and rotation of the gp120 subunits resulting in up to \(\sim 40\) - Å shift in the positioning of the V1V2 base (Figure 1B)<sup>4</sup>. Env opening is accompanied by internal rearrangements within gp120 that involve disruption of inter- protomer interactions formed by the gp120 V1V2 and V3 regions, release of the V3 loop, and formation and exposure of the bridging sheet. The V3 loop and bridging sheet are the structural elements that form the binding site for a GPCR coreceptor, either CCR5 or CXCR4<sup>15- 17</sup>. The structural signatures of CD4- induced Env opening include the bridging sheet and the \(\alpha 0\) helix in gp120 that were first defined in crystal structures of CD4- bound monomeric gp120<sup>18</sup>. CD4- induction of Env also
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 88, 853, 140]]<|/det|>
+re- organizes the gp41 subunit<sup>4,19</sup> resulting in burial of FP within a gp41 cavity such that it is no longer accessible for antibody binding (Figure 1 A, B).  
+
+<|ref|>text<|/ref|><|det|>[[111, 161, 881, 533]]<|/det|>
+While high- resolution structural details have been elucidated for FP in an antibody- accessible conformation (the closed configuration of Env prior to receptor engagement) and in an antibody- inaccessible conformation after CD4 receptor- induced opening of Env, mechanistic details of this FP relocation remain unclear. Here, we use conformation- sensitive antibodies as molecular probes to simultaneously track the trajectories of Env opening and of FP accessibility. For FP accessibility, we used the prototype FP- directed antibody VRC34.01<sup>3</sup>, isolated from a chronically HIV- 1- infected individual, which binds at an epitope comprised primarily of the gp41 FP residues 512- 519 (contributing \(\sim 55\%\) of total interactive surface area) and gp120 glycan N88 ( \(\sim 26\%\) of the total interactive surface area). For Env opening, we used the CD4- induced (CD4i) antibody 17b to assess the formation and exposure of the bridging sheet upon CD4- triggering of Env. As the formation of the bridging sheet requires disruption of the V1V2 cap at the trimer apex and at least partial Env opening, binding to 17b was also an indicator of Env opening<sup>18,20</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[111, 555, 875, 896]]<|/det|>
+Using Env ectodomains stabilized by an intraprotomer gp120/gp41 disulfide and an Ile to Pro change in gp41 (SOSIP)<sup>5</sup>, we performed cryogenic electron microscopy (cryo- EM) to define intermediates where FP remains accessible to antibody binding despite substantial Env opening. Among these conformations were populations with their gp120 promoters either partially rotated from the pre- receptor closed Env conformation or more substantially rotated to resemble the geometry observed in the CD4- induced fully open conformation described previously<sup>4,21</sup>. The partially rotated gp120 were associated with antibody- accessible FP, whereas further gp120 displacement along an axis orthogonal to the central trimer axis resulted in FP burial, suggesting an association of FP burial with the extent of gp120 displacement. Taken together, our data provide evidence that accessibility of FP to antibody binding persists post- receptor engagement despite substantial Env opening. Furthermore, we define the mechanistic steps that lead to FP burial and antibody inaccessibility upon further Env opening. Our
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 88, 825, 139]]<|/det|>
+results resolve several gaps in our knowledge HIV- 1 entry and provide information relevant to the development of vaccines and therapeutics.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 163, 789, 183]]<|/det|>
+## Fusion peptide remains exposed even after substantial opening of the HIV-1 Env trimer  
+
+<|ref|>text<|/ref|><|det|>[[112, 203, 883, 415]]<|/det|>
+To assess CD4- induced changes in FP accessibility, we measured binding to the FP- targeted antibody VRC34.01 at different time points following incubation of BG505.SOSIP Env with either CD4 alone, or together with the fragment antigen binding (Fab) of the coreceptor- mimicking antibody 17b that recognizes an epitope presented upon CD4- induced Env opening (Figure 1C). VRC34.01 binding decreased after CD4- induction, and the decrease was more profound in the presence of 17b Fab. A control experiment without addition of CD4 or 17b showed no change in VRC34.01 binding to BG505.SOSIP Env.  
+
+<|ref|>text<|/ref|><|det|>[[112, 438, 881, 650]]<|/det|>
+We next assessed simultaneous changes in FP exposure measured by binding to VRC34.01, and Env opening measured by binding to 17b (Figure 1D). 17b binding increased post CD4 addition, indicating Env opening and exposure of the bridging sheet, while VRC34.01 binding decreased. Although these overall trends of increase in 17b binding and decrease in VRC34.01 binding upon CD4 induction were as expected, we noted the retention of substantial VRC34.01 binding at the time- points where near- saturation 17b binding had occurred, suggesting that the FP remained exposed and available for antibody binding despite substantial Env opening.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 672, 697, 692]]<|/det|>
+## FP remains accessible for antibody binding after CD4-induced Env opening  
+
+<|ref|>text<|/ref|><|det|>[[112, 713, 886, 893]]<|/det|>
+To visualize the impact of CD4- bound Env conformations on FP positioning, we incubated BG505 SOSIP Env with CD4 and 17b Fab, and performed single particle cryo- EM on the Env complexes at selected time- points, 1.3 hour (hr), 20 hr, and 3 days, post CD4/17b addition, with VRC34.01 Fab added 30 minutes before the samples were vitrified for cryo- EM analysis (Figures 1E, 2, S1- S4, Table S1). We identified three particle populations across the three cryo- EM datasets that differed in their stoichiometries of bound VRC34.01 Fab (Figure 1E and Supplemental Figure S4). Population 1 dominated at all three time points
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 884, 330]]<|/det|>
+and consisted of partially open Env in which each of the three gp120- gp41 protomers was bound to one each of CD4, 17b Fab and VRC34.01 Fab. Another population, named Population 2, was detected at all three time points, albeit in smaller proportions relative to Population 1 (Supplemental Figure S4). In Population 2, each of the three gp120 subunits were bound to one each of CD4 and 17b Fab, while only two protomers were bound to VRC34.01 Fab. The proportion of Population 2 relative to Population 1 increased with longer incubation times. At the 3- day time point, a third population, named Population 3, was detected that resembled Populations 1 and 2 in their bound CD4 and 17b stoichiometries but only had a single VRC34.01 Fab bound, leaving two protomers not bound to VRC34.01.  
+
+<|ref|>text<|/ref|><|det|>[[112, 354, 884, 469]]<|/det|>
+In summary, we identified three populations of CD4- induced Env in our cryo- EM datasets with differing stoichiometries of bound VRC34.01. These results confirmed that FP remained accessible to VRC34.01 binding despite substantial Env opening and suggested FP accessibility to antibody binding to be hindered at the sites in Populations 2 and 3 that were not bound by VRC34.01.  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 493, 864, 542]]<|/det|>
+## FP remains antibody-accessible in a CD4- triggered partially open Env intermediate on the HIV-1 entry pathway  
+
+<|ref|>text<|/ref|><|det|>[[111, 565, 885, 872]]<|/det|>
+Two distinct structural configurations of the FP have thus far been defined in the literature, one that is antibody accessible in the pre- fusion closed \(\mathrm{Env}^{22}\) and a second that is sequestered within a gp41- gp120 pocket in a partially or fully open CD4- induced \(\mathrm{Env}^{4,19}\) . In this study, we have discovered new CD4- induced Env intermediates that are open enough to bind the bridging sheet- directed antibody 17b and yet retain the ability to bind an FP- targeting antibody. To understand Env- structural changes that enable CD4- induced opening, while the FP remains in an antibody- accessible configuration, we first examined Population 1, which was the dominant population in all the cryo- EM datasets (Figures 1D, 2, S1- S4 and Table S1). We selected the Population 1 reconstruction from the 1.3- hr time- point for our analysis as it contained the largest number of particles and the highest resolution among the Population 1 structures from the three datasets.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 879, 398]]<|/det|>
+In Population 1, the gp120 subunits exhibited known structural markers of the CD4- induced conformation<sup>1,4,18</sup>, including the bridging sheet at the 17b- binding interface and residues 63- 73 assembled into the \(\alpha 0\) helix (Figure 2A). The gp41 subunit appeared conformationally less perturbed. FP bound VRC34.01, with a similar interaction interface as previously observed in the structure with closed BG505 SOSIP dominated by the FP and the glycan at gp120 residue position N88 (Figure S5)<sup>3</sup>. Although no symmetry had been applied during the cryo- EM data processing, the three protomers were highly similar in the symmetrically open Population 1 intermediate (Figure S5). Our Population 1 structure revealed a similar gp120 opening geometry as the cryo- ET structure of membrane- associated HIV- 1<sub>ADA,CM</sub> Env bound to three membrane- associated CD4 molecules (Figure 2B)<sup>11</sup>, suggesting that Population 1 represents a physiologically relevant entry intermediate.  
+
+<|ref|>text<|/ref|><|det|>[[111, 418, 881, 888]]<|/det|>
+We next studied the Population 1 structure using a previously defined set of vectors that report on structural rearrangements associated with rigid body movements in gp120 relative to gp41 (Figure 2C)<sup>23</sup>. These vectors describe the orientation of gp120 relative to the gp41 three- helix bundle, capturing rotation of gp120 away from the trimer central axis and rotation orthogonal to the trimer central axis. These measures effectively capture differences between closed, open, and intermediate state Envs. All three Population 1 protomers clustered together in all measures examined and were similar to previously published structures of BG505 (PDB: 6CM3) or B41 (PDB: 6EDU) SOSIP bound to CD4, 17b and 8ANC195 Fab<sup>19</sup>. The Population 1 structures were distinct from previously published open and open occluded state structures (PDBs: 5VN3 and 5VN8, respectively) in gp120 rotations described by a dihedral angle \((\phi)\) that defines orthogonal rotation and angles \((\theta_{1}\) and \(\theta_{2}\) ) describing rotations relative to the trimer central axis (Figure 2C). However, the distance between the gp120 core and W571 was similar between the open and intermediate structures. Contrasting each with the closed state structure clusters indicates the open and open occluded structures occupy distinct angles in the dihedral and the angle between the gp41 three- helix bundle and gp120 termini while the Population 1 and CD4, 17b, 8ANC195- bound Envs differ in the angle describing gp120 rotation away from the central trimer axis. In summary,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 88, 860, 171]]<|/det|>
+our vector analysis indicates that the partially open Population 1 intermediate described here shifts the gp120 domains away from the central axis, while the open and open- occluded shift the gp120 domains orthogonal to the trimer central axis.  
+
+<|ref|>text<|/ref|><|det|>[[112, 194, 881, 632]]<|/det|>
+We compared the configuration of the FP in the previously published partially open CD4, 17b, 8ANC195- bound structure (PDB: 6CM3) and the partially open CD4,17b, VRC34.01- bound Population 1 structure resolved in this study (PDB: 9D90) (Figure 2D). In the Population 1 structure (VRC34.01- bound), the FP was extended out of the Env core to bind the VRC34.01 antibody, whereas, in the 8ANC195- bound structures, the FP was buried in an intra- protomer gp120/gp41 hydrophobic pocket. The formation of the pocket for FP sequestration was facilitated by a shift in the position of the FP proximal region (FPPR) with the major contribution coming from straightening of the FPPR helix that created space for FP burial. The distance between the Cα atoms of FPPR residue Gln540 and the gp120 residue Phe233 measured at 13 Å for the Population 1 structure and at 19 Å for the CD4,17b, 8ANC195 bound BG505 SOSIP structure (Figure 2D). A similar FP configuration was observed in the B41- complex with CD4, 17b and 8ANC195, suggesting that this is an isolate- independent conformational state (Figure S6). The difference in FP accessibility while maintaining overall similar protomer geometry suggested that in this intermediate state the FP can either be antibody- accessible or it can be occluded. While VRC34.01 binds FP and stabilizes its accessible configuration, 8ANC195 stabilizes the FP occluded configuration.  
+
+<|ref|>text<|/ref|><|det|>[[113, 652, 857, 736]]<|/det|>
+In summary, we identified a CD4- triggered partially open Env intermediate on the HIV- 1 entry pathway, with a protomer geometry that accommodates an antibody- accessible or a buried FP, with a conformational change in the FPPR being the major facilitator for this conformational switch of FP.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 758, 669, 778]]<|/det|>
+## Mechanism of downstream FP sequestering and antibody inaccessibility  
+
+<|ref|>text<|/ref|><|det|>[[113, 800, 870, 883]]<|/det|>
+In addition to the near symmetric, partially open Population 1 state where VRC34.01 was bound at each of the three FP sites, we also identified populations that were bound to either one or two VRC34.01 Fabs, leaving two and one FP sites unbound, respectively, despite saturating amounts of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 875, 268]]<|/det|>
+VRC34.01 Fab being used for sample preparation (Figures 1D, 3A-B, S4 and Table S1). As expected, based on binding to antibody 17b, the bridging sheet and the \(\alpha 0\) helix were formed in all gp120 protomers in the Population 2 and Population 3 structures (Figures 3A and B). Examination of these unbound sites revealed FP sequestered within a gp120/gp41 pocket in an antibody-inaccessible configuration (Figure 3C and D), thus providing a structural explanation for the lack of antibody binding to these FP sites.  
+
+<|ref|>text<|/ref|><|det|>[[111, 288, 880, 888]]<|/det|>
+Unlike the near- symmetric Population 1 structure, Populations 2 and 3 displayed marked asymmetry. To quantify Env opening, we measured interprotomer distances between the CD4- binding site gp120- residue Asp368 and gp120- residue Pro124 at the Env trimer apex (Figure 3D and F). As previously recognized, the closed and open Env structures showed substantial differences in these distances<sup>21</sup> (Figure 3D). In the closed Env trimer (PDB: 5ACO), the distance between the Asp368 residues and Pro124 residues measured 14.7 Å and 56 Å, respectively. These distances were much larger in the CD4,17b- bound open Env trimer (PDB: 5VN3) at 76.7 Å and 84.3 Å, respectively. By contrast, in the CD4,17b,8ANC195- bound partially open Env (PDB: 6CM3), these distances were intermediate between the open and closed, at 65 Å and 79.4 Å, respectively. Since all three structures were reconstructed by imposing C3 symmetry during cryo- EM map refinement, each of these interprotomer distances was identical within each structure. These distances measured in the Population 1 structure were similar to the distances in the CD4,17b,8ANC195- bound partially open structure (PDB: 6CM3) (Figure 3D and E). Since no symmetry was applied during the reconstruction of the Population 1 map, three distances were noted for each measure: 63 Å, 66 Å, and 66.9 Å for the interprotomer distances between residue Asp368, and 75 Å, 77.6 Å and 78.2 Å for the interprotomer distances between residue Pro124. In Population 2, the two protomers that were bound to VRC34.01 showed similar separation as observed in Population 1, 61.4 Å and 74.6 Å between residues Asp368 and Pro124, respectively. The protomer that was not bound to VRC34.01 showed greater gp120 displacement where these distances approached closer to those observed in the CD4,17b- bound open Env trimer (PDB: 5VN3). In Population 3, the two
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 868, 140]]<|/det|>
+protomers that were not bound to VRC34.01 had buried FPs and showed gp120 geometries closer to the fully open Env conformation.  
+
+<|ref|>text<|/ref|><|det|>[[111, 161, 880, 567]]<|/det|>
+Taken together, our results demonstrate that FP burial that renders it inaccessible to an FP- directed antibody requires further Env opening, involving transition of Env geometry past an entry intermediate that occurs earlier on CD4- induced Env opening trajectory. This early intermediate resembles the one previously captured in the cryo- ET structure of the CD4- bound HIV- 1 \(\mathrm{ADACM}\) Env, in the single particle cryo- EM structures of CD4,17b,8ANC195- bound BG505 SOSIP (PDB: 6CM3), CD4,17b,8ANC195- bound B41 SOSIP (PDB: 6EDU), and the CD4,17b, VRC34.01- bound BG505 SOSIP Population 1 resolved in this study (PDB: 9D90). In this early intermediate structure, FP can adopt both a buried (antibody- inaccessible) or an antibody- accessible conformation. This intermediate is characterized by gp120 protomers opening like the petals of a tulip where the Env trimer apex separates and gp120 is displaced from the trimer central axis, as a rigid body hinging about the gp120 N/C termini at the trimer base. For stable sequestration of FP that renders it unavailable for antibody binding, further displacement of the gp120 protomers is needed, in the form of a lateral rotation in a plane roughly parallel to the viral membrane and about an axis orthogonal to the central trimer axis.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 589, 642, 609]]<|/det|>
+## Conformational changes in gp41 required for stable FP sequestering  
+
+<|ref|>text<|/ref|><|det|>[[112, 630, 861, 810]]<|/det|>
+To elucidate the gp41 structural features involved in stable FP sequestration, we examined differences in the vicinity of the FP between the partially open intermediate Population 1 conformation (PDB: 9D90; this study) and the previously described fully open Env conformation (PDB 5VN3) \(^4\) . The fully open structure was characterized by a greater displacement of the gp120 subunits that could be visualized by the clear separation of signature \(\alpha 0\) helix from gp41, while this region in the Population 1 structure, although helical, remained associated with the gp41 subunit (Figure 4A).  
+
+<|ref|>text<|/ref|><|det|>[[112, 832, 880, 884]]<|/det|>
+At the FP site, the most striking difference was observed in the gp120/gp41 pocket where FP was buried in the fully open structure versus this region in the partially open intermediate (Figures 4B and C).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 872, 397]]<|/det|>
+In the CD4,17b- bound fully open structure (PDB; 5VN3), this pocket was much larger and measured at \(\sim 26 \mathrm{\AA}\) between FPPR residue Gln540 and gp120 residue Phe233, with the buried FP adopting an extended loop conformation to fill the larger space of the pocket. By contrast, in the partially open intermediate, this distance measured \(13 \mathrm{\AA}\) in Population 1 (CD4,17b, VRC34.01- bound structure) where the FP was exposed and \(19 \mathrm{\AA}\) in the CD4,17b, 8ANC195- bound structure where the FP was buried and assumed a helical conformation. Progressive straightening of FPPR along with changes in both HR1 and HR2 regions of gp41 orchestrated the enlargement of this pocket, which in the fully open structure assumes an interprotomer character with one of its walls lined with the HR1 helix of the adjacent protomer. Thus, the concerted gp120/gp41 re- organizations that resulted in the formation of a larger FP- binding pocket may be responsible for the stable sequestration of the FP in the fully open structure.  
+
+<|ref|>text<|/ref|><|det|>[[112, 418, 881, 629]]<|/det|>
+We observed that the Population 2 gp41 conformation with buried FP was most similar to that of the fully open Env (PDB: 5VN3), with HR1 helices showing close overlap, and the FPPR straightened out further compared to the partially open Population 1 and the 6CM3 structures, albeit not to the extent of the fully open structure (Figure 4D). In both Population 2 and fully open 5VN3 structures, the movement of the HR2 region around residues 638- 662 (indicated by red arrow in Figure 4D) creates room for the FPPR unbending. The gp120/gp41 pocket in this Population 2 protomer measured \(21 \mathrm{\AA}\) , with the cavity size approaching that of the cavity measured in the fully open structure.  
+
+<|ref|>text<|/ref|><|det|>[[112, 652, 876, 830]]<|/det|>
+In summary, our data show that the FP undergoes stepwise changes in its configuration as a consequence of CD4- induced movements in gp120 and gp41. From a closed, pre- receptor state where FP is accessible to antibodies, Env proceeds to partially open states where FP remains available to the FP- directed antibody VRC34.01. Only upon more extensive rotation of gp120 and widespread changes in gp41 does FP become fully buried within a gp120/gp41 pocket and, as a result, no longer accessible to antibody binding.  
+
+<|ref|>text<|/ref|><|det|>[[112, 853, 536, 872]]<|/det|>
+smFRET analysis of Env opening on the virion surface
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 90, 881, 912]]<|/det|>
+smFRET analysis of Env on the surface of intact virions has revealed conformational shifts of virus Env from a pre- triggered (PT) state through a pre- receptor closed (PC) state to a fully open CD4- bound conformational state (CO) in response to CD4 activation \(^{24,25}\) . The pre- fusion, pre- receptor closed Env on virions resembles the FP- accessible Env structure complexed with three VRC34.01 (PDB: 5I8H), while the fully open Env was associated with the symmetric Env structure bound with three CD4 and three 17b (PDB: 5VN3), and a pre- triggered state was suggested that is undefined in currently available structures \(^{24,25}\) . We asked whether the structural differences between partially open VRC34.01- bound Env structures characterized in this study and the fully open FP- sequestered Env would be reflected at the global population level of Env conformations presented on virions. We performed smFRET experiments of two different fluorescently click- labeled Env \(_{\mathrm{BG505}}\) on intact HIV- \(1_{Q23}\) virions \(^{26}\) , in which donor/acceptor fluorescent probes were placed between gp120 V1 and V4 or between gp120 V4 and gp41 \(\alpha 6\) , respectively (Figure S7A). Placing FRET probes at different paired structural elements of Env allowed us to visualize global conformational changes of Env from two different structural perspectives, gp120 V1- V4 and gp120- gp41 (Figure 4E, S7B and S7C, Table S2). Using these two imaging systems, we observed distinct FRET histograms of virus Env in the ligand- free and presence of ligands VRC34.01, VRC34.01+ sCD4+17b, and sCD4+17b (Figure 4E and S7B- E, Table S2). We applied the previously well- defined three- state (PT, PC, CO) Gaussian distributions \(^{26}\) to describe the FRET histograms, which reflect the overall conformational landscape of Env on virions. As expected, ligand- free Env exhibited predominance of the pre- triggered conformation, and Env, in the presence of sCD4 and 17b, prevailed in the fully open CD4- bound state. In the presence of VRC34.01, a decrease was observed in the PT population with an increase in the PC population, consistent with previously published results with the JR- FL Env \(^{3}\) . For the VRC34.01+ sCD4+17b sample, the smFRET histograms suggested that the Env conformational distributions resided between the PC and CO conformations (Figures 4E, S7B- F). Quantifying and comparing the propensity of each primary conformational state occupied by virus Env under ligand- free and different ligand- bound conditions, we observed distinct conformational effect on Env by VRC34.01 in the presence of sCD4+17b, positioned on the Env activation pathway between the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 881, 235]]<|/det|>
+effect of VRC34.01 alone and the sCD4+17b CO state (Figures 4E, S7). These results were consistent between the observations from the gp120- gp41 (Figure 4E and S7E) and the gp120 V1- V4 structural perspectives (Figure S7B- D). Thus, smFRET analysis of the impact of VRC34.01 on the CD4,17b- bound Env was consistent with VRC34.01 stabilizing an intermediate state on the path of CD4- induced Env opening.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 258, 586, 277]]<|/det|>
+## Visualizing Preceding CD4-induced transitions of HIV-1 Env  
+
+<|ref|>text<|/ref|><|det|>[[111, 297, 880, 900]]<|/det|>
+We next sought to visualize CD4- induced Env conformational transitions that occur upstream to Population 1 by performing single particle cryo- EM analysis on a sample of BG505 SOSIP that was incubated with sCD4 for 2 hr, followed by the addition of VRC34.01 Fab 30 minutes before sample vitrification. As antibody 17b works synergistically with CD4 to open Env (Figure 1C), we rationalized that this strategy where we excluded 17b may allow us to capture earlier stages of CD4- induced Env conformational changes. Two distinct particle populations were revealed in the cryo- EM dataset, which yielded reconstructions of 4.08 Å (Population 4) and 4.14 Å (Population 5) global resolutions. For both populations, all three protomers were bound to one each of CD4 and VRC34.01 Fab (Figures 5A, 5B, S8 and Table S1). The two populations differed in the extent of rotation of their gp120 subunits. In Population 4, one of the three gp120 protomers was rotated roughly to the extent observed in the Population 1 structure, while the other two protomers were minimally rotated from their pre- receptor conformation. The distances between the CD4 binding site residue Asp 368 in the two minimally rotated protomers measured 60.2 Å and was thus closer to the distance observed in the closed, pre- receptor Env ( \(\sim 56\) Å) (Figure 3E) than to the distances measured in the partially open Population 1 intermediate ( \(\sim 75\) - 78.2 Å) (Figure 3F). The third protomer that had rotated was separated in this measure from the two other protomers by 61.7 Å and 66.9 Å. The bridging sheet and the α0 helix, which are the structural components associated with CD4- induced Env opening, were only seen in the rotated gp120 protomers. Although two protomers were minimally rotated, their V1V2 regions were unstructured, thus suggesting that CD4- induced disruption of the V1V2 cap of the closed, pre- receptor Env precedes the rotation of the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 881, 300]]<|/det|>
+protomers. This is consistent with time- resolved, temperature- jump small- angle x- ray scattering studies that suggest an order- to- disorder transition in the trimer apex precedes Env transitions involving protomer rotation<sup>27</sup>. In Population 5, two gp120 protomers were rotated and showed formation of the bridging sheet and \(\alpha 0\) helix, while the third protomer remained minimally rotated. Both Populations 4 and 5 were bound to VRC34.01 Fab at all the three FP sites, as expected, since the gp120 subunits had not rotated far enough to allow the gp41 conformational changes required for FP burial and steric inaccessibility to antibodies.  
+
+<|ref|>text<|/ref|><|det|>[[112, 322, 876, 470]]<|/det|>
+In summary, Population 4 and 5 structures represent conformational states preceding the partially opened Population 1 conformation. Taken together, our results demonstrate sequential CD4- induced opening of the gp120 protomers and are consistent with previous studies that show initiation of CD4 binding to the closed Env begins with a single CD4, which induces opening of gp120 protomers needed for binding of additional CD4 molecules<sup>14,28</sup>.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 493, 199, 509]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[111, 533, 876, 907]]<|/det|>
+HIV- 1 entry involves sequential receptor mediated conformational changes. Knowledge of how each part of the Env moves, in synergy with other parts, provides clues that have enabled the design of immunogens<sup>4,29,30</sup>. Despite years of intense research, critical gaps remain in our knowledge of the HIV- 1 entry mechanism. In this study, we have addressed one such gap related to the fate of the FP during receptor- mediated Env conformational changes. The HIV- 1 FP is accessible and a target for bnAbs in the closed Env and a target for bnAbs but becomes buried within a gp120/gp41 cavity in the receptor- bound fully open Env. Here, we add new knowledge by elucidating the structure and conformations of a receptor- bound, partially open, functional Env intermediate where the FP remains accessible to FP- directed antibodies (Step 5 of Figure 5C). We elucidate the stepwise mechanism that leads to the formation of this intermediate (Steps 1- 4, Figure 5C), and the stepwise process that leads from this intermediate to the fully open Env with a stably sequestered FP (Steps 6- 8, Figure 5C). We elucidate the gp120 subunit motions associated with an exposed, transiently buried or a stably sequestered FP. From
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 88, 881, 172]]<|/det|>
+its receptor- bound, fully open conformation, further conformational changes are needed for Env to release its fusion peptide for insertion into the host membrane. Future studies will reveal the mechanistic details of these downstream steps in the HIV- 1 entry and virus- cell membrane fusion processes.  
+
+<|ref|>text<|/ref|><|det|>[[112, 194, 857, 405]]<|/det|>
+The FP site of vulnerability on HIV- 1 is one of the few sites of Env vulnerability against which vaccination has succeeded in elicited antibodies of over \(50\%\) neutralization breadth<sup>8</sup>. Like VRC34.01, these antibodies recognize the accessible conformation of FP in the prefusion- closed state, with the revealed mechanistic details indicating antibody recognition of FP extends into the early entry intermediates identified in this study. Thus, in both the prefusion- closed state as well as early entry intermediates, FP appears to have considerable conformational flexibility, which is crucial for vaccine priming with flexible peptides linked to carriers<sup>7,31</sup>  
+
+<|ref|>text<|/ref|><|det|>[[112, 428, 877, 832]]<|/det|>
+From being solvent exposed and accessible to antibodies to becoming transiently, then stably sequestered in an antibody inaccessible conformation, to finally being released from this sequestered state to insert into the host membrane and mediate fusion, the HIV- 1 FP follows a trajectory that is unique among Type 1 fusion proteins. The typical norm for other viruses is to hold their FP in a partially or wholly occluded, sometimes metastable conformation in the pre- receptor state, which receptor binding mediated conformational changes then release for insertion into the host membrane<sup>32</sup>. For SARS- CoV- 2, for example, FP- directed antibodies have been described but these antibodies cannot access their FP epitopes in the pre- receptor conformation of the spike protein<sup>33,36</sup>. Receptor binding mediated conformational changes reveal these cryptic epitopes to allow antibody binding<sup>34,37,38</sup>. In summary, the HIV- 1 FP tracks a unique trajectory among Type 1 fusion proteins with its multistep receptor- induced conformational transitions. Despite these differences, the central role of receptor- induced conformational changes in controlling and maneuvering the FP through its pre- receptor conformation to its fusion competent state is a common feature amongst all Type 1 fusion proteins.  
+
+<|ref|>text<|/ref|><|det|>[[113, 854, 875, 904]]<|/det|>
+One apparent conundrum that this study elucidates is the genesis of the FP site of in HIV- 1. Why in a virus that has evolved exemplary defenses to shield its vulnerabilities from the immune system,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 881, 428]]<|/det|>
+would the highly conserved FP be exposed and a focus for targeting by broadly neutralizing antibodies? Why would it not be hidden from the immune system within the pre- receptor, closed Env? In this study, we have shown that at least partial opening of the HIV- 1 Env trimer is required for burial of FP within a gp41 cavity that forms because of this opening. While a partially open Env can accommodate FP burial, this study demonstrated that such burial is not stable and only upon more substantial Env opening does FP become stably sequestered. Since opening of Env, even partial, exposes epitopes that make the virus more susceptible to neutralization, on balance, it may have been advantageous for maintaining the neutralization resistant compact and closed HIV- 1 Env conformation to leave the FP exposed, and shield it with a few strategically placed glycans. The exposure of the FP, and thus the creation of the site of vulnerability to antibody, may be a mitigating step towards preventing greater vulnerability due to Env opening that may accompany burial of the flexible hydrophobic FP.  
+
+<|ref|>text<|/ref|><|det|>[[112, 450, 886, 723]]<|/det|>
+Another partially open Env conformation that has been described is the open- occluded conformation that is sampled in a receptor- independent manner by both the native Env as well as Env ectodomain constructs used for vaccine applications<sup>39</sup>. A recent preprint combining MD simulations and smFRET analysis has proposed the open- occluded conformation as neutralization- relevant intermediate of Env on the transition trajectory<sup>40</sup>. Structures of the open- occluded Env ectodomain show FP to be buried within a gp41 cavity<sup>39</sup>. Indeed, the hydrophobic solvent- exposed FP may be a source of Env metastability and its proclivity to shield itself, even if transiently, within a hydrophobic pocket may provide the underlying rationale for the accessibility of partially open Env conformations that allow such FP occlusion.  
+
+<|ref|>text<|/ref|><|det|>[[112, 747, 880, 896]]<|/det|>
+In summary, our study provides a stepwise mechanism for the receptor- induced opening of the HIV- 1 Env and elucidates the trajectory of the fusion peptide from its solvent exposed configuration in the pre- receptor, closed Env to its buried, antibody- inaccessible configuration in the receptor- bound, fully open Env. Collective evidence in our study suggests that the structure of Population 1 represents a general intermediate on the HIV- 1 entry pathway and that this intermediate is accessible for
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 883, 330]]<|/det|>
+binding to broadly neutralizing antibodies such as VRC34.01 and 8ANC195. The evidence includes the recurrence of similar geometry in structures obtained from different isolates (BG505 or B41) and in complex with different gp120/gp41 targeting antibodies (8ANC195 and VRC34.01), and their concurrence with the cryo- ET resolved structure of the CD4- bound HIV- 1ADA.CM Env in the membrane context. Our work thus reveals a functional intermediate with subunit geometry compatible with both a solvent exposed, antibody accessible and an occluded/buried, antibody inaccessible FP configurations. By elucidating the accessibility of FP, a major target for bnAbs, during receptor- induced Env conformational changes our study reveals key insights into this site of vaccine focus.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 384, 266, 400]]<|/det|>
+## Acknowledgements  
+
+<|ref|>text<|/ref|><|det|>[[111, 424, 872, 666]]<|/det|>
+Cryo- EM data were collected at the Duke Krios at the Duke University Shared Materials Instrumentation Facility (SMIF), a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), which is supported by the National Science Foundation (award number ECCS- 2025064) as part of the National Nanotechnology Coordinated Infrastructure (NNCI). This study utilized the computational resources offered by Duke Research Computing (http://rc.duke.edu; NIH 1S10OD018164- 01) at Duke University. This work was supported by NIH grants R01 AI145687 (P.A.), U54 AI170752 (P.A., R.H. and M.L.), R01 AI181600 from NIH/NIAID, an R35 GM151169 from NIH/NIGMS to M.L., and the Vaccine Research Center, an Intramural Division of NIAID, NIH.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 734, 281, 750]]<|/det|>
+## Author contributions  
+
+<|ref|>text<|/ref|><|det|>[[111, 775, 864, 888]]<|/det|>
+P.A. conceived the project and oversaw the study. B.T. and P.A. designed binding studies and cryo- EM experiments. B.T. expressed and purified proteins, performed SPR assays, optimized specimen, and prepared cryo- EM grids, collected cryo- EM data, performed map and coordinate refinement, and performed structural analysis. R.K., W.X., and M.L. performed smFRET analysis of virus Env, K.J.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 88, 876, 172]]<|/det|>
+419 assisted with cryo- EM data collection. S.S. assisted with protein purification. R.H. performed vector analysis. P.D.K. assist with initial SPR experiments. B.T. and P.A. wrote the first draft of the manuscript. 421 All authors reviewed and commented on the manuscript.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[60, 90, 177, 107]]<|/det|>
+## 422 Methods  
+
+<|ref|>text<|/ref|><|det|>[[115, 131, 376, 149]]<|/det|>
+Protein expression and purification  
+
+<|ref|>text<|/ref|><|det|>[[112, 161, 875, 508]]<|/det|>
+HIV- 1 Env ectodomain constructs used in this study were purified form HEK293S GnTI cells (Thermo Fisher Scientific) diluted at the time of transfection to \(1.25 \times 10^{6}\) cells/mL. Before transfection, cells were diluted in Freestyle™ 293 Expression Medium (Cat No. 12338018) to \(1.25 \times 10^{6}\) cells/mL at a volume of \(950 \mathrm{mL}\) . Plasmid DNA expressing the Env ectodomain and furin were co- transfected at a 4:1 ratio ( \(650 \mu \mathrm{g}\) and \(150 \mu \mathrm{g}\) per transfection liter, respectively) and incubated with 293fectin™ transfection reagent (ThermoFisher Cat No. 12347019) in Opti- MEM I Reduced Serum Medium (ThermoFisher Cat No. 31985062). The diluted mixture was added to the cell culture which was incubated at \(37^{\circ}\mathrm{C}\) , \(9\%\) \(\mathrm{CO_2}\) on a shaker at \(120 \mathrm{rpm}\) for 6 days. On day 6 the cell supernatant was harvested by centrifuging the cell culture at \(4000 \mathrm{xg}\) for 30 minutes. The supernatant was filtered with a \(0.45 \mu \mathrm{m}\) PES filter and concentrated to approximately \(100 \mathrm{mL}\) using a Vivaflow® 200 cross- flow cassette (Sartorius Cat No. VF20P2).  
+
+<|ref|>text<|/ref|><|det|>[[112, 528, 881, 744]]<|/det|>
+The cell culture supernatant was passed through \(10 \mathrm{mL}\) PGT145 IgG- conjugated affinity column equilibrated in \(20 \mathrm{mM}\) PBS, pH 7.5. Following loading and washing, Env trimers were eluted using \(3 \mathrm{M}\) \(\mathrm{MgCl}_2\) , pH 7.2. The eluted Env were concentrated to \(\sim 1 \mathrm{mL}\) with a Centricon- 70 \(100 \mathrm{kDa}\) filter (Millipore Sigma). After concentrating, Env were filtered through \(0.22 \mu \mathrm{M}\) filter to remove any aggregates before loading on Superose 10/300 GL column (Cytiva) size exclusion column pre- equilibrated in PBS on an AKTA Pure (Cytiva) system. The fractions corresponding to the Env trimers were pooled, concentrated, flash frozen in liquid nitrogen frozen for long- term storage at \(- 80^{\circ}\mathrm{C}\) .  
+
+<|ref|>text<|/ref|><|det|>[[112, 766, 872, 850]]<|/det|>
+Antibodies were produced in Expi293 cells and purified using a Protein A affinity column followed by size exclusion chromatography using a HiLoad Superdex 200 column equilibrated in \(20 \mathrm{mM}\) PBS, pH 7.5, \(0.002\%\) w/v Azide.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 874, 170]]<|/det|>
+4- domain CD4 was produced in Expj293 cells and purified by Q425- affinity chromatography, followed by size exclusion chromatography using a HiLoad Superdex 200 column equilibrated in 20mM PBS, pH 7.5, 0.002% w/v Azide.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 203, 323, 220]]<|/det|>
+## Surface Plasmon Resonance  
+
+<|ref|>text<|/ref|><|det|>[[111, 234, 881, 543]]<|/det|>
+Surface Plasmon ResonanceSurface Plasmon Resonance binding assays were performed on a T- 200 Biacore system (GE- Healthcare) operating at \(25^{\circ}\mathrm{C}\) . HBS- EP+ (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA and 0.05% surfactant P- 20) was used as running buffer. A 40 nM solution of BG505.SOSIP prepared in the running buffer was incubated at \(25^{\circ}\mathrm{C}\) with either 200 nM of CD4, or with 200 nM CD4 and 200 nM 17b Fab. The binding surface was prepared by flowing 100 nM each of, 17b IgG and VRC34.01 IgG over each flow cells 2 and 4, respectively at \(10~\mu \mathrm{l} / \mathrm{min}\) flow rate for 30 seconds with the \(1^{\mathrm{st}}\) and \(3^{\mathrm{rd}}\) flow cells serving as reference for \(2^{\mathrm{nd}}\) and \(4^{\mathrm{th}}\) flow cells, respectively. After surface preparation, the analyte (either BG505 SOSIP alone or BG505 SOSIP with CD4 or BG505 SOSIP with CD4 and 17b Fab) was flowed at \(30~\mu \mathrm{l} / \mathrm{min}\) flow rate for 60 seconds. The same injections were carried out using HBS- EP buffer in order to obtain a reference curve. The sensorgrams were blank corrected in Biacore T- 200 evaluation software.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 574, 187, 591]]<|/det|>
+## Cryo-EM  
+
+<|ref|>text<|/ref|><|det|>[[111, 605, 886, 905]]<|/det|>
+Purified BG505 SOSIP.664 trimer were diluted to a concentration of \(1.3\mathrm{mg / mL}\) and were incubated with five molar excess of 4D CD4 and 5- molar excess of 17b Fab. After mixing, the samples were incubated at \(25^{\circ}\mathrm{C}\) for different incubation times. VRC34.01 Fab in 5- fold molar excess concentration was added 30 minutes before freezing grids. To prevent interaction of the trimer complexes with the air- water interface during vitrification, the samples were incubated in \(0.085\mathrm{mM}\) n- dodecyl \(\beta\) - D- maltoside (DDM). A \(3.5\mathrm{- }\mu \mathrm{L}\) drop of protein was deposited on a Quantifoil- 1.2/1.3 grid (Electron Microscopy Sciences, PA) that had been glow discharged for 10 s using a PELCO easiGlow Cleaning System (Ted Pella). After a 30 seconds incubation in \(>95\%\) humidity, excess protein was blotted away for 2.5 s before being plunge frozen into liquid ethane
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 884, 494]]<|/det|>
+using a Leica EM GP2 plunge freezer (Leica Microsystems). Frozen grids were imaged using a Titan Krios (Thermo Fisher) equipped with a K3 detector (Gatan). The cryoSPARC (Punjani et al., 2017) software was used for data processing<sup>41</sup>. Raw movies were motion corrected using Patch Motion Correction and Contrast Transfer Function (CTF) were estimated. Micrographs with CTF estimates greater than 8 Å were discarded. Automated blob picker software was used to assign the particle position, and the particles were extracted with the 320- pixel extraction box size Fourier cropped to 80 pixels. Following particle extraction, multiple rounds of 2D classification was performed to remove junk particles and re- extraction of clean particles with 320 pixel box size. A reference free ab- initio 3D reconstruction was used to create 3D reconstructions representing diverse conformational states of the Env. Further, multiple rounds of heterogeneous refinement was performed to get rid of the noise. Finally, non- uniform refinement was used on the pure set of particles to get high resolution cryo- EM map.  
+
+<|ref|>text<|/ref|><|det|>[[115, 507, 883, 553]]<|/det|>
+Phenix, Coot, Pymol, Chimera, ChimeraX and Isolde were used for model building and refinement<sup>42- 47</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[115, 586, 416, 604]]<|/det|>
+Virus packaging and fluorescent labeling  
+
+<|ref|>text<|/ref|><|det|>[[112, 626, 885, 904]]<|/det|>
+The methods of packaging and fluorescent labeling of replication- incompetent amber- free HIV- \(1_{Q23}\) viral particles with incorporated EnvBG505 have been described previously<sup>48</sup>. HIV- 1 virions that lack reverse transcriptase (ΔRT) were prepared and used for imaging. Amber- free HIV- \(1_{Q23}\) virions incorporated with two different double- tagged Env were used in this study, including dual- amber N136TAG S401TAG and hybrid click/peptide V4- A1 R542TAG. Amber- free V1V4 N136\* S401\* (\\*, unnatural amino acid - ncAA) viruses carrying click- chemistry- reactive ncAA at 136 in V1 and 401 in V4 were produced by co- transfecting HEK293T cells with a tag- free ΔRT plasmid, an Env- tagged variant N136TAG S401TAG (TAG, amber stop codon) ΔRT plasmid, and an amber suppressor plasmid tRNA<sup>Py1</sup>/NESPylRS<sup>AF</sup>. The amber suppressor can express tRNA and its cognate amino acid acyl- tRNA- synthetase in HEK293T cells. ncAA TCO\* (250 μM)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 886, 590]]<|/det|>
+was added to the transfection system. Similarly, V4A1 R542\* viruses were prepared using the Env- tagged V4A1 (peptide A1 tag: DSLDMLEM in V4 loop) R542TAG \(\Delta\) RT plasmid. The ratio of tag- free vs. tagged Env plasmids used during transfection was adjusted based on previously characterized Env expression levels \(^{48}\) to ensure that, statistically, on average, one tagged protomer within an Env trimer on a virion was available for fluorescent labeling (enzymatically or click) \(^{48 - 51}\) . 40 hours post- transfection, the supernatant was harvested and filtered, then viruses were concentrated at 25000 rpm for 2 hours using an ultracentrifuge. Next, the virus pellet was resuspended using the labeling buffer containing 50 mM HEPES, 10 mM MgCl₂, and 10 mM CaCl₂. The fluorescent labeling of prepared virus Env was similar to the previously described \(^{48 - 51}\) . For the amber- free V1V4 N136\* S401\* viruses, two TCO\* were fluorescently labeled by 0.1 μM tetrazine- conjugated LD555- TTZ and LD655- TTZ by click chemistry. For the amber- free V4A1 R542\* viruses, the A1 peptide in V4 was labeled by LD655- CoA, 0.65 μM in the presence of enzyme AcpS (5 μM), and the TCO\* in gp41 R542 were click labeled by LD555- TTZ. Dyes were customized by Lumidyne Technologies. The above reaction mixture was incubated at room temperature overnight in the dark. PEG2000- biotin was then added at a final concentration of 0.1 mg/ml to the labeled viruses, followed by 30 min incubation at room temperature. Then, the labeled viruses were further purified using a 6%- 18% gradient of Opti- prep (Sigma- Aldrich) and centrifuged at 40,000 rpm for 1 hour at 4°C.  
+
+<|ref|>text<|/ref|><|det|>[[113, 609, 490, 627]]<|/det|>
+smFRET data acquisition and analysis of virus Env  
+
+<|ref|>text<|/ref|><|det|>[[112, 650, 886, 896]]<|/det|>
+All single- molecule fluorescence resonance energy transfer (smFRET) data of fluorescently labeled viruses were collected using a custom- made prism- based total internal reflection fluorescence (prism- TIRF) microscope equipped with a fluorescence signal detection system. The detailed operating manual has been described previously \(^{48}\) . Briefly, the sample loading module, a streptavidin- coated PEG passivated biotin quartz imaging chamber, was cleaned with the imaging buffer, and the background fluorescence signal was removed using the high- intensity laser. The imaging buffer contains 50 mM Tris pH 7.4, 50 mM NaCl, a cocktail of triplet- state quenchers, and oxygen scavenger: 2 mM protocatechuic acid and 8 nM protocatechuic- 3,4- dioxygenase. The labeled viruses were then loaded into the sample loading module. Un
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 884, 300]]<|/det|>
+immobilized viruses were removed using the imaging buffer, and the fluorescence signals were collected. Under the ligand- present experimental conditions, fluorescently labeled viruses were incubated with the indicated \(0.1\mathrm{mg / ml}\) antibody/ligand ( \(>5\mathrm{x}\) above IC95) for 30 mins at room temperature before imaging. All fluorescence signals were recorded simultaneously on two synchronized sCMOS cameras (Hamamatsu ORCA- Flash4.0 V3) at \(25\mathrm{Hz}\) for 80 seconds. The smFRET data were viewed, processed, and analyzed using the SPARTAN software package \(^{52}\) shared by the Scott Blanchard lab and custom MATLAB- based scripts.  
+
+<|ref|>text<|/ref|><|det|>[[111, 321, 886, 893]]<|/det|>
+Recorded 80- second movies (2000 frames/movie) were extracted as donor/acceptor fluorescence traces (time series) with background subtracted and crosstalk corrected. The energy transfer efficiency (FRET efficiency values or simplified as FRET in figures) from the donor fluorophore to the acceptor fluorophore was calculated using \(\mathrm{FRET} = \mathrm{I}_{\mathrm{A}} / (\gamma \mathrm{I}_{\mathrm{D}} + \mathrm{I}_{\mathrm{A}})\) , in which \(\mathrm{I}_{\mathrm{D}}\) and \(\mathrm{I}_{\mathrm{A}}\) represent the fluorescence of the donor and acceptor, respectively, and \(\gamma\) is the correlation coefficient compensating for variations in detection efficiencies. FRET traces (FRET efficiency traces) were further derived. FRET traces reflect real- time relative distance changes between donor and acceptor, resulting from the global conformational dynamics of Env. Under each experimental condition, approximately more than 200 individual traces were included in the final FRET histogram. These included traces meet the following filter settings: 1) a single photo bleaching point (ruling out cases of multiple labeled protomers in a trimer, multiple labeled Envs on one virion, no- labeled Env on one virion; 2) sufficient signal- to- noise ratio; 3) anti- correlated feature between donor and acceptor fluorescence (indicating active Env undergoing conformational changes, ruling out inactive Env as well as Env lacking either donor or acceptor or both). We used automatic filters in combination with manual visualization to ensure that traces of molecules with only one Cy3/Cy5- labeled protomer in a trimer on a viral particle were included for further data processing. FRET traces that meet all the above- mentioned criteria were included to compile FRET histograms/distributions. FRET histograms (conformational distributions) were presented as mean \(\pm\) s.e.m. and fitted into a sum of three distinct Gaussian/Normal distributions using the least- squares fitting algorithm in MATLAB. Parameters were
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 884, 203]]<|/det|>
+determined based on visual inspection of all traces that exhibit state- to- state transitions and the idealization of individual traces using three- state hidden Markov modeling. Each Gaussian represented one conformational state of virus Env. The area under each Gaussian curve was further calculated as an estimation of relative state occupancy, which is the probability of the corresponding state Env occupies.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[55, 90, 90, 103]]<|/det|>
+549
+
+<|ref|>text<|/ref|><|det|>[[59, 122, 90, 133]]<|/det|>
+550
+
+<|ref|>text<|/ref|><|det|>[[59, 150, 203, 161]]<|/det|>
+551 **References**
+
+<|ref|>text<|/ref|><|det|>[[59, 190, 855, 203]]<|/det|>
+552 1 Pancera, M. et al. Structure and immune recognition of trimeric pre-fusion HIV-1 Env.
+
+<|ref|>text<|/ref|><|det|>[[59, 207, 413, 219]]<|/det|>
+553 Nature 514, 455-461 (2014).
+
+<|ref|>text<|/ref|><|det|>[[59, 223, 822, 252]]<|/det|>
+554 2 Lyumkis, D. et al. Cryo-EM structure of a fully glycosylated soluble cleaved HIV-1 envelope trimer. Science 342, 1484-1490 (2013).
+
+<|ref|>text<|/ref|><|det|>[[59, 253, 872, 282]]<|/det|>
+555 Kong, R. et al. Fusion peptide of HIV-1 as a site of vulnerability to neutralizing antibody. Science 352, 828-833 (2016).
+
+<|ref|>text<|/ref|><|det|>[[59, 285, 878, 314]]<|/det|>
+556 Ozorowski, G. et al. Open and closed structures reveal allostery and pliability in the HIV-1 envelope spike. Nature 547, 360-363 (2017).
+
+<|ref|>text<|/ref|><|det|>[[59, 317, 830, 342]]<|/det|>
+557 Sanders, R. W. et al. A next-generation cleaved, soluble HIV-1 Env trimer, BG505
+
+<|ref|>text<|/ref|><|det|>[[59, 337, 840, 366]]<|/det|>
+561 SOSIP.664 gp140, expresses multiple epitopes for broadly neutralizing but not non-neutralizing antibodies. PLoS Pathog 9, e1003618 (2013).
+
+<|ref|>text<|/ref|><|det|>[[59, 366, 532, 380]]<|/det|>
+562 https://doi.org:10.1371/journal.ppat.1003618
+
+<|ref|>text<|/ref|><|det|>[[59, 382, 726, 396]]<|/det|>
+564 Harrison, S. C. Viral membrane fusion. Virology 479, 498-507 (2015).
+
+<|ref|>text<|/ref|><|det|>[[59, 398, 878, 427]]<|/det|>
+565 Xu, K. et al. Epitope-based vaccine design yields fusion peptide-directed antibodies that neutralize diverse strains of HIV-1. Nature medicine 24, 857-867 (2018).
+
+<|ref|>text<|/ref|><|det|>[[59, 430, 878, 459]]<|/det|>
+567 Kong, R. et al. Antibody lineages with vaccine-induced antigen-binding hotspots develop broad HIV neutralization. Cell 178, 567-584. e519 (2019).
+
+<|ref|>text<|/ref|><|det|>[[59, 461, 805, 490]]<|/det|>
+569 Liu, J., Bartesaghi, A., Borgnia, M. J., Sapiro, G. & Subramaniam, S. Molecular architecture of native HIV-1 gp120 trimers. Nature 455, 109-113 (2008).
+
+<|ref|>text<|/ref|><|det|>[[59, 492, 881, 521]]<|/det|>
+571 Li, Z. et al. Subnanometer structures of HIV-1 envelope trimers on aldrithiol-2-inactivated virus particles. Nature Structural & Molecular Biology 27, 726-734 (2020).
+
+<|ref|>text<|/ref|><|det|>[[59, 524, 846, 553]]<|/det|>
+573 Li, W. et al. HIV-1 Env trimers asymmetrically engage CD4 receptors in membranes. Nature 623, 1026-1033 (2023).
+
+<|ref|>text<|/ref|><|det|>[[59, 556, 840, 619]]<|/det|>
+575 Harris, A. et al. Trimeric HIV-1 glycoprotein gp140 immunogens and native HIV-1 envelope glycoproteins display the same closed and open quaternary molecular architectures. Proceedings of the National Academy of Sciences 108, 11440-11445 (2011).
+
+<|ref|>text<|/ref|><|det|>[[59, 622, 881, 666]]<|/det|>
+579 13 Stadtmueller, B. M. et al. DEER spectroscopy measurements reveal multiple conformations of HIV-1 SOSIP envelopes that show similarities with envelopes on native virions. Immunity 49, 235-246. e234 (2018).
+
+<|ref|>text<|/ref|><|det|>[[59, 669, 870, 699]]<|/det|>
+582 Dam, K.-M. A., Fan, C., Yang, Z. & Bjorkman, P. J. Intermediate conformations of CD4-bound HIV-1 Env heterotrimers. Nature 623, 1017-1025 (2023).
+
+<|ref|>text<|/ref|><|det|>[[59, 701, 860, 730]]<|/det|>
+584 Huang, C.-c. et al. Structure of a V3-containing HIV-1 gp120 core. Science 310, 1025-1028 (2005).
+
+<|ref|>text<|/ref|><|det|>[[59, 733, 857, 762]]<|/det|>
+586 Shaik, M. M. et al. Structural basis of coreceptor recognition by HIV-1 envelope spike. Nature 565, 318-323 (2019).
+
+<|ref|>text<|/ref|><|det|>[[59, 764, 781, 779]]<|/det|>
+587 Hoffman, T. L. et al. Stable exposure of the coreceptor-binding site in a CD4-
+
+<|ref|>text<|/ref|><|det|>[[59, 780, 875, 809]]<|/det|>
+589 independent HIV-1 envelope protein. Proceedings of the National Academy of Sciences 96, 6359-6364 (1999).
+
+<|ref|>text<|/ref|><|det|>[[59, 812, 875, 841]]<|/det|>
+591 Kwong, P. D. et al. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393, 648-659 (1998).
+
+<|ref|>text<|/ref|><|det|>[[59, 843, 875, 888]]<|/det|>
+593 19 Wang, H., Barnes, C. O., Yang, Z., Nussenzweig, M. C. & Bjorkman, P. J. Partially open HIV-1 envelope structures exhibit conformational changes relevant for coreceptor binding and fusion. Cell Host & Microbe 24, 579-592. e574 (2018).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 90, 880, 904]]<|/det|>
+596 20 Thali, M. et al. Characterization of conserved human immunodeficiency virus type 1 597 gp120 neutralization epitopes exposed upon gp120- CD4 binding. Journal of virology 67, 598 3978- 3988 (1993). 599 21 Yang, Z., Wang, H., Liu, A. Z., Gristick, H. B. & Bjorkman, P. J. Asymmetric opening of 600 HIV- 1 Env bound to CD4 and a coreceptor- mimicking antibody. Nature structural & 601 molecular biology 26, 1167- 1175 (2019). 602 Lee, J. H., De Val, N., Lyumkis, D. & Ward, A. B. Model building and refinement of a 603 natively glycosylated HIV- 1 Env protein by high- resolution cryoelectron microscopy. 604 Structure 23, 1943- 1951 (2015). 605 Henderson, R. et al. Disruption of the HIV- 1 Envelope allosteric network blocks CD4- 606 induced rearrangements. Nature communications 11, 520 (2020). 607 24 Lu, M. et al. Associating HIV- 1 envelope glycoprotein structures with states on the virus 608 observed by smFRET. Nature 568, 415- 419 (2019). 609 25 Munro, J. B. et al. Conformational dynamics of single HIV- 1 envelope trimers on the 610 surface of native virions. Science 346, 759- 763 (2014). 611 26 Ao, Y. et al. Bioorthogonal click labeling of an amber- free HIV- 1 provirus for in- virus 612 single molecule imaging. Cell Chemical Biology 31, 487- 501. e487 (2024). 613 27 Bennett, A. L. et al. Microsecond dynamics control the HIV- 1 Envelope conformation. Sci 614 Adv 10, eadj0396 (2024). https://doi.org:10.1126/sciadv.adj0396 615 28 Liu, Q. et al. Quaternary contact in the initial interaction of CD4 with the HIV- 1 envelope 616 trimer. Nature structural & molecular biology 24, 370- 378 (2017). 617 29 Henderson, R. et al. (2020). 618 Joyce, M. G. et al. Soluble prefusion closed DS- SOSIP. 664- Env trimers of diverse HIV- 619 1 strains. Cell reports 21, 2992- 3002 (2017). 620 31 Ou, L. et al. Preclinical development of a fusion peptide conjugate as an HIV vaccine 621 immunogen. Scientific reports 10, 3032 (2020). 622 32 May, A. J., Pothula, K. R., Janowska, K. & Acharya, P. Structures of Langya Virus 623 Fusion Protein Ectodomain in Pre- and Postfusion Conformation. J Virol 97, e0043323 624 (2023). https://doi.org:10.1128/jvi.00433- 23 625 33 Jackson, C. B., Farzan, M., Chen, B. & Choe, H. Mechanisms of SARS- CoV- 2 entry into 626 cells. Nature reviews Molecular cell biology 23, 3- 20 (2022). 627 34 Low, J. S. et al. ACE2- binding exposes the SARS- CoV- 2 fusion peptide to broadly 628 neutralizing coronavirus antibodies. Science 377, 735- 742 (2022). 629 35 Gobeil, S. M.- C. et al. Effect of natural mutations of SARS- CoV- 2 on spike structure, 630 conformation, and antigenicity. Science 373, eabi6226 (2021). 631 36 Gobeil, S. M. et al. Structural diversity of the SARS- CoV- 2 Omicron spike. Mol Cell 82, 632 2050- 2068 e2056 (2022). https://doi.org:10.1016/j.molcel.2022.03.028 633 37 Aguilar, H. C., Henderson, B. A., Zamora, J. L. & Johnston, G. P. Paramyxovirus 634 glycoproteins and the membrane fusion process. Current clinical microbiology reports 3, 635 142- 154 (2016). 636 Baker, K. A., Dutch, R. E., Lamb, R. A. & Jardetzky, T. S. Structural basis for 637 paramyxovirus- mediated membrane fusion. Molecular cell 3, 309- 319 (1999). 638 39 Yang, Z. et al. Neutralizing antibodies induced in immunized macaques recognize the 639 CD4- binding site on an occluded- open HIV- 1 envelope trimer. Nature communications 640 13, 732 (2022). 641 40 Lee, M. et al. HIV- 1- envelope trimer transitions from prefusion- closed to CD4- bound- 642 open conformations through an occluded- intermediate state. bioRxiv, 2024.2007. 643 2015.603531 (2024). 644 41 Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for 645 rapid unsupervised cryo- EM structure determination. Nature methods 14, 290- 296 646 (2017).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 90, 880, 523]]<|/det|>
+647 42 Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and 648 electrons: recent developments in Phenix. Acta Crystallographica Section D: Structural 649 Biology 75, 861- 877 (2019). 650 43 Afonine, P. V. et al. Real- space refinement in PHENIX for cryo- EM and crystallography. 651 Acta Crystallographica Section D: Structural Biology 74, 531- 544 (2018). 652 Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. 653 Acta Crystallographica Section D: Biological Crystallography 66, 486- 501 (2010). 654 Schrödinger, L. The PyMOL Molecular Graphics System, Version 1.8. (No Title) (2015). 655 Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research 656 and analysis. Journal of computational chemistry 25, 1605- 1612 (2004). 657 Goddard, T. D. et al. UCSF ChimeraX: Meeting modern challenges in visualization and 658 analysis. Protein science 27, 14- 25 (2018). 659 Ao, Y. et al. Bioorthogonal click labeling of an amber- free HIV- 1 provirus for in- virus 660 single molecule imaging. Cell Chem Biol 31, 487- 501 e487 (2024). 661 https://doi.org:10.1016/j.chembiol.2023.12.017 662 Lu, M. et al. Associating HIV- 1 envelope glycoprotein structures with states on the virus 663 observed by smFRET. Nature 568, 415- 419 (2019). https://doi.org:10.1038/s41586- 019- 664 1101- y 665 50 Ma, X. et al. HIV- 1 Env trimer opens through an asymmetric intermediate in which 666 individual protomers adopt distinct conformations. Elife 7, e34271 (2018). 667 https://doi.org:10.7554/eLife.34271 668 Munro, J. B. et al. Conformational dynamics of single HIV- 1 envelope trimers on the 669 surface of native virions. Science 346, 759- 763 (2014). 670 https://doi.org:10.1126/science.1254426 671 52 Juette, M. F. et al. Single- Molecule imaging of non- equilibrium molecular ensembles on 672 the millisecond timescale. Nat Methods 13, 341- 344 (2016). 673 https://doi.org:10.1038/nmeth.3769
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[95, 88, 899, 275]]<|/det|>  
+
+<|ref|>image<|/ref|><|det|>[[98, 290, 895, 608]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[99, 624, 743, 642]]<|/det|>
+E Cryo- EM identified partially open populations with antibody accessible FP  
+
+<|ref|>image<|/ref|><|det|>[[113, 644, 860, 910]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[105, 664, 888, 886]]<|/det|>
+Figure 1. Time- dependent conformational changes in HIV- 1 BG505 SOSIP Env upon incubation with CD4. A. Structure of pre- fusion, pre- receptor, closed HIV- 1 Env (PDB: 518H) bound to broadly neutralizing, fusion peptide- directed antibody VRC34.01. The Env is shown in surface representation with the gp120 subunits colored light gray, and within the gp120 subunits, the V1V2 loop colored wheat, V3 loop olive and the residues contributing to the bridging sheet in the open Env colored in red. The gp41 subunits are colored black with the fusion peptide (FP) colored cyan. The antibody VRC34.01 is shown in ribbon representation bound to its FP- centered epitope. B. Structure of pre- fusion, CD4- bound open HIV- 1 Env bound to CD4- induced antibody 17b. The Env is colored similarly as in panel A. CD4 is shown as a yellow ribbon and 17b Fab is shown as an orange ribbon. C. Surface plasmon based binding (SPR) analysis monitoring FP burial. Env was incubated with either sCD4 alone or with CD4 and the coreceptor mimicking antibody 17b. At different time- points after incubation, binding was measured to the fusion peptide targeting antibody VRC34.01. D. Simultaneous Env opening and fusion peptide burial were measured by incubating Env with CD4 and at different time- points injecting over a VRC34.01 IgG or a 17b IgG surface. E. Cryo- EM reconstructions of three distinct populations of CD4/17b- bound, partially open Env bound to VRC34.01. Population 1 is bound to VRC34.01 at all three sites. The blue arrows indicate sites unoccupied by VRC34.01 in Populations 2 and 3.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[95, 88, 835, 110]]<|/det|>
+A Population 1 bound to CD4, 17b, and FP- directed antibody VRC34.01  
+
+<|ref|>image<|/ref|><|det|>[[95, 112, 900, 275]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[100, 289, 813, 310]]<|/det|>
+B Population 1 fit into in situ cryo- ET of CD4- bound Env intermediate  
+
+<|ref|>image<|/ref|><|det|>[[105, 310, 902, 460]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[102, 475, 718, 496]]<|/det|>
+C Population 1 comparison to Env in diverse conformations  
+
+<|ref|>image<|/ref|><|det|>[[102, 497, 880, 679]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[102, 688, 743, 711]]<|/det|>
+D Two distinct intermediate configurations of the HIV- 1 Env FP  
+
+<|ref|>image<|/ref|><|det|>[[98, 720, 901, 904]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[105, 627, 891, 907]]<|/det|>
+Figure 2: A partially open intermediate on the HIV- 1 entry pathway retains FP accessibility to antibody binding. A. Three views of the Population 1 structure shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. Within gp120, the bridging sheet is colored red and the α0 helix green. B. Population 1 coordinates including Env (gp120 in light gray, gp41 in black) and CD4 (yellow) fitted into the in situ cryo- ET reconstruction of a partially open CD4- bound Env (EMD- 29294). C. (Left to right) Vectors describing the position of gp120 relative to gp41. The gp120 structure (blue), gp120 V1/V2 region (green), and gp41 (orange) in the closed state overlayed with the centroid locations depicting the dihedral, angles, and distances describing the position of gp120 relative to gp41. Dihedral, angle, and distance values for closed, intermediate, and open state structures. D. (Left) Population 1 protomer shown in surface representation zoomed- in at the location of the FP. FP is shown in cartoon representation. The gp120 subunit is colored light grey, gp41 black, FP cyan and FPPR pale green. (Middle) One protomer of the partially open Env bound to CD4, 17b Fab and 8ANC195 Fab (PDB ID: 6CM3) shown in surface representation zoomed- in at the location of the FP. FP is shown in cartoon representation. The gp120 subunit is colored dark grey, gp41 magenta, FP dark teal and FPPR light pink. (Right) Overlay of a Population 1 protomer with a protomer of a partially open Env bound to CD4, 17b Fab and 8ANC195 Fab (PDB ID: 6CM3). The gp120 subunits of each structure were used for the superposition. Inset zooms in on the FP and FPPR. Zoomed- in panel is slightly rotated compared to the zoomed- out view for better visualization. The solid lines (colored pale green for Population 1 and light pink for 6CM3) show the distance between FPPR residues Gln 540 and gp120 residue Phe 223.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[95, 90, 870, 111]]<|/det|>
+A Population 2 bound to CD4, 17b, and 2X FP- directed antibody VRC34.01  
+
+<|ref|>image<|/ref|><|det|>[[95, 113, 870, 270]]<|/det|>  
+
+<|ref|>image<|/ref|><|det|>[[95, 284, 905, 444]]<|/det|>  
+
+<|ref|>sub_title<|/ref|><|det|>[[95, 455, 351, 474]]<|/det|>
+## E Extent of Env opening  
+
+<|ref|>text<|/ref|><|det|>[[118, 481, 375, 520]]<|/det|>
+Env: BG505 SOSIP Ligands: CD4, 17b, 8ANC195 Conformation: Closed (PDB:5ACO)  
+
+<|ref|>text<|/ref|><|det|>[[468, 480, 592, 520]]<|/det|>
+B41 SOSIP CD4, 17b Open (PDB:5VN3)  
+
+<|ref|>text<|/ref|><|det|>[[690, 480, 873, 520]]<|/det|>
+BG505 SOSIP CD4, 17b, 8ANC195 Partially open (PDB:6CM3)  
+
+<|ref|>image<|/ref|><|det|>[[199, 522, 360, 644]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[110, 648, 303, 675]]<|/det|>
+V1- V2 base: 14.7 Å CD4bs: 56 Å  
+
+<|ref|>image<|/ref|><|det|>[[457, 522, 616, 644]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[480, 648, 540, 675]]<|/det|>
+76.7 Å 84.3 Å  
+
+<|ref|>image<|/ref|><|det|>[[692, 522, 861, 644]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[732, 648, 787, 675]]<|/det|>
+65 Å 79.4 Å  
+
+<|ref|>sub_title<|/ref|><|det|>[[98, 685, 116, 701]]<|/det|>
+## F  
+
+<|ref|>text<|/ref|><|det|>[[110, 708, 403, 747]]<|/det|>
+Env: BG505 SOSIP Ligands: CD4, 17b, (3X) VRC34.01 Conformation: Partially open (Population 1)  
+
+<|ref|>text<|/ref|><|det|>[[460, 708, 654, 747]]<|/det|>
+BG505 SOSIP CD4, 17b, (2X) VRC34.01 Partially open (Population 2)  
+
+<|ref|>text<|/ref|><|det|>[[684, 708, 875, 747]]<|/det|>
+BG505 SOSIP CD4, 17b, (1X) VRC34.01 Partially open (Population 3)  
+
+<|ref|>image<|/ref|><|det|>[[225, 750, 375, 870]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[104, 877, 378, 907]]<|/det|>
+V1- V2 base: 63 Å, 66 Å, 66.9 CD4bs: 75 Å, 77.6 Å, 78.2 Å  
+
+<|ref|>image<|/ref|><|det|>[[455, 750, 645, 870]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[460, 877, 630, 907]]<|/det|>
+61.4 Å, 71.8 Å, 70.1 Å, 74.6 Å, 82.8 Å, 80.2 Å  
+
+<|ref|>image<|/ref|><|det|>[[686, 750, 880, 870]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[691, 877, 861, 907]]<|/det|>
+73.6 Å, 71.4 Å, 69.6 Å 61.6 Å, 79.5 Å, 82.8 Å
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[97, 696, 881, 907]]<|/det|>
+Figure 3: Burial of FP upon gp120 opening. A. Three views of the Population 2 structure shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. Within gp120, the bridging sheet is colored red and the α0 helix green. Blue circle headed arrows indicate the gp41 positions that are not bound to VRC4.01 Fab. B. View of Population 3 coordinates from the viral membrane shown in cartoon representation with gp41 colored black, gp120 light gray, CD4 yellow, VRC34.01 Fab blue, 17b Fab orange. Glycans are shown in stick representation. The FP within the gp41 subunit is colored cyan. C. Population 2 structure zoomed-in view of gp41 subunit that was not bound to VRC34.01 showing the buried FP in cyan and the FPPR in light green. The cryo-EM reconstruction is shown as a transparent surface with fitted coordinates shown in cartoon representation. D. Population 3 structure zoomed-in view of its two gp41 subunits that were not bound to VRC34.01 showing the buried FP in cyan and the FPPR in light green. The cryo-EM reconstruction is shown as a transparent surface with fitted coordinates shown in cartoon representation. E and F. Extent of Env openness measured as the distance between residue 368 (blue spheres) and residue 124 (red spheres) in E. previously published Env conformational states and F. Env conformational states defined in this study.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[100, 96, 612, 115]]<|/det|>
+## A State of the FP along CD4-induced Env opening pathway  
+
+<|ref|>image<|/ref|><|det|>[[95, 120, 900, 460]]<|/det|>  
+
+<|ref|>sub_title<|/ref|><|det|>[[102, 475, 570, 493]]<|/det|>
+## D gp41 reorganization required to stably sequester FP  
+
+<|ref|>image<|/ref|><|det|>[[105, 498, 884, 701]]<|/det|>  
+
+<|ref|>sub_title<|/ref|><|det|>[[102, 717, 763, 737]]<|/det|>
+## E smFRET analysis of impact of VRC34.01 binding on Env opening trajectory  
+
+<|ref|>image<|/ref|><|det|>[[135, 765, 881, 910]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[105, 604, 891, 899]]<|/det|>
+Figure 4: FP trajectory upon CD4- induced Env opening. A. HIV- 1 Env structures organized based on extent of opening. Left to right: closed Env bound to VRC34.01 (PDB: 5I8H) with gp41 colored olive, FPPR orange and FP cyan; partially open, bound to CD4, 17b and VRC34.01 (PDB:9D90; This study) with gp41 colored black, FPPR light green and FP cyan; partially open, bound to CD4, 17b and 8ANC195 (PDB:6CM3) with gp41 colored magenta, FPPR light pink and FP teal; partially open, bound to CD4, 17b and VRC34.01 (EMD- 46671; This study) with gp41 colored black, FPPR light green and FP cyan; fully open, bound to CD4 and 17b (PDB:5VN3) with gp41 colored blue, FPPR light blue and FP cyan. The gp120 subunit is colored light grey. Inset shows a zoomed in view of the region around the \(\alpha 0\) helix (green). B. 180° rotated views of structures are shown in A. with a brown square indicating the FP region. The FP is colored cyan except in the partially open CD,17b,8ANC195 bound BG505 structure where the FP is colored teal. C. zoomed in views of the region around the FP. The red arrows indicate the direction of CD4- induced Env opening from the pre- CD4, closed Env to the fully CD4- induced fully open Env. D. Comparisons of gp41 organization between fully open Env with sequestered and inaccessible FP (PDB: 5VN3), and (left) P1 (transiently exposed FP), (middle) partially open Env with transiently buried FP (PDB: 6CM3), and (right) P2 (inaccessible FP). The red arrows in the left and middle panels indicate the HR2 movement that occurs between the partially open and the full open Envs allowing the FPPR to reorient creating space for FP burial. E- F. Three- dimensional presentation (E) and quantification (F) of conformational distribution- indicated FRET histograms shown in from the gp120- gp41 perspective in Figure S7. Virus EnvBG505 samples three primary conformational states (PT: Pre- triggered, PC: Prefusion Closed, and CO: CD4- bound open). PT predominates in the ligand- free condition, while VRC34 shifts the conformational landscape differently from that of the CD4- bound opening.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[92, 102, 900, 275]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[95, 93, 781, 113]]<|/det|>
+<center>A Population 4 bound to CD4 and FP-directed antibody VRC34.01 </center>  
+
+<|ref|>image<|/ref|><|det|>[[102, 303, 900, 472]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[95, 281, 785, 303]]<|/det|>
+<center>B Population 5 bound to CD4 and FP-directed antibody VRC34.01 </center>  
+
+<|ref|>image<|/ref|><|det|>[[102, 512, 902, 820]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[98, 483, 498, 504]]<|/det|>
+<center>C Step-wise CD4-induced Env opening. </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[106, 604, 890, 899]]<|/det|>
+Figure 5: Partially open early intermediates and a step- wise mechanism for CD4- induced Env opening. A, CD4, VRC34.01- bound BG505 SOSIP Env with a single gp120 rotated and showing the formation of the bridging sheet (red) and \(\alpha 0\) helix that are the markers for CD4- induced Env. B, CD4, VRC34.01- bound BG505 SOSIP Env with two gp120s rotated and showing the formation of the bridging sheet (red) and \(\alpha 0\) helix that are the markers for CD4- induced Env. C, A structure- guided mechanism for stepwise Env opening along the HIV- 1 entry pathway. Top panel structures were determined previously, bottom panel structures were determined in this study. Stepwise transitions are marked with numbers within a circle on top of each structure starting from (1) the binding of a single CD4 to a closed Env trimer (PDB: 5U1F, 8FYI). This is followed by (2) opening of the Env trimer (EMD- 29292) that allows additional CD4 molecules to bind. (5) A partially open Env conformation was described bound to CD4, a coreceptor mimicking antibody, and the gp120/gp41 interface targeting antibody 8ANC195 (PDB: 6CM3, 6EDU) where the FP was buried within a gp41 cavity. The CD4- induced opening of the HIV- 1 Env culminates in the complete rotation of all the gp120 subunits that are accompanied by gp41 conformational changes and resulting in the burial of FP. This state is numbered (8) in this schematic. This study showed that the geometry of the functional entry intermediate (5) that was also visualized on membrane- associated Env (EMD- 29294), was compatible with the FP being either buried or exposed, and thus, in this conformation the FP was accessible to antibodies. Further, this study filled in mechanistic gaps between (2) and (5) by showing stepwise gp120 rotation to reach this functional entry intermediate. Finally, this study visualized a stepwise mechanism for how the functional entry intermediate (5) may transition to the fully open Env (8), yet again by stepwise opening of the each gp120 subunit from its partially rotated to the fully rotated conformation, which was accompanied by burial of the FP in the corresponding protomer.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[42, 43, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[42, 93, 768, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 131, 463, 178]]<|/det|>
+- TableS1.xlsx- FPpaperSupplementfigures14Sept2024.pdf
+
+<--- Page Split --->

preprint/preprint__98a3eb377368a4beb38d1aca76d6144c971ad88eff639624195d30516e45948c/images_list.json

@@ -0,0 +1,63 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "a Wildtype (10 mm)",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 19
+  },
+  {
+    "type": "image",
+    "img_path": "images/Extended_Data_Figure_3.jpg",
+    "caption": "Extended Data Fig. 3: Spinal curvature in 21dpf \\(sspo^{dmh4 / + }\\) mutant fish.",
+    "footnote": [],
+    "bbox": [
+      [
+        114,
+        303,
+        708,
+        465
+      ]
+    ],
+    "page_idx": 21
+  },
+  {
+    "type": "image",
+    "img_path": "images/Extended_Data_Figure_5.jpg",
+    "caption": "Extended Data Fig. 5: Shear wave elastography (SWE) of the zebrafish spine. a, Photograph of SWE experimental apparatus setup. b, Schematic of SWE mechanics in zebrafish, imaging with ultrafast ultrasound at 28000 frames per second (fps). High frequency linear ultrasound probe induces an acoustic radiation force (ARF) into the center of the zebrafish spine,",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 23
+  },
+  {
+    "type": "image",
+    "img_path": "images/Extended_Data_Figure_6.jpg",
+    "caption": "Extended Data Fig. 6: No significant difference in tissue mineral density is observed in vertebral centra of \\(sspo^{dmh4 / +}\\) and wildtype sibling zebrafish.",
+    "footnote": [],
+    "bbox": [
+      [
+        120,
+        180,
+        858,
+        528
+      ]
+    ],
+    "page_idx": 25
+  },
+  {
+    "type": "image",
+    "img_path": "images/Extended_Data_Figure_7.jpg",
+    "caption": "Extended Data Fig. 7: Increased spine stiffness is observed across diverse zebrafish AIS models.",
+    "footnote": [],
+    "bbox": [
+      [
+        125,
+        200,
+        876,
+        420
+      ]
+    ],
+    "page_idx": 28
+  }
+]

preprint/preprint__98a3eb377368a4beb38d1aca76d6144c971ad88eff639624195d30516e45948c/preprint__98a3eb377368a4beb38d1aca76d6144c971ad88eff639624195d30516e45948c.mmd

@@ -0,0 +1,506 @@
+
+# Oxidative stress induces intervertebral ECM remodelling, elevated tissue stiffness and idiopathic-like scoliosis  
+
+Brian Ciruna ciruna@sickkids.ca  
+
+The Hospital for Sick Children https://orcid.org/0000- 0001- 7918- 0953  
+
+The Hospital for Sick Children https://orcid.org/0000- 0003- 4346- 9434  
+
+Ran Xu The Hospital for Sick Children  
+
+Josh Gopaul The Hospital for Sick Children  
+
+Arash Panahifar Canadian Light Source https://orcid.org/0000- 0001- 9483- 6949  
+
+Vida Erfani The Hospital for Sick Children  
+
+Jenica Van Gennip The Hospital for Sick Children https://orcid.org/0000- 0002- 4230- 7508  
+
+B Eames University of Saskatchewan  
+
+Nikan Fakhari The Hospital for Sick Children  
+
+Jerome Baranger Institute Physics for Medicine Paris,Inserm, ESPCI PSL Paris, CNRS https://orcid.org/0000- 0002- 2311- 716X  
+
+David Lebel The Hospital for Sick Children  
+
+Olivier Villemain The Hospital for Sick Children  
+
+Article  
+
+Keywords:
+
+<--- Page Split --->
+
+**Posted Date:** September 9th, 2024  
+
+**DOI:** https://doi.org/10.21203/rs.3.rs-4978808/v1  
+
+**License:** © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+**Additional Declarations:** There is **NO** Competing Interest.  
+
+**Version of Record:** A version of this preprint was published at Nature Communications on September 30th, 2025. See the published version at https://doi.org/10.1038/s41467-025-63742-2.
+
+<--- Page Split --->
+
+## Title:  
+
+## Oxidative stress induces intervertebral ECM remodeling, elevated tissue stiffness and idiopathic-like scoliosis  
+
+## Authors:  
+
+Patrick G. Pumputis \(^{1,2\dagger}\) , Ran Xu \(^{1,2\dagger}\) , Josh Gopau \(^{1\dagger}\) , Arash Panahifar \(^{4,5}\) , Vida Erfani \(^{1,2}\) , Jenica Van Gennip \(^{1}\) , B. Frank Eames \(^{6}\) , Nikan Fakhari \(^{3,7}\) , Jerome Baranger \(^{3}\) , David E. Lebel \(^{8}\) , Olivier Villemain \(^{3,7*}\) , Brian Ciruna \(^{1,2*}\)  
+
+## Affiliations:  
+
+\(^{1}\) Developmental & Stem Cell Biology Program, The Hospital for Sick Children; \(^{2}\) Toronto, Ontario, Canada \(^{3}\) Department of Molecular Genetics, University of Toronto; Toronto, Ontario, Canada \(^{3}\) Translational Medicine Program, The Hospital for Sick Children; Toronto, Ontario, Canada \(^{4}\) BioMedical Imaging and Therapy Beamline, Canadian Light Source; Saskatoon, Canada \(^{5}\) Department of Medical Imaging, College of Medicine, University of Saskatchewan; Saskatoon, Canada \(^{6}\) Department of Anatomy, Physiology, and Pharmacology, University of Saskatchewan; Saskatoon, Canada \(^{7}\) Department of Medical Biophysics, University of Toronto; Toronto, Ontario, Canada \(^{8}\) Division of Orthopaedic Surgery, The Hospital for Sick Children; Toronto, Ontario, Canada \(^{+}\) These authors contributed equally to this work \(*\) Co- senior and co- corresponding authors. Emails: ciruna@sickkids.ca & olivier.villemain@chubordeaux.fr  
+
+## Abstract:  
+
+Adolescent idiopathic scoliosis (AIS) is the most prevalent pediatric spine disorder, developing in the absence of obvious congenital or physiological defects \(^{1}\) . Patient genetic sequencing and mouse functional studies have demonstrated association of musculoskeletal collagen variants and cartilaginous extracellular matrix (ECM) defects in a subset of cases \(^{2 - 4}\) . However, the underlying biological causes of AIS are poorly understood, thus treatment options remain limited to physical bracing or invasive corrective surgery \(^{1}\) . Here we interrogate the biological causes of scoliosis in zebrafish preclinical models of AIS. We demonstrate that neuroinflammation- associated reduction- oxidation (redox) imbalance induces cell stress and collagen remodeling defects within intervertebral segments of the developing spine. Mutant spines are consequently stiffer, as measured by shear wave elastography, and exhibit deformations of intervertebral structures. Remarkably, both elevated spine stiffness and intervertebral ECM phenotypes are detectable prior to scoliosis onset in zebrafish models, suggesting a causal relationship, and can be suppressed by antioxidant treatment. Together, our studies implicate oxidative stress- induced intervertebral deformations in the pathogenesis of AIS
+
+<--- Page Split --->
+
+and identify elevated spine stiffness and redox imbalance as plausible first- in- kind prognostic biomarkers and therapeutic targets.  
+
+## Main Text:  
+
+## Introduction:  
+
+Adolescent idiopathic scoliosis (AIS) is a common pediatric disorder characterized by lateral deviations in the spinal column (with a rotational component) that develop in the absence of obvious congenital or physiological defects<sup>1</sup>. AIS afflicts 3- 4% of children worldwide yet underlying causes are poorly understood, in part due to genetic heterogeneity and suspected environmental influences<sup>1</sup>. Patient genetic sequencing and mouse functional studies have demonstrated association of musculoskeletal collagen variants and cartilaginous extracellular matrix (ECM) defects in a subset of cases<sup>2- 4</sup>. However, over 95% of total causative genetic variance in AIS is thought to be unknown<sup>5</sup>. Because most AIS etiological studies involve patients with established scoliotic curves, determining a biological mechanism of cause and effect is particularly challenging, and treatment options remain limited to mechanical interventions like bracing or invasive corrective surgery<sup>1</sup>.  
+
+Zebrafish have emerged as powerful experimental models for dissecting complex biological mechanisms associated with AIS<sup>6,7</sup>. Remarkably, it was discovered that neuroinflammatory signals and oxidative stress, arising from imbalances in cerebrospinal fluid (CSF) homeostasis, are both necessary and sufficient to cause AIS- like spinal curvatures in zebrafish<sup>8- 10</sup>. Furthermore, treatment with antioxidant and immunomodulating compounds like N- acetylcysteine (NAC) or NAC- ethyl ester (NACET, a more bioavailable form of NAC) can suppress scoliosis onset and severe spinal curve progression in zebrafish models<sup>9- 11</sup>. While this provides proof- of- principle that scoliosis might be managed therapeutically, undetermined mechanisms by which oxidative stress leads to spinal curvature and their relevance to human AIS pose a barrier to clinical translation.  
+
+Here we interrogate the downstream causes of scoliosis in the dominant SCO- spondin \((ssp^{dmhd / + })\) mutant zebrafish model of AIS<sup>10,12</sup>. These fish develop idiopathic- like spinal curvatures in response to neuroinflammatory signals that are linked to the disruption of Reissner's fiber (RF), a proteinaceous filament that threads through ventricular cavities of the spinal cord and brain. We demonstrate that abnormal axial reduction- oxidation (redox) states in \(ssp^{dmhd / + }\) mutants induce both intervertebral collagen ECM remodelling defects and elevated spine stiffness that appear causally linked to the onset and progression of scoliosis. Furthermore, we characterize resulting intervertebral disc deformations as a pathophysiological mechanism shared between zebrafish scoliosis models and human AIS patients. Finally, we provide evidence that elevated spine stiffness and oxidative stress may prove to be valuable, first- in- kind, prognostic biomarkers and therapeutic targets for idiopathic- like scoliosis.
+
+<--- Page Split --->
+
+## Results:  
+
+sspo<sup>dmh4/+</sup> mutants exhibit an elevated oxidative state in proximity to the developing spine NACET, a potent anti- oxidant, can suppress the onset and severe progression of scoliosis in sspo<sup>dmh4/+</sup> AIS models<sup>10</sup>. To investigate how oxidative stress may influence spine development, we characterized the spatial distribution of redox imbalance in larval zebrafish using the cell permeable fluorescent redox probe 2',7'- dichlorofluorescein diacetate (DCFH- DA)<sup>13</sup>. While the probe cannot attribute redox changes to specific reactive oxygen species (ROS), it can be used as a general indicator of redox state<sup>13</sup>. Staining was performed on wildtype (WT) and sspo<sup>dmh4/+</sup> zebrafish at 10 days post fertilization (dpf) and 6 mm standard length (SL), prior to scoliosis onset. At this stage, sspo<sup>dmh4/+</sup> zebrafish appear morphologically indistinguishable from WT (Fig. 1a, b). However, significantly stronger oxidation levels were detected in the dorsal telencephalon and cranial vessels of sspo<sup>dmh4/+</sup> mutants (Fig. 1c, d, g), which spatially correlate with sites of neuroinflammation previously identified in these animals<sup>10</sup>. Strikingly, elevated oxidative signals were also observed along the trunk of sspo<sup>dmh4/+</sup> larva in the spinal cord, trunk vasculature and dorsal aorta, as well as in discrete peri- notochordal segments (Fig. 1e, f, h- k). These notochordal segments correspond to developing cartilaginous intervertebral domains (IVDs) of the spine, as demonstrated by the complementary pattern of TgBAC(entpd5a::pkRed) reporter expression in alternating mineralizing domains (Fig. 11- n)<sup>14</sup>. In contrast, no difference in DCFH- DA signals were observed in the trunk muscle (myotome) of 10 dpf WT and sspo<sup>dmh4/+</sup> zebrafish larva (Extended Data Fig. 1a- c). Overall, these data suggest that neuroinflammatory responses observed in sspo<sup>dmh4/+</sup> mutants associate with tissue non- autonomous elevation of ROS in close juxtaposition to the notochord and developing spine.  
+
+## Redox imbalance induces endoplasmic reticulum (ER) stress within intervertebral segments  
+
+The zebrafish notochord, comprised of inner vacuolated cells and outer peri- notochordal sheath cells, provides structural support to the embryo and serves as a template for spine development<sup>6</sup>. Sheath cells, abundant in rough ER, are secretory cells that contribute to the notochord's ECM- rich lamellar sheath<sup>6</sup>, which segments into developing vertebrae and IVDs<sup>14</sup>. Since dysregulation of redox homeostasis in cells can initiate ER stress and an unfolded protein response (UPR)<sup>15</sup>, we hypothesized that peri- notochordal accumulation of ROS in sspo<sup>dmh4/+</sup> IVD segments may induce ER stress and disrupt sheath cell function.  
+
+To assess ER stress in sspo<sup>dmh4/+</sup> mutants, we modified an established transgenic assay that incorporates an IRE1 endonuclease- catalyzed splicing cassette that reports on XBP1 activation via expression of a functionally inert XBP1Δ- GFP fusion protein (Extended Data Fig. 2a)<sup>16</sup>. Specifically, the xbp1δ- eGFP ER stress reporter was cloned downstream of a col2a1a promoter<sup>17</sup> to direct expression within cartilaginous IVD segments and the site of observed oxidation (arrowheads, Fig. 1f, l- n). Basal levels of Tg(col2a1a::xbp1δ- eGFP) reporter activity and ER stress could be detected within developing IVDs of WT fish (10 dpf, 6 mm SL) (Fig. 1o) in between alternating mineralizing domains labeled by Tg(entpd5a::pkRed)<sup>14</sup>. Strikingly, a significant increase in Tg(col2a1a::xbp1δ- eGFP) reporter activity was observed within IVDs of
+
+<--- Page Split --->
+
+\(s s p o^{d m h4 / + }\) mutant zebrafish (10 dpf, 6 mm SL), indicating an active ER stress response prior to the onset of scoliosis (Fig. 1p, q). Notably, this ER stress response was transient, as no difference in \(T g(c o l2a1a::x b p1\delta - e G F P)\) reporter activity was observed between \(s s p o^{d m h4 / + }\) mutant and wildtype zebrafish at 14 dpf (6.5 mm SL; Extended Data Fig. 2g- i).  
+
+To determine whether observed ER stress was linked to oxidative stress, larval zebrafish were treated with \(500~\mu \mathrm{M}\) of the anti- oxidant NACET from 5 to 10 dpf. NACET treatment had no effect on \(T g(c o l2a1a::x b p1\delta - e G F P)\) reporter activity in wildtype zebrafish, but fully suppressed ER stress responses in \(s s p o^{d m h4 / + }\) mutant zebrafish to wildtype levels (Extended Data Fig. 2b- f). Together, these data indicate that redox imbalances induce ER stress within the IVDs of \(s s p o^{d m h4 / + }\) mutant animals preceding scoliosis onset, and may thus be functionally linked to spinal curvature.  
+
+## \(s s p o^{d m h4 / + }\) mutants exhibit IVD deformations and collagen ECM defects  
+
+To further investigate the functional consequences of oxidative/ER stress on zebrafish IVD development, we performed high- resolution synchrotron- based X- ray micro- computed tomography (synchrotron \(\mu \mathrm{CT}\) ) on juvenile fish at 21dpf (10 mm SL), when spinal curvature is well- established in \(s s p o^{d m h4 / + }\) AIS models (Extended Data Fig. 3). Samples were stained using a diffusive iodine- based contrast enhancement, and 4- 5 segments of the caudal spine were imaged at \(0.36~\mu \mathrm{m}\) voxel size to produce high- resolution 3D datasets of the mineralized skeleton and surrounding soft tissues (Fig. 2a,e; Supplementary Videos 1- 4). At stages analyzed, zebrafish IVDs have differentiated into two principal structures (Extended Data Fig. 4a, b). These include an inner domain comprised of vacuolated and sheath notochordal cells; and an outer intervertebral ligament (IVL) that physically connects two adjacent vertebral bodies, comprised of layered collagen type II and elastin matrix, dense collagen type 1 matrix and collagen type 1 bundle fibers \(^{18}\) . These inner and outer intervertebral structures are functionally analogous to the human nucleus pulposus (NP) and annulus fibrosus (AF), respectively \(^{18,19}\) .  
+
+Remarkably, the IVL (AF) of \(s s p o^{d m h4 / + }\) mutants exhibited obvious structural defects and appeared highly deformable. In wildtype zebrafish, the IVL could be identified connecting neighbouring vertebral bodies, looping outward in the coronal plane away from notochordal cells and towards neighbouring muscle fibres (arrowhead, Fig. 2b). This organization of IVL structure, previously described for wildtype adult zebrafish \(^{18,19}\) , was stable across the circumference of the IVD (Fig. 2c- c'"; Supplementary Video 5) with its dense collagen layer appearing in transverse sections as smooth, concentric bands of high- contrast stain encircling the vertebral endplates (arrowhead, Fig. 2d). In contrast, coronal images of \(s s p o^{d m h4 / + }\) mutants often demonstrated an inversion of typical AF structure, with sections of \(s s p o^{d m h4 / + }\) IVL folding interiorly towards the notochordal cell layers (arrowhead, Fig. 2f). Strikingly, the contour of \(s s p o^{d m h4 / + }\) IVL layers exhibited significant deformations over short distances (Fig. 2g- g'") and, in some instances, could be observed transitioning between wildtype and inverted orientations within \(30~\mu \mathrm{m}\) intervals (Supplementary Videos 6, 7). As a result, transverse sections of \(s s p o^{d m h4 / + }\)
+
+<--- Page Split --->
+
+1 mutants exhibited irregular and twisted IVL borders (arrows, Fig. 2h) displaced towards the interior surface of the vertebral endplates (arrowhead, Fig. 2h).  
+
+To probe the structural integrity of the IVD matrix in \(sspo^{dmh4 / + }\) mutants, we performed collagen hybridizing peptide (CHP) stains on axial skeleton preparations of juvenile fish at 21 dpf (10 mm SL). CHP is a synthetic construct that mimics the helical structure of endogenous collagen peptides, and can bind to open regions in collagen triple helices to report on collagen ECM damage and remodelling (Extended Data Fig. 4g) \(^{20}\) . CHP staining appeared stronger in \(sspo^{dmh4 / + }\) mutant IVLs compared to wildtype, and highlighted irregularities in IVL structure (arrows, Extended Data Fig. 4c-f) previously observed in microCT images. Furthermore, brighter and ectopic CHP signals were also observed within the NP layer of \(sspo^{dmh4 / + }\) mutants (arrowheads, Extended Data Fig. 4c-f) indicating extensive damage to the collagen matrix, dysregulated ECM remodelling, or both. These data demonstrate abnormalities in the structural properties of \(sspo^{dmh4 / + }\) IVDs, both at the morphological and molecular level.  
+
+## Zebrafish AIS models exhibit stiffer spines  
+
+The viscoelastic properties of human intervertebral disc tissue are also significantly altered in AIS patients. Specifically, the AF is stiffer in AIS patients compared to healthy controls, as measured by shear wave elastography (SWE) \(^{21,22}\) . SWE uses an acoustic radiation force to induce mechanical tissue displacement and calculates the velocity of shear waves propagating in the orthogonal direction. As tissue stiffness is directly proportional to shear wave speed, this provides a quantitative measure of tissue elasticity \(^{23,24}\) . To further characterize physiological defects associated with oxidative stress and scoliosis in our models, we utilized SWE to quantitatively assess zebrafish spine stiffness (Extended Data Fig. 5a- c).  
+
+To begin, we conducted SWE at 16 dpf (average \(\mathrm{SL} = 7.1 + / - 0.2 \mathrm{mm}\) ) and 21 dpf (average \(\mathrm{SL} = 8.9 + / - 0.6 \mathrm{mm}\) ) stages, corresponding to the onset and progression of scoliosis in \(sspo^{dmh4 / + }\) mutants, respectively (Fig. 3a, b). Euthanized zebrafish were embedded in low melt agarose, and SWE was performed on the abdominal segment of the spine (note that variations in SWE ‘push’ locations and agarose density did not significantly alter shear wave velocities; Extended Data Fig. 5d, e). Remarkably, \(sspo^{dmh4 / + }\) shear wave velocities were significantly faster than wildtype controls at both time points (Fig. 3c, d), indicating an increase in spine stiffness in scoliotic \(sspo^{dmh4 / + }\) mutant zebrafish. Observed increases in spine stiffness were not likely caused by changes in vertebral bone density, as microCT analysis of the mineralized axial skeleton of adult zebrafish did not reveal significant differences in \(sspo^{dmh4 / + }\) tissue mineral density compared to wildtype siblings (Extended Data Fig. 6).  
+
+To determine whether elevated spine stiffness represents a generalizable feature of zebrafish scoliosis, or a property specific to \(sspo^{dmh4 / + }\) mutants, we performed SWE on two additional AIS models (katnb1 and vangl2) that share upstream CSF homeostasis defects but differ in underlying molecular genetic causes \(^{25,26}\) . Again, significantly faster shear wave velocities were observed in scoliotic katnb1 and vangl2 mutants compared to control siblings (Extended Data Fig. 7). Therefore, our results demonstrate altered viscoelastic properties of the
+
+<--- Page Split --->
+
+1 zebrafish scoliotic spine that parallel elevated intervertebral disc stiffness observed in human 2 AIS patients \(^{21,22}\) .  
+
+## Oxidative stress-induced spine stiffness is a prognostic biomarker of scoliosis  
+
+Shear wave velocities varied noticeably within \(sspo^{dmh4 / + }\) mutant cohorts (Fig. 3c, d). Since the penetrance and severity of spinal curvature is also variable among \(sspo^{dmh4 / + }\) zebrafish \(^{10}\) , we hypothesized that differences in shear wave velocity/spine stiffness may directly reflect scoliosis severity. Using a second cohort of 21 dpf \(sspo^{dmh4 / + }\) mutants (average \(\mathrm{SL} = 8.2\) \(+ / - 0.7 \mathrm{mm}\) ), we performed microCT imaging after SWE and measured total Cobb angle values, which is a biomarker of spinal curve severity. Remarkably, a significant and positive correlation between shear wave velocity and total Cobb- angle/curve severity was observed ( \(\mathrm{R} = 0.831\) ; Fig. 3e).  
+
+Elevated spine stiffness was measured after scoliosis onset and could therefore have developed as a consequence of spinal curvature. To determine the relationship between stiffness and scoliosis, we next conducted SWE on \(sspo^{dmh4 / + }\) and wildtype fish at 10 dpf (average \(\mathrm{SL} = 4.7 + / - 0.4 \mathrm{mm}\) ), prior to obvious morphological phenotypes. Remarkably, \(sspo^{dmh4 / + }\) mutant spines were significantly stiffer than wildtype controls before scoliosis onset, suggesting a causal relationship (Fig. 3f). To determine whether elevated tissue stiffness was linked to elevated oxidative stress, larval zebrafish were treated with \(500 \mu \mathrm{M}\) of NACET from 5 to 10 dpf, prior to SWE. Strikingly, NACET treatment had no effect on wildtype control animals, but fully suppressed abnormal spine stiffness in \(sspo^{dmh4 / + }\) zebrafish to wildtype levels (Fig. 3f). Together, these results directly link oxidative stress to altered viscoelastic properties of the scoliotic spine and identify elevated spine stiffness as a prognostic biomarker of spinal curvature in zebrafish AIS models.  
+
+## IVD matrix defects precede scoliosis, and are directly linked to oxidative stress  
+
+IVD deformations and collagen remodelling defects were also observed after scoliosis onset in \(sspo^{dmh4 / + }\) mutants and may therefore reflect the cause or the consequences of spinal curvature. Although SWE indicates structural defects likely precede scoliosis, it lacks sufficient spatial resolution to identify affected tissues. To further investigate whether ultrastructural ECM defects drive spinal curvature in \(sspo^{dmh4 / + }\) models, we performed transmission electron microscopy (TEM) on 10 dpf zebrafish (5 mm SL) prior to an obvious morphological phenotype. Specifically, we focussed on mineralized and cartilaginous domains of the peri- notochordal ECM that template vertebral centra and IVD development, respectively (Fig. 4a).  
+
+In wildtype animals, a uniform layer of collagen ECM was observed within cartilaginous domains, with bundles of collagen fibres organized circumferentially around the notochord projecting out of the sagittal plane of sectioning (Fig. 4b, b'). In contrast, \(sspo^{dmh4 / + }\) cartilaginous segments exhibited irregularities in the thickness and orientation of collagen ECM layers (Fig. 4c, c'). Quantification of collagen ECM thickness revealed that while the local minimum was consistent between WT and \(sspo^{dmh4 / + }\) IVD segments, the maximum thickness and local max/min
+
+<--- Page Split --->
+
+ratios were significantly higher in \(sspo^{dmh4 / + }\) mutants (Fig. 4f- h). Furthermore, the ECM appeared disorganized and misoriented in developing \(sspo^{dmh4 / + }\) IVD segments, with collagen fibre bundles observed running parallel to the sagittal plane (Fig 4. c, c'). Although the penetrance and severity of \(sspo^{dmh4 / + }\) ECM phenotypes varied among IVD segments (Fig. 4f- h), all \(sspo^{dmh4 / + }\) fish \((N = 9)\) exhibited IVD defects. Notably, collagen ECM defects were specific to IVDs, and were not observed in mineralized domains around the notochord sheath (Extended Data Fig. 8).  
+
+To determine whether collagen ECM defects are also linked to the elevated oxidative state observed around the IVD of \(sspo^{dmh4 / + }\) mutants, larval zebrafish were treated with \(500~\mu \mathrm{M}\) NAcET from 5 to 10 dpf, prior to TEM imaging. Strikingly, NAcET treatment had no effect on collagen matrix organization in wildtype animals but restored collagen fibre orientation and suppressed ECM thickness in \(sspo^{dmh4 / + }\) mutants to wildtype levels (Fig. 4d- h). Together, these results demonstrate that oxidative stress- induced collagen ECM defects, specific to IVD segments, precede scoliosis onset in \(sspo^{dmh4 / + }\) animals and may therefore be functionally linked to IVL deformations and idiopathic- like spinal curvature.  
+
+## Discussion  
+
+## Unconventional AIS models corroborate conventional theories  
+
+Despite the neuroinflammatory origins of scoliosis in zebrafish models \(^{9 - 11}\) , we demonstrate that pathogenic mechanisms ultimately converge on a historical focus of human AIS studies – the IVD. Trueta (1968) first reported that morphological deformations of the IVD contribute greatly to AIS spine deformity \(^{27}\) , and Harrington (1977) theorized loss of physical properties (collagen) of the IVD might thus present as a common denominator in AIS \(^{28}\) . These ideas have since been explored, in part, by longitudinal radiographic studies of AIS patients associating IVD wedging with the early presentation and progression of spinal curvature \(^{29,30}\) , and mathematical modelling of AF collagen fibre imbalance as an etiological factor in scoliosis \(^{31}\) . Although, biochemical and histological studies have also identified irregular collagen and elastic properties in IVDs isolated from AIS patients \(^{32 - 34}\) , these defects were observed only after scoliosis onset. Here we provide direct evidence that ROS- induced IVD ER stress/UPR responses, altered viscoelastic properties, and collagen ECM defects all precede spinal curvature in \(sspo^{dmh4 / + }\) zebrafish. Although it remains to be determined how oxidative stress ultimately perturbs collagen matrix development (i.e. via abnormal protein translation, trafficking, cross- linking, etc.) our data, together with published reports, strongly support a causal role for IVD ECM remodelling defects in the etiopathogenesis of scoliosis and highlight the relevance of zebrafish AIS models for pathomechanism discovery and therapeutic development.  
+
+## Spine stiffness as a prognostic biomarker for AIS  
+
+At present, genetic screening for AIS has proven ineffective \(^{35}\) . Diagnosis can only be made after scoliosis onset, and high variability in severe curve progression makes it difficult to predict which patients may ultimately require bracing or surgical intervention. Here we demonstrate the prognostic capabilities of SWE in zebrafish AIS models. Strikingly, we have
+
+<--- Page Split --->
+
+1 determined that elevated spine stiffness can be detected prior to scoliosis onset. As our data 2 complement clinical SWE studies demonstrating elevated AF stiffness in affected AIS 3 patients21,22, spine stiffness in \(sspo^{dmh4 / + }\) zebrafish might similarly be caused by observed IVD 4 collagen remodelling defects. Indeed, spine stiffness and IVD matrix defects are both oxidative 5 stress- dependent and mechanistically linked in \(sspo^{dmh4 / + }\) models, and increases in the thickness 6 and isotropy of a medium (as observed for \(sspo^{dmh4 / + }\) collagen IVD matrix) are predicted to 7 elevate medium stiffness36,37. Although the spatial and molecular origins of elevated stiffness in 8 \(sspo^{dmh4 / + }\) zebrafish remains to be determined, our data provide proof- of- concept that spine 9 stiffness might be explored as a prognostic biomarker for AIS development. Given that SWE is a 10 non- invasive imaging technique that is widely applied in the clinic, translation of these findings 11 holds tremendous potential to identify patients at high risk of severe scoliosis for earlier, non- 12 surgical intervention.  
+
+## A new framework for considering AIS origins and therapies  
+
+The oxidative stress- induced origin of ECM defects identified in zebrafish AIS models is equally deserving of clinical consideration. AIS is a common pediatric disorder'. Nevertheless, it is estimated that \(< 5\%\) total genetic variance in AIS has been determined, and genetic associations with cartilaginous ECM and IVD development account for only a subset of these cases2- 4. However, environmental variables like infection, diet, lifestyle and microbiome composition can profoundly influence ROS levels38- 40 and could therefore play a pervasive, yet largely unexplored role in AIS pathogenesis. In addition, neuroinflammation and oxidative stress can also have genetic origins (as in \(sspo^{dmh4 / + }\) mutant zebrafish) and should be explored as etiological factors driving human spinal curvature. This includes not only AIS, but other genetically defined disorders that present with both CNS oxi- inflammation and high incidences of developmental scoliosis like DiGeorge Syndrome41,42, Rett Syndrome43,44, CDKL5 Deficiency Disorder45,46, and Cerebral Cavernous Malformation47,48.  
+
+Finally, regardless of the origins of oxidative stress, prophylactic and therapeutic administration of antioxidants has proven remarkably effective in preventing the onset and severe progression of scoliosis in zebrafish AIS models9- 11. Furthermore, we have demonstrated that ER stress responses, elevated spine stiffness and collagen ECM defects in \(sspo^{dmh4 / + }\) mutants, which are all functionally linked to spinal curvature, can be fully suppressed by NACET treatment. If oxidative stress- induced ECM remodeling defects prove relevant to patient populations, then translation of these findings could have profound impact on the future management and prevention of human spinal curvature.
+
+<--- Page Split --->
+
+## 1 Methods:  
+
+2 Zebrafish 3 Zebrafish husbandry and experimental protocols were approved by the Hospital for Sick 4 Children's Animal Care Committee, and all protocols were performed in accordance with 5 Canadian Council on Animal Care guidelines. Wildtype zebrafish from TU strains were used. 6 \(sspo^{dmh4}(10)\) , \(katnb1^{mh102}(26)\) , \(vangl2^{sfGFP}\) (hsc170Tg) \(^{25}\) , \(Tg(\beta actin\colon \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \mathrm{loop - mCherry - STOP - loop - }\) zGrad) (hsc185Tg) \(^{25}\) , and \(Tg(foxj1a\colon \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \mathrm{cre})\) (hsc105Tg) \(^{25}\) mutant and transgenic fish used in this study have been previously described. Information on newly developed transgenic fish 9 \(Tg(col2a1a\colon xbp1\delta - eGFP)\) can be found below in the "Transgenesis" method section. Embryos from natural matings were grown at \(28.5^{\circ}\mathrm{C}\) . When required, experimental animals were 11 euthanized with tricaine (500 mg/L; MS- 222/MESAB), followed by submersion of anesthetized 12 fish in ice water for several minutes. As laboratory zebrafish strains do not utilize a chromosomal 13 sex determination mechanism and sex differentiation does not initiate until after \(\sim 3\) weeks post 14 fertilization \(^{49}\) , we cannot report sex for our embryonic and larval studies.  
+
+## 16 Zebrafish genotyping  
+
+16 Zebrafish genotyping 17 Genotyping protocols have been described previously and can be found in the following 18 publications: \(sspo^{dmh4 / + }(10)\) , \(katnb1^{mh102 / mh102}(26)\) , \(vangl2^{sfGFP / sfGFP}(25)\) . Primer sequences used for genotyping can be found in Extended Data Table 1.  
+
+## 21 Transgenesis  
+
+21 Transgenesis 22 Entry plasmids were generated through BP recombination into respective pDONR plasmids 23 (Invitrogen) and then cloned into standard Tol2 kit Gateway compatible vectors using LR 24 recombination methods to create the final transgenes \(^{50}\) . To generate \(Tg(col2a1a\colon xbp1\delta - eGFP)\) , a previously generated p5E- \(col2a1a^{51}\) , pME- \(xbp1\delta - eGFP\) (obtained by cloning the \(xbp1\) partial sequence using Gateway primers, detailed in table 1, from \(ef1a\colon xbp1\delta - gfp\) plasmid \(^{16}\) graciously provided by Dr. Shao Jun Du), and p3E- polyA \(^{50}\) were recombined into pDEST Tol2 HR2 28 transgenesis vector. Embryos were injected at the one cell stage with 25 pg of Tol2 transposase 29 RNA and 25 pg of the transgene (plasmid). Injected embryos were then screened at 48 hpf for 30 transgenesis marker expression. Imaging of reporter expression was performed on an Axio 31 Zoom.V16 (Zeiss). Embryos displaying strong reporter expression were grown to adulthood and 32 crossed to wildtype fish to establish stable F1 lines. F1 lines were then bred into \(sspo^{dmh4 / + }\) and 33 maintained in \(sspo^{dmh4 / + }\) fish.  
+
+## 35 DCFH-DA staining  
+
+35 DCFH-DA staining 36 Zebrafish larva were collected and placed in 12 well plates containing \(2\mathrm{mL}\) of \(5\mu \mathrm{M}\) DCFH- DA (Sigma) in E3 media. Plates were kept in the dark and incubated for 30 minutes at a temperature of \(28.5^{\circ}\mathrm{C}\) . Larva were then washed three times with E3 media for 5 minutes each wash in the dark and kept at \(28.5^{\circ}\mathrm{C}\) until imaged.
+
+<--- Page Split --->
+
+1 DCFH- DA, Tg(col2a1a::xbp1&eGFP), and TgBAC(entpd5a::pkRed) imaging  2 Experimental zebrafish larva were euthanized with tricaine (500 mg/L) before imaging. 10 dpf  3 larva were mounted dorsally or laterally in \(0.8\%\) low melt agarose (BioShop), measured for  4 standard length, and imaged on an LSM 710 confocal microscope (Zeiss).  
+
+5 Collagen hybridizing peptide  7 Juvenile zebrafish were collected at 21dpf (10mm SL) and anesthetized with tricaine (500mg/L).  8 Anesthetized fish were fixed with \(100\%\) methanol at \(- 20^{\circ}\mathrm{C}\) for 48 hours, then rehydrated with  9 1xPBS before the skin was gently removed with a pair of dissection forceps. Collagen  10 hybridizing peptide (F- CHP from 3Helix, diluted in 1xPBS to \(20\mu \mathrm{M}\) ) was then heated to \(80^{\circ}\mathrm{C}\)  11 for activation, followed by quenching on an ice block down to room temperature. Fish were then  12 immersed in CHP for 48 hours at \(4^{\circ}\mathrm{C}\) in dark. After CHP staining, fish were transferred into  13 individual tubes for soft tissue lysing in \(2\% \mathrm{KOH}\) /ethylene glycol/1xPBS solution, then switched  14 to \(1\% \mathrm{KOH}\) /ethylene glycol/1xPBS when the vertebrae were exposed. Samples were placed on a  15 nutator during tissue lysing until most of the muscle, spinal cord and ventral vasculature had  16 been digested, then cleared with glycerol before imaged on an LSM 710 confocal microscope  17 (Zeiss).  
+
+19 Transmission electron microscopy  20 Fish were collected at 10dpf (5mm SL) and anesthetized with tricaine (500mg/L). Anesthetized  21 samples were then fixed in \(2\%\) paraformaldehyde and \(2.5\%\) glutaraldehyde in \(0.1\mathrm{M}\) sodium  22 cacodylate buffer for 2 hours. Rinsed in buffer, post- fixed in \(1\%\) osmium tetroxide in buffer for  23 90 min, dehydrated in a graded ethanol series ( \(50\% ,70\% ,90\%\) and \(100\%\) , 20 minutes each step)  24 followed by two propylene oxide changes for 30 min, and embedded in Quetol- Spurr resin.  25 Blocks were cured overnight in the oven at \(60^{\circ}\mathrm{C}\) . Fish were cut along the sagittal plane \(100\mu \mathrm{m}\)  26 deep into the sample. Sections \(70\mathrm{nm}\) thick were cut on a Leica EM UC7 ultramicrotome, and  27 post- stained with \(2\%\) uranyl acetate and \(3\%\) lead citrate for 20 minutes each and washed for 5  28 minutes after each staining. Sections were air dried at room temperature before imaged with a  29 Hitachi HT7800 transmission electron microscope.  
+
+30 Synchrotron \(\mu \mathrm{CT}\) imaging, reconstruction and segmentation  31 Juvenile zebrafish were collected at 21dpf (10mm SL) and anesthetized with tricaine (500mg/L).  32 Anesthetized fish were fixed with \(4\%\) PFA at \(4^{\circ}\mathrm{C}\) overnight, then washed with 1xPBS before  33 stained in I2E solution ( \(1\%\) iodine metal in \(100\%\) ethanol) overnight at room temperature.  34 Samples were then washed again with 1xPBS to remove excess I2E stain, then mounted in \(1.5\%\)  35 agarose in R.O. water in \(0.2\mathrm{mL}\) tubes. Synchrotron \(\mu \mathrm{CT}\) were performed at the bending magnet  36 beamline of BioMedical Imaging and Therapy beamlines (BMIT- BM) at the Canadian Light  37 Source \(^{52}\) . The beamline is operated in filtered white mode, therefor, it is essential to filter the  38 beam sufficiently to prevent radiation damage to the sample and agarose gel. Filters of \(1.76\mathrm{mm}\)  39 Al and \(0.275\mathrm{mm}\) Sn were used. Detector was an indirect X- ray microscope (Optique Peter,
+
+<--- Page Split --->
+
+France) combined with a sCMOS PCO Edge 5.5 camera (PCO, Germany), a 20x objective, and a \(4\mu \mathrm{m}\) LSO scintillator to obtain an effective pixel size of \(0.360\mu \mathrm{m}\) . Generally, the dimension of scanned samples was \(400\times 600\times 750\mu \mathrm{m}^3\) . 1500 projections over 180 degrees were collected at exposure time of 1.5s. Sample was \(25.7\mathrm{m}\) from the source, and propagation distance was \(3\mathrm{cm}\) . Phase retrieval and image reconstructions were done using tofu reconstruction package \(^{53}\) .  
+
+## NACET treatments  
+
+Fish were housed off- system in \(500\mathrm{mL}\) of \(300\mathrm{mg / L}\) Instant Ocean Sea Salt treated water in 1.8 L tanks with 12 fish per tank. Water was changed once per day. N- acetyl- L- cysteine ethyl ester (NACET, BOC Sciences Cat# B0689- 029481) was prepared as a \(250\mathrm{mM}\) stock solution in 300 mg/L Instant Ocean Sea Salt dissolved in MiliQ water. NACET was administered with water changes once per day at a final concentration of \(500\mu \mathrm{M}\) . NACET treatment occurred over the course of 5 days starting at 5 dpf. Fish were fed in accordance with their regular feed schedule throughout the treatment. At 10 dpf, fish were euthanized for either ultrasound shear wave elastography imaging, microcomputed tomography, confocal microscopy, or transmission election microscopy.  
+
+## Shear wave elastography (SWE)  
+
+The spinal tissue stiffness of the zebrafish was assessed using a SWE protocol (as described in Figure S1) programmed into a research ultrafast ultrasound system (Vantage 256, Verasonics Inc., Kirkland, WA, USA) with a linear ultrasound probe (L22- 14vX, 18MHz center frequency). Euthanized zebrafish were mounted ventrally in a \(35\mathrm{mm}\times 10\mathrm{mm}\) dish (Corning) filled with \(1.5\%\) low melt agarose (BioShop). The probe was positioned longitudinally, dorsal to the zebrafish, in contact with ultrasound gel applied to agarose- embedded samples. The SWE protocol for a single acquisition consisted of two steps: 1) the induction of an acoustic radiation force through a focused ultrasound beam at a push depth between \(4 - 9\mathrm{mm}\) aimed at the center of the zebrafish spine, resulting in a transient perturbation of the tissue and shear waves propagation; 2) an ultrafast Doppler sequence was created with coherent compound plane wave imaging consisting of 3 plane waves (range - 3 degrees to +3 degrees, 3- degree steps) with a pulse repetition frequency of \(28\mathrm{kHz}\) . Coherent compounding of the plane waves was applied using a sliding window method as in Kang et al. (window size 3, corresponding to the 3 different titled plane waves, and window step of 1), enabling a virtual framerate of \(28\mathrm{kHz}^{54}\) . Each acquisition lasted 10 milliseconds. Three acquisitions were collected per zebrafish. After scans were complete, the ultrafast imaging data was streamed to an internal network and then post- processed offline using MATLAB 2019a (The MathWorks Inc., Natick, MA, USA). Tissue velocity data was computed using a Doppler- based autocorrelation estimator, from which tissue velocity maps were reconstructed and presented as a space- time matrix showing shear waves propagation within a specific region of interest. The mean of these shear wave velocities was computed and used as surrogate for spine stiffness (derived from the shear modulus) \(^{55}\) .
+
+<--- Page Split --->
+
+MicroCT scanning, reconstruction, and density analysisZebrafish were euthanized at 4mpf and fixed in \(10\%\) neutral- buffered formalin (Sigma- Aldrich). Fish were mounted in tubes using \(1\%\) agarose. Scanning was performed with a SkyScan 1275 microCT (Bruker, Kontich, Belgium) using \(50\mathrm{kV}\) and \(80\mu \mathrm{A}\) , sample rotation of \(180^{\circ}\) , image rotation steps of \(0.2^{\circ}\) , frame averaging of 10, exposure time of \(55\mathrm{ms}\) , camera binning of \(1\times 1\) , and using a pixel size of \(18\mathrm{um}\) . Projection images were reconstructed into cross- sections using SkyScan’s NRecon v.1.7.4.6 software (Bruker, Kontich, Belgium) in a range of attenuation coefficients 0–0.25, with a beam- hardening correction of \(40\%\) . The reconstructed images were stored as 16- bit TIFF images. Maximum Intensity Projections were generated (imageJ software) for tissue mineral density analysis. The line selection tool in ImageJ was then used to manually measure the mean gray value within each vertebral centrum, which can be used as a relative measure for tissue mineral density.  
+
+## Cobb angle statistical analysis  
+
+Lines were drawn parallel to the top and bottom most displaced vertebrae for each curve. The Cobb angle was then measured as the angle of intersection between lines drawn perpendicular to the original 2 lines<sup>9</sup>. Analysis was conducted using ImageJ<sup>56,57</sup>. Cobb angle measurements for lateral and dorsal curvatures were summed to obtain a combined Cobb angle measurement for each fish. Results were graphed and statistical significance was calculated using GraphPad Prism 10.1.1 (GraphPad Software).  
+
+Quantification of DCFH- DA staining, \(Tg(col2a1a::xbp1\delta - eGFP)\) and \(TgBAC(entpd5a::pkRed)\) Using ImageJ FIJI, regions of interest (ROI) were selected to measure the area and integrated density of the ROI. The mean grey value of the background was also measured. The corrected total fluorescence (CTF) of the ROI were calculated as followed: Integrated Density - (Area of ROI x Mean Grey Value of Background). CTF values were then calculated relative to wildtype controls by dividing CTF values to the average CTF wildtype value. Statistical analysis was performed using two- way ANOVA with Tukey’s multiple comparisons test or unpaired twotailed student’s \(t\) - test in GraphPad Prism version 10.1.1 (GraphPad Software).  
+
+## Notochord ECM measurements  
+
+TEM scans at \(6000\mathrm{x}\) and \(10000\mathrm{x}\) were selected for measurements. The ECM widths were taken with ImageJ and defined as the distance from the basal boundary of the notochord ECM to the basal side of the notochord sheath (yellow dash line, Fig. 4B- E). Two measurements were taken from a single scanned image: a local maximum width (the thickest area along the ECM) and a local minimum (thinnest area along the ECM). Local ratio was obtained from dividing the maximum width by its paired minimum width.  
+
+Data availability: Due to their large size, microCT imaging files are available upon request. All other data are available in the main text or the supplementary materials.
+
+<--- Page Split --->
+
+## Main Text References  
+
+1. Cheng, J. C. et al. Adolescent idiopathic scoliosis. Nature Reviews Disease Primers 1, 1-21 (2015).  
+
+2. Wise, C. A. et al. The cartilage matrisome in adolescent idiopathic scoliosis. Bone Res 8, 1-13 (2020).  
+
+3. Haller, G. et al. A polygenic burden of rare variants across extracellular matrix genes among individuals with adolescent idiopathic scoliosis. Human Molecular Genetics 25, 202-209 (2016).  
+
+4. Karner, C. M., Long, F., Solnica-Krezel, L., Monk, K. R. & Gray, R. S. Gpr126/Adgrg6 deletion in cartilage models idiopathic scoliosis and pectus excavatum in mice. Human Molecular Genetics 24, 4365-4373 (2015).  
+
+5. Kou, I. et al. Genome-wide association study identifies 14 previously unreported susceptibility loci for adolescent idiopathic scoliosis in Japanese. Nat Commun 10, 3685 (2019).  
+
+6. Bagnat, M. & Gray, R. S. Development of a straight vertebrate body axis. Development 147, (2020).  
+
+7. Boswell, C. W. & Ciruna, B. Understanding Idiopathic Scoliosis: A New Zebrafish School of Thought. Trends in Genetics 33, 183-196 (2017).  
+
+8. Grimes, D. T. et al. Zebrafish models of idiopathic scoliosis link cerebrospinal fluid flow defects to spine curvature. Science 352, 1341-1344 (2016).  
+
+9. Van Gennip, J. L. M., Boswell, C. W. & Ciruna, B. Neuroinflammatory signals drive spinal curve formation in zebrafish models of idiopathic scoliosis. Science Advances 4, eaav1781 (2018).  
+
+10. Rose, C. D. et al. SCO-Spondin Defects and Neuroinflammation Are Conserved Mechanisms Driving Spinal Deformity across Genetic Models of Idiopathic Scoliosis. Current Biology 30, 2363-2373.e6 (2020).  
+
+11. Djebar, M. et al. Astrogliosis And Neuroinflammation Underlie Scoliosis Upon Cilia Dysfunction. 2024.02.22.581530 Preprint at https://doi.org/10.1101/2024.02.22.581530 (2024).  
+
+12. Henke, K. et al. Genetic Screen for Postembryonic Development in the Zebrafish (Danio rerio): Dominant Mutations Affecting Adult Form. Genetics 207, 609-623 (2017).  
+
+13. Soh, N. Recent advances in fluorescent probes for the detection of reactive oxygen species. Anal Bioanal Chem 386, 532-543 (2006).  
+
+14. Wopat, S. et al. Spine Patterning Is Guided by Segmentation of the Notochord Sheath. Cell Reports 22, 2026-2038 (2018).  
+
+15. Zhang, Z. et al. Redox signaling and unfolded protein response coordinate cell fate decisions under ER stress. Redox Biology 25, 101047 (2019).  
+
+16. Li, J. et al. A transgenic zebrafish model for monitoring xbp1 splicing and endoplasmic reticulum stress in vivo. Mechanisms of Development 137, 33-44 (2015).
+
+<--- Page Split --->
+
+17. Dale, R. M. & Topczewski, J. Identification of an evolutionarily conserved regulatory element of the zebrafish col2a1a gene. Developmental Biology 357, 518–531 (2011).  
+18. Kague, E. et al. 3D assessment of intervertebral disc degeneration in zebrafish identifies changes in bone density that prime disc disease. Bone Res 9, 39 (2021).  
+19. Gistelinck, C. et al. Zebrafish type I collagen mutants faithfully recapitulate human type I collagenopathies. Proceedings of the National Academy of Sciences 115, E8037–E8046 (2018).  
+20. Zitnay, J. L. et al. Molecular level detection and localization of mechanical damage in collagen enabled by collagen hybridizing peptides. Nat Commun 8, 14913 (2017).  
+21. Langlais, T. et al. Shear-wave elastography can evaluate annulus fibrosis alteration in adolescent scoliosis. Eur Radiol 28, 2830–2837 (2018).  
+22. Vergari, C. et al. Shear wave elastography of lumbar annulus fibrosis in adolescent idiopathic scoliosis before and after surgical intervention. European Radiology 30, 1980–1985 (2020).  
+23. Villemain, O. et al. Ultrafast Ultrasound Imaging in Pediatric and Adult Cardiology: Techniques, Applications, and Perspectives. JACC: Cardiovascular Imaging 13, 1771–1791 (2020).  
+24. Villalobos Lizardi, J. C. et al. A guide for assessment of myocardial stiffness in health and disease. Nat Cardiovasc Res 1, 8–22 (2022).  
+25. Jussila, M., Boswell, C. W., Griffiths, N. W., Pumputis, P. G. & Ciruna, B. Live imaging and conditional disruption of native PCP activity using endogenously tagged zebrafish sfGFP-Vangl2. Nat Commun 13, 5598 (2022).  
+26. Meyer-Miner, A., Van Gennip, J. L. M., Henke, K., Harris, M. P. & Ciruna, B. Resolving primary pathomechanisms driving idiopathic-like spinal curvature using a new katnb1 scoliosis model. iScience 25, 105028 (2022).  
+27. Trueta, J. in chapter 35 The Aetio-Pathology and Management of Scoliosis. in Studies of the Development and Decay of the Human Frame 306–315 (William Heinemann Medical Books LTD., 1968).  
+28. Harrington, P. R. The etiology of idiopathic scoliosis. Clin Orthop Relat Res 17–25 (1977).  
+29. de Reuver, S. et al. Anterior lengthening in scoliosis occurs only in the disc and is similar in different types of scoliosis. Spine J 20, 1653–1658 (2020).  
+30. Will, R. E., Stokes, I. A., Qiu, X., Walker, M. R. & Sanders, J. O. Cobb angle progression in adolescent scoliosis begins at the intervertebral disc. Spine (Phila Pa 1976) 34, 2782–2786 (2009).  
+31. Heidari, B., FitzPatrick, D., Synnott, K. & McCormack, D. Modelling of annulus fibrosis imbalance as an aetiological factor in adolescent idiopathic scoliosis. Clin Biomech (Bristol, Avon) 19, 217–224 (2004).  
+32. Duance, V. C. et al. Changes in collagen cross-linking in degenerative disc disease and scoliosis. Spine (Phila Pa 1976) 23, 2545–2551 (1998).
+
+<--- Page Split --->
+
+1 33. Taylor, T. K., Ghosh, P. & Bushell, G. R. The contribution of the intervertebral disk to the scoliotic deformity. Clin Orthop Relat Res 79- 90 (1981). 2 34. Yu, J., Fairbank, J. C. T., Roberts, S. & Urban, J. P. G. The elastic fiber network of the anulus fibrosus of the normal and scoliotic human intervertebral disc. Spine (Phila Pa 1976) 5 30, 1815- 1820 (2005). 6 35. Cheng, T., Einarsdottir, E., Kere, J. & Gerdhem, P. Idiopathic scoliosis: a systematic review and meta-analysis of heritability. EFORT Open Rev 7, 414- 421 (2022). 7 36. Savage, M. K. Seismic anisotropy and mantle deformation: What have we learned from shear wave splitting? Reviews of Geophysics 37, 65- 106 (1999). 8 37. Villemain, O. et al. Myocardial Stiffness Evaluation Using Noninvasive Shear Wave Imaging in Healthy and Hypertrophic Cardiomyopathic Adults. JACC Cardiovasc Imaging 12, 1135- 1145 (2019). 9 38. Bostick, J. W., Schonhoff, A. M. & Mazmanian, S. K. Gut microbiome-mediated regulation of neuroinflammation. Curr Opin Immunol 76, 102177 (2022). 10 39. Foo, J., Bellot, G., Pervaiz, S. & Alonso, S. Mitochondria- mediated oxidative stress during viral infection. Trends in Microbiology 30, 679- 692 (2022). 11 40. Peters, A., Nawrot, T. S. & Baccarelli, A. A. Hallmarks of environmental insults. Cell 184, 1455- 1468 (2021). 12 41. Homans, J. F. et al. Scoliosis in association with the 22q11.2 deletion syndrome: an observational study. Arch Dis Child 104, 19- 24 (2019). 13 42. Menghi, M. et al. Neuroinflammation and Oxidative Stress in Individuals Affected by DiGeorge Syndrome. Int J Mol Sci 24, 4242 (2023). 14 43. Huang, T. J., Lubicky, J. P. & Hammerberg, K. W. Scoliosis in Rett syndrome. Orthop Rev 23, 931- 937 (1994). 15 44. Pecorelli, A., Cervellati, C., Hayek, J. & Valacchi, G. OxInflammation in Rett syndrome. Int J Biochem Cell Biol 81, 246- 253 (2016). 16 45. Jagtap, S. et al. Aberrant mitochondrial function in patient- derived neural cells from CDKL5 deficiency disorder and Rett syndrome. Hum Mol Genet 28, 3625- 3636 (2019). 17 46. Mangatt, M. et al. Prevalence and onset of comorbidities in the CDKL5 disorder differ from Rett syndrome. Orphanet J Rare Dis 11, 39 (2016). 18 47. Retta, S. F. & Glading, A. J. Oxidative stress and inflammation in cerebral cavernous malformation disease pathogenesis: Two sides of the same coin. Int J Biochem Cell Biol 81, 254- 270 (2016). 19 48. Shenkar, R. et al. EXCEPTIONAL AGGRESSIVENESS OF CEREBRAL CAVERNOUS MALFORMATION DISEASE ASSOCIATED WITH PDCD10 MUTATIONS. Genet Med 17, 188- 196 (2015). 20 49. Kossack, M. E. & Draper, B. W. Genetic regulation of sex determination and maintenance in zebrafish (Danio rerio). Curr Top Dev Biol 134, 119- 149 (2019). 21 50. Kwan, K. M. et al. The Tol2kit: A multisite gateway- based construction kit for Tol2 transposon transgenesis constructs. Developmental Dynamics 236, 3088- 3099 (2007).
+
+<--- Page Split --->
+
+51. Rebello, D. et al. COL11A2 as a candidate gene for vertebral malformations and congenital scoliosis. Hum Mol Genet 32, 2913-2928 (2023).  
+52. Gasilov, S. et al. Hard X-ray imaging and tomography at the Biomedical Imaging and Therapy beamlines of Canadian Light Source. J Synchrotron Rad 31, (2024).  
+53. Faragó, T. et al. Tofu: a fast, versatile and user-friendly image processing toolkit for computed tomography. J Synchrotron Rad 29, 916-927 (2022).  
+54. Kang, J., Jang, W. S. & Yoo, Y. High PRF ultrafast sliding compound doppler imaging: fully qualitative and quantitative analysis of blood flow. Phys Med Biol 63, 045004 (2018).  
+55. Taljanovic, M. S. et al. Shear-Wave Elastography: Basic Physics and Musculoskeletal Applications. Radiographics 37, 855-870 (2017).  
+56. Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9, 671-675 (2012).  
+57. Schindelin, J., Rueden, C. T., Hiner, M. C. & Eliceiri, K. W. The ImageJ ecosystem: An open platform for biomedical image analysis. Mol Reprod Dev 82, 518-529 (2015).  
+
+58. Hu, W. F. et al. Katanin p80 Regulates Human Cortical Development by Limiting Centriole and Cilia Number. Neuron 84, 1240-1257 (2014).  
+59. Hur, M. et al. MicroCT-based phenomics in the zebrafish skeleton reveals virtues of deep phenotyping in a distributed organ system. eLife 6, e26014 (2017).  
+
+Acknowledgments: We gratefully acknowledge the SickKids' Nanoscale Biomedical Imaging Facility for assistance with TEM sample preparation and imaging, the SickKids' Zebrafish Facility technicians for excellent zebrafish care, Nigel Griffiths for technical assistance, and Dr. Ronald Kwon support in microCT TMD analyses. Part of the research described in this paper was performed at the Canadian Light Source, a national research facility of the University of Saskatchewan, which is supported by the Canada Foundation for Innovation (CFI), the Natural Sciences and Engineering Research Council (NSERC), the National Research Council (NRC), the Canadian Institutes of Health Research (CIHR), the Government of Saskatchewan, and the University of Saskatchewan.  
+
+The following grants supported this work: Canadian Institutes of Health Research Foundation grant FDN- 167285 (B.C.) Canada Research Chair Program (B.C.) Canadian Institute of Health and Research Canada Graduate Scholarship FBD- 181495 (P.G.P.) Natural Sciences and Engineering Research Council of Canada (NSERC) grant RGPIN- 2021- 03539 (O.V.) Canadian Institutes of Health Research operating grant 148683 (B.F.E.) The Hospital for Sick Children Research Institute funding (B.C., O.V., D.E.L.)
+
+<--- Page Split --->
+
+1 Authors contributions: 2 Conceptualization: B.C. and O.V. conceptualized, administered, and supervised the project. 3 P.G.P., R.X., J.G., A.P., V.E., J.V.G. performed experiments, data analysis, and visualized the 4 data. P.G.P., R.X., J.G., A.P., V.E., J.V.G., B.F.E., N.F., J.B., O.V., B.C. contributed to the 5 methodology of the project. N.F., J.B. created the software for zebrafish SWE analysis. B.C., 6 O.V., P.G.P., B.F.E., D.E.L. acquired funding. P.G.P., R.X., J.G., O.V., B.C. wrote the original 7 draft of the paper. P.G.P., R.X., J.G., A.P., V.E., J.V.G., B.F.E., N.F., J.B., D.E.L., O.V., B.C. 8 reviewed and edited the paper.  
+
+Competing Interests: Authors declare that they have no competing interests  
+
+Materials and correspondence:  
+
+Correspond to Brian Ciruna (ciruna@sickkids.ca).
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+
+<--- Page Split --->
+
+Fig. 1: IVD redox imbalance and ER stress precede idiopathic-like scoliosis. a, b, Lateral and dorsal wholemount images of 10 dpf (6 mm SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva. Red boxes indicate regions imaged by confocal microscopy. Scale bars \(= 1 \mathrm{mm}\) . c-f, Confocal images of DCFH-DA oxidation staining of the dorsal telecephalon (c, d) and lateral trunk (e, f) of 10 dpf (6 mm SL) wildtype (c, e) and \(sspo^{dmh4 / + }\) mutant (d, f) zebrafish larva. Scale bars \(= 100 \mu \mathrm{m}\) . Lookup tables arbitrary units (A.U.) are 0 to 145. \(\mathrm{DT} =\) dorsal telecephalon; IVD = intervertebral domain; DA = dorsal aorta; CC = central canal; SC = spinal cord. Arrowheads indicate increased oxidative state along developing intervertebral domains. g-k, Relative fluorescence quantification of DCFH-DA oxidation staining in dorsal telecephalon (g; \(\mathrm{p} = 0.0160\) , \(\mathrm{N} = 3\) , \(\mathrm{n} = 5\) for each genotype), intervertebral domains (h; \(\mathrm{p} = 0.0042\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), dorsal aorta (i; \(\mathrm{p} = 0.0098\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), central canal (j; \(\mathrm{p} = 0.0035\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), and spinal cord (k; \(\mathrm{p} = 0.0032\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), all showing significant increased oxidative state in \(sspo^{dmh4 / + }\) mutant zebrafish larva. Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test. l-n, Confocal images of \(Tg(entpd5a::pkRed)\) reporter expression within mineralizing notochord segments and DCFH-DA staining, demonstrating greater oxidation in and around the intervertebral domains (IVD, arrowhead) in 10 dpf (6 mm SL) \(sspo^{dmh4 / + }\) larva. Scale bar \(= 50 \mu \mathrm{m}\) . o-p, Confocal images of \(Tg(col2al1::xbp1\delta -gfp)\) and \(Tg(entpd5a::pkRed)\) reporter expression in 10 dpf (6 mm SL) wildtype (o) and \(sspo^{dmh4 / + }\) (p) zebrafish larva, demonstrating an elevated ER stress response within the intervertebral domains of \(sspo^{dmh4 / + }\) mutants, prior to the onset of scoliosis. Scale bar \(= 100 \mu \mathrm{m}\) . q, Relative fluorescence quantification of \(Tg(col2al1::xbp1\delta -gfp)\) reporter expression within intervertebral domains ( \(\mathrm{p} = 0.0127\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 11\) for each genotype), demonstrating elevated ER stress in \(sspo^{dmh4 / + }\) mutants ( \(\mathrm{p} \leq 0.05\) ). Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test ( \(\ast \mathrm{p} \leq 0.05\) , \(\ast \ast \mathrm{p} \leq 0.01\) ).
+
+<--- Page Split --->
+
+<center>Figure 2</center> 
+
+![](images/Extended_Data_Figure_3.jpg)
+
+
+<--- Page Split --->
+
+## Fig. 2: Scoliotic fish exhibit intervertebral ligament deformations.  
+
+a-g, High-resolution synchrotron-based X-ray micro-computed tomography (micro-CT) imaging of iodine contrast- stained wildtype (a-d) and \(sspo^{dmh4 / + }\) mutant (e-g) juvenile fish at 21 dpf (10 mm SL). Representative sections from coronal (a-c, e-f) and transverse (d, g) planes are shown. Dashed boxes (a, e) indicate featured intervertebral segments. Arrowheads (b-d, f-h) indicate the intervertebral ligament, which loops outward toward neighbouring musculature in wildtype animals (N = 3 fish, n = 13/13 IVDs), but often exhibited structural deformations and even inverted orientations in \(sspo^{dmh4 / + }\) mutants (N = 8 fish; n = 22/29 IVDs). Arrows (h) highlight sites of \(sspo^{dmh4 / + }\) IVL deformation in the transverse plane. Sequential images (3.6 μm step) through the coronal plane of wildtype (c-c'') and \(sspo^{dmh4 / + }\) (g-g'') zebrafish highlight the highly deformable structure of \(sspo^{dmh4 / + }\) mutant IVLs. Scale bars = 20 μm (c, g); 50 μm (b, d, f, h); 100 μm (a, e). IVL = intervertebral ligament; AF = annulus fibrosus; NP = nucleus pulposus.
+
+<--- Page Split --->
+![](images/Extended_Data_Figure_5.jpg)
+
+
+<--- Page Split --->
+
+1 Fig. 3: Elevated spine stiffness is observed prior to scoliosis onset, positively correlates with curve severity, and is fully suppressed by antioxidant therapy.  
+
+2 a, b, Lateral wholemount images of wildtype and scoliotic \(sspo^{dmh4 / + }\) mutant zebrafish at 16 dpf (a, SL = 7.1 +/- 0.2 mm) and 21 dpf (b, SL = 8.9 +/- 0.6 mm). Scale bar = 2 mm. c, d, Quantification of shear wave velocity measured by SWE along the spine of 16 dpf (c) and 21 dpf (d) zebrafish, demonstrating significantly greater shear wave velocity and thus spinal stiffness in \(sspo^{dmh4 / + }\) mutant animals. c, p = 0.0040, N = 3, n = 15 for each genotype. d, p = 0.0027, N = 3, n = 15 for each genotype. Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test. e, Positive correlation between scoliosis severity (Total Cobb angle) and spinal stiffness (shear wave velocity) in 21 dpf \(sspo^{dmh4 / + }\) zebrafish, post-curve onset (N = 3, n = 20 for each genotype). Data was analyzed by a simple linear regression (R = 0.831, R² = 0.690, p = 5.73 x 10⁻⁶). f, Quantification of shear wave velocity measured by SWE along the spine of 10 dpf (SL = 4.7 +/- 0.4 mm) demonstrating significant elevation of spine stiffness in \(sspo^{dmh4 / + }\) mutant zebrafish, prior to scoliosis (p < 0.0001, N = 3, n = 15 for each genotype and treatment). Antioxidant treatment (500μM NACET) fully suppressed spine stiffness in \(sspo^{dmh4 / + }\) mutant zebrafish to wildtype levels. Box plots show all data points, median, and min to max whiskers. Data were analyzed by two-way ANOVA and Tukey test (ns: non-significant (p > 0.05), **p ≤ 0.01, ****p ≤ 0.0001).
+
+<--- Page Split --->
+![](images/Extended_Data_Figure_6.jpg)
+
+
+<--- Page Split --->
+
+Fig. 4: Oxidative stress-induced ECM matrix defects precede scoliosis onset. a, Schematic of zebrafish notochord development at 10 dpf (5mm SL) showing cartilaginous ECM domains (IVD), mineralized domains (vertebra), notochord vacuole and notochord sheath cells. b, c, Representative transmission electron micrographs of the collagenous ECM in developing IVDs of 10 dpf (5mm SL) wildtype (b) and \(sspo^{dmh4 / + }\) (c) larvae. d, e, Antioxidant treatment (500μM NACET) had no effect on wildtype ECM development (d) but suppressed irregular collagen ECM thickness and fiber orientation phenotypes in \(sspo^{dmh4 / + }\) mutant animals (e). f-h, Quantification of matrix thickness showed no difference in local minimum ECM widths (g), but significant differences between wildtype (N = 8, n = 36) and \(sspo^{dmh4 / + }\) mutants (N = 9, n = 47) in maximum local thickness (f) and local thickness ratios (h). The maximum matrix thickness and ratio of \(sspo^{dmh4 / + }\) mutant IVDs (N = 7, n = 35) were restored to wildtype level upon NACET treatment, while NACET treatment had no effect on wildtype larval IVDs (N = 6, n = 22). Box plots show all data points, median, and min to max whiskers. Points of the same colour belong to the same fish, demonstrating variations in phenotype severity among IVD segments of individual \(sspo^{dmh4 / + }\) fish. Data were analyzed by two-way ANOVA and Tukey test (ns: non-significant p > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).
+
+<--- Page Split --->
+
+1
+
+2
+
+3
+
+4
+
+5
+
+6
+
+7
+
+**8** **Extended Data Table 1: Oligonucleotide sequences used for genotyping and cloning.**
+
+<table><tr><td>Oligonucleotides</td><td>Source</td></tr><tr><td>Forward HRM primer for sspo Exon 17:<br>5'-GGAGCTGTGAGCCGGAC-3'</td><td>10</td></tr><tr><td>Reverse HRM primer for sspo Exon<br>17:5'-TCCTGCGCCAGTCCATCA3'</td><td>10</td></tr><tr><td>Forward sequencing primer for katnb1<sup>mh102</sup>:<br>5'-ACACAGACTTCATGTTTCTGACAGGC-3'</td><td>58</td></tr><tr><td>Reverse sequencing primer for katnb1<sup>mh102</sup>:<br>5'-TGAGCTCAGACACAACTGAGGGTT-3'</td><td>58</td></tr><tr><td>(attb1) Myc-xbp1-eGFP Forward for middle entry<br>5'-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCGCCACCATG<br>GAGCAAAAGCTCATTTCTG-3'</td><td>This<br>paper</td></tr><tr><td>(attb2) Myc-xbp1-eGFP-taa Reverse for middle entry<br>5'-GGGGACCACTTTGTACAAGAAAGCTGGGTGTTACTTGTACAGCTCGTCACTGC-3'</td><td>This<br>paper</td></tr></table>
+
+9
+
+10
+
+11
+
+12
+
+13
+
+14
+
+15
+
+16
+
+17
+
+18
+
+19
+
+20
+
+21
+
+22
+
+23
+
+24
+
+<--- Page Split --->
+![](images/Extended_Data_Figure_7.jpg)
+  
+
+Extended Data Fig. 1: No change in oxidative state within trunk myotome of \(sspo^{dmh4 / + }\) mutant zebrafish.  
+
+a, b, Confocal images of DCFH- DA oxidation staining of lateral trunk myotomes of \(10\mathrm{dpf}(6\mathrm{mm}\) SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva. Scale bars \(= 100\mu \mathrm{m}\) . Lookup tables arbitrary units (A.U.) are 0 to 145. c, Relative fluorescence quantification of DCFH- DA oxidation staining shows no significant difference \((\mathrm{p} > 0.05)\) between wildtype and \(sspo^{dmh4 / + }\) larva (10 dpf, \(6\mathrm{mmSL}\) ). \(\mathrm{N} = 4\) \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 9\) \(sspo^{dmh4 / + }\) . ns: non- significant \((\mathrm{p} > 0.05)\) . Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test.
+
+<--- Page Split --->
+![PLACEHOLDER_29_0]
+  
+
+![PLACEHOLDER_29_1]
+  
+
+Extended Data Fig. 2: Transient ER stress response within the IVDs of \(sspo^{dmh4 / +}\) mutant larvae.  
+
+a, Schematic of the ER stress reporter \(Tg(col2a1a::xbp1\delta - gfp)\) . Under normal conditions, splicing does not occur within the \(xbp1\delta - gfp\) transcript, eGFP coding sequence is out of frame, and translation is prematurely terminated. During ER stress, IRE1 splices the \(xbp1\delta - gfp\) transcript leading to a frameshift and translation of a spliced and functional eGFP fusion protein. Created
+
+<--- Page Split --->
+
+1 with BioRender.com. b- e, Imaging and (f) quantification of \(Tg(col2a1l::xbp1\delta - gfp)\) reporter 2 expression within the IVDs of 10 dpf (5 mm SL) zebrafish larva. \(sspo^{dmh4 / + }\) mutants (c) exhibit a 3 significantly increased ER stress response over wildtype (b) animals. Treatment with antioxidant 4 (500 \(\mu \mathrm{M}\) NACET) from 5 to 10 dpf results in a significant reduction in ER stress reporter 5 expression \((\mathrm{p} = 0.0440)\) within the IVD of \(sspo^{dmh4 / + }\) mutant zebrafish (e) larva, similar to wildtype 6 levels (d). Scale bar \(= 100 \mu \mathrm{m}\) . Box plots show all data points, median, and min to max whiskers. 7 \(\mathrm{N} = 4\) , \(\mathrm{n} = 10\) WT, \(\mathrm{n} = 12\) \(sspo^{dmh4 / + }\) , \(\mathrm{n} = 11\) WT+NACET, \(\mathrm{n} = 11\) \(sspo^{dmh4 / + }\) +NACET. Data were 8 analyzed by two- way ANOVA and Tukey test. ns: non- significant, \(^{*}\mathrm{p}\leq 0.05\) , \(^{**}p\leq 0.01\) . g, h, 9 Imaging and (i) quantification of \(Tg(col2a1l::xbp1\delta - gfp)\) reporter expression in 14 dpf (6.5 mm 10 SL) wildtype (g) and \(sspo^{dmh4 / + }\) mutant (h) zebrafish larva. In contrast to earlier stages of 11 development, no significant difference in ER stress reporter expression is observed in the 12 intervertebral domains of \(sspo^{dmh4 / + }\) mutants, after the onset of scoliosis. Scale bar \(= 100 \mu \mathrm{m}\) . \(\mathrm{N} =\) 13 \(3\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 11\) \(sspo^{dmh4 / + }\) . Box plots show all data points, median, and min to max whiskers. 14 Data were analyzed by unpaired Student's \(t\) test. 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
+
+<--- Page Split --->
+![PLACEHOLDER_31_0]
+
+<center>a Wildtype (10 mm) </center>  
+
+![PLACEHOLDER_31_1]
+
+<center>Extended Data Fig. 3: Spinal curvature in 21dpf \(sspo^{dmh4 / + }\) mutant fish. </center>  
+
+a, b, Lateral wholemount images of 21 dpf (10 mm SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva, indicating regions selected for high-resolution synchrotron- based X- ray micro- computed tomography (micro- CT) imaging (red box). Scale bars = 1 mm.
+
+<--- Page Split --->
+![PLACEHOLDER_32_0]
+  
+
+## Extended Data Fig. 4: \(sspo^{dmh4 / +}\) IVDs exhibit collagenous ECM damage and matrix remodeling.  
+
+a, b, Schematic representing the structure and organization of zebrafish IVDs at 21 dpf (10mm SL) in the sagittal (a) and transverse (b) planes. VB = vertebra body, NV=notochord vacuole, NS = notochord sheath cell, CE = collagen type II and elastin matrix, DC = dense collagen type I matrix, CF =collagen type I bundle fibre, NP = nucleus pulposus, IVL = intervertebral ligament, AF = annulus fibrosus. c-f, Confocal images of CHP- stained vertebral and IVD segments at 21 dpf (10mm SL). c, d, CHP staining in wildtype animals demonstrates normal remodelling of the collagen ECM matrix within the NP (arrowheads) and AF (arrows), as seen in sagittal (c, \(\mathrm{n} = 12\) ; scale bar \(= 20\mu \mathrm{m}\) ) and transverse (d, \(\mathrm{n} = 13\) ; scale bar \(= 10\mu \mathrm{m}\) ) planes. e, f, In contrast, \(sspo^{dmh4 / + }\) mutant animals exhibit ectopic CHP staining within NP (arrowheads) and AF (arrows) structures, as seen in sagittal (e, \(\mathrm{n} = 12\) , scale bar \(= 20\mu \mathrm{m}\) ) and transverse (f, \(\mathrm{n} = 15\) , scale bar \(= 10\mu \mathrm{m}\) ) planes, indicating significant damage to the collagenous ECM matrix of \(sspo^{dmh4 / + }\) mutant IVDs. g, Schematic depicting the mechanism by which collagen hybridizing peptide (CHP, green probe) integrates into open regions (yellow stars) of collagen fibril triple helices (purple) to report on collagen ECM damage and remodeling.
+
+<--- Page Split --->
+![PLACEHOLDER_33_0]
+
+<center>Extended Data Fig. 5: Shear wave elastography (SWE) of the zebrafish spine. a, Photograph of SWE experimental apparatus setup. b, Schematic of SWE mechanics in zebrafish, imaging with ultrafast ultrasound at 28000 frames per second (fps). High frequency linear ultrasound probe induces an acoustic radiation force (ARF) into the center of the zebrafish spine, </center>
+
+<--- Page Split --->
+
+1 and two resulting shear wave velocities (rostral direction and caudal direction) are measured in a 2 10 millisecond acquisition. c, Data processing procedure for every acquisition using a custom 3 MATLAB program. 1. Zebrafish region of interest identified. 2. Shear wave propagation 4 visualized over every video frame and averaged along identified axis of interest along zebrafish 5 spine. 3. Space- time map retrieved for axis of interest and slopes of wave transforms estimated to 6 determine SWV. d, Quantification of shear wave velocity measured by SWE for 11 dpf wildtype 7 zebrafish testing different AFR push locations, demonstrating that they do not significantly alter 8 measured SWV (p > 0.05, n = 10 location 1, n = 12 location 2, n = 8 location 3). e, Quantification 9 of shear wave velocity measured by SWE for 11 dpf wildtype zebrafish testing low melt agarose 10 concentrations for mounting zebrafish, demonstrating that low melt agarose concentration does 11 not impact SWV measurements. (p > 0.05, n = 8 for both low melt agarose concentrations). Box 12 plots show all data points, median, and min to max whiskers. Data were analyzed by either one- 13 way ANOVA and Tukey post- hoc test (d) or unpaired Student's t test (e).
+
+<--- Page Split --->
+![PLACEHOLDER_35_0]
+
+<center>Extended Data Fig. 6: No significant difference in tissue mineral density is observed in vertebral centra of \(sspo^{dmh4 / +}\) and wildtype sibling zebrafish. </center>  
+
+a, b, MicroCT scans of 4 mpf fish were acquired (A, \(\mathrm{n} = 12\) wildtype; B, \(\mathrm{n} = 16\) \(sspo^{dmh4 / +}\) ). c, Quantification of relative centrum density was manually performed on two- dimensional maximum intensity projections of reconstructed microCT scans. Multiple unpaired t test with Welch correction was performed for each vertebra. There was no significant difference in centrum density at any vertebral position (adjusted p- values \(>0.05\) ). Error bars indicate standard deviation. d, This manual method for quantification of relative density was validated by comparing to relative density values obtained from a previously established software, FishCuT59. Manually measured mean gray values of vertebral centra are closely correlated to values obtained using FishCuT ( \(\mathrm{R} = 0.868\) , \(\mathrm{R}^2 = 0.754\) , \(\mathrm{p} = 2.23 \times 10^{- 18}\) ).
+
+<--- Page Split --->
+![PLACEHOLDER_36_0]
+
+<center>Extended Data Fig. 7: Increased spine stiffness is observed across diverse zebrafish AIS models. </center>  
+
+a, Lateral wholemount images and quantification of shear wave velocity measured by SWE along the spine of \(20\mathrm{dpf}(8.18 + / - 0.46\mathrm{mm}\mathrm{SL})\) scoliotic katnb1mh102 mutant zebrafish, and katnb1mh102/sibling controls. katnb1mh102 mutant zebrafish have disrupted katanin microtubule severing activity that results in ependymal cell and choroid plexus cilia differentiation defects, CSF flow defects and oxidative stress responses, with little disruption to Reissner's fiber polymerization26. Scoliotic katnb1mh102 mutant zebrafish also demonstrate significantly elevated shear wave velocity and thus spine stiffness, compared to their control katnb1mh102/siblings \((\mathrm{p} = 0.0047\) \(\mathrm{N} = 3\) \(\mathrm{n} = 11\) for each genotype). Data were analyzed by unpaired Student's \(t\) test. Scale bar \(= 2\mathrm{mm}\) . b, Lateral wholemount images and quantification of shear wave velocity measured by SWE along the spine of \(10\mathrm{dpf}(5.10 + / - 0.48\mathrm{mm}\mathrm{SL})\) scoliotic vangl2sGFP;foxjla::icre;zGrad mutant zebrafish, and their non- scoliotic vangl2sGFP;foxjla::icre control siblings. Conditional degradation of the planar cell polarity effector protein vangl2 in foxjla- positive motile ciliated lineages (vangl2sGFP;foxjla::icre;zGrad), results in loss of brain ependymal cell cilia, CSF flow defects, and ectopic accumulations of Reissner's fiber25. Scoliotic vangl2sGFP;foxjla::icre;zGrad mutants also demonstrate significantly greater shear wave velocity and thus spine stiffness, compared to their control siblings \((\mathrm{p}\leq 0.0001\) \(\mathrm{N} = 3\) \(\mathrm{n} = 15\) for each group). Data were analyzed by unpaired Student's \(t\) test. Scale bar \(= 2\mathrm{mm}\) . \(^{**}p\leq 0.01\) , \(^{***}p\leq 0.0001\)
+
+<--- Page Split --->
+![PLACEHOLDER_37_0]
+
+
+<--- Page Split --->
+
+1 Extended Data Fig. 8: ECM of mineralized notochord domains appear normal in \(sspo^{dmh4 / + }\) mutants.  
+
+3 a, Schematic of zebrafish notochord development at 10 dpf (5mm SL) showing cartilaginous ECM 4 domains (IVD), mineralized domains (vertebra), notochord vacuole and notochord sheath cells. b, 5 c, Representative transmission electron micrographs of the collagenous ECM in mineralized 6 domains of 10 dpf (5mm SL) wildtype (b, \(\mathrm{N} = 7\) , \(\mathrm{n} = 24\) ) and \(sspo^{dmh4 / + }\) (c, \(\mathrm{N} = 7\) , \(\mathrm{n} = 23\) ) larvae. 7 d, e, Antioxidant treatment (500μM NACET) had no obvious effect on mineralized ECM 8 development in wildtype (d, \(\mathrm{N} = 5\) , \(\mathrm{n} = 17\) ) or \(sspo^{dmh4 / + }\) mutant (e, \(\mathrm{N} = 6\) , \(\mathrm{n} = 19\) ) animals. f-h, 9 Quantification of mineralized matrix thickness showed no significant difference in local maximum 10 matrix thickness (f), local minimum matrix thickness (g), and local matrix thickness ratio (h) 11 across the groups. Box plots show all data points, median, and min to max whiskers; points of the 12 same colour belong to the same fish. Data were analyzed by two- way ANOVA (n.s.: nonsignificant \(\mathrm{p} > 0.05\) , \*: \(\mathrm{p} \leq 0.05\) , \*\*: \(\mathrm{p} \leq 0.01\) , \*\*\*: \(\mathrm{p} \leq 0.001\) , \*\*\*\*: \(\mathrm{p} \leq 0.0001\) ).
+
+<--- Page Split --->
+
+## Supplementary Information  
+
+Supplementary Video 1. Wildtype microCT images, transverse plane.  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 744μm series, transverse plane.  
+
+Supplementary Video 2. Wildtype microCT images, coronal plane.  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 640μm series, coronal plane.  
+
+Supplementary Video 3. \(ssp^{dmh4 / +}\) microCT images, transverse plane.  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 693μm series, transverse plane.  
+
+Supplementary Video 4. \(ssp^{dmh4 / +}\) microCT images, coronal plane.  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 771μm series, coronal plane.  
+
+Supplementary Video 5. Wildtype zebrafish exhibit stable IVL structures (example 2).  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.  
+
+Supplementary Video 6. Scoliotic zebrafish exhibit IVL deformations (example 2).  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.  
+
+Supplementary Video 7. Scoliotic zebrafish exhibit IVL deformations (example 3).  
+
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- MovieS1WTtransverse.mov- MovieS2WTcoronal.mov- MovieS3sspotransverse.mov- MovieS4sspocoronal.mov- MovieS5WTIVL2coronal.mov- MovieS6sspolVL2coronal.mov- MovieS7sspolVL3coronal.mov
+
+<--- Page Split --->

preprint/preprint__98a3eb377368a4beb38d1aca76d6144c971ad88eff639624195d30516e45948c/preprint__98a3eb377368a4beb38d1aca76d6144c971ad88eff639624195d30516e45948c_det.mmd

@@ -0,0 +1,651 @@
+<|ref|>title<|/ref|><|det|>[[44, 108, 803, 208]]<|/det|>
+# Oxidative stress induces intervertebral ECM remodelling, elevated tissue stiffness and idiopathic-like scoliosis  
+
+<|ref|>text<|/ref|><|det|>[[44, 230, 256, 275]]<|/det|>
+Brian Ciruna ciruna@sickkids.ca  
+
+<|ref|>text<|/ref|><|det|>[[44, 301, 678, 345]]<|/det|>
+The Hospital for Sick Children https://orcid.org/0000- 0001- 7918- 0953  
+
+<|ref|>text<|/ref|><|det|>[[44, 349, 677, 368]]<|/det|>
+The Hospital for Sick Children https://orcid.org/0000- 0003- 4346- 9434  
+
+<|ref|>text<|/ref|><|det|>[[44, 374, 316, 415]]<|/det|>
+Ran Xu The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 421, 316, 461]]<|/det|>
+Josh Gopaul The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 467, 614, 507]]<|/det|>
+Arash Panahifar Canadian Light Source https://orcid.org/0000- 0001- 9483- 6949  
+
+<|ref|>text<|/ref|><|det|>[[44, 512, 316, 553]]<|/det|>
+Vida Erfani The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 559, 677, 600]]<|/det|>
+Jenica Van Gennip The Hospital for Sick Children https://orcid.org/0000- 0002- 4230- 7508  
+
+<|ref|>text<|/ref|><|det|>[[44, 606, 300, 645]]<|/det|>
+B Eames University of Saskatchewan  
+
+<|ref|>text<|/ref|><|det|>[[44, 652, 316, 692]]<|/det|>
+Nikan Fakhari The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 698, 947, 761]]<|/det|>
+Jerome Baranger Institute Physics for Medicine Paris,Inserm, ESPCI PSL Paris, CNRS https://orcid.org/0000- 0002- 2311- 716X  
+
+<|ref|>text<|/ref|><|det|>[[44, 767, 316, 808]]<|/det|>
+David Lebel The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 814, 316, 854]]<|/det|>
+Olivier Villemain The Hospital for Sick Children  
+
+<|ref|>text<|/ref|><|det|>[[44, 896, 103, 913]]<|/det|>
+Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 934, 135, 952]]<|/det|>
+Keywords:
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[42, 45, 344, 64]]<|/det|>
+**Posted Date:** September 9th, 2024  
+
+<|ref|>text<|/ref|><|det|>[[42, 83, 475, 103]]<|/det|>
+**DOI:** https://doi.org/10.21203/rs.3.rs-4978808/v1  
+
+<|ref|>text<|/ref|><|det|>[[42, 120, 916, 164]]<|/det|>
+**License:** © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 181, 535, 201]]<|/det|>
+**Additional Declarations:** There is **NO** Competing Interest.  
+
+<|ref|>text<|/ref|><|det|>[[42, 236, 920, 280]]<|/det|>
+**Version of Record:** A version of this preprint was published at Nature Communications on September 30th, 2025. See the published version at https://doi.org/10.1038/s41467-025-63742-2.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 163, 107]]<|/det|>
+## Title:  
+
+<|ref|>sub_title<|/ref|><|det|>[[256, 109, 740, 148]]<|/det|>
+## Oxidative stress induces intervertebral ECM remodeling, elevated tissue stiffness and idiopathic-like scoliosis  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 172, 193, 188]]<|/det|>
+## Authors:  
+
+<|ref|>text<|/ref|><|det|>[[115, 190, 852, 250]]<|/det|>
+Patrick G. Pumputis \(^{1,2\dagger}\) , Ran Xu \(^{1,2\dagger}\) , Josh Gopau \(^{1\dagger}\) , Arash Panahifar \(^{4,5}\) , Vida Erfani \(^{1,2}\) , Jenica Van Gennip \(^{1}\) , B. Frank Eames \(^{6}\) , Nikan Fakhari \(^{3,7}\) , Jerome Baranger \(^{3}\) , David E. Lebel \(^{8}\) , Olivier Villemain \(^{3,7*}\) , Brian Ciruna \(^{1,2*}\)  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 272, 216, 288]]<|/det|>
+## Affiliations:  
+
+<|ref|>text<|/ref|><|det|>[[112, 290, 878, 572]]<|/det|>
+\(^{1}\) Developmental & Stem Cell Biology Program, The Hospital for Sick Children; \(^{2}\) Toronto, Ontario, Canada \(^{3}\) Department of Molecular Genetics, University of Toronto; Toronto, Ontario, Canada \(^{3}\) Translational Medicine Program, The Hospital for Sick Children; Toronto, Ontario, Canada \(^{4}\) BioMedical Imaging and Therapy Beamline, Canadian Light Source; Saskatoon, Canada \(^{5}\) Department of Medical Imaging, College of Medicine, University of Saskatchewan; Saskatoon, Canada \(^{6}\) Department of Anatomy, Physiology, and Pharmacology, University of Saskatchewan; Saskatoon, Canada \(^{7}\) Department of Medical Biophysics, University of Toronto; Toronto, Ontario, Canada \(^{8}\) Division of Orthopaedic Surgery, The Hospital for Sick Children; Toronto, Ontario, Canada \(^{+}\) These authors contributed equally to this work \(*\) Co- senior and co- corresponding authors. Emails: ciruna@sickkids.ca & olivier.villemain@chubordeaux.fr  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 592, 196, 608]]<|/det|>
+## Abstract:  
+
+<|ref|>text<|/ref|><|det|>[[111, 610, 875, 891]]<|/det|>
+Adolescent idiopathic scoliosis (AIS) is the most prevalent pediatric spine disorder, developing in the absence of obvious congenital or physiological defects \(^{1}\) . Patient genetic sequencing and mouse functional studies have demonstrated association of musculoskeletal collagen variants and cartilaginous extracellular matrix (ECM) defects in a subset of cases \(^{2 - 4}\) . However, the underlying biological causes of AIS are poorly understood, thus treatment options remain limited to physical bracing or invasive corrective surgery \(^{1}\) . Here we interrogate the biological causes of scoliosis in zebrafish preclinical models of AIS. We demonstrate that neuroinflammation- associated reduction- oxidation (redox) imbalance induces cell stress and collagen remodeling defects within intervertebral segments of the developing spine. Mutant spines are consequently stiffer, as measured by shear wave elastography, and exhibit deformations of intervertebral structures. Remarkably, both elevated spine stiffness and intervertebral ECM phenotypes are detectable prior to scoliosis onset in zebrafish models, suggesting a causal relationship, and can be suppressed by antioxidant treatment. Together, our studies implicate oxidative stress- induced intervertebral deformations in the pathogenesis of AIS
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 89, 853, 129]]<|/det|>
+and identify elevated spine stiffness and redox imbalance as plausible first- in- kind prognostic biomarkers and therapeutic targets.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 152, 210, 168]]<|/det|>
+## Main Text:  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 172, 230, 188]]<|/det|>
+## Introduction:  
+
+<|ref|>text<|/ref|><|det|>[[112, 191, 872, 409]]<|/det|>
+Adolescent idiopathic scoliosis (AIS) is a common pediatric disorder characterized by lateral deviations in the spinal column (with a rotational component) that develop in the absence of obvious congenital or physiological defects<sup>1</sup>. AIS afflicts 3- 4% of children worldwide yet underlying causes are poorly understood, in part due to genetic heterogeneity and suspected environmental influences<sup>1</sup>. Patient genetic sequencing and mouse functional studies have demonstrated association of musculoskeletal collagen variants and cartilaginous extracellular matrix (ECM) defects in a subset of cases<sup>2- 4</sup>. However, over 95% of total causative genetic variance in AIS is thought to be unknown<sup>5</sup>. Because most AIS etiological studies involve patients with established scoliotic curves, determining a biological mechanism of cause and effect is particularly challenging, and treatment options remain limited to mechanical interventions like bracing or invasive corrective surgery<sup>1</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[112, 409, 866, 608]]<|/det|>
+Zebrafish have emerged as powerful experimental models for dissecting complex biological mechanisms associated with AIS<sup>6,7</sup>. Remarkably, it was discovered that neuroinflammatory signals and oxidative stress, arising from imbalances in cerebrospinal fluid (CSF) homeostasis, are both necessary and sufficient to cause AIS- like spinal curvatures in zebrafish<sup>8- 10</sup>. Furthermore, treatment with antioxidant and immunomodulating compounds like N- acetylcysteine (NAC) or NAC- ethyl ester (NACET, a more bioavailable form of NAC) can suppress scoliosis onset and severe spinal curve progression in zebrafish models<sup>9- 11</sup>. While this provides proof- of- principle that scoliosis might be managed therapeutically, undetermined mechanisms by which oxidative stress leads to spinal curvature and their relevance to human AIS pose a barrier to clinical translation.  
+
+<|ref|>text<|/ref|><|det|>[[112, 610, 883, 828]]<|/det|>
+Here we interrogate the downstream causes of scoliosis in the dominant SCO- spondin \((ssp^{dmhd / + })\) mutant zebrafish model of AIS<sup>10,12</sup>. These fish develop idiopathic- like spinal curvatures in response to neuroinflammatory signals that are linked to the disruption of Reissner's fiber (RF), a proteinaceous filament that threads through ventricular cavities of the spinal cord and brain. We demonstrate that abnormal axial reduction- oxidation (redox) states in \(ssp^{dmhd / + }\) mutants induce both intervertebral collagen ECM remodelling defects and elevated spine stiffness that appear causally linked to the onset and progression of scoliosis. Furthermore, we characterize resulting intervertebral disc deformations as a pathophysiological mechanism shared between zebrafish scoliosis models and human AIS patients. Finally, we provide evidence that elevated spine stiffness and oxidative stress may prove to be valuable, first- in- kind, prognostic biomarkers and therapeutic targets for idiopathic- like scoliosis.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 92, 185, 107]]<|/det|>
+## Results:  
+
+<|ref|>text<|/ref|><|det|>[[111, 110, 880, 530]]<|/det|>
+sspo<sup>dmh4/+</sup> mutants exhibit an elevated oxidative state in proximity to the developing spine NACET, a potent anti- oxidant, can suppress the onset and severe progression of scoliosis in sspo<sup>dmh4/+</sup> AIS models<sup>10</sup>. To investigate how oxidative stress may influence spine development, we characterized the spatial distribution of redox imbalance in larval zebrafish using the cell permeable fluorescent redox probe 2',7'- dichlorofluorescein diacetate (DCFH- DA)<sup>13</sup>. While the probe cannot attribute redox changes to specific reactive oxygen species (ROS), it can be used as a general indicator of redox state<sup>13</sup>. Staining was performed on wildtype (WT) and sspo<sup>dmh4/+</sup> zebrafish at 10 days post fertilization (dpf) and 6 mm standard length (SL), prior to scoliosis onset. At this stage, sspo<sup>dmh4/+</sup> zebrafish appear morphologically indistinguishable from WT (Fig. 1a, b). However, significantly stronger oxidation levels were detected in the dorsal telencephalon and cranial vessels of sspo<sup>dmh4/+</sup> mutants (Fig. 1c, d, g), which spatially correlate with sites of neuroinflammation previously identified in these animals<sup>10</sup>. Strikingly, elevated oxidative signals were also observed along the trunk of sspo<sup>dmh4/+</sup> larva in the spinal cord, trunk vasculature and dorsal aorta, as well as in discrete peri- notochordal segments (Fig. 1e, f, h- k). These notochordal segments correspond to developing cartilaginous intervertebral domains (IVDs) of the spine, as demonstrated by the complementary pattern of TgBAC(entpd5a::pkRed) reporter expression in alternating mineralizing domains (Fig. 11- n)<sup>14</sup>. In contrast, no difference in DCFH- DA signals were observed in the trunk muscle (myotome) of 10 dpf WT and sspo<sup>dmh4/+</sup> zebrafish larva (Extended Data Fig. 1a- c). Overall, these data suggest that neuroinflammatory responses observed in sspo<sup>dmh4/+</sup> mutants associate with tissue non- autonomous elevation of ROS in close juxtaposition to the notochord and developing spine.  
+
+<|ref|>sub_title<|/ref|><|det|>[[112, 550, 860, 569]]<|/det|>
+## Redox imbalance induces endoplasmic reticulum (ER) stress within intervertebral segments  
+
+<|ref|>text<|/ref|><|det|>[[112, 570, 879, 707]]<|/det|>
+The zebrafish notochord, comprised of inner vacuolated cells and outer peri- notochordal sheath cells, provides structural support to the embryo and serves as a template for spine development<sup>6</sup>. Sheath cells, abundant in rough ER, are secretory cells that contribute to the notochord's ECM- rich lamellar sheath<sup>6</sup>, which segments into developing vertebrae and IVDs<sup>14</sup>. Since dysregulation of redox homeostasis in cells can initiate ER stress and an unfolded protein response (UPR)<sup>15</sup>, we hypothesized that peri- notochordal accumulation of ROS in sspo<sup>dmh4/+</sup> IVD segments may induce ER stress and disrupt sheath cell function.  
+
+<|ref|>text<|/ref|><|det|>[[112, 709, 877, 890]]<|/det|>
+To assess ER stress in sspo<sup>dmh4/+</sup> mutants, we modified an established transgenic assay that incorporates an IRE1 endonuclease- catalyzed splicing cassette that reports on XBP1 activation via expression of a functionally inert XBP1Δ- GFP fusion protein (Extended Data Fig. 2a)<sup>16</sup>. Specifically, the xbp1δ- eGFP ER stress reporter was cloned downstream of a col2a1a promoter<sup>17</sup> to direct expression within cartilaginous IVD segments and the site of observed oxidation (arrowheads, Fig. 1f, l- n). Basal levels of Tg(col2a1a::xbp1δ- eGFP) reporter activity and ER stress could be detected within developing IVDs of WT fish (10 dpf, 6 mm SL) (Fig. 1o) in between alternating mineralizing domains labeled by Tg(entpd5a::pkRed)<sup>14</sup>. Strikingly, a significant increase in Tg(col2a1a::xbp1δ- eGFP) reporter activity was observed within IVDs of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 88, 881, 169]]<|/det|>
+\(s s p o^{d m h4 / + }\) mutant zebrafish (10 dpf, 6 mm SL), indicating an active ER stress response prior to the onset of scoliosis (Fig. 1p, q). Notably, this ER stress response was transient, as no difference in \(T g(c o l2a1a::x b p1\delta - e G F P)\) reporter activity was observed between \(s s p o^{d m h4 / + }\) mutant and wildtype zebrafish at 14 dpf (6.5 mm SL; Extended Data Fig. 2g- i).  
+
+<|ref|>text<|/ref|><|det|>[[110, 169, 875, 310]]<|/det|>
+To determine whether observed ER stress was linked to oxidative stress, larval zebrafish were treated with \(500~\mu \mathrm{M}\) of the anti- oxidant NACET from 5 to 10 dpf. NACET treatment had no effect on \(T g(c o l2a1a::x b p1\delta - e G F P)\) reporter activity in wildtype zebrafish, but fully suppressed ER stress responses in \(s s p o^{d m h4 / + }\) mutant zebrafish to wildtype levels (Extended Data Fig. 2b- f). Together, these data indicate that redox imbalances induce ER stress within the IVDs of \(s s p o^{d m h4 / + }\) mutant animals preceding scoliosis onset, and may thus be functionally linked to spinal curvature.  
+
+<|ref|>sub_title<|/ref|><|det|>[[111, 329, 694, 349]]<|/det|>
+## \(s s p o^{d m h4 / + }\) mutants exhibit IVD deformations and collagen ECM defects  
+
+<|ref|>text<|/ref|><|det|>[[110, 350, 879, 609]]<|/det|>
+To further investigate the functional consequences of oxidative/ER stress on zebrafish IVD development, we performed high- resolution synchrotron- based X- ray micro- computed tomography (synchrotron \(\mu \mathrm{CT}\) ) on juvenile fish at 21dpf (10 mm SL), when spinal curvature is well- established in \(s s p o^{d m h4 / + }\) AIS models (Extended Data Fig. 3). Samples were stained using a diffusive iodine- based contrast enhancement, and 4- 5 segments of the caudal spine were imaged at \(0.36~\mu \mathrm{m}\) voxel size to produce high- resolution 3D datasets of the mineralized skeleton and surrounding soft tissues (Fig. 2a,e; Supplementary Videos 1- 4). At stages analyzed, zebrafish IVDs have differentiated into two principal structures (Extended Data Fig. 4a, b). These include an inner domain comprised of vacuolated and sheath notochordal cells; and an outer intervertebral ligament (IVL) that physically connects two adjacent vertebral bodies, comprised of layered collagen type II and elastin matrix, dense collagen type 1 matrix and collagen type 1 bundle fibers \(^{18}\) . These inner and outer intervertebral structures are functionally analogous to the human nucleus pulposus (NP) and annulus fibrosus (AF), respectively \(^{18,19}\) .  
+
+<|ref|>text<|/ref|><|det|>[[110, 609, 881, 870]]<|/det|>
+Remarkably, the IVL (AF) of \(s s p o^{d m h4 / + }\) mutants exhibited obvious structural defects and appeared highly deformable. In wildtype zebrafish, the IVL could be identified connecting neighbouring vertebral bodies, looping outward in the coronal plane away from notochordal cells and towards neighbouring muscle fibres (arrowhead, Fig. 2b). This organization of IVL structure, previously described for wildtype adult zebrafish \(^{18,19}\) , was stable across the circumference of the IVD (Fig. 2c- c'"; Supplementary Video 5) with its dense collagen layer appearing in transverse sections as smooth, concentric bands of high- contrast stain encircling the vertebral endplates (arrowhead, Fig. 2d). In contrast, coronal images of \(s s p o^{d m h4 / + }\) mutants often demonstrated an inversion of typical AF structure, with sections of \(s s p o^{d m h4 / + }\) IVL folding interiorly towards the notochordal cell layers (arrowhead, Fig. 2f). Strikingly, the contour of \(s s p o^{d m h4 / + }\) IVL layers exhibited significant deformations over short distances (Fig. 2g- g'") and, in some instances, could be observed transitioning between wildtype and inverted orientations within \(30~\mu \mathrm{m}\) intervals (Supplementary Videos 6, 7). As a result, transverse sections of \(s s p o^{d m h4 / + }\)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[67, 90, 861, 129]]<|/det|>
+1 mutants exhibited irregular and twisted IVL borders (arrows, Fig. 2h) displaced towards the interior surface of the vertebral endplates (arrowhead, Fig. 2h).  
+
+<|ref|>text<|/ref|><|det|>[[110, 130, 884, 350]]<|/det|>
+To probe the structural integrity of the IVD matrix in \(sspo^{dmh4 / + }\) mutants, we performed collagen hybridizing peptide (CHP) stains on axial skeleton preparations of juvenile fish at 21 dpf (10 mm SL). CHP is a synthetic construct that mimics the helical structure of endogenous collagen peptides, and can bind to open regions in collagen triple helices to report on collagen ECM damage and remodelling (Extended Data Fig. 4g) \(^{20}\) . CHP staining appeared stronger in \(sspo^{dmh4 / + }\) mutant IVLs compared to wildtype, and highlighted irregularities in IVL structure (arrows, Extended Data Fig. 4c-f) previously observed in microCT images. Furthermore, brighter and ectopic CHP signals were also observed within the NP layer of \(sspo^{dmh4 / + }\) mutants (arrowheads, Extended Data Fig. 4c-f) indicating extensive damage to the collagen matrix, dysregulated ECM remodelling, or both. These data demonstrate abnormalities in the structural properties of \(sspo^{dmh4 / + }\) IVDs, both at the morphological and molecular level.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 370, 463, 388]]<|/det|>
+## Zebrafish AIS models exhibit stiffer spines  
+
+<|ref|>text<|/ref|><|det|>[[112, 390, 875, 550]]<|/det|>
+The viscoelastic properties of human intervertebral disc tissue are also significantly altered in AIS patients. Specifically, the AF is stiffer in AIS patients compared to healthy controls, as measured by shear wave elastography (SWE) \(^{21,22}\) . SWE uses an acoustic radiation force to induce mechanical tissue displacement and calculates the velocity of shear waves propagating in the orthogonal direction. As tissue stiffness is directly proportional to shear wave speed, this provides a quantitative measure of tissue elasticity \(^{23,24}\) . To further characterize physiological defects associated with oxidative stress and scoliosis in our models, we utilized SWE to quantitatively assess zebrafish spine stiffness (Extended Data Fig. 5a- c).  
+
+<|ref|>text<|/ref|><|det|>[[112, 551, 881, 770]]<|/det|>
+To begin, we conducted SWE at 16 dpf (average \(\mathrm{SL} = 7.1 + / - 0.2 \mathrm{mm}\) ) and 21 dpf (average \(\mathrm{SL} = 8.9 + / - 0.6 \mathrm{mm}\) ) stages, corresponding to the onset and progression of scoliosis in \(sspo^{dmh4 / + }\) mutants, respectively (Fig. 3a, b). Euthanized zebrafish were embedded in low melt agarose, and SWE was performed on the abdominal segment of the spine (note that variations in SWE ‘push’ locations and agarose density did not significantly alter shear wave velocities; Extended Data Fig. 5d, e). Remarkably, \(sspo^{dmh4 / + }\) shear wave velocities were significantly faster than wildtype controls at both time points (Fig. 3c, d), indicating an increase in spine stiffness in scoliotic \(sspo^{dmh4 / + }\) mutant zebrafish. Observed increases in spine stiffness were not likely caused by changes in vertebral bone density, as microCT analysis of the mineralized axial skeleton of adult zebrafish did not reveal significant differences in \(sspo^{dmh4 / + }\) tissue mineral density compared to wildtype siblings (Extended Data Fig. 6).  
+
+<|ref|>text<|/ref|><|det|>[[113, 771, 866, 890]]<|/det|>
+To determine whether elevated spine stiffness represents a generalizable feature of zebrafish scoliosis, or a property specific to \(sspo^{dmh4 / + }\) mutants, we performed SWE on two additional AIS models (katnb1 and vangl2) that share upstream CSF homeostasis defects but differ in underlying molecular genetic causes \(^{25,26}\) . Again, significantly faster shear wave velocities were observed in scoliotic katnb1 and vangl2 mutants compared to control siblings (Extended Data Fig. 7). Therefore, our results demonstrate altered viscoelastic properties of the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[67, 90, 850, 130]]<|/det|>
+1 zebrafish scoliotic spine that parallel elevated intervertebral disc stiffness observed in human 2 AIS patients \(^{21,22}\) .  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 151, 740, 170]]<|/det|>
+## Oxidative stress-induced spine stiffness is a prognostic biomarker of scoliosis  
+
+<|ref|>text<|/ref|><|det|>[[112, 171, 876, 328]]<|/det|>
+Shear wave velocities varied noticeably within \(sspo^{dmh4 / + }\) mutant cohorts (Fig. 3c, d). Since the penetrance and severity of spinal curvature is also variable among \(sspo^{dmh4 / + }\) zebrafish \(^{10}\) , we hypothesized that differences in shear wave velocity/spine stiffness may directly reflect scoliosis severity. Using a second cohort of 21 dpf \(sspo^{dmh4 / + }\) mutants (average \(\mathrm{SL} = 8.2\) \(+ / - 0.7 \mathrm{mm}\) ), we performed microCT imaging after SWE and measured total Cobb angle values, which is a biomarker of spinal curve severity. Remarkably, a significant and positive correlation between shear wave velocity and total Cobb- angle/curve severity was observed ( \(\mathrm{R} = 0.831\) ; Fig. 3e).  
+
+<|ref|>text<|/ref|><|det|>[[112, 330, 880, 570]]<|/det|>
+Elevated spine stiffness was measured after scoliosis onset and could therefore have developed as a consequence of spinal curvature. To determine the relationship between stiffness and scoliosis, we next conducted SWE on \(sspo^{dmh4 / + }\) and wildtype fish at 10 dpf (average \(\mathrm{SL} = 4.7 + / - 0.4 \mathrm{mm}\) ), prior to obvious morphological phenotypes. Remarkably, \(sspo^{dmh4 / + }\) mutant spines were significantly stiffer than wildtype controls before scoliosis onset, suggesting a causal relationship (Fig. 3f). To determine whether elevated tissue stiffness was linked to elevated oxidative stress, larval zebrafish were treated with \(500 \mu \mathrm{M}\) of NACET from 5 to 10 dpf, prior to SWE. Strikingly, NACET treatment had no effect on wildtype control animals, but fully suppressed abnormal spine stiffness in \(sspo^{dmh4 / + }\) zebrafish to wildtype levels (Fig. 3f). Together, these results directly link oxidative stress to altered viscoelastic properties of the scoliotic spine and identify elevated spine stiffness as a prognostic biomarker of spinal curvature in zebrafish AIS models.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 590, 750, 610]]<|/det|>
+## IVD matrix defects precede scoliosis, and are directly linked to oxidative stress  
+
+<|ref|>text<|/ref|><|det|>[[112, 610, 880, 770]]<|/det|>
+IVD deformations and collagen remodelling defects were also observed after scoliosis onset in \(sspo^{dmh4 / + }\) mutants and may therefore reflect the cause or the consequences of spinal curvature. Although SWE indicates structural defects likely precede scoliosis, it lacks sufficient spatial resolution to identify affected tissues. To further investigate whether ultrastructural ECM defects drive spinal curvature in \(sspo^{dmh4 / + }\) models, we performed transmission electron microscopy (TEM) on 10 dpf zebrafish (5 mm SL) prior to an obvious morphological phenotype. Specifically, we focussed on mineralized and cartilaginous domains of the peri- notochordal ECM that template vertebral centra and IVD development, respectively (Fig. 4a).  
+
+<|ref|>text<|/ref|><|det|>[[112, 771, 881, 890]]<|/det|>
+In wildtype animals, a uniform layer of collagen ECM was observed within cartilaginous domains, with bundles of collagen fibres organized circumferentially around the notochord projecting out of the sagittal plane of sectioning (Fig. 4b, b'). In contrast, \(sspo^{dmh4 / + }\) cartilaginous segments exhibited irregularities in the thickness and orientation of collagen ECM layers (Fig. 4c, c'). Quantification of collagen ECM thickness revealed that while the local minimum was consistent between WT and \(sspo^{dmh4 / + }\) IVD segments, the maximum thickness and local max/min
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 881, 207]]<|/det|>
+ratios were significantly higher in \(sspo^{dmh4 / + }\) mutants (Fig. 4f- h). Furthermore, the ECM appeared disorganized and misoriented in developing \(sspo^{dmh4 / + }\) IVD segments, with collagen fibre bundles observed running parallel to the sagittal plane (Fig 4. c, c'). Although the penetrance and severity of \(sspo^{dmh4 / + }\) ECM phenotypes varied among IVD segments (Fig. 4f- h), all \(sspo^{dmh4 / + }\) fish \((N = 9)\) exhibited IVD defects. Notably, collagen ECM defects were specific to IVDs, and were not observed in mineralized domains around the notochord sheath (Extended Data Fig. 8).  
+
+<|ref|>text<|/ref|><|det|>[[112, 210, 876, 369]]<|/det|>
+To determine whether collagen ECM defects are also linked to the elevated oxidative state observed around the IVD of \(sspo^{dmh4 / + }\) mutants, larval zebrafish were treated with \(500~\mu \mathrm{M}\) NAcET from 5 to 10 dpf, prior to TEM imaging. Strikingly, NAcET treatment had no effect on collagen matrix organization in wildtype animals but restored collagen fibre orientation and suppressed ECM thickness in \(sspo^{dmh4 / + }\) mutants to wildtype levels (Fig. 4d- h). Together, these results demonstrate that oxidative stress- induced collagen ECM defects, specific to IVD segments, precede scoliosis onset in \(sspo^{dmh4 / + }\) animals and may therefore be functionally linked to IVL deformations and idiopathic- like spinal curvature.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 392, 206, 408]]<|/det|>
+## Discussion  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 411, 624, 428]]<|/det|>
+## Unconventional AIS models corroborate conventional theories  
+
+<|ref|>text<|/ref|><|det|>[[110, 430, 881, 770]]<|/det|>
+Despite the neuroinflammatory origins of scoliosis in zebrafish models \(^{9 - 11}\) , we demonstrate that pathogenic mechanisms ultimately converge on a historical focus of human AIS studies – the IVD. Trueta (1968) first reported that morphological deformations of the IVD contribute greatly to AIS spine deformity \(^{27}\) , and Harrington (1977) theorized loss of physical properties (collagen) of the IVD might thus present as a common denominator in AIS \(^{28}\) . These ideas have since been explored, in part, by longitudinal radiographic studies of AIS patients associating IVD wedging with the early presentation and progression of spinal curvature \(^{29,30}\) , and mathematical modelling of AF collagen fibre imbalance as an etiological factor in scoliosis \(^{31}\) . Although, biochemical and histological studies have also identified irregular collagen and elastic properties in IVDs isolated from AIS patients \(^{32 - 34}\) , these defects were observed only after scoliosis onset. Here we provide direct evidence that ROS- induced IVD ER stress/UPR responses, altered viscoelastic properties, and collagen ECM defects all precede spinal curvature in \(sspo^{dmh4 / + }\) zebrafish. Although it remains to be determined how oxidative stress ultimately perturbs collagen matrix development (i.e. via abnormal protein translation, trafficking, cross- linking, etc.) our data, together with published reports, strongly support a causal role for IVD ECM remodelling defects in the etiopathogenesis of scoliosis and highlight the relevance of zebrafish AIS models for pathomechanism discovery and therapeutic development.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 791, 516, 810]]<|/det|>
+## Spine stiffness as a prognostic biomarker for AIS  
+
+<|ref|>text<|/ref|><|det|>[[113, 811, 863, 889]]<|/det|>
+At present, genetic screening for AIS has proven ineffective \(^{35}\) . Diagnosis can only be made after scoliosis onset, and high variability in severe curve progression makes it difficult to predict which patients may ultimately require bracing or surgical intervention. Here we demonstrate the prognostic capabilities of SWE in zebrafish AIS models. Strikingly, we have
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 88, 880, 330]]<|/det|>
+1 determined that elevated spine stiffness can be detected prior to scoliosis onset. As our data 2 complement clinical SWE studies demonstrating elevated AF stiffness in affected AIS 3 patients21,22, spine stiffness in \(sspo^{dmh4 / + }\) zebrafish might similarly be caused by observed IVD 4 collagen remodelling defects. Indeed, spine stiffness and IVD matrix defects are both oxidative 5 stress- dependent and mechanistically linked in \(sspo^{dmh4 / + }\) models, and increases in the thickness 6 and isotropy of a medium (as observed for \(sspo^{dmh4 / + }\) collagen IVD matrix) are predicted to 7 elevate medium stiffness36,37. Although the spatial and molecular origins of elevated stiffness in 8 \(sspo^{dmh4 / + }\) zebrafish remains to be determined, our data provide proof- of- concept that spine 9 stiffness might be explored as a prognostic biomarker for AIS development. Given that SWE is a 10 non- invasive imaging technique that is widely applied in the clinic, translation of these findings 11 holds tremendous potential to identify patients at high risk of severe scoliosis for earlier, non- 12 surgical intervention.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 350, 602, 368]]<|/det|>
+## A new framework for considering AIS origins and therapies  
+
+<|ref|>text<|/ref|><|det|>[[112, 370, 881, 607]]<|/det|>
+The oxidative stress- induced origin of ECM defects identified in zebrafish AIS models is equally deserving of clinical consideration. AIS is a common pediatric disorder'. Nevertheless, it is estimated that \(< 5\%\) total genetic variance in AIS has been determined, and genetic associations with cartilaginous ECM and IVD development account for only a subset of these cases2- 4. However, environmental variables like infection, diet, lifestyle and microbiome composition can profoundly influence ROS levels38- 40 and could therefore play a pervasive, yet largely unexplored role in AIS pathogenesis. In addition, neuroinflammation and oxidative stress can also have genetic origins (as in \(sspo^{dmh4 / + }\) mutant zebrafish) and should be explored as etiological factors driving human spinal curvature. This includes not only AIS, but other genetically defined disorders that present with both CNS oxi- inflammation and high incidences of developmental scoliosis like DiGeorge Syndrome41,42, Rett Syndrome43,44, CDKL5 Deficiency Disorder45,46, and Cerebral Cavernous Malformation47,48.  
+
+<|ref|>text<|/ref|><|det|>[[113, 608, 882, 769]]<|/det|>
+Finally, regardless of the origins of oxidative stress, prophylactic and therapeutic administration of antioxidants has proven remarkably effective in preventing the onset and severe progression of scoliosis in zebrafish AIS models9- 11. Furthermore, we have demonstrated that ER stress responses, elevated spine stiffness and collagen ECM defects in \(sspo^{dmh4 / + }\) mutants, which are all functionally linked to spinal curvature, can be fully suppressed by NACET treatment. If oxidative stress- induced ECM remodeling defects prove relevant to patient populations, then translation of these findings could have profound impact on the future management and prevention of human spinal curvature.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[68, 91, 197, 108]]<|/det|>
+## 1 Methods:  
+
+<|ref|>text<|/ref|><|det|>[[66, 111, 881, 370]]<|/det|>
+2 Zebrafish 3 Zebrafish husbandry and experimental protocols were approved by the Hospital for Sick 4 Children's Animal Care Committee, and all protocols were performed in accordance with 5 Canadian Council on Animal Care guidelines. Wildtype zebrafish from TU strains were used. 6 \(sspo^{dmh4}(10)\) , \(katnb1^{mh102}(26)\) , \(vangl2^{sfGFP}\) (hsc170Tg) \(^{25}\) , \(Tg(\beta actin\colon \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \mathrm{loop - mCherry - STOP - loop - }\) zGrad) (hsc185Tg) \(^{25}\) , and \(Tg(foxj1a\colon \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \mathrm{cre})\) (hsc105Tg) \(^{25}\) mutant and transgenic fish used in this study have been previously described. Information on newly developed transgenic fish 9 \(Tg(col2a1a\colon xbp1\delta - eGFP)\) can be found below in the "Transgenesis" method section. Embryos from natural matings were grown at \(28.5^{\circ}\mathrm{C}\) . When required, experimental animals were 11 euthanized with tricaine (500 mg/L; MS- 222/MESAB), followed by submersion of anesthetized 12 fish in ice water for several minutes. As laboratory zebrafish strains do not utilize a chromosomal 13 sex determination mechanism and sex differentiation does not initiate until after \(\sim 3\) weeks post 14 fertilization \(^{49}\) , we cannot report sex for our embryonic and larval studies.  
+
+<|ref|>sub_title<|/ref|><|det|>[[66, 390, 287, 409]]<|/det|>
+## 16 Zebrafish genotyping  
+
+<|ref|>text<|/ref|><|det|>[[66, 410, 875, 470]]<|/det|>
+16 Zebrafish genotyping 17 Genotyping protocols have been described previously and can be found in the following 18 publications: \(sspo^{dmh4 / + }(10)\) , \(katnb1^{mh102 / mh102}(26)\) , \(vangl2^{sfGFP / sfGFP}(25)\) . Primer sequences used for genotyping can be found in Extended Data Table 1.  
+
+<|ref|>sub_title<|/ref|><|det|>[[66, 490, 220, 508]]<|/det|>
+## 21 Transgenesis  
+
+<|ref|>text<|/ref|><|det|>[[66, 510, 875, 750]]<|/det|>
+21 Transgenesis 22 Entry plasmids were generated through BP recombination into respective pDONR plasmids 23 (Invitrogen) and then cloned into standard Tol2 kit Gateway compatible vectors using LR 24 recombination methods to create the final transgenes \(^{50}\) . To generate \(Tg(col2a1a\colon xbp1\delta - eGFP)\) , a previously generated p5E- \(col2a1a^{51}\) , pME- \(xbp1\delta - eGFP\) (obtained by cloning the \(xbp1\) partial sequence using Gateway primers, detailed in table 1, from \(ef1a\colon xbp1\delta - gfp\) plasmid \(^{16}\) graciously provided by Dr. Shao Jun Du), and p3E- polyA \(^{50}\) were recombined into pDEST Tol2 HR2 28 transgenesis vector. Embryos were injected at the one cell stage with 25 pg of Tol2 transposase 29 RNA and 25 pg of the transgene (plasmid). Injected embryos were then screened at 48 hpf for 30 transgenesis marker expression. Imaging of reporter expression was performed on an Axio 31 Zoom.V16 (Zeiss). Embryos displaying strong reporter expression were grown to adulthood and 32 crossed to wildtype fish to establish stable F1 lines. F1 lines were then bred into \(sspo^{dmh4 / + }\) and 33 maintained in \(sspo^{dmh4 / + }\) fish.  
+
+<|ref|>sub_title<|/ref|><|det|>[[66, 771, 272, 789]]<|/det|>
+## 35 DCFH-DA staining  
+
+<|ref|>text<|/ref|><|det|>[[66, 791, 875, 870]]<|/det|>
+35 DCFH-DA staining 36 Zebrafish larva were collected and placed in 12 well plates containing \(2\mathrm{mL}\) of \(5\mu \mathrm{M}\) DCFH- DA (Sigma) in E3 media. Plates were kept in the dark and incubated for 30 minutes at a temperature of \(28.5^{\circ}\mathrm{C}\) . Larva were then washed three times with E3 media for 5 minutes each wash in the dark and kept at \(28.5^{\circ}\mathrm{C}\) until imaged.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 90, 872, 170]]<|/det|>
+1 DCFH- DA, Tg(col2a1a::xbp1&eGFP), and TgBAC(entpd5a::pkRed) imaging  2 Experimental zebrafish larva were euthanized with tricaine (500 mg/L) before imaging. 10 dpf  3 larva were mounted dorsally or laterally in \(0.8\%\) low melt agarose (BioShop), measured for  4 standard length, and imaged on an LSM 710 confocal microscope (Zeiss).  
+
+<|ref|>text<|/ref|><|det|>[[67, 190, 876, 430]]<|/det|>
+5 Collagen hybridizing peptide  7 Juvenile zebrafish were collected at 21dpf (10mm SL) and anesthetized with tricaine (500mg/L).  8 Anesthetized fish were fixed with \(100\%\) methanol at \(- 20^{\circ}\mathrm{C}\) for 48 hours, then rehydrated with  9 1xPBS before the skin was gently removed with a pair of dissection forceps. Collagen  10 hybridizing peptide (F- CHP from 3Helix, diluted in 1xPBS to \(20\mu \mathrm{M}\) ) was then heated to \(80^{\circ}\mathrm{C}\)  11 for activation, followed by quenching on an ice block down to room temperature. Fish were then  12 immersed in CHP for 48 hours at \(4^{\circ}\mathrm{C}\) in dark. After CHP staining, fish were transferred into  13 individual tubes for soft tissue lysing in \(2\% \mathrm{KOH}\) /ethylene glycol/1xPBS solution, then switched  14 to \(1\% \mathrm{KOH}\) /ethylene glycol/1xPBS when the vertebrae were exposed. Samples were placed on a  15 nutator during tissue lysing until most of the muscle, spinal cord and ventral vasculature had  16 been digested, then cleared with glycerol before imaged on an LSM 710 confocal microscope  17 (Zeiss).  
+
+<|ref|>text<|/ref|><|det|>[[67, 450, 870, 670]]<|/det|>
+19 Transmission electron microscopy  20 Fish were collected at 10dpf (5mm SL) and anesthetized with tricaine (500mg/L). Anesthetized  21 samples were then fixed in \(2\%\) paraformaldehyde and \(2.5\%\) glutaraldehyde in \(0.1\mathrm{M}\) sodium  22 cacodylate buffer for 2 hours. Rinsed in buffer, post- fixed in \(1\%\) osmium tetroxide in buffer for  23 90 min, dehydrated in a graded ethanol series ( \(50\% ,70\% ,90\%\) and \(100\%\) , 20 minutes each step)  24 followed by two propylene oxide changes for 30 min, and embedded in Quetol- Spurr resin.  25 Blocks were cured overnight in the oven at \(60^{\circ}\mathrm{C}\) . Fish were cut along the sagittal plane \(100\mu \mathrm{m}\)  26 deep into the sample. Sections \(70\mathrm{nm}\) thick were cut on a Leica EM UC7 ultramicrotome, and  27 post- stained with \(2\%\) uranyl acetate and \(3\%\) lead citrate for 20 minutes each and washed for 5  28 minutes after each staining. Sections were air dried at room temperature before imaged with a  29 Hitachi HT7800 transmission electron microscope.  
+
+<|ref|>text<|/ref|><|det|>[[66, 690, 876, 890]]<|/det|>
+30 Synchrotron \(\mu \mathrm{CT}\) imaging, reconstruction and segmentation  31 Juvenile zebrafish were collected at 21dpf (10mm SL) and anesthetized with tricaine (500mg/L).  32 Anesthetized fish were fixed with \(4\%\) PFA at \(4^{\circ}\mathrm{C}\) overnight, then washed with 1xPBS before  33 stained in I2E solution ( \(1\%\) iodine metal in \(100\%\) ethanol) overnight at room temperature.  34 Samples were then washed again with 1xPBS to remove excess I2E stain, then mounted in \(1.5\%\)  35 agarose in R.O. water in \(0.2\mathrm{mL}\) tubes. Synchrotron \(\mu \mathrm{CT}\) were performed at the bending magnet  36 beamline of BioMedical Imaging and Therapy beamlines (BMIT- BM) at the Canadian Light  37 Source \(^{52}\) . The beamline is operated in filtered white mode, therefor, it is essential to filter the  38 beam sufficiently to prevent radiation damage to the sample and agarose gel. Filters of \(1.76\mathrm{mm}\)  39 Al and \(0.275\mathrm{mm}\) Sn were used. Detector was an indirect X- ray microscope (Optique Peter,
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 88, 880, 191]]<|/det|>
+France) combined with a sCMOS PCO Edge 5.5 camera (PCO, Germany), a 20x objective, and a \(4\mu \mathrm{m}\) LSO scintillator to obtain an effective pixel size of \(0.360\mu \mathrm{m}\) . Generally, the dimension of scanned samples was \(400\times 600\times 750\mu \mathrm{m}^3\) . 1500 projections over 180 degrees were collected at exposure time of 1.5s. Sample was \(25.7\mathrm{m}\) from the source, and propagation distance was \(3\mathrm{cm}\) . Phase retrieval and image reconstructions were done using tofu reconstruction package \(^{53}\) .  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 212, 270, 229]]<|/det|>
+## NACET treatments  
+
+<|ref|>text<|/ref|><|det|>[[112, 231, 872, 410]]<|/det|>
+Fish were housed off- system in \(500\mathrm{mL}\) of \(300\mathrm{mg / L}\) Instant Ocean Sea Salt treated water in 1.8 L tanks with 12 fish per tank. Water was changed once per day. N- acetyl- L- cysteine ethyl ester (NACET, BOC Sciences Cat# B0689- 029481) was prepared as a \(250\mathrm{mM}\) stock solution in 300 mg/L Instant Ocean Sea Salt dissolved in MiliQ water. NACET was administered with water changes once per day at a final concentration of \(500\mu \mathrm{M}\) . NACET treatment occurred over the course of 5 days starting at 5 dpf. Fish were fed in accordance with their regular feed schedule throughout the treatment. At 10 dpf, fish were euthanized for either ultrasound shear wave elastography imaging, microcomputed tomography, confocal microscopy, or transmission election microscopy.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 430, 374, 448]]<|/det|>
+## Shear wave elastography (SWE)  
+
+<|ref|>text<|/ref|><|det|>[[110, 450, 883, 870]]<|/det|>
+The spinal tissue stiffness of the zebrafish was assessed using a SWE protocol (as described in Figure S1) programmed into a research ultrafast ultrasound system (Vantage 256, Verasonics Inc., Kirkland, WA, USA) with a linear ultrasound probe (L22- 14vX, 18MHz center frequency). Euthanized zebrafish were mounted ventrally in a \(35\mathrm{mm}\times 10\mathrm{mm}\) dish (Corning) filled with \(1.5\%\) low melt agarose (BioShop). The probe was positioned longitudinally, dorsal to the zebrafish, in contact with ultrasound gel applied to agarose- embedded samples. The SWE protocol for a single acquisition consisted of two steps: 1) the induction of an acoustic radiation force through a focused ultrasound beam at a push depth between \(4 - 9\mathrm{mm}\) aimed at the center of the zebrafish spine, resulting in a transient perturbation of the tissue and shear waves propagation; 2) an ultrafast Doppler sequence was created with coherent compound plane wave imaging consisting of 3 plane waves (range - 3 degrees to +3 degrees, 3- degree steps) with a pulse repetition frequency of \(28\mathrm{kHz}\) . Coherent compounding of the plane waves was applied using a sliding window method as in Kang et al. (window size 3, corresponding to the 3 different titled plane waves, and window step of 1), enabling a virtual framerate of \(28\mathrm{kHz}^{54}\) . Each acquisition lasted 10 milliseconds. Three acquisitions were collected per zebrafish. After scans were complete, the ultrafast imaging data was streamed to an internal network and then post- processed offline using MATLAB 2019a (The MathWorks Inc., Natick, MA, USA). Tissue velocity data was computed using a Doppler- based autocorrelation estimator, from which tissue velocity maps were reconstructed and presented as a space- time matrix showing shear waves propagation within a specific region of interest. The mean of these shear wave velocities was computed and used as surrogate for spine stiffness (derived from the shear modulus) \(^{55}\) .
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 90, 881, 316]]<|/det|>
+MicroCT scanning, reconstruction, and density analysisZebrafish were euthanized at 4mpf and fixed in \(10\%\) neutral- buffered formalin (Sigma- Aldrich). Fish were mounted in tubes using \(1\%\) agarose. Scanning was performed with a SkyScan 1275 microCT (Bruker, Kontich, Belgium) using \(50\mathrm{kV}\) and \(80\mu \mathrm{A}\) , sample rotation of \(180^{\circ}\) , image rotation steps of \(0.2^{\circ}\) , frame averaging of 10, exposure time of \(55\mathrm{ms}\) , camera binning of \(1\times 1\) , and using a pixel size of \(18\mathrm{um}\) . Projection images were reconstructed into cross- sections using SkyScan’s NRecon v.1.7.4.6 software (Bruker, Kontich, Belgium) in a range of attenuation coefficients 0–0.25, with a beam- hardening correction of \(40\%\) . The reconstructed images were stored as 16- bit TIFF images. Maximum Intensity Projections were generated (imageJ software) for tissue mineral density analysis. The line selection tool in ImageJ was then used to manually measure the mean gray value within each vertebral centrum, which can be used as a relative measure for tissue mineral density.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 333, 355, 350]]<|/det|>
+## Cobb angle statistical analysis  
+
+<|ref|>text<|/ref|><|det|>[[113, 352, 880, 472]]<|/det|>
+Lines were drawn parallel to the top and bottom most displaced vertebrae for each curve. The Cobb angle was then measured as the angle of intersection between lines drawn perpendicular to the original 2 lines<sup>9</sup>. Analysis was conducted using ImageJ<sup>56,57</sup>. Cobb angle measurements for lateral and dorsal curvatures were summed to obtain a combined Cobb angle measurement for each fish. Results were graphed and statistical significance was calculated using GraphPad Prism 10.1.1 (GraphPad Software).  
+
+<|ref|>text<|/ref|><|det|>[[112, 490, 867, 652]]<|/det|>
+Quantification of DCFH- DA staining, \(Tg(col2a1a::xbp1\delta - eGFP)\) and \(TgBAC(entpd5a::pkRed)\) Using ImageJ FIJI, regions of interest (ROI) were selected to measure the area and integrated density of the ROI. The mean grey value of the background was also measured. The corrected total fluorescence (CTF) of the ROI were calculated as followed: Integrated Density - (Area of ROI x Mean Grey Value of Background). CTF values were then calculated relative to wildtype controls by dividing CTF values to the average CTF wildtype value. Statistical analysis was performed using two- way ANOVA with Tukey’s multiple comparisons test or unpaired twotailed student’s \(t\) - test in GraphPad Prism version 10.1.1 (GraphPad Software).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 672, 366, 690]]<|/det|>
+## Notochord ECM measurements  
+
+<|ref|>text<|/ref|><|det|>[[112, 692, 872, 813]]<|/det|>
+TEM scans at \(6000\mathrm{x}\) and \(10000\mathrm{x}\) were selected for measurements. The ECM widths were taken with ImageJ and defined as the distance from the basal boundary of the notochord ECM to the basal side of the notochord sheath (yellow dash line, Fig. 4B- E). Two measurements were taken from a single scanned image: a local maximum width (the thickest area along the ECM) and a local minimum (thinnest area along the ECM). Local ratio was obtained from dividing the maximum width by its paired minimum width.  
+
+<|ref|>text<|/ref|><|det|>[[112, 831, 875, 872]]<|/det|>
+Data availability: Due to their large size, microCT imaging files are available upon request. All other data are available in the main text or the supplementary materials.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 90, 303, 108]]<|/det|>
+## Main Text References  
+
+<|ref|>text<|/ref|><|det|>[[80, 110, 875, 152]]<|/det|>
+1. Cheng, J. C. et al. Adolescent idiopathic scoliosis. Nature Reviews Disease Primers 1, 1-21 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[80, 152, 875, 191]]<|/det|>
+2. Wise, C. A. et al. The cartilage matrisome in adolescent idiopathic scoliosis. Bone Res 8, 1-13 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[80, 191, 876, 250]]<|/det|>
+3. Haller, G. et al. A polygenic burden of rare variants across extracellular matrix genes among individuals with adolescent idiopathic scoliosis. Human Molecular Genetics 25, 202-209 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[80, 250, 840, 310]]<|/det|>
+4. Karner, C. M., Long, F., Solnica-Krezel, L., Monk, K. R. & Gray, R. S. Gpr126/Adgrg6 deletion in cartilage models idiopathic scoliosis and pectus excavatum in mice. Human Molecular Genetics 24, 4365-4373 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[80, 310, 844, 370]]<|/det|>
+5. Kou, I. et al. Genome-wide association study identifies 14 previously unreported susceptibility loci for adolescent idiopathic scoliosis in Japanese. Nat Commun 10, 3685 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[80, 370, 875, 410]]<|/det|>
+6. Bagnat, M. & Gray, R. S. Development of a straight vertebrate body axis. Development 147, (2020).  
+
+<|ref|>text<|/ref|><|det|>[[80, 410, 866, 450]]<|/det|>
+7. Boswell, C. W. & Ciruna, B. Understanding Idiopathic Scoliosis: A New Zebrafish School of Thought. Trends in Genetics 33, 183-196 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[80, 450, 852, 490]]<|/det|>
+8. Grimes, D. T. et al. Zebrafish models of idiopathic scoliosis link cerebrospinal fluid flow defects to spine curvature. Science 352, 1341-1344 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[80, 490, 870, 550]]<|/det|>
+9. Van Gennip, J. L. M., Boswell, C. W. & Ciruna, B. Neuroinflammatory signals drive spinal curve formation in zebrafish models of idiopathic scoliosis. Science Advances 4, eaav1781 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[80, 550, 825, 610]]<|/det|>
+10. Rose, C. D. et al. SCO-Spondin Defects and Neuroinflammation Are Conserved Mechanisms Driving Spinal Deformity across Genetic Models of Idiopathic Scoliosis. Current Biology 30, 2363-2373.e6 (2020).  
+
+<|ref|>text<|/ref|><|det|>[[80, 610, 840, 670]]<|/det|>
+11. Djebar, M. et al. Astrogliosis And Neuroinflammation Underlie Scoliosis Upon Cilia Dysfunction. 2024.02.22.581530 Preprint at https://doi.org/10.1101/2024.02.22.581530 (2024).  
+
+<|ref|>text<|/ref|><|det|>[[80, 670, 860, 710]]<|/det|>
+12. Henke, K. et al. Genetic Screen for Postembryonic Development in the Zebrafish (Danio rerio): Dominant Mutations Affecting Adult Form. Genetics 207, 609-623 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[80, 710, 866, 750]]<|/det|>
+13. Soh, N. Recent advances in fluorescent probes for the detection of reactive oxygen species. Anal Bioanal Chem 386, 532-543 (2006).  
+
+<|ref|>text<|/ref|><|det|>[[80, 750, 870, 790]]<|/det|>
+14. Wopat, S. et al. Spine Patterning Is Guided by Segmentation of the Notochord Sheath. Cell Reports 22, 2026-2038 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[80, 790, 876, 830]]<|/det|>
+15. Zhang, Z. et al. Redox signaling and unfolded protein response coordinate cell fate decisions under ER stress. Redox Biology 25, 101047 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[80, 830, 839, 870]]<|/det|>
+16. Li, J. et al. A transgenic zebrafish model for monitoring xbp1 splicing and endoplasmic reticulum stress in vivo. Mechanisms of Development 137, 33-44 (2015).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 90, 884, 875]]<|/det|>
+17. Dale, R. M. & Topczewski, J. Identification of an evolutionarily conserved regulatory element of the zebrafish col2a1a gene. Developmental Biology 357, 518–531 (2011).  
+18. Kague, E. et al. 3D assessment of intervertebral disc degeneration in zebrafish identifies changes in bone density that prime disc disease. Bone Res 9, 39 (2021).  
+19. Gistelinck, C. et al. Zebrafish type I collagen mutants faithfully recapitulate human type I collagenopathies. Proceedings of the National Academy of Sciences 115, E8037–E8046 (2018).  
+20. Zitnay, J. L. et al. Molecular level detection and localization of mechanical damage in collagen enabled by collagen hybridizing peptides. Nat Commun 8, 14913 (2017).  
+21. Langlais, T. et al. Shear-wave elastography can evaluate annulus fibrosis alteration in adolescent scoliosis. Eur Radiol 28, 2830–2837 (2018).  
+22. Vergari, C. et al. Shear wave elastography of lumbar annulus fibrosis in adolescent idiopathic scoliosis before and after surgical intervention. European Radiology 30, 1980–1985 (2020).  
+23. Villemain, O. et al. Ultrafast Ultrasound Imaging in Pediatric and Adult Cardiology: Techniques, Applications, and Perspectives. JACC: Cardiovascular Imaging 13, 1771–1791 (2020).  
+24. Villalobos Lizardi, J. C. et al. A guide for assessment of myocardial stiffness in health and disease. Nat Cardiovasc Res 1, 8–22 (2022).  
+25. Jussila, M., Boswell, C. W., Griffiths, N. W., Pumputis, P. G. & Ciruna, B. Live imaging and conditional disruption of native PCP activity using endogenously tagged zebrafish sfGFP-Vangl2. Nat Commun 13, 5598 (2022).  
+26. Meyer-Miner, A., Van Gennip, J. L. M., Henke, K., Harris, M. P. & Ciruna, B. Resolving primary pathomechanisms driving idiopathic-like spinal curvature using a new katnb1 scoliosis model. iScience 25, 105028 (2022).  
+27. Trueta, J. in chapter 35 The Aetio-Pathology and Management of Scoliosis. in Studies of the Development and Decay of the Human Frame 306–315 (William Heinemann Medical Books LTD., 1968).  
+28. Harrington, P. R. The etiology of idiopathic scoliosis. Clin Orthop Relat Res 17–25 (1977).  
+29. de Reuver, S. et al. Anterior lengthening in scoliosis occurs only in the disc and is similar in different types of scoliosis. Spine J 20, 1653–1658 (2020).  
+30. Will, R. E., Stokes, I. A., Qiu, X., Walker, M. R. & Sanders, J. O. Cobb angle progression in adolescent scoliosis begins at the intervertebral disc. Spine (Phila Pa 1976) 34, 2782–2786 (2009).  
+31. Heidari, B., FitzPatrick, D., Synnott, K. & McCormack, D. Modelling of annulus fibrosis imbalance as an aetiological factor in adolescent idiopathic scoliosis. Clin Biomech (Bristol, Avon) 19, 217–224 (2004).  
+32. Duance, V. C. et al. Changes in collagen cross-linking in degenerative disc disease and scoliosis. Spine (Phila Pa 1976) 23, 2545–2551 (1998).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[67, 90, 879, 900]]<|/det|>
+1 33. Taylor, T. K., Ghosh, P. & Bushell, G. R. The contribution of the intervertebral disk to the scoliotic deformity. Clin Orthop Relat Res 79- 90 (1981). 2 34. Yu, J., Fairbank, J. C. T., Roberts, S. & Urban, J. P. G. The elastic fiber network of the anulus fibrosus of the normal and scoliotic human intervertebral disc. Spine (Phila Pa 1976) 5 30, 1815- 1820 (2005). 6 35. Cheng, T., Einarsdottir, E., Kere, J. & Gerdhem, P. Idiopathic scoliosis: a systematic review and meta-analysis of heritability. EFORT Open Rev 7, 414- 421 (2022). 7 36. Savage, M. K. Seismic anisotropy and mantle deformation: What have we learned from shear wave splitting? Reviews of Geophysics 37, 65- 106 (1999). 8 37. Villemain, O. et al. Myocardial Stiffness Evaluation Using Noninvasive Shear Wave Imaging in Healthy and Hypertrophic Cardiomyopathic Adults. JACC Cardiovasc Imaging 12, 1135- 1145 (2019). 9 38. Bostick, J. W., Schonhoff, A. M. & Mazmanian, S. K. Gut microbiome-mediated regulation of neuroinflammation. Curr Opin Immunol 76, 102177 (2022). 10 39. Foo, J., Bellot, G., Pervaiz, S. & Alonso, S. Mitochondria- mediated oxidative stress during viral infection. Trends in Microbiology 30, 679- 692 (2022). 11 40. Peters, A., Nawrot, T. S. & Baccarelli, A. A. Hallmarks of environmental insults. Cell 184, 1455- 1468 (2021). 12 41. Homans, J. F. et al. Scoliosis in association with the 22q11.2 deletion syndrome: an observational study. Arch Dis Child 104, 19- 24 (2019). 13 42. Menghi, M. et al. Neuroinflammation and Oxidative Stress in Individuals Affected by DiGeorge Syndrome. Int J Mol Sci 24, 4242 (2023). 14 43. Huang, T. J., Lubicky, J. P. & Hammerberg, K. W. Scoliosis in Rett syndrome. Orthop Rev 23, 931- 937 (1994). 15 44. Pecorelli, A., Cervellati, C., Hayek, J. & Valacchi, G. OxInflammation in Rett syndrome. Int J Biochem Cell Biol 81, 246- 253 (2016). 16 45. Jagtap, S. et al. Aberrant mitochondrial function in patient- derived neural cells from CDKL5 deficiency disorder and Rett syndrome. Hum Mol Genet 28, 3625- 3636 (2019). 17 46. Mangatt, M. et al. Prevalence and onset of comorbidities in the CDKL5 disorder differ from Rett syndrome. Orphanet J Rare Dis 11, 39 (2016). 18 47. Retta, S. F. & Glading, A. J. Oxidative stress and inflammation in cerebral cavernous malformation disease pathogenesis: Two sides of the same coin. Int J Biochem Cell Biol 81, 254- 270 (2016). 19 48. Shenkar, R. et al. EXCEPTIONAL AGGRESSIVENESS OF CEREBRAL CAVERNOUS MALFORMATION DISEASE ASSOCIATED WITH PDCD10 MUTATIONS. Genet Med 17, 188- 196 (2015). 20 49. Kossack, M. E. & Draper, B. W. Genetic regulation of sex determination and maintenance in zebrafish (Danio rerio). Curr Top Dev Biol 134, 119- 149 (2019). 21 50. Kwan, K. M. et al. The Tol2kit: A multisite gateway- based construction kit for Tol2 transposon transgenesis constructs. Developmental Dynamics 236, 3088- 3099 (2007).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[67, 88, 880, 410]]<|/det|>
+51. Rebello, D. et al. COL11A2 as a candidate gene for vertebral malformations and congenital scoliosis. Hum Mol Genet 32, 2913-2928 (2023).  
+52. Gasilov, S. et al. Hard X-ray imaging and tomography at the Biomedical Imaging and Therapy beamlines of Canadian Light Source. J Synchrotron Rad 31, (2024).  
+53. Faragó, T. et al. Tofu: a fast, versatile and user-friendly image processing toolkit for computed tomography. J Synchrotron Rad 29, 916-927 (2022).  
+54. Kang, J., Jang, W. S. & Yoo, Y. High PRF ultrafast sliding compound doppler imaging: fully qualitative and quantitative analysis of blood flow. Phys Med Biol 63, 045004 (2018).  
+55. Taljanovic, M. S. et al. Shear-Wave Elastography: Basic Physics and Musculoskeletal Applications. Radiographics 37, 855-870 (2017).  
+56. Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9, 671-675 (2012).  
+57. Schindelin, J., Rueden, C. T., Hiner, M. C. & Eliceiri, K. W. The ImageJ ecosystem: An open platform for biomedical image analysis. Mol Reprod Dev 82, 518-529 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[66, 390, 872, 469]]<|/det|>
+58. Hu, W. F. et al. Katanin p80 Regulates Human Cortical Development by Limiting Centriole and Cilia Number. Neuron 84, 1240-1257 (2014).  
+59. Hur, M. et al. MicroCT-based phenomics in the zebrafish skeleton reveals virtues of deep phenotyping in a distributed organ system. eLife 6, e26014 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[66, 508, 870, 686]]<|/det|>
+Acknowledgments: We gratefully acknowledge the SickKids' Nanoscale Biomedical Imaging Facility for assistance with TEM sample preparation and imaging, the SickKids' Zebrafish Facility technicians for excellent zebrafish care, Nigel Griffiths for technical assistance, and Dr. Ronald Kwon support in microCT TMD analyses. Part of the research described in this paper was performed at the Canadian Light Source, a national research facility of the University of Saskatchewan, which is supported by the Canada Foundation for Innovation (CFI), the Natural Sciences and Engineering Research Council (NSERC), the National Research Council (NRC), the Canadian Institutes of Health Research (CIHR), the Government of Saskatchewan, and the University of Saskatchewan.  
+
+<|ref|>text<|/ref|><|det|>[[66, 708, 866, 870]]<|/det|>
+The following grants supported this work: Canadian Institutes of Health Research Foundation grant FDN- 167285 (B.C.) Canada Research Chair Program (B.C.) Canadian Institute of Health and Research Canada Graduate Scholarship FBD- 181495 (P.G.P.) Natural Sciences and Engineering Research Council of Canada (NSERC) grant RGPIN- 2021- 03539 (O.V.) Canadian Institutes of Health Research operating grant 148683 (B.F.E.) The Hospital for Sick Children Research Institute funding (B.C., O.V., D.E.L.)
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[71, 90, 866, 250]]<|/det|>
+1 Authors contributions: 2 Conceptualization: B.C. and O.V. conceptualized, administered, and supervised the project. 3 P.G.P., R.X., J.G., A.P., V.E., J.V.G. performed experiments, data analysis, and visualized the 4 data. P.G.P., R.X., J.G., A.P., V.E., J.V.G., B.F.E., N.F., J.B., O.V., B.C. contributed to the 5 methodology of the project. N.F., J.B. created the software for zebrafish SWE analysis. B.C., 6 O.V., P.G.P., B.F.E., D.E.L. acquired funding. P.G.P., R.X., J.G., O.V., B.C. wrote the original 7 draft of the paper. P.G.P., R.X., J.G., A.P., V.E., J.V.G., B.F.E., N.F., J.B., D.E.L., O.V., B.C. 8 reviewed and edited the paper.  
+
+<|ref|>text<|/ref|><|det|>[[113, 270, 722, 290]]<|/det|>
+Competing Interests: Authors declare that they have no competing interests  
+
+<|ref|>text<|/ref|><|det|>[[113, 310, 377, 327]]<|/det|>
+Materials and correspondence:  
+
+<|ref|>text<|/ref|><|det|>[[113, 330, 512, 348]]<|/det|>
+Correspond to Brian Ciruna (ciruna@sickkids.ca).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[111, 125, 884, 870]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 90, 886, 526]]<|/det|>
+Fig. 1: IVD redox imbalance and ER stress precede idiopathic-like scoliosis. a, b, Lateral and dorsal wholemount images of 10 dpf (6 mm SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva. Red boxes indicate regions imaged by confocal microscopy. Scale bars \(= 1 \mathrm{mm}\) . c-f, Confocal images of DCFH-DA oxidation staining of the dorsal telecephalon (c, d) and lateral trunk (e, f) of 10 dpf (6 mm SL) wildtype (c, e) and \(sspo^{dmh4 / + }\) mutant (d, f) zebrafish larva. Scale bars \(= 100 \mu \mathrm{m}\) . Lookup tables arbitrary units (A.U.) are 0 to 145. \(\mathrm{DT} =\) dorsal telecephalon; IVD = intervertebral domain; DA = dorsal aorta; CC = central canal; SC = spinal cord. Arrowheads indicate increased oxidative state along developing intervertebral domains. g-k, Relative fluorescence quantification of DCFH-DA oxidation staining in dorsal telecephalon (g; \(\mathrm{p} = 0.0160\) , \(\mathrm{N} = 3\) , \(\mathrm{n} = 5\) for each genotype), intervertebral domains (h; \(\mathrm{p} = 0.0042\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), dorsal aorta (i; \(\mathrm{p} = 0.0098\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), central canal (j; \(\mathrm{p} = 0.0035\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), and spinal cord (k; \(\mathrm{p} = 0.0032\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 10\) \(sspo^{dmh4 / + }\) ), all showing significant increased oxidative state in \(sspo^{dmh4 / + }\) mutant zebrafish larva. Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test. l-n, Confocal images of \(Tg(entpd5a::pkRed)\) reporter expression within mineralizing notochord segments and DCFH-DA staining, demonstrating greater oxidation in and around the intervertebral domains (IVD, arrowhead) in 10 dpf (6 mm SL) \(sspo^{dmh4 / + }\) larva. Scale bar \(= 50 \mu \mathrm{m}\) . o-p, Confocal images of \(Tg(col2al1::xbp1\delta -gfp)\) and \(Tg(entpd5a::pkRed)\) reporter expression in 10 dpf (6 mm SL) wildtype (o) and \(sspo^{dmh4 / + }\) (p) zebrafish larva, demonstrating an elevated ER stress response within the intervertebral domains of \(sspo^{dmh4 / + }\) mutants, prior to the onset of scoliosis. Scale bar \(= 100 \mu \mathrm{m}\) . q, Relative fluorescence quantification of \(Tg(col2al1::xbp1\delta -gfp)\) reporter expression within intervertebral domains ( \(\mathrm{p} = 0.0127\) , \(\mathrm{N} = 4\) , \(\mathrm{n} = 11\) for each genotype), demonstrating elevated ER stress in \(sspo^{dmh4 / + }\) mutants ( \(\mathrm{p} \leq 0.05\) ). Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test ( \(\ast \mathrm{p} \leq 0.05\) , \(\ast \ast \mathrm{p} \leq 0.01\) ).
+
+<--- Page Split --->
+<|ref|>image_caption<|/ref|><|det|>[[68, 92, 189, 108]]<|/det|>
+<center>Figure 2</center> 
+
+<|ref|>image<|/ref|><|det|>[[123, 111, 900, 895]]<|/det|>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 106, 666, 125]]<|/det|>
+## Fig. 2: Scoliotic fish exhibit intervertebral ligament deformations.  
+
+<|ref|>text<|/ref|><|det|>[[112, 124, 883, 320]]<|/det|>
+a-g, High-resolution synchrotron-based X-ray micro-computed tomography (micro-CT) imaging of iodine contrast- stained wildtype (a-d) and \(sspo^{dmh4 / + }\) mutant (e-g) juvenile fish at 21 dpf (10 mm SL). Representative sections from coronal (a-c, e-f) and transverse (d, g) planes are shown. Dashed boxes (a, e) indicate featured intervertebral segments. Arrowheads (b-d, f-h) indicate the intervertebral ligament, which loops outward toward neighbouring musculature in wildtype animals (N = 3 fish, n = 13/13 IVDs), but often exhibited structural deformations and even inverted orientations in \(sspo^{dmh4 / + }\) mutants (N = 8 fish; n = 22/29 IVDs). Arrows (h) highlight sites of \(sspo^{dmh4 / + }\) IVL deformation in the transverse plane. Sequential images (3.6 μm step) through the coronal plane of wildtype (c-c'') and \(sspo^{dmh4 / + }\) (g-g'') zebrafish highlight the highly deformable structure of \(sspo^{dmh4 / + }\) mutant IVLs. Scale bars = 20 μm (c, g); 50 μm (b, d, f, h); 100 μm (a, e). IVL = intervertebral ligament; AF = annulus fibrosus; NP = nucleus pulposus.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 115, 777, 890]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 90, 884, 120]]<|/det|>
+1 Fig. 3: Elevated spine stiffness is observed prior to scoliosis onset, positively correlates with curve severity, and is fully suppressed by antioxidant therapy.  
+
+<|ref|>text<|/ref|><|det|>[[70, 120, 884, 405]]<|/det|>
+2 a, b, Lateral wholemount images of wildtype and scoliotic \(sspo^{dmh4 / + }\) mutant zebrafish at 16 dpf (a, SL = 7.1 +/- 0.2 mm) and 21 dpf (b, SL = 8.9 +/- 0.6 mm). Scale bar = 2 mm. c, d, Quantification of shear wave velocity measured by SWE along the spine of 16 dpf (c) and 21 dpf (d) zebrafish, demonstrating significantly greater shear wave velocity and thus spinal stiffness in \(sspo^{dmh4 / + }\) mutant animals. c, p = 0.0040, N = 3, n = 15 for each genotype. d, p = 0.0027, N = 3, n = 15 for each genotype. Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test. e, Positive correlation between scoliosis severity (Total Cobb angle) and spinal stiffness (shear wave velocity) in 21 dpf \(sspo^{dmh4 / + }\) zebrafish, post-curve onset (N = 3, n = 20 for each genotype). Data was analyzed by a simple linear regression (R = 0.831, R² = 0.690, p = 5.73 x 10⁻⁶). f, Quantification of shear wave velocity measured by SWE along the spine of 10 dpf (SL = 4.7 +/- 0.4 mm) demonstrating significant elevation of spine stiffness in \(sspo^{dmh4 / + }\) mutant zebrafish, prior to scoliosis (p < 0.0001, N = 3, n = 15 for each genotype and treatment). Antioxidant treatment (500μM NACET) fully suppressed spine stiffness in \(sspo^{dmh4 / + }\) mutant zebrafish to wildtype levels. Box plots show all data points, median, and min to max whiskers. Data were analyzed by two-way ANOVA and Tukey test (ns: non-significant (p > 0.05), **p ≤ 0.01, ****p ≤ 0.0001).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[135, 110, 825, 911]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 90, 886, 393]]<|/det|>
+Fig. 4: Oxidative stress-induced ECM matrix defects precede scoliosis onset. a, Schematic of zebrafish notochord development at 10 dpf (5mm SL) showing cartilaginous ECM domains (IVD), mineralized domains (vertebra), notochord vacuole and notochord sheath cells. b, c, Representative transmission electron micrographs of the collagenous ECM in developing IVDs of 10 dpf (5mm SL) wildtype (b) and \(sspo^{dmh4 / + }\) (c) larvae. d, e, Antioxidant treatment (500μM NACET) had no effect on wildtype ECM development (d) but suppressed irregular collagen ECM thickness and fiber orientation phenotypes in \(sspo^{dmh4 / + }\) mutant animals (e). f-h, Quantification of matrix thickness showed no difference in local minimum ECM widths (g), but significant differences between wildtype (N = 8, n = 36) and \(sspo^{dmh4 / + }\) mutants (N = 9, n = 47) in maximum local thickness (f) and local thickness ratios (h). The maximum matrix thickness and ratio of \(sspo^{dmh4 / + }\) mutant IVDs (N = 7, n = 35) were restored to wildtype level upon NACET treatment, while NACET treatment had no effect on wildtype larval IVDs (N = 6, n = 22). Box plots show all data points, median, and min to max whiskers. Points of the same colour belong to the same fish, demonstrating variations in phenotype severity among IVD segments of individual \(sspo^{dmh4 / + }\) fish. Data were analyzed by two-way ANOVA and Tukey test (ns: non-significant p > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[75, 95, 88, 106]]<|/det|>
+1
+
+<|ref|>text<|/ref|><|det|>[[75, 115, 88, 125]]<|/det|>
+2
+
+<|ref|>text<|/ref|><|det|>[[75, 135, 88, 145]]<|/det|>
+3
+
+<|ref|>text<|/ref|><|det|>[[75, 155, 88, 165]]<|/det|>
+4
+
+<|ref|>text<|/ref|><|det|>[[75, 175, 88, 185]]<|/det|>
+5
+
+<|ref|>text<|/ref|><|det|>[[75, 194, 88, 204]]<|/det|>
+6
+
+<|ref|>text<|/ref|><|det|>[[75, 214, 88, 224]]<|/det|>
+7
+
+<|ref|>text<|/ref|><|det|>[[75, 234, 818, 247]]<|/det|>
+**8** **Extended Data Table 1: Oligonucleotide sequences used for genotyping and cloning.**
+
+<|ref|>table<|/ref|><|det|>[[117, 247, 880, 480]]<|/det|>
+
+<table><tr><td>Oligonucleotides</td><td>Source</td></tr><tr><td>Forward HRM primer for sspo Exon 17:<br>5'-GGAGCTGTGAGCCGGAC-3'</td><td>10</td></tr><tr><td>Reverse HRM primer for sspo Exon<br>17:5'-TCCTGCGCCAGTCCATCA3'</td><td>10</td></tr><tr><td>Forward sequencing primer for katnb1<sup>mh102</sup>:<br>5'-ACACAGACTTCATGTTTCTGACAGGC-3'</td><td>58</td></tr><tr><td>Reverse sequencing primer for katnb1<sup>mh102</sup>:<br>5'-TGAGCTCAGACACAACTGAGGGTT-3'</td><td>58</td></tr><tr><td>(attb1) Myc-xbp1-eGFP Forward for middle entry<br>5'-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCGCCACCATG<br>GAGCAAAAGCTCATTTCTG-3'</td><td>This<br>paper</td></tr><tr><td>(attb2) Myc-xbp1-eGFP-taa Reverse for middle entry<br>5'-GGGGACCACTTTGTACAAGAAAGCTGGGTGTTACTTGTACAGCTCGTCACTGC-3'</td><td>This<br>paper</td></tr></table>
+
+<|ref|>text<|/ref|><|det|>[[75, 470, 88, 480]]<|/det|>
+9
+
+<|ref|>text<|/ref|><|det|>[[75, 490, 88, 500]]<|/det|>
+10
+
+<|ref|>text<|/ref|><|det|>[[75, 510, 88, 520]]<|/det|>
+11
+
+<|ref|>text<|/ref|><|det|>[[75, 530, 88, 540]]<|/det|>
+12
+
+<|ref|>text<|/ref|><|det|>[[75, 550, 88, 560]]<|/det|>
+13
+
+<|ref|>text<|/ref|><|det|>[[75, 569, 88, 579]]<|/det|>
+14
+
+<|ref|>text<|/ref|><|det|>[[75, 589, 88, 599]]<|/det|>
+15
+
+<|ref|>text<|/ref|><|det|>[[75, 608, 88, 618]]<|/det|>
+16
+
+<|ref|>text<|/ref|><|det|>[[75, 628, 88, 638]]<|/det|>
+17
+
+<|ref|>text<|/ref|><|det|>[[75, 647, 88, 657]]<|/det|>
+18
+
+<|ref|>text<|/ref|><|det|>[[75, 667, 88, 677]]<|/det|>
+19
+
+<|ref|>text<|/ref|><|det|>[[75, 686, 88, 696]]<|/det|>
+20
+
+<|ref|>text<|/ref|><|det|>[[75, 705, 88, 715]]<|/det|>
+21
+
+<|ref|>text<|/ref|><|det|>[[75, 725, 88, 735]]<|/det|>
+22
+
+<|ref|>text<|/ref|><|det|>[[75, 744, 88, 754]]<|/det|>
+23
+
+<|ref|>text<|/ref|><|det|>[[75, 763, 88, 773]]<|/det|>
+24
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[113, 85, 844, 352]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[113, 372, 836, 409]]<|/det|>
+Extended Data Fig. 1: No change in oxidative state within trunk myotome of \(sspo^{dmh4 / + }\) mutant zebrafish.  
+
+<|ref|>text<|/ref|><|det|>[[113, 409, 884, 523]]<|/det|>
+a, b, Confocal images of DCFH- DA oxidation staining of lateral trunk myotomes of \(10\mathrm{dpf}(6\mathrm{mm}\) SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva. Scale bars \(= 100\mu \mathrm{m}\) . Lookup tables arbitrary units (A.U.) are 0 to 145. c, Relative fluorescence quantification of DCFH- DA oxidation staining shows no significant difference \((\mathrm{p} > 0.05)\) between wildtype and \(sspo^{dmh4 / + }\) larva (10 dpf, \(6\mathrm{mmSL}\) ). \(\mathrm{N} = 4\) \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 9\) \(sspo^{dmh4 / + }\) . ns: non- significant \((\mathrm{p} > 0.05)\) . Box plots show all data points, median, and min to max whiskers. Data were analyzed by unpaired Student's \(t\) test.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[130, 95, 680, 201]]<|/det|>  
+
+<|ref|>image<|/ref|><|det|>[[130, 201, 680, 781]]<|/det|>  
+
+<|ref|>text<|/ref|><|det|>[[113, 787, 857, 821]]<|/det|>
+Extended Data Fig. 2: Transient ER stress response within the IVDs of \(sspo^{dmh4 / +}\) mutant larvae.  
+
+<|ref|>text<|/ref|><|det|>[[113, 822, 883, 893]]<|/det|>
+a, Schematic of the ER stress reporter \(Tg(col2a1a::xbp1\delta - gfp)\) . Under normal conditions, splicing does not occur within the \(xbp1\delta - gfp\) transcript, eGFP coding sequence is out of frame, and translation is prematurely terminated. During ER stress, IRE1 splices the \(xbp1\delta - gfp\) transcript leading to a frameshift and translation of a spliced and functional eGFP fusion protein. Created
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 88, 884, 336]]<|/det|>
+1 with BioRender.com. b- e, Imaging and (f) quantification of \(Tg(col2a1l::xbp1\delta - gfp)\) reporter 2 expression within the IVDs of 10 dpf (5 mm SL) zebrafish larva. \(sspo^{dmh4 / + }\) mutants (c) exhibit a 3 significantly increased ER stress response over wildtype (b) animals. Treatment with antioxidant 4 (500 \(\mu \mathrm{M}\) NACET) from 5 to 10 dpf results in a significant reduction in ER stress reporter 5 expression \((\mathrm{p} = 0.0440)\) within the IVD of \(sspo^{dmh4 / + }\) mutant zebrafish (e) larva, similar to wildtype 6 levels (d). Scale bar \(= 100 \mu \mathrm{m}\) . Box plots show all data points, median, and min to max whiskers. 7 \(\mathrm{N} = 4\) , \(\mathrm{n} = 10\) WT, \(\mathrm{n} = 12\) \(sspo^{dmh4 / + }\) , \(\mathrm{n} = 11\) WT+NACET, \(\mathrm{n} = 11\) \(sspo^{dmh4 / + }\) +NACET. Data were 8 analyzed by two- way ANOVA and Tukey test. ns: non- significant, \(^{*}\mathrm{p}\leq 0.05\) , \(^{**}p\leq 0.01\) . g, h, 9 Imaging and (i) quantification of \(Tg(col2a1l::xbp1\delta - gfp)\) reporter expression in 14 dpf (6.5 mm 10 SL) wildtype (g) and \(sspo^{dmh4 / + }\) mutant (h) zebrafish larva. In contrast to earlier stages of 11 development, no significant difference in ER stress reporter expression is observed in the 12 intervertebral domains of \(sspo^{dmh4 / + }\) mutants, after the onset of scoliosis. Scale bar \(= 100 \mu \mathrm{m}\) . \(\mathrm{N} =\) 13 \(3\) , \(\mathrm{n} = 7\) WT, \(\mathrm{n} = 11\) \(sspo^{dmh4 / + }\) . Box plots show all data points, median, and min to max whiskers. 14 Data were analyzed by unpaired Student's \(t\) test. 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[114, 120, 708, 300]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[114, 110, 260, 124]]<|/det|>
+<center>a Wildtype (10 mm) </center>  
+
+<|ref|>image<|/ref|><|det|>[[114, 303, 708, 465]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[114, 485, 707, 503]]<|/det|>
+<center>Extended Data Fig. 3: Spinal curvature in 21dpf \(sspo^{dmh4 / + }\) mutant fish. </center>  
+
+<|ref|>text<|/ref|><|det|>[[114, 503, 883, 556]]<|/det|>
+a, b, Lateral wholemount images of 21 dpf (10 mm SL) wildtype (a) and \(sspo^{dmh4 / + }\) mutant (b) zebrafish larva, indicating regions selected for high-resolution synchrotron- based X- ray micro- computed tomography (micro- CT) imaging (red box). Scale bars = 1 mm.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[128, 137, 875, 470]]<|/det|>  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 486, 815, 523]]<|/det|>
+## Extended Data Fig. 4: \(sspo^{dmh4 / +}\) IVDs exhibit collagenous ECM damage and matrix remodeling.  
+
+<|ref|>text<|/ref|><|det|>[[111, 521, 884, 767]]<|/det|>
+a, b, Schematic representing the structure and organization of zebrafish IVDs at 21 dpf (10mm SL) in the sagittal (a) and transverse (b) planes. VB = vertebra body, NV=notochord vacuole, NS = notochord sheath cell, CE = collagen type II and elastin matrix, DC = dense collagen type I matrix, CF =collagen type I bundle fibre, NP = nucleus pulposus, IVL = intervertebral ligament, AF = annulus fibrosus. c-f, Confocal images of CHP- stained vertebral and IVD segments at 21 dpf (10mm SL). c, d, CHP staining in wildtype animals demonstrates normal remodelling of the collagen ECM matrix within the NP (arrowheads) and AF (arrows), as seen in sagittal (c, \(\mathrm{n} = 12\) ; scale bar \(= 20\mu \mathrm{m}\) ) and transverse (d, \(\mathrm{n} = 13\) ; scale bar \(= 10\mu \mathrm{m}\) ) planes. e, f, In contrast, \(sspo^{dmh4 / + }\) mutant animals exhibit ectopic CHP staining within NP (arrowheads) and AF (arrows) structures, as seen in sagittal (e, \(\mathrm{n} = 12\) , scale bar \(= 20\mu \mathrm{m}\) ) and transverse (f, \(\mathrm{n} = 15\) , scale bar \(= 10\mu \mathrm{m}\) ) planes, indicating significant damage to the collagenous ECM matrix of \(sspo^{dmh4 / + }\) mutant IVDs. g, Schematic depicting the mechanism by which collagen hybridizing peptide (CHP, green probe) integrates into open regions (yellow stars) of collagen fibril triple helices (purple) to report on collagen ECM damage and remodeling.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[113, 140, 728, 810]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 821, 883, 892]]<|/det|>
+<center>Extended Data Fig. 5: Shear wave elastography (SWE) of the zebrafish spine. a, Photograph of SWE experimental apparatus setup. b, Schematic of SWE mechanics in zebrafish, imaging with ultrafast ultrasound at 28000 frames per second (fps). High frequency linear ultrasound probe induces an acoustic radiation force (ARF) into the center of the zebrafish spine, </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 88, 884, 316]]<|/det|>
+1 and two resulting shear wave velocities (rostral direction and caudal direction) are measured in a 2 10 millisecond acquisition. c, Data processing procedure for every acquisition using a custom 3 MATLAB program. 1. Zebrafish region of interest identified. 2. Shear wave propagation 4 visualized over every video frame and averaged along identified axis of interest along zebrafish 5 spine. 3. Space- time map retrieved for axis of interest and slopes of wave transforms estimated to 6 determine SWV. d, Quantification of shear wave velocity measured by SWE for 11 dpf wildtype 7 zebrafish testing different AFR push locations, demonstrating that they do not significantly alter 8 measured SWV (p > 0.05, n = 10 location 1, n = 12 location 2, n = 8 location 3). e, Quantification 9 of shear wave velocity measured by SWE for 11 dpf wildtype zebrafish testing low melt agarose 10 concentrations for mounting zebrafish, demonstrating that low melt agarose concentration does 11 not impact SWV measurements. (p > 0.05, n = 8 for both low melt agarose concentrations). Box 12 plots show all data points, median, and min to max whiskers. Data were analyzed by either one- 13 way ANOVA and Tukey post- hoc test (d) or unpaired Student's t test (e).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 180, 858, 528]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 550, 839, 586]]<|/det|>
+<center>Extended Data Fig. 6: No significant difference in tissue mineral density is observed in vertebral centra of \(sspo^{dmh4 / +}\) and wildtype sibling zebrafish. </center>  
+
+<|ref|>text<|/ref|><|det|>[[112, 585, 884, 740]]<|/det|>
+a, b, MicroCT scans of 4 mpf fish were acquired (A, \(\mathrm{n} = 12\) wildtype; B, \(\mathrm{n} = 16\) \(sspo^{dmh4 / +}\) ). c, Quantification of relative centrum density was manually performed on two- dimensional maximum intensity projections of reconstructed microCT scans. Multiple unpaired t test with Welch correction was performed for each vertebra. There was no significant difference in centrum density at any vertebral position (adjusted p- values \(>0.05\) ). Error bars indicate standard deviation. d, This manual method for quantification of relative density was validated by comparing to relative density values obtained from a previously established software, FishCuT59. Manually measured mean gray values of vertebral centra are closely correlated to values obtained using FishCuT ( \(\mathrm{R} = 0.868\) , \(\mathrm{R}^2 = 0.754\) , \(\mathrm{p} = 2.23 \times 10^{- 18}\) ).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[125, 200, 876, 420]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 432, 848, 468]]<|/det|>
+<center>Extended Data Fig. 7: Increased spine stiffness is observed across diverse zebrafish AIS models. </center>  
+
+<|ref|>text<|/ref|><|det|>[[111, 468, 884, 763]]<|/det|>
+a, Lateral wholemount images and quantification of shear wave velocity measured by SWE along the spine of \(20\mathrm{dpf}(8.18 + / - 0.46\mathrm{mm}\mathrm{SL})\) scoliotic katnb1mh102 mutant zebrafish, and katnb1mh102/sibling controls. katnb1mh102 mutant zebrafish have disrupted katanin microtubule severing activity that results in ependymal cell and choroid plexus cilia differentiation defects, CSF flow defects and oxidative stress responses, with little disruption to Reissner's fiber polymerization26. Scoliotic katnb1mh102 mutant zebrafish also demonstrate significantly elevated shear wave velocity and thus spine stiffness, compared to their control katnb1mh102/siblings \((\mathrm{p} = 0.0047\) \(\mathrm{N} = 3\) \(\mathrm{n} = 11\) for each genotype). Data were analyzed by unpaired Student's \(t\) test. Scale bar \(= 2\mathrm{mm}\) . b, Lateral wholemount images and quantification of shear wave velocity measured by SWE along the spine of \(10\mathrm{dpf}(5.10 + / - 0.48\mathrm{mm}\mathrm{SL})\) scoliotic vangl2sGFP;foxjla::icre;zGrad mutant zebrafish, and their non- scoliotic vangl2sGFP;foxjla::icre control siblings. Conditional degradation of the planar cell polarity effector protein vangl2 in foxjla- positive motile ciliated lineages (vangl2sGFP;foxjla::icre;zGrad), results in loss of brain ependymal cell cilia, CSF flow defects, and ectopic accumulations of Reissner's fiber25. Scoliotic vangl2sGFP;foxjla::icre;zGrad mutants also demonstrate significantly greater shear wave velocity and thus spine stiffness, compared to their control siblings \((\mathrm{p}\leq 0.0001\) \(\mathrm{N} = 3\) \(\mathrm{n} = 15\) for each group). Data were analyzed by unpaired Student's \(t\) test. Scale bar \(= 2\mathrm{mm}\) . \(^{**}p\leq 0.01\) , \(^{***}p\leq 0.0001\)
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[115, 87, 845, 901]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 88, 882, 130]]<|/det|>
+1 Extended Data Fig. 8: ECM of mineralized notochord domains appear normal in \(sspo^{dmh4 / + }\) mutants.  
+
+<|ref|>text<|/ref|><|det|>[[66, 120, 884, 325]]<|/det|>
+3 a, Schematic of zebrafish notochord development at 10 dpf (5mm SL) showing cartilaginous ECM 4 domains (IVD), mineralized domains (vertebra), notochord vacuole and notochord sheath cells. b, 5 c, Representative transmission electron micrographs of the collagenous ECM in mineralized 6 domains of 10 dpf (5mm SL) wildtype (b, \(\mathrm{N} = 7\) , \(\mathrm{n} = 24\) ) and \(sspo^{dmh4 / + }\) (c, \(\mathrm{N} = 7\) , \(\mathrm{n} = 23\) ) larvae. 7 d, e, Antioxidant treatment (500μM NACET) had no obvious effect on mineralized ECM 8 development in wildtype (d, \(\mathrm{N} = 5\) , \(\mathrm{n} = 17\) ) or \(sspo^{dmh4 / + }\) mutant (e, \(\mathrm{N} = 6\) , \(\mathrm{n} = 19\) ) animals. f-h, 9 Quantification of mineralized matrix thickness showed no significant difference in local maximum 10 matrix thickness (f), local minimum matrix thickness (g), and local matrix thickness ratio (h) 11 across the groups. Box plots show all data points, median, and min to max whiskers; points of the 12 same colour belong to the same fish. Data were analyzed by two- way ANOVA (n.s.: nonsignificant \(\mathrm{p} > 0.05\) , \*: \(\mathrm{p} \leq 0.05\) , \*\*: \(\mathrm{p} \leq 0.01\) , \*\*\*: \(\mathrm{p} \leq 0.001\) , \*\*\*\*: \(\mathrm{p} \leq 0.0001\) ).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 353, 109]]<|/det|>
+## Supplementary Information  
+
+<|ref|>text<|/ref|><|det|>[[115, 128, 697, 146]]<|/det|>
+Supplementary Video 1. Wildtype microCT images, transverse plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 146, 875, 199]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 744μm series, transverse plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 214, 675, 232]]<|/det|>
+Supplementary Video 2. Wildtype microCT images, coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 232, 875, 285]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 640μm series, coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 300, 696, 320]]<|/det|>
+Supplementary Video 3. \(ssp^{dmh4 / +}\) microCT images, transverse plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 320, 875, 373]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 693μm series, transverse plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 388, 675, 407]]<|/det|>
+Supplementary Video 4. \(ssp^{dmh4 / +}\) microCT images, coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 407, 875, 459]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series of iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 771μm series, coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 475, 839, 494]]<|/det|>
+Supplementary Video 5. Wildtype zebrafish exhibit stable IVL structures (example 2).  
+
+<|ref|>text<|/ref|><|det|>[[115, 494, 875, 546]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained wildtype juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 562, 803, 580]]<|/det|>
+Supplementary Video 6. Scoliotic zebrafish exhibit IVL deformations (example 2).  
+
+<|ref|>text<|/ref|><|det|>[[115, 581, 875, 633]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.  
+
+<|ref|>text<|/ref|><|det|>[[115, 648, 803, 667]]<|/det|>
+Supplementary Video 7. Scoliotic zebrafish exhibit IVL deformations (example 3).  
+
+<|ref|>text<|/ref|><|det|>[[115, 667, 875, 720]]<|/det|>
+High- resolution synchrotron- based X- ray micro- computed tomography (micro- CT) image series through the intervertebral domain of an iodine contrast- stained \(ssp^{dmh4 / +}\) juvenile fish at 21dpf (10mm SL). 36μm series, in the coronal plane.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[42, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[42, 93, 768, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[59, 131, 352, 310]]<|/det|>
+- MovieS1WTtransverse.mov- MovieS2WTcoronal.mov- MovieS3sspotransverse.mov- MovieS4sspocoronal.mov- MovieS5WTIVL2coronal.mov- MovieS6sspolVL2coronal.mov- MovieS7sspolVL3coronal.mov
+
+<--- Page Split --->

preprint/preprint__98a7f7399d69c3b07697f2560f446c1e043a7929fa7b8d85ef58b7cf97af6536/images_list.json

@@ -0,0 +1,92 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1. Schematic illustrating how Hybrid Unmixing (HyU) enhances analysis of multiplexed hyperspectral fluorescent signals in vivo. (A) Multicolor fluorescent biological sample (here a zebrafish embryo) is imaged in hyperspectral mode, collecting the fluorescence spectrum of each voxel in the specimen. (B) HyU represents spectral data as a phasor plot, a 2D histogram of the real and imaginary Fourier components (at a single harmonic). (C) Spectral denoising filters reduce the Poisson and instrumental noise on the phasor histogram, providing the first signal improvement. (D) The phasor acts as an encoder, where each histogram-bin corresponds to a number n of",
+    "footnote": [],
+    "bbox": [
+      [
+        121,
+        480,
+        848,
+        800
+      ]
+    ],
+    "page_idx": 3
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2: Hybrid Unmixing outperforms standard Linear Unmixing (LU) in both synthetic and live spectral fluorescence imaging. (A) Hybrid Unmixing (HyU) and (B) Linear Unmixing (LU) tested using a hyperspectral fluorescence simulation that was generated from four fluorescent signatures (emission spectra, Sup fig 5E). (C) Absolute Mean Squared Error (MSE) shows that HyU offers a consistent reduction in error across a broad range of photons per spectra (#photons/independent spectral components, here resulting from 4 reference spectra combined). (D) The performance differences in the MSE of HyU relative to LU persists when applying multiple phasor denoising filters (0 to 5 median filters). The analysis of this synthetic data shows the consistent improvement of HyU at low photon counts with over a 2-fold improvement when 5 denoising filters are applied at a signal level of 16 photons per spectrum. (E) Unmixing of experimental data from a 4-color zebrafish shows increased contrast for",
+    "footnote": [],
+    "bbox": [
+      [
+        122,
+        88,
+        850,
+        765
+      ]
+    ],
+    "page_idx": 6
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3: Hybrid Unmixing enhances unmixing for low-signal in vivo multiplexing and achieves deeper volumetric imaging. (A) Hybrid Unmixing (HyU) volumetric renderings compared to those of (B) Linear Unmixing (LU) for the trunk portion in a 4-color zebrafish demonstrate an increased contrast and reduced residual in HyU results, especially over deeper parts of the sample. The 4 labels in the fish are Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact-mRuby;fli1:mKO2), respectively labeling clathrin-coated pits (green), membrane (yellow),",
+    "footnote": [],
+    "bbox": [
+      [
+        115,
+        87,
+        880,
+        830
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4: HyU reveals the dynamics of developing vasculature by enabling multiplexed volumetric time-lapse. Hybrid Unmixing (HyU) overcomes challenges in performing multiplexed volumetric time-lapse in vivo imaging of a developing embryo. Here we present this (A) HyU rendering for the trunk portion of a 3-color zebrafish Gt/cltca-citrine); Tg(kdrl:mCherry;fli1:mKO2) at timepoint 0. (B) HyU unmixed results allow for quantitative analysis and segmentation, here an example representing the time evolution of the segmented volumes of mCherry (vasculature, magenta) mKO2(endothelial-lymphatics, yellow) and citrine (clathrin-coated pits, cyan). (C1-4) Time lapse imaging of the formation of the vasculature over 300 mins (zoomed-in rendering of the box in A) at 0, 100, 200, 300 minutes. This show that HyU provides good unmixing at low light levels to permit multiplexing to be used in the observation of development of a live embryo.",
+    "footnote": [],
+    "bbox": [
+      [
+        117,
+        103,
+        880,
+        475
+      ]
+    ],
+    "page_idx": 10
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5: HyU enables identification and unmixing of low photon intrinsic signals in conjunction with extrinsic signals. (A) HyU results of a whole zebrafish embryo provide a frame of reference not only for the improved unmixing of extrinsic signals, but also its increased sensitivity which enables identification and unmixing of intrinsic signals which inherently exist in a low-photon environment. (B) HyU results of the head region (box in A) reveal the simplicity of identifying an unknown autofluorescence signal among multiple extrinsic signals using the phasor method for a quadra-transgenic zebrafish Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact-",
+    "footnote": [],
+    "bbox": [
+      [
+        150,
+        99,
+        844,
+        817
+      ]
+    ],
+    "page_idx": 12
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Figure 6: HyU pushes the upper limits of live multiplexed volumetric timelapse imaging of intrinsic and extrinsic signals.",
+    "footnote": [],
+    "bbox": [
+      [
+        118,
+        140,
+        873,
+        630
+      ]
+    ],
+    "page_idx": 14
+  }
+]

preprint/preprint__98a7f7399d69c3b07697f2560f446c1e043a7929fa7b8d85ef58b7cf97af6536/preprint__98a7f7399d69c3b07697f2560f446c1e043a7929fa7b8d85ef58b7cf97af6536.mmd

@@ -0,0 +1,229 @@
+
+# HyU: Hybrid Unmixing for longitudinal in vivo imaging of low signal to noise fluorescence  
+
+Hsiao Chiang  University of Southern California  
+
+Daniel Koo  University of Southern California  
+
+Masahiro Kitano  University of Southern California  
+
+Jay Unruh  Stowers Institute for Medical Research  
+
+Le Trinh  University of Southern California  
+
+Scott Fraser  University of Southern California  
+
+Francesco Cutrale ( francesco.cutrale@gmail.com )  University of Southern California  https://orcid.org/0000- 0003- 0517- 3069  
+
+Methods Article  
+
+Keywords:  
+
+Posted Date: January 12th, 2022  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 1073331/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License.  Read Full License
+
+<--- Page Split --->
+
+# Title: HyU: Hybrid Unmixing for longitudinal in vivo imaging of low signal to noise fluorescence  
+
+Authors: Hsiao Ju Chiang<sup>1,2,†</sup>, Daniel E.S. Koo<sup>1,2,†</sup>, Masahiro Kitano<sup>1,3</sup>, Jay Unruh<sup>4</sup>, Le A. Trinh<sup>1,2,3</sup>, Scott E. Fraser<sup>1,2,3</sup>, Francesco Cutrale<sup>1,2,3\*</sup>  
+
+## Affiliations:  
+
+<sup>1</sup>Translational Imaging Center, University of Southern California, Los Angeles, CA  <sup>2</sup>Department of Biomedical Engineering, University of Southern California, Los Angeles, CA  <sup>3</sup>Molecular and Computational Biology, University of Southern California, Los Angeles, CA  <sup>4</sup>Stowers Institute for Medical Research, Kansas City, MO  
+
+\*Correspondence to: cutrale@usc.edu  †Equal contribution  
+
+Abstract: The expanded application of fluorescence imaging in biomedical and biological research towards more complex systems and geometries requires tools that can analyze a multitude of components at widely varying time- and length- scales. The major challenge in such complex imaging experiments is to cleanly separate multiple fluorescent labels with overlapping spectra from one another and background autofluorescence, without perturbing the sample with high levels of light. Thus, there is a requirement for efficient and robust analysis tools capable of quantitatively separating these signals.  
+
+In response, we have combined multispectral fluorescence microscopy with hyperspectral phasors and linear unmixing to create Hybrid Unmixing (HyU). Here we demonstrate its capabilities in the dynamic imaging of multiple fluorescent labels in live, developing zebrafish embryos. HyU is more sensitive to low light levels of fluorescence compared to conventional linear unmixing approaches, permitting better multiplexed volumetric imaging over time, with less bleaching. HyU can also simultaneously image both bright exogenous and dim endogenous labels because of its high dynamic range. This allows studies of cellular behaviors, tagged components, and cell metabolism within the same specimen, offering a powerful window into the orchestrated complexity of biological systems.  
+
+One Sentence Summary: Hybrid Unmixing offers enhanced imaging of multiplexed fluorescence labels, enabling longitudinal imaging of multiple fluorescent signals with reduced illumination intensities.
+
+<--- Page Split --->
+
+## Main Text:  
+
+## Introduction  
+
+In recent years, high- content imaging approaches have been refined for decoding the complex and dynamical orchestration of biological processes. \(^{1,2,3}\) Fluorescence, with its high contrast, high specificity and multiple parameters, has become the reference technique for imaging. \(^{4,5}\) Continuous improvements in fluorescent microscopes \(^{6 - 9}\) and the ever- expanding palette of genetically- encoded and synthesized fluorophores have enabled the labeling and observation of a large number of molecular species \(^{10,11}\) . This offers the potential of using multiplexed imaging to follow multiple labels simultaneously in the same specimen, but the technologies for this have fallen short of their fully imagined capabilities. Standard fluorescence microscopes collect multiple images sequentially, employing different excitation and detection bandpass filters for each label. Recently developed techniques allow for massive multiplexing by utilizing sequential labeling of fixed samples but are not suitable for in vivo imaging. \(^{12,13}\) Unfortunately, these approaches are ill- suited to separating overlapping fluorescence emission signals, and the narrow bandpass optical filters used to increase selectivity, decrease the photon efficiency of the imaging. (Figs. S1, S2) These limitations have restricted the number of imaged fluorophores per sample (usually 3- 4) and risks exposing the specimen to damaging levels of exciting light. This has been a significant obstacle for the dynamic imaging, and has prevented in vivo imaging from reaching its full potential.  
+
+Hyperspectral Fluorescent Imaging (HFI) potentially overcomes the limitations of overlapping emissions by expanding signal detection into the spectral domain. \(^{14}\) HFI captures a spectral profile from each pixel, resulting in a hyperspectral cube (x,y, wavelength) of data, that can be processed to deduce the labels present in that pixel. Linear unmixing (LU) has been widely utilized to analyze HFI data, and has performed well with bright samples emitting strong signals from fully- characterized, extrinsic fluorophores such as fluorescent proteins and dyes \(^{15 - 17}\) . However, in vivo fluorescence microscopy is almost always limited in the number of photons collected per pixel (due to the expression levels, the bio- physical fluorescent properties, and the sensitivity of the detection system), which reduces the quality of the spectra acquired.  
+
+A further challenge which affects quality of spectra is the presence of multiple forms of noise in the imaging of the sample. Two examples of instrumental noise are photon noise and read noise. Photon noise, also known as Poisson noise, is an inherent property related to the statistical variation of photons emission from a source and of detection. Poisson noise is inevitable when imaging fluorescent dyes and is more pronounced in the low- photon regime. It poses challenges especially in live and time lapse imaging, where the power of the exciting laser is reduced to avoid photo- damage to the sample, decreasing the amount of fluorescent signal. Read noise arises from voltage fluctuations in microscopes operating in analog mode, during the conversion from photon to digital levels intensity and commonly affects fluorescence imaging acquisition. Most biological samples used for in vivo microscopy are labelled using extrinsic signals from fluorescent proteins or probes but often include intrinsic signals (autofluorescence).  
+
+Autofluorescence contributes photons that are undesired, difficult to identify and to account for in LU. The cumulative presence of noise inevitably leads to a degradation of acquired spectra during imaging. As a result, the spectral separation by LU is often compromised, and the Signal to Noise ratio (SNR) of the final unmixing is often reduced by the weakest of the signals
+
+<--- Page Split --->
+
+detected. \(^{16}\) Increasing the amount of laser excitation can partially overcome these challenges, but the higher energy deposition in the sample causes photo- bleaching and - damage, affecting both the integrity of the live sample and the duration of the observation. Traditional unmixing strategies such as LU are computationally demanding, requiring long analyses and often slowing the experiment. Combined, these compromises have reduced both the overall multiplexing capability and the adoption of HFI multiplexing technologies.  
+
+We have developed Hybrid Unmixing (HyU) as an answer to the challenges that have limited the wider acceptance of HFI for in vivo imaging. HyU employs the phasor approach \(^{18}\) merged with traditional unmixing algorithms to more rapidly and more accurately untangle the fluorescent signals from multiple exogenous and endogenous labels. The phasor approach \(^{18}\) , a popular dimensionality reduction approach for the analysis of both fluorescence lifetime and spectral image analysis \(^{19 - 21}\) provides key advantages to HyU, including spectral compression, denoising, and computational reduction. HyU pairs phasor processing with unmixing algorithms, such as LU, to provide unsupervised analysis of HFI data, removing user subjectivity. Our results show that HyU offers three key advantages: (1) improved unmixing over conventional LU, especially for low intensity images, down to 5 photons per spectra; (2) simplified identification of independent spectral components; (3) dramatically faster processing of large datasets, overcoming the typical unmixing bottleneck for in vivo fluorescence microscopy.  
+
+![](images/Figure_1.jpg)
+
+<center>Figure 1. Schematic illustrating how Hybrid Unmixing (HyU) enhances analysis of multiplexed hyperspectral fluorescent signals in vivo. (A) Multicolor fluorescent biological sample (here a zebrafish embryo) is imaged in hyperspectral mode, collecting the fluorescence spectrum of each voxel in the specimen. (B) HyU represents spectral data as a phasor plot, a 2D histogram of the real and imaginary Fourier components (at a single harmonic). (C) Spectral denoising filters reduce the Poisson and instrumental noise on the phasor histogram, providing the first signal improvement. (D) The phasor acts as an encoder, where each histogram-bin corresponds to a number n of </center>
+
+<--- Page Split --->
+
+pixels, each with a relatively similar spectrum (E). Summing these spectra effectively averages the spectra for that phasor position.  
+
+This denoising results in cleaner average spectrum for this set of pixels, which are ideally suited for analytical decomposition through unmixing algorithms (F). (G) Unmixing results in images that separated into spectral components. Here, linear unmixing (LU) is used for unmixing, but HyU is compatible with any unmixing algorithm.  
+
+Note that HyU offers a major reduction in data size and complexity of the LU (or any other unmixing) computation, because the calculation is applied to the \(10^{4}\) histogram bins (D), rather the the \(\sim 10^{7}\) voxels in the specimen (A). This reduces the number of calculations required for LU dramatically.  
+
+## Results  
+
+HyU combines the best features of hyperspectral phasor analysis and linear unmixing (LU), resulting in faster computation speeds and more reliable results, especially at low light levels. Phasor approaches reduce the computational load because they are compressive, reducing the 32 channels of an HFI spectral plot into a position on a 2D- histogram, representing the real and imaginary Fourier components of the spectrum (Fig. 1A,B). Different 32 channel spectra are represented as different positions on the 2D phasor plot, and mixtures of the two spectra will be rendered at a position along a line connecting the pure spectra. Because the spectral content of an entire 2D or 3D image set is rendered on a single phasor plot, there is a dramatic data compression - from a spectrum for each voxel in an image set (up to or even beyond Gigavoxels) to a histogram value on the phasor plot (Megapixels). In addition, because each "bin" on the phasor plot histogram corresponds to multiple voxels with highly similar spectral profiles, the binning itself represents spectral averaging, which reduces the Poisson and instrumental noise (Fig. 1C- E). Poisson noise in the collected light is unavoidable in HFI unless the excitation is turned so high that the statistics of collected fluorescence creates hundreds or thousands of photons per spectral bin. The clear separation of the spectral phasor plot and its referenced imaging data, permits denoising algorithms to be applied to phasor plot with minimal degradation of the image resolution. LU or other unmixing approaches can be applied to the spectra on the phasor plot, offering a dramatic reduction in computational burden for large image data sets (Fig. 1D). To understand this saving, consider the conventional approach of LU applied to image data at the voxel level (Fig. 1A,F). A timelapse volumetric dataset of \(512 \times 768 \times 17\) (x, y, z) pixels, over 6 timepoints, (Sup. table 1), would require 40 million operations. HyU's requires only \(\sim 18\) thousand operations to unmix each bin on the phasor plot, representing more than a thousand- fold saving (Fig. 1F,G).  
+
+To quantitatively assess the relative performance of LU and HyU, we analyzed them on synthetic hyperspectral fluorescent datasets, created by computationally modelling the biophysics of fluorescence spectral emission and microscope performance (Fig 2 A, B, figs. S3- S5). We used this synthetic dataset to evaluate LU and HyU algorithm performance quantitatively by using metrics such as Mean Square Error (MSE) and unmixing residual (see Fig. S6, Methods; for both metrics, a lower value indicates better performance). In addition to the computational efficiency mentioned above, HyU analysis shows better ability to capture spatial features over a wide dynamic range of intensities, when compared with standard LU, in large part due to the denoising created by processing in phasor space (Fig. 2 A, B). The improved accuracy is demonstrated by a lower MSE, in comparing the results of LU and HyU to the image ground truth. The absolute MSE for HyU is consistently up to 2x lower than that of LU, especially at
+
+<--- Page Split --->
+
+low and ultra- low fluorescence levels (Fig. 2C). MSE can be further decreased by the use of denoising filters on the phasor plot, resulting in superiority of HyU relative to LU for HFI at low (5- 20 photons/spectrum) and ultralow (2- 5 photons/spectrum) levels (Fig. 2D). To better characterize the performance in the experimental data without ground truth, we also define the unmixing residual as the difference between the original multichannel hyperspectral images and their unmixed results. Residuals provide a measure of how closely the unmixed results reconstruct the original signal (Fig. S3, Methods). Unmixing residuals are inversely proportional to the performance of the algorithm, with low residuals indicating high similarity between the unmixed and the original signals. Analysis of unmixing residuals in the synthetic data highlights an improved interpretation of the spectral information in HyU with an average unmixing residual reduction of \(21\%\) compared to the standard (Fig. S5C). The reduction in both MSE and average unmixing residual for synthetic data demonstrates the superior performance of HyU, and provides a baseline comparison when demonstrating performance improvements for experimental data.  
+
+We support the enhanced performance of HyU with analysis of experimental data, which reveals comparatively lower unmixing residuals and a higher dynamic range as compared to LU. Data was acquired from a quadra- transgenic zebrafish embryo Tg(ubiq:Lifeact- mRuby); Gt(cltca- citrine); Tg(ubiq:lyn- tdTomato); Tg(fli1:mKO2), labelling actin, clathrin, plasma membrane, and pan- endothelial cells, respectively (Figs. 2E- L, 3, S7- S9, Supplementary Movie 1). HyU unmixing of the data shows minimal signal cross- talk between channels while LU presents noticeable bleed- through (Fig. 2M- P). Consistently with synthetic data, we utilize the unmixing residual as the main indicator for quality of the analysis in experimental data, owing to the absence of a ground truth. The residual images (Fig. 2F, G) depict a striking difference in performance between HyU and LU. The average relative residual of HyU denotes a 7- fold improvement compared to LU (Fig. 2H) in disentangling the fluorescent spectra. We visualize the unmixed channels independently (Fig. 2, I to L), zooming in on details (Fig.2 I to P) to highlight areas affected by bleed- through and which are difficult to unmix. HyU, with contrast 2- fold higher than standard LU, reduces bleed- through effects and produces images with sharper spatial features, leading to better interpretation of the experimental data (Fig. 2 K, L, fig. S7, Methods).
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2: Hybrid Unmixing outperforms standard Linear Unmixing (LU) in both synthetic and live spectral fluorescence imaging. (A) Hybrid Unmixing (HyU) and (B) Linear Unmixing (LU) tested using a hyperspectral fluorescence simulation that was generated from four fluorescent signatures (emission spectra, Sup fig 5E). (C) Absolute Mean Squared Error (MSE) shows that HyU offers a consistent reduction in error across a broad range of photons per spectra (#photons/independent spectral components, here resulting from 4 reference spectra combined). (D) The performance differences in the MSE of HyU relative to LU persists when applying multiple phasor denoising filters (0 to 5 median filters). The analysis of this synthetic data shows the consistent improvement of HyU at low photon counts with over a 2-fold improvement when 5 denoising filters are applied at a signal level of 16 photons per spectrum. (E) Unmixing of experimental data from a 4-color zebrafish shows increased contrast for </center>
+
+<--- Page Split --->
+
+HyU (left) compared to LU (right). Scale bar \(= 50 \mu \mathrm{m}\) . (F, G) The increased accuracy is revealed by residual images of HyU and LU, showing the spatial distribution of unassigned signals after the analysis of data in E. The results show consistently lower residual values for HyU (F) compared to LU (G). (H) Box plots of the residuals in F and G presents values of \(11\%\) for HyU compared to \(77\%\) for LU with \(*(p < 10^{-10})\) . (I-L) Enlarged rendering of HyU results (E, white box) clearly shows low levels of bleed-through between labels (M-P) Similar enlargement of LU results show noticeably worse performance. Note that regions with bright signals (membrane J, N white arrow) bleed through other channels (M) and (O). Scale bar: \(20 \mu \mathrm{m}\) . Tetra-labeled specimen used here was Gt(cltca- citrine); Tg(ubiq:lyn-tdTomato; ubig:Lifeact-mRuby;fli1:mKO2)
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3: Hybrid Unmixing enhances unmixing for low-signal in vivo multiplexing and achieves deeper volumetric imaging. (A) Hybrid Unmixing (HyU) volumetric renderings compared to those of (B) Linear Unmixing (LU) for the trunk portion in a 4-color zebrafish demonstrate an increased contrast and reduced residual in HyU results, especially over deeper parts of the sample. The 4 labels in the fish are Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact-mRuby;fli1:mKO2), respectively labeling clathrin-coated pits (green), membrane (yellow), </center>
+
+<--- Page Split --->
+
+actin (cyan) and endothelial (magenta). (C,E) HyU results have increased spatial resolution and less bleed though comparing to those of (D,F) LU. Scale bar: \(20 \mu \mathrm{m}\) . When observing the zoomed- in visualization of the surface region of the sample, the yellow signal distinctly marks the membrane and the cyan signal clearly labels the actin in (C) HyU. The same signals are not distinct in (D) LU because of multiple incorrectly assigned magenta pixels that bleed through compromising the true signal in other channels. Similarly, for the zoomed- in visualization of the Perivascular region of the embryo, in (E) HyU, the yellow and magenta signals clearly distinguish the membrane and vasculature while in (F) LU, the results are corrupted by greater noise. (G,H) Intensity line plots of each of the four results signals for HyU (solid) and LU (dashed) demonstrate the improved profiles with greatly reduced noise peaks in HyU as compared to LU. Intensities are scaled by the maximum of each unmixed channel. DL: digital level. (I) Box plots of the relative residual values as a function of z depth for HyU and LU highlight the improvements in the unmixing results. HyU has an unmixing residual of \(6.6\% \pm 5.3\%\) compared to LU's \(58\% \pm 17\%\) . The average amount of residual is 9- fold lower in HyU with narrower variance of residual.  
+
+Applying HyU to another HFI dataset further highlights HyU's improvements in noise reduction and reconstitution of spatial features for low- photon unmixing. (Figs. 3, S8). In the zoomed- in image of a single slice of the embryo skin surface, acquired in the trunk region, the HyU image correctly does not display pan- endothelial (magenta) signal in the periderm, an area which should be devoid of endothelial cells and mKO2 signal (Fig. 3C). In contrast, the result from LU shows visually distinctive pan- endothelial signal throughout the tissue plane (Fig. 3D). This incorrect estimation of the relative contribution of mKO2 fluorescence for LU is possibly due to the presence of noise, corrupting the spectral profiles. This is further delineated in the intensity profiles of the mKO2 signal between HyU and LU with much higher individual peaks from noise demonstrated for LU (Fig. 3G, lower left). Intensity profiles for both magnified cross- sections of the volume (Fig. 3C- F) provide a striking visualization of the improvements of HyU. The line intensity profiles in HyU present reduced noise and represent more closely the expected distribution of signals (Fig. 3G,H). The visible micro patterns of actin on the membrane of the periderm suggest that the improvements quantified with synthetic data are maintained in live samples' signals and geometrical patterns of microridges<sup>22</sup>. By contrast, noise corruption and the presence of misplaced signals are characterized in the results from LU, with high frequency intensity variations that mis- match both the labeling and biological patterns.  
+
+HyU is more accurate and results in more reliable unmixing results across the depth of sample with greatly reduced unmixing residuals. The average residual for HyU is 9- fold lower than that of LU with a 3- fold narrower variance. (Figs. 3I, S8). This reduction in the residual is consistent with increasing z- depth where HyU unmixing results stably maintain both lower residuals and variance on average. These reduced residuals correspond both to a mathematically more precise and more uniform decomposition of signals as illustrated by the distribution of residuals versus photons (Figs. S8E, F, S14).
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4: HyU reveals the dynamics of developing vasculature by enabling multiplexed volumetric time-lapse. Hybrid Unmixing (HyU) overcomes challenges in performing multiplexed volumetric time-lapse in vivo imaging of a developing embryo. Here we present this (A) HyU rendering for the trunk portion of a 3-color zebrafish Gt/cltca-citrine); Tg(kdrl:mCherry;fli1:mKO2) at timepoint 0. (B) HyU unmixed results allow for quantitative analysis and segmentation, here an example representing the time evolution of the segmented volumes of mCherry (vasculature, magenta) mKO2(endothelial-lymphatics, yellow) and citrine (clathrin-coated pits, cyan). (C1-4) Time lapse imaging of the formation of the vasculature over 300 mins (zoomed-in rendering of the box in A) at 0, 100, 200, 300 minutes. This show that HyU provides good unmixing at low light levels to permit multiplexing to be used in the observation of development of a live embryo. </center>  
+
+We utilized HyU's increased sensitivity to overcome common challenges of multiplexed imaging such as poor photon yield and spectral cross- talk and were able to visualize dynamics in a developing zebrafish embryo. We used a triple- transgenic zebrafish embryo with labeled pan- endothelial cells, vasculature, and clathrin- coated pits (Tg(fli1:mKO2); Tg(kdrl:mCherry); Gt(cltca- Citrine)). Multiplexing these spectrally close fluorescent proteins is enabled by HyU's increased sensitivity at lower photon counts. The increased performance at lower SNR allowed us to maintain high quality results (Fig. 4, Supplementary Movie 2) while performing faster acquisitions and reducing photon- damage through lower excitation laser power and pixel dwell time. Decreased experimental requirements allow for tiling of larger volumes, extending the field- of- view while still providing enough time resolution for developmental events, even with a high number of multiplexed fluorescent signals. The time- lapses visualize the formation of ventral vasculo- endothelial protrusions acquired in parallel to the development of clathrin and kdrl. HyU enables comparative quantifications of spatio- temporal features, allowing for the
+
+<--- Page Split --->
+
+determination of volumetric changes over lengthy timelapses, in this case, over the course of 300 minutes (Fig. 4B) \(^{23,24}\) .  
+
+HyU provides the ability to combine the information from intrinsic and extrinsic signals during live imaging of samples, at both single (Fig. 5) and multiple time points (Fig. 6). The graphical representation of phasors allows identification of unexpected intrinsic fluorescence signatures in a quadra- transgenic zebrafish embryo \(Gt(cltca- citrine);Tg(ubiq:lyn- tdTomato;ubiq:Lifeact- mRuby;fli1:mKO2)\) , imaged with single photon (488 and 561nm excitation) (Fig. 5A- D). The elongated distribution on the phasor (Fig. 5C) highlights the presence of an additional, unexpected spectral signature, related to strong sample autofluorescence (Fig. 5D blue). HyU analysis of the sample, inclusive of this additional signal, provides separation of the contributions of 5 different fluorescent spectra with residual \(3.9\% \pm 0.3\%\) . HyU allows for reduced energy load, tiled imaging of the entire embryo without perturbing its development or depleting its fluorescence signal (Fig. 5A). The higher speed, lower power imaging allows for subsequent re- imaging of the same sample, as we report in the zoomed high- resolution acquisitions of the head section (Fig. 5B,E).  
+
+With the ability to unmix low photon signals, HyU enables imaging and decoding of intrinsic signals, which are inherently low light. Two photon lasers are ideal for exciting and imaging blue- shifted intrinsic fluorescence from samples \(^{25,26}\) . Here, the same quadra- transgenic sample is imaged using 740 nm excitation to access both intrinsic and extrinsic signals (Fig 5 E- G, sup Note 2). HyU enables unmixing of at least 9 intrinsic and transgenic fluorescent signals (Fig. 5), recovering fluorescent intensities from labels illuminated at a sub- optimal excitation wavelength (Fig. 5E). The spectra for intrinsic fluorescence were obtained from in vitro measurements and values reported in literature (Methods). For this sample the intrinsic signals arise from events related mainly with metabolic activity (NADH and Retinoids) \(^{27 - 31}\) , tissue structure (elastin) \(^{32}\) , and illumination (laser reflection) (Fig. 5E). These results confirm our conclusion that HyU is a powerful tool for allowing the imaging and analysis of endogenous labels.
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 5: HyU enables identification and unmixing of low photon intrinsic signals in conjunction with extrinsic signals. (A) HyU results of a whole zebrafish embryo provide a frame of reference not only for the improved unmixing of extrinsic signals, but also its increased sensitivity which enables identification and unmixing of intrinsic signals which inherently exist in a low-photon environment. (B) HyU results of the head region (box in A) reveal the simplicity of identifying an unknown autofluorescence signal among multiple extrinsic signals using the phasor method for a quadra-transgenic zebrafish Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact- </center>
+
+<--- Page Split --->
+
+mRuby;fli1:mKO2) imaged over multiple tiles. Scale bar: \(80\mu \mathrm{m}\) . (C) The input spectra required to perform the unmixing are easily identified on (D) the phasor plot when visualizing each spectrum as a spatial location. Phasors offer a simplified identification and selection of independent and unexpected spectral components in the encoded HyU approach. Intrinsic signals are notoriously low in emitted photons leading to an inability to unmix using traditional unmixing algorithms. (E) The zoomed- in acquisition of the head region of the embryo (box in A) displays HyU's unmixing results of many intrinsic and extrinsic signals when in an environment of very low photon output, a previously highly difficult experimental condition to unmix. Scale bar: \(70\mu \mathrm{m}\) . (F) The phasor plot representation provides the easily identifiable eight independent fluorescent fingerprint locations. (G) The spectra corresponding to each of the eight independent spectral components are also provided a reference. Colors in (F) match renderings in (E) and (G): NADH bound (red), NADH free (yellow), retinoid (magenta), retinoic acid (cyan), reflection (green), elastin (purple) and extrinsic signals: mKO2 (blue), and mRuby (orange). All signals were excited with a (A- D) single photon laser at both \(488~\mathrm{nm}\) and \(561~\mathrm{nm}\) or a (E- G) two photon laser at \(740~\mathrm{nm}\) .
+
+<--- Page Split --->
+![](images/Figure_6.jpg)
+
+<center>Figure 6: HyU pushes the upper limits of live multiplexed volumetric timelapse imaging of intrinsic and extrinsic signals. </center>  
+
+HyU's increased sensitivity provides a simple solution for the challenging task of imaging timelapse data at 6 time points (125 mins) for both intrinsic signals and extrinsic signals of a quadra- transgenic zebrafish: Tg(cltca- Citrine);(ubiq:lyn- tdTomato);(ubiq:Lifecat- mRuby);(fli1:mKO2)). (A) - (F) Volumetric renderings of HyU results for time points acquired at 25 min intervals reveal the high- contrast and - multiplexed labels of NADH bound (red), NADH free (yellow), retinoid (magenta), retinoic acid (cyan), mKO2 (green), and autofluorescence from blood cells (blue) when excited @740nm. Further extrinsic signals for mKO2 (yellow), tdTomato (magenta), mRuby (cyan), Citrine (green) and blood cells autofluorescence (blue) are also readily unmixed using HyU when exciting the sample @ 488/561nm. HyU provides the capacity to simultaneously multiplex 9 signals in a live sample over long periods of time, a previously unexplored task. Scale bar: 50 μm.  
+
+Finally, we exploited the HyU capabilities to multiplex volumetric timelapse of extrinsic and intrinsic signals by imaging the tail region of the same quadra- transgenic zebrafish embryo. We excite extrinsic labels at 488/561 nm and the intrinsic signals with 740 nm two photon, collecting
+
+<--- Page Split --->
+
+6 tiled volumes over 125 mins (Figs. 6, S9- S11, Supplementary Movie 3). HyU unmixing in this sample allows for distinction of 9 signals, separating their contributions with sufficiently low requirements to allow repeated imaging of notoriously low SNR intrinsic fluorescence.  
+
+## Discussion  
+
+Our results reveal the advantages of Hybrid Unmixing (HyU) over more conventional Linear Unmixing (LU) in performing complex multiplexing experiments. HyU overcomes the significant challenges of separating multiple fluorescent and autofluorescent labels with overlapping spectra while minimally perturbing the sample with excitation light.  
+
+The chief advantage of HyU is its multiplexing capability when imaging in the presence of biological and instrumental noise, especially at low signal levels. HyU increased sensitivity improves multiplexing in photon limited applications (Fig. 2F- L), in deeper volumetric acquisitions (Fig. 3I) and in signal starved imaging of autofluorescence (Fig. 5E, Fig. 6). Our simulation results (Fig. 2) demonstrate that HyU improves unmixing of spatially and spectrally overlapping fluorophores excited simultaneously. The increased robustness at low photon imaging conditions reduces the imaging requirements for excitation levels and detector integration time, allowing for imaging with reduced photo- toxicity. Live imaging on multi- color samples performed at high sampling frequency enables improved tiling to increase the field- of- view (Fig. 3, 4) while maximizing the usage of the finite fluorescent signals over time. Two- photon imaging of intrinsic and extrinsic signals suggests the ability of HyU to multiplex signals with large dynamic range differences (Fig. 5) extending multiplexed volumetric imaging into the time dimension (Fig. 6). Although improved, images with particularly low signal still present corruption (Fig. S4), setting a reasonable range of utilization above 8 photons/spectrum.  
+
+Simplicity of use and versatility are other key advantages of HyU, inherited from both the phasor approach<sup>33</sup> and traditional unmixing algorithms. Phasors here operate as a spectral encoder, reducing computational load and integrating similar spectral signatures in histogram bins of the phasor plot. This representation simplifies identification of independent spectral signatures (Fig. 5, Supplementary Note1) through both phasor plot selection and phasor residual mapping (fig. S11), accounting for unexpected intrinsic signals (Figs. 5, 6, S12, Supplementary Note2) in a semi- automated manner, while still allowing fully- automated analysis by means of spectral libraries.  
+
+The simplicity of this approach is especially helpful in live imaging where identifying independent spectral components remains an open challenge, owing to the presence of intrinsic signals (Fig. S12, Supplementary Note 1). High- SNR reference spectra can be derived from other experimental data or identified directly on the phasor. Selection of portions on the phasor plot allows for visualization of the corresponding spectra in the wavelength domain (Fig. 5C,D,F,G). This intuitive versatility allows for identification of both the number of unexpected signatures and their spectra, a task previously difficult to perform due to noise and lack of global visualization tools. In single photon imaging (Fig. 5A- D), HyU phasor allowed identification of a fifth distinct spectral component arising from general autofluorescent background, thereby improving the unmixed results. In two photon imaging, HyU enabled identification and multiplexing of 8 highly overlapping signals possessing a wide dynamic range of intensities, between intrinsic and extrinsic markers (Fig. 5F,G). Combination of single and two photon imaging increased the number of multiplexed fluorophores to 9 (Fig. 6), considering some of the
+
+<--- Page Split --->
+
+extrinsic labels being excited at two photons. Multiplexing of signals may be further improved by implementing HyU on fluorescent dyes.  
+
+HyU performs better than standard algorithms both in the presence and absence of phasor noise reduction filters<sup>33</sup>. Compared with LU, the unmixing enhancement when such filters<sup>33</sup> are applied is demonstrated by a decrease of the MSE of up to \(21\%\) (Fig 2C), with a reduction of the average amount of residuals by 7- fold. Even in the absence of phasor denoising filters, HyU performs up to \(7.3\%\) better than the standard (Fig. 2D) based on Mean Squared Error of synthetic data unmixing. This base improvement is due to the averaging of similarly shaped spectra in each phasor histogram bin, which reduces the statistical variability within the spectra used for the unmixing calculations (Fig. 1E). This averaging strategy works well for general fluorescence spectra owing to their broad and mostly unique spectral shape.  
+
+In the absence of noise, for example in the ground truth simulations, LU produces an MSE 6- fold lower than HyU (Fig. S5, B, C, S6G). In these noiseless conditions, the binning and averaging of spectra in the phasor histogram, without denoising, provides statistically indifferent values of error respect to LU, suggesting results of similar quality.  
+
+HyU can interface with different unmixing algorithms, adapting to existing experimental pipelines. We successfully tested hybridization with iterative approaches such as non- negative matrix factorization<sup>34</sup>, fully constrained and non- negative least- squares<sup>35</sup> (Methods). Speed tests with iterative fitting unmixing algorithms demonstrate a speed increase of up to 500- fold when the HyU compressive strategy is applied. (Fig. S13, Supplementary Note 3). Due to the initial computational overhead for encoding spectra in phasors, there is a 2- fold speed reduction for HyU in comparison to standard LU. However, this may be improved with further optimizations of the HyU implementation.  
+
+One restriction of HyU derives from the mathematics of linear unmixing, where linear equations representing the unmixed channels need to be solved for the unknown contributions of each analyzed fluorophore. To obtain a unique solution from these equations and to avoid an underdetermined equation system, the maximum number of spectra for unmixing may not exceed the number of channels acquired<sup>36</sup>, generally 32 for commercial microscopes. This number could be increased; however, due to the broad and photon- starved nature of fluorescence spectra, acquisition of a larger number of channels could negatively affect the sample, imaging time and intensities. Depending on the number of labels in the specimen of interest, extending the number of labels to simultaneously unmix beyond 32 will likely require spectral resolution upsampling strategies.  
+
+HyU improvement is related to the presence of various types of noise in microscopy images, such as Gaussian, Poisson and digital as well as unidentified sources of spectral signatures (Fig. S5B,C, S6G). In the multiplexing of fluorescent signals, HyU offers improved performance, quality- and speed- wise in the low- signal regime. HyU is poised to be used in the context of in vivo imaging, harvesting information from samples labeled at endogenous- level.  
+
+In conclusion, the results presented in this paper quantitatively show that HyU, a phasor based, computational unmixing framework, is well suited for tackling the many challenges present in live imaging of multiple fluorescence labels. HyU's reduced requirements in amount of
+
+<--- Page Split --->
+
+fluorescent signal permit a reduction of laser excitation load and imaging time. These factors enable multiplexed imaging of biological events with longer duration, higher speed and lower photo- toxicity while providing access to information- rich imaging across different spatiotemporal scales. The reduced requirements of HyU make it fully compatible with any commercial and common microscopes capable of spectral detection, facilitating access to the technology. Our analysis demonstrates HyU's robustness, simplicity and improvement in identifying both new and known spectral signatures, and vastly improved unmixing outputs, providing a much- needed tool for delving into the many questions still surrounding studies with live imaging.
+
+<--- Page Split --->
+
+## References and Notes:  
+
+1. Valm, A. M. et al. Applying systems-level spectral imaging and analysis to reveal the organelle interactome. Nature 546, 162-167 (2017).  
+2. Tsurui, H. et al. Seven-color fluorescence imaging of tissue samples based on fourier spectroscopy and singular value decomposition. Journal of Histochemistry and Cytochemistry 48, 653-662 (2000).  
+3. Amat, F. et al. Fast, accurate reconstruction of cell lineages from large-scale fluorescence microscopy data. Nature Methods 11, 951-958 (2014).  
+4. Ueno, T. & Nagano, T. Fluorescent probes for sensing and imaging. Nature Methods vol. 8 642-645 (2011).  
+5. Lichtman, J. W. & Conchello, J. A. Fluorescence microscopy. Nature Methods vol. 2 910-919 (2005).  
+6. Sinclair, M. B., Haaland, D. M., Timlin, J. A. & Jones, H. D. T. Hyperspectral confocal microscope. 45, 6283-6291 (2006).  
+7. Jahr, W., Schmid, B., Schmied, C., Fahrbach, F. O. & Huisken, J. Hyperspectral light sheet microscopy. Nature Communications 6, 1-7 (2015).  
+8. Truong, T. V., Supatto, W., Koos, D. S., Choi, J. M. & Fraser, S. E. Deep and fast live imaging with two-photon scanned light-sheet microscopy. 8, (2011).  
+9. Chen, B. C. et al. Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution. Science 346, (2014).  
+10. Kredel, S. et al. mRuby, a bright monomeric red fluorescent protein for labeling of subcellular structures. PLoS ONE 4, 1-7 (2009).  
+11. Sakaue-Sawano, A. et al. Visualizing Spatiotemporal Dynamics of Multicellular Cell-Cycle Progression. Cell 132, 487-498 (2008).  
+12. Wade, O. K. et al. 124-Color Super-resolution Imaging by Engineering DNA-PAINT Blinking Kinetics. Nano Letters 19, 2641-2646 (2019).  
+13. Strauss, S. & Jungmann, R. Up to 100-fold speed-up and multiplexing in optimized DNA-PAINT. Nature Methods 17, 789-791 (2020).  
+14. Zimmermann, T., Rietdorf, J. & Pepperkok, R. Spectral imaging and its applications in live cell microscopy. FEBS Letters 546, 87-92 (2003).  
+15. Paddock, S. Multi-Spectral Imaging and Linear Unmixing Add a Whole New Dimension to. BioTechniques 31, 1272-1278 (2001).  
+16. Zimmermann, T. Spectral imaging and linear unmixing in light microscopy. Advances in Biochemical Engineering/Biotechnology 95, 245-265 (2005).  
+17. Garini, Y., Young, I. T. & McNamara, G. Spectral imaging: Principles and applications. Cytometry Part A vol. 69 (2006).  
+18. Digman, M. A., Caiolfa, V. R., Zamai, M. & Gratton, E. The Phasor Approach to Fluorescence Lifetime Imaging Analysis. Biophysical Journal 94, L14-L16 (2008).  
+19. Fereidouni, F., Bader, A. N., Colonna, A. & Gerritsen, H. C. Phasor analysis of multiphoton spectral images distinguishes autofluorescence components of in vivo human skin. Journal of Biophotonics 7, 589-596 (2014).  
+20. Scipioni, L., Rossetta, A., Tedeschi, G. & Gratton, E. Phasor S-FLIM: a new paradigm for fast and robust spectral fluorescence lifetime imaging. Nature Methods 18, 542-550 (2021).  
+21. Ranjit, S., Malacrida, L., Jameson, D. M. & Gratton, E. Fit-free analysis of fluorescence lifetime imaging data using the phasor approach. Nature Protocols 13, 1979-2004 (2018).
+
+<--- Page Split --->
+
+22. Depasquale, J. A. Actin Microridges. Anatomical record (Hoboken, N.J. : 2007) 301, 2037-2050 (2018).23. Okuda, K. S., Hogan, B. M., Cantelmo, A. R. & Hogan, B. M. Endothelial Cell Dynamics in Vascular Development : Insights From Live-Imaging in Zebrafish. 11, (2020).24. Isogai, S., Lawson, N. D., Torrealday, S., Horiguchi, M. & Weinstein, B. M. Angiogenic network formation in the developing vertebrate trunk. (2003) doi:10.1242/dev.00733.25. Denk, W., Strickler, J. H. & Webb, W. W. Two-photon laser scanning fluorescence microscopy. Science 248, (1990).26. Zipfel, W. R. et al. Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation. Proceedings of the National Academy of Sciences 100, 7075-7080 (2003).27. Bird, D. K. et al. Metabolic mapping of MCF10A human breast cells via multiphoton fluorescence lifetime imaging of the coenzyme NADH. Cancer Research 65, (2005).28. Lakowicz, J. R., Szmacinski, H., Nowaczyk, K. & Johnson, M. L. Fluorescence lifetime imaging of free and protein-bound NADH. Proceedings of the National Academy of Sciences of the United States of America 89, (1992).29. Skala, M. C. et al. In vivo multiphoton microscopy of NADH and FAD redox states, fluorescence lifetimes, and cellular morphology in precancerous epithelia. Proceedings of the National Academy of Sciences of the United States of America 104, 19494-19499 (2007).30. Sharick, J. T. et al. Protein-bound NAD(P)H Lifetime is Sensitive to Multiple Fates of Glucose Carbon. Scientific Reports 8, (2018).31. Stringari, C. et al. Phasor approach to fluorescence lifetime microscopy distinguishes different metabolic states of germ cells in a live tissue. Proceedings of the National Academy of Sciences of the United States of America 108, (2011).32. Wagnieres, G. A., Star, W. M. & Wilson, B. C. Invited Review In Vivo Fluorescence Spectroscopy and Imaging for Oncological Applications. 68, 603-632 (1998).33. Cutrale, F. et al. Hyperspectral phasor analysis enables multi-plexed 5D in vivo imaging. Nature Publishing Group (2017) doi:10.1038/nmeth.4134.34. Févotte, C. & Dobigeon, N. Nonlinear hyperspectral unmixing with robust nonnegative matrix factorization. IEEE Transactions on Image Processing 24, (2015).35. Heslop, D., von Dobeneck, T. & Höcker, M. Using non-negative matrix factorization in the "unmixing" of diffuse reflectance spectra. Marine Geology 241, 63-78 (2007).36. Paddock, S. W. Confocal Microscopy, Methods and Protocols, Second Edition. Humana Press 1075, 388 (2014).  
+
+Acknowledgments: The authors would like to thank F. Schneider, S. Restrepo (Translational Imaging Center, University of Southern California), Chi- Li Chiu (Bitplane Inc.) and Samuel Ojosnegros (Institute for Bioengineering of Catalonia) for helpful discussions.  
+
+Funding: This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant DGE- 1842487, Department of Defense PR150666 and University of Southern California.
+
+<--- Page Split --->
+
+Author contributions: H.J.C., D.E.S.K., and F.C. analyzed the results and wrote the software. H.J.C., D.E.S.K., J.U. and F.C. provided conceptualization. H.J.C., D.E.S.K., M.K., F.C, and L.A.T. helped in the experimental design and data analysis. M.K. generated the inducible and mKO2 zebrafish lines. H.J.C. and F.C. acquired data. S.E.F. provided supervision. H.J.C., D.E.S.K., and F.C. wrote the paper. L.A.T., J.U., M.K. supported review and editing.  
+
+Competing interests: The University of Southern California has filed a provisional patent application covering this method.  
+
+Data and materials availability: All the relevant data are available from the corresponding author upon reasonable request. Datasets for Figs. 1–6 and simulations are available for download at http://bioimaging.usc.edu/software.html#sampledatasets in the samples section. All the relevant code is available from the corresponding author upon reasonable request. Software and instructions can be downloaded from http://bioimaging.usc.edu/software.html.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+SupplementaryMovie1v3. mp4SupplementaryMovie24. mp4SupplementaryMovie3v3. mp4SupplementaryMovieCaptions211108DKFC.pdfHybUnmSupplementaryMaterialsNM20211111FC.pdfREADME2021201FC.pdf
+
+<--- Page Split --->

preprint/preprint__98a7f7399d69c3b07697f2560f446c1e043a7929fa7b8d85ef58b7cf97af6536/preprint__98a7f7399d69c3b07697f2560f446c1e043a7929fa7b8d85ef58b7cf97af6536_det.mmd

@@ -0,0 +1,285 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 825, 176]]<|/det|>
+# HyU: Hybrid Unmixing for longitudinal in vivo imaging of low signal to noise fluorescence  
+
+<|ref|>text<|/ref|><|det|>[[44, 196, 339, 237]]<|/det|>
+Hsiao Chiang  University of Southern California  
+
+<|ref|>text<|/ref|><|det|>[[44, 243, 339, 283]]<|/det|>
+Daniel Koo  University of Southern California  
+
+<|ref|>text<|/ref|><|det|>[[44, 290, 339, 330]]<|/det|>
+Masahiro Kitano  University of Southern California  
+
+<|ref|>text<|/ref|><|det|>[[44, 336, 392, 376]]<|/det|>
+Jay Unruh  Stowers Institute for Medical Research  
+
+<|ref|>text<|/ref|><|det|>[[44, 382, 339, 422]]<|/det|>
+Le Trinh  University of Southern California  
+
+<|ref|>text<|/ref|><|det|>[[44, 428, 339, 468]]<|/det|>
+Scott Fraser  University of Southern California  
+
+<|ref|>text<|/ref|><|det|>[[44, 473, 697, 515]]<|/det|>
+Francesco Cutrale ( francesco.cutrale@gmail.com )  University of Southern California  https://orcid.org/0000- 0003- 0517- 3069  
+
+<|ref|>text<|/ref|><|det|>[[44, 556, 182, 575]]<|/det|>
+Methods Article  
+
+<|ref|>text<|/ref|><|det|>[[44, 595, 137, 613]]<|/det|>
+Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 632, 331, 652]]<|/det|>
+Posted Date: January 12th, 2022  
+
+<|ref|>text<|/ref|><|det|>[[44, 670, 475, 690]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 1073331/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 707, 910, 750]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License.  Read Full License
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[120, 99, 880, 140]]<|/det|>
+# Title: HyU: Hybrid Unmixing for longitudinal in vivo imaging of low signal to noise fluorescence  
+
+<|ref|>text<|/ref|><|det|>[[140, 145, 860, 183]]<|/det|>
+Authors: Hsiao Ju Chiang<sup>1,2,†</sup>, Daniel E.S. Koo<sup>1,2,†</sup>, Masahiro Kitano<sup>1,3</sup>, Jay Unruh<sup>4</sup>, Le A. Trinh<sup>1,2,3</sup>, Scott E. Fraser<sup>1,2,3</sup>, Francesco Cutrale<sup>1,2,3\*</sup>  
+
+<|ref|>sub_title<|/ref|><|det|>[[118, 205, 218, 222]]<|/det|>
+## Affiliations:  
+
+<|ref|>text<|/ref|><|det|>[[117, 229, 857, 322]]<|/det|>
+<sup>1</sup>Translational Imaging Center, University of Southern California, Los Angeles, CA  <sup>2</sup>Department of Biomedical Engineering, University of Southern California, Los Angeles, CA  <sup>3</sup>Molecular and Computational Biology, University of Southern California, Los Angeles, CA  <sup>4</sup>Stowers Institute for Medical Research, Kansas City, MO  
+
+<|ref|>text<|/ref|><|det|>[[117, 354, 418, 397]]<|/det|>
+\*Correspondence to: cutrale@usc.edu  †Equal contribution  
+
+<|ref|>text<|/ref|><|det|>[[116, 446, 884, 570]]<|/det|>
+Abstract: The expanded application of fluorescence imaging in biomedical and biological research towards more complex systems and geometries requires tools that can analyze a multitude of components at widely varying time- and length- scales. The major challenge in such complex imaging experiments is to cleanly separate multiple fluorescent labels with overlapping spectra from one another and background autofluorescence, without perturbing the sample with high levels of light. Thus, there is a requirement for efficient and robust analysis tools capable of quantitatively separating these signals.  
+
+<|ref|>text<|/ref|><|det|>[[115, 576, 876, 735]]<|/det|>
+In response, we have combined multispectral fluorescence microscopy with hyperspectral phasors and linear unmixing to create Hybrid Unmixing (HyU). Here we demonstrate its capabilities in the dynamic imaging of multiple fluorescent labels in live, developing zebrafish embryos. HyU is more sensitive to low light levels of fluorescence compared to conventional linear unmixing approaches, permitting better multiplexed volumetric imaging over time, with less bleaching. HyU can also simultaneously image both bright exogenous and dim endogenous labels because of its high dynamic range. This allows studies of cellular behaviors, tagged components, and cell metabolism within the same specimen, offering a powerful window into the orchestrated complexity of biological systems.  
+
+<|ref|>text<|/ref|><|det|>[[115, 815, 866, 868]]<|/det|>
+One Sentence Summary: Hybrid Unmixing offers enhanced imaging of multiplexed fluorescence labels, enabling longitudinal imaging of multiple fluorescent signals with reduced illumination intensities.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[116, 91, 211, 108]]<|/det|>
+## Main Text:  
+
+<|ref|>sub_title<|/ref|><|det|>[[116, 141, 225, 158]]<|/det|>
+## Introduction  
+
+<|ref|>text<|/ref|><|det|>[[115, 165, 881, 463]]<|/det|>
+In recent years, high- content imaging approaches have been refined for decoding the complex and dynamical orchestration of biological processes. \(^{1,2,3}\) Fluorescence, with its high contrast, high specificity and multiple parameters, has become the reference technique for imaging. \(^{4,5}\) Continuous improvements in fluorescent microscopes \(^{6 - 9}\) and the ever- expanding palette of genetically- encoded and synthesized fluorophores have enabled the labeling and observation of a large number of molecular species \(^{10,11}\) . This offers the potential of using multiplexed imaging to follow multiple labels simultaneously in the same specimen, but the technologies for this have fallen short of their fully imagined capabilities. Standard fluorescence microscopes collect multiple images sequentially, employing different excitation and detection bandpass filters for each label. Recently developed techniques allow for massive multiplexing by utilizing sequential labeling of fixed samples but are not suitable for in vivo imaging. \(^{12,13}\) Unfortunately, these approaches are ill- suited to separating overlapping fluorescence emission signals, and the narrow bandpass optical filters used to increase selectivity, decrease the photon efficiency of the imaging. (Figs. S1, S2) These limitations have restricted the number of imaged fluorophores per sample (usually 3- 4) and risks exposing the specimen to damaging levels of exciting light. This has been a significant obstacle for the dynamic imaging, and has prevented in vivo imaging from reaching its full potential.  
+
+<|ref|>text<|/ref|><|det|>[[115, 469, 875, 627]]<|/det|>
+Hyperspectral Fluorescent Imaging (HFI) potentially overcomes the limitations of overlapping emissions by expanding signal detection into the spectral domain. \(^{14}\) HFI captures a spectral profile from each pixel, resulting in a hyperspectral cube (x,y, wavelength) of data, that can be processed to deduce the labels present in that pixel. Linear unmixing (LU) has been widely utilized to analyze HFI data, and has performed well with bright samples emitting strong signals from fully- characterized, extrinsic fluorophores such as fluorescent proteins and dyes \(^{15 - 17}\) . However, in vivo fluorescence microscopy is almost always limited in the number of photons collected per pixel (due to the expression levels, the bio- physical fluorescent properties, and the sensitivity of the detection system), which reduces the quality of the spectra acquired.  
+
+<|ref|>text<|/ref|><|det|>[[115, 633, 878, 825]]<|/det|>
+A further challenge which affects quality of spectra is the presence of multiple forms of noise in the imaging of the sample. Two examples of instrumental noise are photon noise and read noise. Photon noise, also known as Poisson noise, is an inherent property related to the statistical variation of photons emission from a source and of detection. Poisson noise is inevitable when imaging fluorescent dyes and is more pronounced in the low- photon regime. It poses challenges especially in live and time lapse imaging, where the power of the exciting laser is reduced to avoid photo- damage to the sample, decreasing the amount of fluorescent signal. Read noise arises from voltage fluctuations in microscopes operating in analog mode, during the conversion from photon to digital levels intensity and commonly affects fluorescence imaging acquisition. Most biological samples used for in vivo microscopy are labelled using extrinsic signals from fluorescent proteins or probes but often include intrinsic signals (autofluorescence).  
+
+<|ref|>text<|/ref|><|det|>[[115, 825, 875, 895]]<|/det|>
+Autofluorescence contributes photons that are undesired, difficult to identify and to account for in LU. The cumulative presence of noise inevitably leads to a degradation of acquired spectra during imaging. As a result, the spectral separation by LU is often compromised, and the Signal to Noise ratio (SNR) of the final unmixing is often reduced by the weakest of the signals
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 880, 197]]<|/det|>
+detected. \(^{16}\) Increasing the amount of laser excitation can partially overcome these challenges, but the higher energy deposition in the sample causes photo- bleaching and - damage, affecting both the integrity of the live sample and the duration of the observation. Traditional unmixing strategies such as LU are computationally demanding, requiring long analyses and often slowing the experiment. Combined, these compromises have reduced both the overall multiplexing capability and the adoption of HFI multiplexing technologies.  
+
+<|ref|>text<|/ref|><|det|>[[115, 203, 880, 413]]<|/det|>
+We have developed Hybrid Unmixing (HyU) as an answer to the challenges that have limited the wider acceptance of HFI for in vivo imaging. HyU employs the phasor approach \(^{18}\) merged with traditional unmixing algorithms to more rapidly and more accurately untangle the fluorescent signals from multiple exogenous and endogenous labels. The phasor approach \(^{18}\) , a popular dimensionality reduction approach for the analysis of both fluorescence lifetime and spectral image analysis \(^{19 - 21}\) provides key advantages to HyU, including spectral compression, denoising, and computational reduction. HyU pairs phasor processing with unmixing algorithms, such as LU, to provide unsupervised analysis of HFI data, removing user subjectivity. Our results show that HyU offers three key advantages: (1) improved unmixing over conventional LU, especially for low intensity images, down to 5 photons per spectra; (2) simplified identification of independent spectral components; (3) dramatically faster processing of large datasets, overcoming the typical unmixing bottleneck for in vivo fluorescence microscopy.  
+
+<|ref|>image<|/ref|><|det|>[[121, 480, 848, 800]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 810, 880, 900]]<|/det|>
+<center>Figure 1. Schematic illustrating how Hybrid Unmixing (HyU) enhances analysis of multiplexed hyperspectral fluorescent signals in vivo. (A) Multicolor fluorescent biological sample (here a zebrafish embryo) is imaged in hyperspectral mode, collecting the fluorescence spectrum of each voxel in the specimen. (B) HyU represents spectral data as a phasor plot, a 2D histogram of the real and imaginary Fourier components (at a single harmonic). (C) Spectral denoising filters reduce the Poisson and instrumental noise on the phasor histogram, providing the first signal improvement. (D) The phasor acts as an encoder, where each histogram-bin corresponds to a number n of </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 872, 122]]<|/det|>
+pixels, each with a relatively similar spectrum (E). Summing these spectra effectively averages the spectra for that phasor position.  
+
+<|ref|>text<|/ref|><|det|>[[115, 121, 840, 180]]<|/det|>
+This denoising results in cleaner average spectrum for this set of pixels, which are ideally suited for analytical decomposition through unmixing algorithms (F). (G) Unmixing results in images that separated into spectral components. Here, linear unmixing (LU) is used for unmixing, but HyU is compatible with any unmixing algorithm.  
+
+<|ref|>text<|/ref|><|det|>[[115, 178, 881, 225]]<|/det|>
+Note that HyU offers a major reduction in data size and complexity of the LU (or any other unmixing) computation, because the calculation is applied to the \(10^{4}\) histogram bins (D), rather the the \(\sim 10^{7}\) voxels in the specimen (A). This reduces the number of calculations required for LU dramatically.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 244, 179, 261]]<|/det|>
+## Results  
+
+<|ref|>text<|/ref|><|det|>[[113, 282, 884, 685]]<|/det|>
+HyU combines the best features of hyperspectral phasor analysis and linear unmixing (LU), resulting in faster computation speeds and more reliable results, especially at low light levels. Phasor approaches reduce the computational load because they are compressive, reducing the 32 channels of an HFI spectral plot into a position on a 2D- histogram, representing the real and imaginary Fourier components of the spectrum (Fig. 1A,B). Different 32 channel spectra are represented as different positions on the 2D phasor plot, and mixtures of the two spectra will be rendered at a position along a line connecting the pure spectra. Because the spectral content of an entire 2D or 3D image set is rendered on a single phasor plot, there is a dramatic data compression - from a spectrum for each voxel in an image set (up to or even beyond Gigavoxels) to a histogram value on the phasor plot (Megapixels). In addition, because each "bin" on the phasor plot histogram corresponds to multiple voxels with highly similar spectral profiles, the binning itself represents spectral averaging, which reduces the Poisson and instrumental noise (Fig. 1C- E). Poisson noise in the collected light is unavoidable in HFI unless the excitation is turned so high that the statistics of collected fluorescence creates hundreds or thousands of photons per spectral bin. The clear separation of the spectral phasor plot and its referenced imaging data, permits denoising algorithms to be applied to phasor plot with minimal degradation of the image resolution. LU or other unmixing approaches can be applied to the spectra on the phasor plot, offering a dramatic reduction in computational burden for large image data sets (Fig. 1D). To understand this saving, consider the conventional approach of LU applied to image data at the voxel level (Fig. 1A,F). A timelapse volumetric dataset of \(512 \times 768 \times 17\) (x, y, z) pixels, over 6 timepoints, (Sup. table 1), would require 40 million operations. HyU's requires only \(\sim 18\) thousand operations to unmix each bin on the phasor plot, representing more than a thousand- fold saving (Fig. 1F,G).  
+
+<|ref|>text<|/ref|><|det|>[[115, 715, 884, 908]]<|/det|>
+To quantitatively assess the relative performance of LU and HyU, we analyzed them on synthetic hyperspectral fluorescent datasets, created by computationally modelling the biophysics of fluorescence spectral emission and microscope performance (Fig 2 A, B, figs. S3- S5). We used this synthetic dataset to evaluate LU and HyU algorithm performance quantitatively by using metrics such as Mean Square Error (MSE) and unmixing residual (see Fig. S6, Methods; for both metrics, a lower value indicates better performance). In addition to the computational efficiency mentioned above, HyU analysis shows better ability to capture spatial features over a wide dynamic range of intensities, when compared with standard LU, in large part due to the denoising created by processing in phasor space (Fig. 2 A, B). The improved accuracy is demonstrated by a lower MSE, in comparing the results of LU and HyU to the image ground truth. The absolute MSE for HyU is consistently up to 2x lower than that of LU, especially at
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 90, 882, 336]]<|/det|>
+low and ultra- low fluorescence levels (Fig. 2C). MSE can be further decreased by the use of denoising filters on the phasor plot, resulting in superiority of HyU relative to LU for HFI at low (5- 20 photons/spectrum) and ultralow (2- 5 photons/spectrum) levels (Fig. 2D). To better characterize the performance in the experimental data without ground truth, we also define the unmixing residual as the difference between the original multichannel hyperspectral images and their unmixed results. Residuals provide a measure of how closely the unmixed results reconstruct the original signal (Fig. S3, Methods). Unmixing residuals are inversely proportional to the performance of the algorithm, with low residuals indicating high similarity between the unmixed and the original signals. Analysis of unmixing residuals in the synthetic data highlights an improved interpretation of the spectral information in HyU with an average unmixing residual reduction of \(21\%\) compared to the standard (Fig. S5C). The reduction in both MSE and average unmixing residual for synthetic data demonstrates the superior performance of HyU, and provides a baseline comparison when demonstrating performance improvements for experimental data.  
+
+<|ref|>text<|/ref|><|det|>[[115, 342, 877, 622]]<|/det|>
+We support the enhanced performance of HyU with analysis of experimental data, which reveals comparatively lower unmixing residuals and a higher dynamic range as compared to LU. Data was acquired from a quadra- transgenic zebrafish embryo Tg(ubiq:Lifeact- mRuby); Gt(cltca- citrine); Tg(ubiq:lyn- tdTomato); Tg(fli1:mKO2), labelling actin, clathrin, plasma membrane, and pan- endothelial cells, respectively (Figs. 2E- L, 3, S7- S9, Supplementary Movie 1). HyU unmixing of the data shows minimal signal cross- talk between channels while LU presents noticeable bleed- through (Fig. 2M- P). Consistently with synthetic data, we utilize the unmixing residual as the main indicator for quality of the analysis in experimental data, owing to the absence of a ground truth. The residual images (Fig. 2F, G) depict a striking difference in performance between HyU and LU. The average relative residual of HyU denotes a 7- fold improvement compared to LU (Fig. 2H) in disentangling the fluorescent spectra. We visualize the unmixed channels independently (Fig. 2, I to L), zooming in on details (Fig.2 I to P) to highlight areas affected by bleed- through and which are difficult to unmix. HyU, with contrast 2- fold higher than standard LU, reduces bleed- through effects and produces images with sharper spatial features, leading to better interpretation of the experimental data (Fig. 2 K, L, fig. S7, Methods).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[122, 88, 850, 765]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 767, 875, 900]]<|/det|>
+<center>Figure 2: Hybrid Unmixing outperforms standard Linear Unmixing (LU) in both synthetic and live spectral fluorescence imaging. (A) Hybrid Unmixing (HyU) and (B) Linear Unmixing (LU) tested using a hyperspectral fluorescence simulation that was generated from four fluorescent signatures (emission spectra, Sup fig 5E). (C) Absolute Mean Squared Error (MSE) shows that HyU offers a consistent reduction in error across a broad range of photons per spectra (#photons/independent spectral components, here resulting from 4 reference spectra combined). (D) The performance differences in the MSE of HyU relative to LU persists when applying multiple phasor denoising filters (0 to 5 median filters). The analysis of this synthetic data shows the consistent improvement of HyU at low photon counts with over a 2-fold improvement when 5 denoising filters are applied at a signal level of 16 photons per spectrum. (E) Unmixing of experimental data from a 4-color zebrafish shows increased contrast for </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 884, 207]]<|/det|>
+HyU (left) compared to LU (right). Scale bar \(= 50 \mu \mathrm{m}\) . (F, G) The increased accuracy is revealed by residual images of HyU and LU, showing the spatial distribution of unassigned signals after the analysis of data in E. The results show consistently lower residual values for HyU (F) compared to LU (G). (H) Box plots of the residuals in F and G presents values of \(11\%\) for HyU compared to \(77\%\) for LU with \(*(p < 10^{-10})\) . (I-L) Enlarged rendering of HyU results (E, white box) clearly shows low levels of bleed-through between labels (M-P) Similar enlargement of LU results show noticeably worse performance. Note that regions with bright signals (membrane J, N white arrow) bleed through other channels (M) and (O). Scale bar: \(20 \mu \mathrm{m}\) . Tetra-labeled specimen used here was Gt(cltca- citrine); Tg(ubiq:lyn-tdTomato; ubig:Lifeact-mRuby;fli1:mKO2)
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[115, 87, 880, 830]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 831, 884, 904]]<|/det|>
+<center>Figure 3: Hybrid Unmixing enhances unmixing for low-signal in vivo multiplexing and achieves deeper volumetric imaging. (A) Hybrid Unmixing (HyU) volumetric renderings compared to those of (B) Linear Unmixing (LU) for the trunk portion in a 4-color zebrafish demonstrate an increased contrast and reduced residual in HyU results, especially over deeper parts of the sample. The 4 labels in the fish are Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact-mRuby;fli1:mKO2), respectively labeling clathrin-coated pits (green), membrane (yellow), </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 877, 268]]<|/det|>
+actin (cyan) and endothelial (magenta). (C,E) HyU results have increased spatial resolution and less bleed though comparing to those of (D,F) LU. Scale bar: \(20 \mu \mathrm{m}\) . When observing the zoomed- in visualization of the surface region of the sample, the yellow signal distinctly marks the membrane and the cyan signal clearly labels the actin in (C) HyU. The same signals are not distinct in (D) LU because of multiple incorrectly assigned magenta pixels that bleed through compromising the true signal in other channels. Similarly, for the zoomed- in visualization of the Perivascular region of the embryo, in (E) HyU, the yellow and magenta signals clearly distinguish the membrane and vasculature while in (F) LU, the results are corrupted by greater noise. (G,H) Intensity line plots of each of the four results signals for HyU (solid) and LU (dashed) demonstrate the improved profiles with greatly reduced noise peaks in HyU as compared to LU. Intensities are scaled by the maximum of each unmixed channel. DL: digital level. (I) Box plots of the relative residual values as a function of z depth for HyU and LU highlight the improvements in the unmixing results. HyU has an unmixing residual of \(6.6\% \pm 5.3\%\) compared to LU's \(58\% \pm 17\%\) . The average amount of residual is 9- fold lower in HyU with narrower variance of residual.  
+
+<|ref|>text<|/ref|><|det|>[[115, 281, 884, 580]]<|/det|>
+Applying HyU to another HFI dataset further highlights HyU's improvements in noise reduction and reconstitution of spatial features for low- photon unmixing. (Figs. 3, S8). In the zoomed- in image of a single slice of the embryo skin surface, acquired in the trunk region, the HyU image correctly does not display pan- endothelial (magenta) signal in the periderm, an area which should be devoid of endothelial cells and mKO2 signal (Fig. 3C). In contrast, the result from LU shows visually distinctive pan- endothelial signal throughout the tissue plane (Fig. 3D). This incorrect estimation of the relative contribution of mKO2 fluorescence for LU is possibly due to the presence of noise, corrupting the spectral profiles. This is further delineated in the intensity profiles of the mKO2 signal between HyU and LU with much higher individual peaks from noise demonstrated for LU (Fig. 3G, lower left). Intensity profiles for both magnified cross- sections of the volume (Fig. 3C- F) provide a striking visualization of the improvements of HyU. The line intensity profiles in HyU present reduced noise and represent more closely the expected distribution of signals (Fig. 3G,H). The visible micro patterns of actin on the membrane of the periderm suggest that the improvements quantified with synthetic data are maintained in live samples' signals and geometrical patterns of microridges<sup>22</sup>. By contrast, noise corruption and the presence of misplaced signals are characterized in the results from LU, with high frequency intensity variations that mis- match both the labeling and biological patterns.  
+
+<|ref|>text<|/ref|><|det|>[[115, 610, 879, 735]]<|/det|>
+HyU is more accurate and results in more reliable unmixing results across the depth of sample with greatly reduced unmixing residuals. The average residual for HyU is 9- fold lower than that of LU with a 3- fold narrower variance. (Figs. 3I, S8). This reduction in the residual is consistent with increasing z- depth where HyU unmixing results stably maintain both lower residuals and variance on average. These reduced residuals correspond both to a mathematically more precise and more uniform decomposition of signals as illustrated by the distribution of residuals versus photons (Figs. S8E, F, S14).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[117, 103, 880, 475]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 488, 881, 620]]<|/det|>
+<center>Figure 4: HyU reveals the dynamics of developing vasculature by enabling multiplexed volumetric time-lapse. Hybrid Unmixing (HyU) overcomes challenges in performing multiplexed volumetric time-lapse in vivo imaging of a developing embryo. Here we present this (A) HyU rendering for the trunk portion of a 3-color zebrafish Gt/cltca-citrine); Tg(kdrl:mCherry;fli1:mKO2) at timepoint 0. (B) HyU unmixed results allow for quantitative analysis and segmentation, here an example representing the time evolution of the segmented volumes of mCherry (vasculature, magenta) mKO2(endothelial-lymphatics, yellow) and citrine (clathrin-coated pits, cyan). (C1-4) Time lapse imaging of the formation of the vasculature over 300 mins (zoomed-in rendering of the box in A) at 0, 100, 200, 300 minutes. This show that HyU provides good unmixing at low light levels to permit multiplexing to be used in the observation of development of a live embryo. </center>  
+
+<|ref|>text<|/ref|><|det|>[[115, 652, 880, 881]]<|/det|>
+We utilized HyU's increased sensitivity to overcome common challenges of multiplexed imaging such as poor photon yield and spectral cross- talk and were able to visualize dynamics in a developing zebrafish embryo. We used a triple- transgenic zebrafish embryo with labeled pan- endothelial cells, vasculature, and clathrin- coated pits (Tg(fli1:mKO2); Tg(kdrl:mCherry); Gt(cltca- Citrine)). Multiplexing these spectrally close fluorescent proteins is enabled by HyU's increased sensitivity at lower photon counts. The increased performance at lower SNR allowed us to maintain high quality results (Fig. 4, Supplementary Movie 2) while performing faster acquisitions and reducing photon- damage through lower excitation laser power and pixel dwell time. Decreased experimental requirements allow for tiling of larger volumes, extending the field- of- view while still providing enough time resolution for developmental events, even with a high number of multiplexed fluorescent signals. The time- lapses visualize the formation of ventral vasculo- endothelial protrusions acquired in parallel to the development of clathrin and kdrl. HyU enables comparative quantifications of spatio- temporal features, allowing for the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 880, 127]]<|/det|>
+determination of volumetric changes over lengthy timelapses, in this case, over the course of 300 minutes (Fig. 4B) \(^{23,24}\) .  
+
+<|ref|>text<|/ref|><|det|>[[115, 157, 875, 386]]<|/det|>
+HyU provides the ability to combine the information from intrinsic and extrinsic signals during live imaging of samples, at both single (Fig. 5) and multiple time points (Fig. 6). The graphical representation of phasors allows identification of unexpected intrinsic fluorescence signatures in a quadra- transgenic zebrafish embryo \(Gt(cltca- citrine);Tg(ubiq:lyn- tdTomato;ubiq:Lifeact- mRuby;fli1:mKO2)\) , imaged with single photon (488 and 561nm excitation) (Fig. 5A- D). The elongated distribution on the phasor (Fig. 5C) highlights the presence of an additional, unexpected spectral signature, related to strong sample autofluorescence (Fig. 5D blue). HyU analysis of the sample, inclusive of this additional signal, provides separation of the contributions of 5 different fluorescent spectra with residual \(3.9\% \pm 0.3\%\) . HyU allows for reduced energy load, tiled imaging of the entire embryo without perturbing its development or depleting its fluorescence signal (Fig. 5A). The higher speed, lower power imaging allows for subsequent re- imaging of the same sample, as we report in the zoomed high- resolution acquisitions of the head section (Fig. 5B,E).  
+
+<|ref|>text<|/ref|><|det|>[[115, 392, 880, 586]]<|/det|>
+With the ability to unmix low photon signals, HyU enables imaging and decoding of intrinsic signals, which are inherently low light. Two photon lasers are ideal for exciting and imaging blue- shifted intrinsic fluorescence from samples \(^{25,26}\) . Here, the same quadra- transgenic sample is imaged using 740 nm excitation to access both intrinsic and extrinsic signals (Fig 5 E- G, sup Note 2). HyU enables unmixing of at least 9 intrinsic and transgenic fluorescent signals (Fig. 5), recovering fluorescent intensities from labels illuminated at a sub- optimal excitation wavelength (Fig. 5E). The spectra for intrinsic fluorescence were obtained from in vitro measurements and values reported in literature (Methods). For this sample the intrinsic signals arise from events related mainly with metabolic activity (NADH and Retinoids) \(^{27 - 31}\) , tissue structure (elastin) \(^{32}\) , and illumination (laser reflection) (Fig. 5E). These results confirm our conclusion that HyU is a powerful tool for allowing the imaging and analysis of endogenous labels.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[150, 99, 844, 817]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 820, 880, 909]]<|/det|>
+<center>Figure 5: HyU enables identification and unmixing of low photon intrinsic signals in conjunction with extrinsic signals. (A) HyU results of a whole zebrafish embryo provide a frame of reference not only for the improved unmixing of extrinsic signals, but also its increased sensitivity which enables identification and unmixing of intrinsic signals which inherently exist in a low-photon environment. (B) HyU results of the head region (box in A) reveal the simplicity of identifying an unknown autofluorescence signal among multiple extrinsic signals using the phasor method for a quadra-transgenic zebrafish Gt(cltca-citrine); Tg(ubiq:lyn-tdTomato;ubiq:Lifeact- </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 91, 880, 270]]<|/det|>
+mRuby;fli1:mKO2) imaged over multiple tiles. Scale bar: \(80\mu \mathrm{m}\) . (C) The input spectra required to perform the unmixing are easily identified on (D) the phasor plot when visualizing each spectrum as a spatial location. Phasors offer a simplified identification and selection of independent and unexpected spectral components in the encoded HyU approach. Intrinsic signals are notoriously low in emitted photons leading to an inability to unmix using traditional unmixing algorithms. (E) The zoomed- in acquisition of the head region of the embryo (box in A) displays HyU's unmixing results of many intrinsic and extrinsic signals when in an environment of very low photon output, a previously highly difficult experimental condition to unmix. Scale bar: \(70\mu \mathrm{m}\) . (F) The phasor plot representation provides the easily identifiable eight independent fluorescent fingerprint locations. (G) The spectra corresponding to each of the eight independent spectral components are also provided a reference. Colors in (F) match renderings in (E) and (G): NADH bound (red), NADH free (yellow), retinoid (magenta), retinoic acid (cyan), reflection (green), elastin (purple) and extrinsic signals: mKO2 (blue), and mRuby (orange). All signals were excited with a (A- D) single photon laser at both \(488~\mathrm{nm}\) and \(561~\mathrm{nm}\) or a (E- G) two photon laser at \(740~\mathrm{nm}\) .
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[118, 140, 873, 630]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 633, 840, 662]]<|/det|>
+<center>Figure 6: HyU pushes the upper limits of live multiplexed volumetric timelapse imaging of intrinsic and extrinsic signals. </center>  
+
+<|ref|>text<|/ref|><|det|>[[115, 661, 880, 794]]<|/det|>
+HyU's increased sensitivity provides a simple solution for the challenging task of imaging timelapse data at 6 time points (125 mins) for both intrinsic signals and extrinsic signals of a quadra- transgenic zebrafish: Tg(cltca- Citrine);(ubiq:lyn- tdTomato);(ubiq:Lifecat- mRuby);(fli1:mKO2)). (A) - (F) Volumetric renderings of HyU results for time points acquired at 25 min intervals reveal the high- contrast and - multiplexed labels of NADH bound (red), NADH free (yellow), retinoid (magenta), retinoic acid (cyan), mKO2 (green), and autofluorescence from blood cells (blue) when excited @740nm. Further extrinsic signals for mKO2 (yellow), tdTomato (magenta), mRuby (cyan), Citrine (green) and blood cells autofluorescence (blue) are also readily unmixed using HyU when exciting the sample @ 488/561nm. HyU provides the capacity to simultaneously multiplex 9 signals in a live sample over long periods of time, a previously unexplored task. Scale bar: 50 μm.  
+
+<|ref|>text<|/ref|><|det|>[[115, 849, 882, 902]]<|/det|>
+Finally, we exploited the HyU capabilities to multiplex volumetric timelapse of extrinsic and intrinsic signals by imaging the tail region of the same quadra- transgenic zebrafish embryo. We excite extrinsic labels at 488/561 nm and the intrinsic signals with 740 nm two photon, collecting
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[116, 90, 882, 144]]<|/det|>
+6 tiled volumes over 125 mins (Figs. 6, S9- S11, Supplementary Movie 3). HyU unmixing in this sample allows for distinction of 9 signals, separating their contributions with sufficiently low requirements to allow repeated imaging of notoriously low SNR intrinsic fluorescence.  
+
+<|ref|>sub_title<|/ref|><|det|>[[116, 176, 207, 193]]<|/det|>
+## Discussion  
+
+<|ref|>text<|/ref|><|det|>[[116, 201, 850, 272]]<|/det|>
+Our results reveal the advantages of Hybrid Unmixing (HyU) over more conventional Linear Unmixing (LU) in performing complex multiplexing experiments. HyU overcomes the significant challenges of separating multiple fluorescent and autofluorescent labels with overlapping spectra while minimally perturbing the sample with excitation light.  
+
+<|ref|>text<|/ref|><|det|>[[115, 278, 880, 524]]<|/det|>
+The chief advantage of HyU is its multiplexing capability when imaging in the presence of biological and instrumental noise, especially at low signal levels. HyU increased sensitivity improves multiplexing in photon limited applications (Fig. 2F- L), in deeper volumetric acquisitions (Fig. 3I) and in signal starved imaging of autofluorescence (Fig. 5E, Fig. 6). Our simulation results (Fig. 2) demonstrate that HyU improves unmixing of spatially and spectrally overlapping fluorophores excited simultaneously. The increased robustness at low photon imaging conditions reduces the imaging requirements for excitation levels and detector integration time, allowing for imaging with reduced photo- toxicity. Live imaging on multi- color samples performed at high sampling frequency enables improved tiling to increase the field- of- view (Fig. 3, 4) while maximizing the usage of the finite fluorescent signals over time. Two- photon imaging of intrinsic and extrinsic signals suggests the ability of HyU to multiplex signals with large dynamic range differences (Fig. 5) extending multiplexed volumetric imaging into the time dimension (Fig. 6). Although improved, images with particularly low signal still present corruption (Fig. S4), setting a reasonable range of utilization above 8 photons/spectrum.  
+
+<|ref|>text<|/ref|><|det|>[[115, 529, 881, 669]]<|/det|>
+Simplicity of use and versatility are other key advantages of HyU, inherited from both the phasor approach<sup>33</sup> and traditional unmixing algorithms. Phasors here operate as a spectral encoder, reducing computational load and integrating similar spectral signatures in histogram bins of the phasor plot. This representation simplifies identification of independent spectral signatures (Fig. 5, Supplementary Note1) through both phasor plot selection and phasor residual mapping (fig. S11), accounting for unexpected intrinsic signals (Figs. 5, 6, S12, Supplementary Note2) in a semi- automated manner, while still allowing fully- automated analysis by means of spectral libraries.  
+
+<|ref|>text<|/ref|><|det|>[[115, 675, 879, 904]]<|/det|>
+The simplicity of this approach is especially helpful in live imaging where identifying independent spectral components remains an open challenge, owing to the presence of intrinsic signals (Fig. S12, Supplementary Note 1). High- SNR reference spectra can be derived from other experimental data or identified directly on the phasor. Selection of portions on the phasor plot allows for visualization of the corresponding spectra in the wavelength domain (Fig. 5C,D,F,G). This intuitive versatility allows for identification of both the number of unexpected signatures and their spectra, a task previously difficult to perform due to noise and lack of global visualization tools. In single photon imaging (Fig. 5A- D), HyU phasor allowed identification of a fifth distinct spectral component arising from general autofluorescent background, thereby improving the unmixed results. In two photon imaging, HyU enabled identification and multiplexing of 8 highly overlapping signals possessing a wide dynamic range of intensities, between intrinsic and extrinsic markers (Fig. 5F,G). Combination of single and two photon imaging increased the number of multiplexed fluorophores to 9 (Fig. 6), considering some of the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 867, 127]]<|/det|>
+extrinsic labels being excited at two photons. Multiplexing of signals may be further improved by implementing HyU on fluorescent dyes.  
+
+<|ref|>text<|/ref|><|det|>[[115, 133, 883, 291]]<|/det|>
+HyU performs better than standard algorithms both in the presence and absence of phasor noise reduction filters<sup>33</sup>. Compared with LU, the unmixing enhancement when such filters<sup>33</sup> are applied is demonstrated by a decrease of the MSE of up to \(21\%\) (Fig 2C), with a reduction of the average amount of residuals by 7- fold. Even in the absence of phasor denoising filters, HyU performs up to \(7.3\%\) better than the standard (Fig. 2D) based on Mean Squared Error of synthetic data unmixing. This base improvement is due to the averaging of similarly shaped spectra in each phasor histogram bin, which reduces the statistical variability within the spectra used for the unmixing calculations (Fig. 1E). This averaging strategy works well for general fluorescence spectra owing to their broad and mostly unique spectral shape.  
+
+<|ref|>text<|/ref|><|det|>[[115, 290, 882, 360]]<|/det|>
+In the absence of noise, for example in the ground truth simulations, LU produces an MSE 6- fold lower than HyU (Fig. S5, B, C, S6G). In these noiseless conditions, the binning and averaging of spectra in the phasor histogram, without denoising, provides statistically indifferent values of error respect to LU, suggesting results of similar quality.  
+
+<|ref|>text<|/ref|><|det|>[[115, 392, 879, 533]]<|/det|>
+HyU can interface with different unmixing algorithms, adapting to existing experimental pipelines. We successfully tested hybridization with iterative approaches such as non- negative matrix factorization<sup>34</sup>, fully constrained and non- negative least- squares<sup>35</sup> (Methods). Speed tests with iterative fitting unmixing algorithms demonstrate a speed increase of up to 500- fold when the HyU compressive strategy is applied. (Fig. S13, Supplementary Note 3). Due to the initial computational overhead for encoding spectra in phasors, there is a 2- fold speed reduction for HyU in comparison to standard LU. However, this may be improved with further optimizations of the HyU implementation.  
+
+<|ref|>text<|/ref|><|det|>[[115, 538, 879, 715]]<|/det|>
+One restriction of HyU derives from the mathematics of linear unmixing, where linear equations representing the unmixed channels need to be solved for the unknown contributions of each analyzed fluorophore. To obtain a unique solution from these equations and to avoid an underdetermined equation system, the maximum number of spectra for unmixing may not exceed the number of channels acquired<sup>36</sup>, generally 32 for commercial microscopes. This number could be increased; however, due to the broad and photon- starved nature of fluorescence spectra, acquisition of a larger number of channels could negatively affect the sample, imaging time and intensities. Depending on the number of labels in the specimen of interest, extending the number of labels to simultaneously unmix beyond 32 will likely require spectral resolution upsampling strategies.  
+
+<|ref|>text<|/ref|><|det|>[[115, 720, 868, 809]]<|/det|>
+HyU improvement is related to the presence of various types of noise in microscopy images, such as Gaussian, Poisson and digital as well as unidentified sources of spectral signatures (Fig. S5B,C, S6G). In the multiplexing of fluorescent signals, HyU offers improved performance, quality- and speed- wise in the low- signal regime. HyU is poised to be used in the context of in vivo imaging, harvesting information from samples labeled at endogenous- level.  
+
+<|ref|>text<|/ref|><|det|>[[115, 841, 864, 894]]<|/det|>
+In conclusion, the results presented in this paper quantitatively show that HyU, a phasor based, computational unmixing framework, is well suited for tackling the many challenges present in live imaging of multiple fluorescence labels. HyU's reduced requirements in amount of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 872, 250]]<|/det|>
+fluorescent signal permit a reduction of laser excitation load and imaging time. These factors enable multiplexed imaging of biological events with longer duration, higher speed and lower photo- toxicity while providing access to information- rich imaging across different spatiotemporal scales. The reduced requirements of HyU make it fully compatible with any commercial and common microscopes capable of spectral detection, facilitating access to the technology. Our analysis demonstrates HyU's robustness, simplicity and improvement in identifying both new and known spectral signatures, and vastly improved unmixing outputs, providing a much- needed tool for delving into the many questions still surrounding studies with live imaging.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[116, 91, 305, 109]]<|/det|>
+## References and Notes:  
+
+<|ref|>text<|/ref|><|det|>[[112, 115, 879, 901]]<|/det|>
+1. Valm, A. M. et al. Applying systems-level spectral imaging and analysis to reveal the organelle interactome. Nature 546, 162-167 (2017).  
+2. Tsurui, H. et al. Seven-color fluorescence imaging of tissue samples based on fourier spectroscopy and singular value decomposition. Journal of Histochemistry and Cytochemistry 48, 653-662 (2000).  
+3. Amat, F. et al. Fast, accurate reconstruction of cell lineages from large-scale fluorescence microscopy data. Nature Methods 11, 951-958 (2014).  
+4. Ueno, T. & Nagano, T. Fluorescent probes for sensing and imaging. Nature Methods vol. 8 642-645 (2011).  
+5. Lichtman, J. W. & Conchello, J. A. Fluorescence microscopy. Nature Methods vol. 2 910-919 (2005).  
+6. Sinclair, M. B., Haaland, D. M., Timlin, J. A. & Jones, H. D. T. Hyperspectral confocal microscope. 45, 6283-6291 (2006).  
+7. Jahr, W., Schmid, B., Schmied, C., Fahrbach, F. O. & Huisken, J. Hyperspectral light sheet microscopy. Nature Communications 6, 1-7 (2015).  
+8. Truong, T. V., Supatto, W., Koos, D. S., Choi, J. M. & Fraser, S. E. Deep and fast live imaging with two-photon scanned light-sheet microscopy. 8, (2011).  
+9. Chen, B. C. et al. Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution. Science 346, (2014).  
+10. Kredel, S. et al. mRuby, a bright monomeric red fluorescent protein for labeling of subcellular structures. PLoS ONE 4, 1-7 (2009).  
+11. Sakaue-Sawano, A. et al. Visualizing Spatiotemporal Dynamics of Multicellular Cell-Cycle Progression. Cell 132, 487-498 (2008).  
+12. Wade, O. K. et al. 124-Color Super-resolution Imaging by Engineering DNA-PAINT Blinking Kinetics. Nano Letters 19, 2641-2646 (2019).  
+13. Strauss, S. & Jungmann, R. Up to 100-fold speed-up and multiplexing in optimized DNA-PAINT. Nature Methods 17, 789-791 (2020).  
+14. Zimmermann, T., Rietdorf, J. & Pepperkok, R. Spectral imaging and its applications in live cell microscopy. FEBS Letters 546, 87-92 (2003).  
+15. Paddock, S. Multi-Spectral Imaging and Linear Unmixing Add a Whole New Dimension to. BioTechniques 31, 1272-1278 (2001).  
+16. Zimmermann, T. Spectral imaging and linear unmixing in light microscopy. Advances in Biochemical Engineering/Biotechnology 95, 245-265 (2005).  
+17. Garini, Y., Young, I. T. & McNamara, G. Spectral imaging: Principles and applications. Cytometry Part A vol. 69 (2006).  
+18. Digman, M. A., Caiolfa, V. R., Zamai, M. & Gratton, E. The Phasor Approach to Fluorescence Lifetime Imaging Analysis. Biophysical Journal 94, L14-L16 (2008).  
+19. Fereidouni, F., Bader, A. N., Colonna, A. & Gerritsen, H. C. Phasor analysis of multiphoton spectral images distinguishes autofluorescence components of in vivo human skin. Journal of Biophotonics 7, 589-596 (2014).  
+20. Scipioni, L., Rossetta, A., Tedeschi, G. & Gratton, E. Phasor S-FLIM: a new paradigm for fast and robust spectral fluorescence lifetime imaging. Nature Methods 18, 542-550 (2021).  
+21. Ranjit, S., Malacrida, L., Jameson, D. M. & Gratton, E. Fit-free analysis of fluorescence lifetime imaging data using the phasor approach. Nature Protocols 13, 1979-2004 (2018).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 90, 880, 707]]<|/det|>
+22. Depasquale, J. A. Actin Microridges. Anatomical record (Hoboken, N.J. : 2007) 301, 2037-2050 (2018).23. Okuda, K. S., Hogan, B. M., Cantelmo, A. R. & Hogan, B. M. Endothelial Cell Dynamics in Vascular Development : Insights From Live-Imaging in Zebrafish. 11, (2020).24. Isogai, S., Lawson, N. D., Torrealday, S., Horiguchi, M. & Weinstein, B. M. Angiogenic network formation in the developing vertebrate trunk. (2003) doi:10.1242/dev.00733.25. Denk, W., Strickler, J. H. & Webb, W. W. Two-photon laser scanning fluorescence microscopy. Science 248, (1990).26. Zipfel, W. R. et al. Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation. Proceedings of the National Academy of Sciences 100, 7075-7080 (2003).27. Bird, D. K. et al. Metabolic mapping of MCF10A human breast cells via multiphoton fluorescence lifetime imaging of the coenzyme NADH. Cancer Research 65, (2005).28. Lakowicz, J. R., Szmacinski, H., Nowaczyk, K. & Johnson, M. L. Fluorescence lifetime imaging of free and protein-bound NADH. Proceedings of the National Academy of Sciences of the United States of America 89, (1992).29. Skala, M. C. et al. In vivo multiphoton microscopy of NADH and FAD redox states, fluorescence lifetimes, and cellular morphology in precancerous epithelia. Proceedings of the National Academy of Sciences of the United States of America 104, 19494-19499 (2007).30. Sharick, J. T. et al. Protein-bound NAD(P)H Lifetime is Sensitive to Multiple Fates of Glucose Carbon. Scientific Reports 8, (2018).31. Stringari, C. et al. Phasor approach to fluorescence lifetime microscopy distinguishes different metabolic states of germ cells in a live tissue. Proceedings of the National Academy of Sciences of the United States of America 108, (2011).32. Wagnieres, G. A., Star, W. M. & Wilson, B. C. Invited Review In Vivo Fluorescence Spectroscopy and Imaging for Oncological Applications. 68, 603-632 (1998).33. Cutrale, F. et al. Hyperspectral phasor analysis enables multi-plexed 5D in vivo imaging. Nature Publishing Group (2017) doi:10.1038/nmeth.4134.34. Févotte, C. & Dobigeon, N. Nonlinear hyperspectral unmixing with robust nonnegative matrix factorization. IEEE Transactions on Image Processing 24, (2015).35. Heslop, D., von Dobeneck, T. & Höcker, M. Using non-negative matrix factorization in the "unmixing" of diffuse reflectance spectra. Marine Geology 241, 63-78 (2007).36. Paddock, S. W. Confocal Microscopy, Methods and Protocols, Second Edition. Humana Press 1075, 388 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[115, 747, 860, 802]]<|/det|>
+Acknowledgments: The authors would like to thank F. Schneider, S. Restrepo (Translational Imaging Center, University of Southern California), Chi- Li Chiu (Bitplane Inc.) and Samuel Ojosnegros (Institute for Bioengineering of Catalonia) for helpful discussions.  
+
+<|ref|>text<|/ref|><|det|>[[115, 825, 861, 878]]<|/det|>
+Funding: This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant DGE- 1842487, Department of Defense PR150666 and University of Southern California.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 90, 861, 180]]<|/det|>
+Author contributions: H.J.C., D.E.S.K., and F.C. analyzed the results and wrote the software. H.J.C., D.E.S.K., J.U. and F.C. provided conceptualization. H.J.C., D.E.S.K., M.K., F.C, and L.A.T. helped in the experimental design and data analysis. M.K. generated the inducible and mKO2 zebrafish lines. H.J.C. and F.C. acquired data. S.E.F. provided supervision. H.J.C., D.E.S.K., and F.C. wrote the paper. L.A.T., J.U., M.K. supported review and editing.  
+
+<|ref|>text<|/ref|><|det|>[[115, 186, 833, 222]]<|/det|>
+Competing interests: The University of Southern California has filed a provisional patent application covering this method.  
+
+<|ref|>text<|/ref|><|det|>[[115, 228, 880, 316]]<|/det|>
+Data and materials availability: All the relevant data are available from the corresponding author upon reasonable request. Datasets for Figs. 1–6 and simulations are available for download at http://bioimaging.usc.edu/software.html#sampledatasets in the samples section. All the relevant code is available from the corresponding author upon reasonable request. Software and instructions can be downloaded from http://bioimaging.usc.edu/software.html.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 93, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 130, 550, 283]]<|/det|>
+SupplementaryMovie1v3. mp4SupplementaryMovie24. mp4SupplementaryMovie3v3. mp4SupplementaryMovieCaptions211108DKFC.pdfHybUnmSupplementaryMaterialsNM20211111FC.pdfREADME2021201FC.pdf
+
+<--- Page Split --->

preprint/preprint__98e188d197f798d498e66156851783607bd993dcbb8ac0d0cb28a16b4678777a/images_list.json

@@ -0,0 +1,62 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Fig. 1. Element specific XAS \\(L_{3}\\) curves (right side of graphs as white lines) and corresponding ResPES EDC for the elements of the \\(\\mathrm{CrMnFeCoNi}\\) HEA. In the EDC the superimposed white line corresponds to the EDC intensities at the XAS \\(L_{3}\\) maximum. Energetic positions of the XAS measurements and EDCs are marked by arrows.",
+    "footnote": [],
+    "bbox": [
+      [
+        80,
+        259,
+        470,
+        655
+      ]
+    ],
+    "page_idx": 3
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Fig. 2 Element resolved electronic spectra of CrMnFeCoNi. (a) Calculated pDOS for elements in CrMnFeCoNi with LDA and LDA+DMFT as yellow and blue solid lines, respectively. Measured EDCs in the XAS \\(L_{3}\\) absorption maximum as black solid lines. The self-convolution of the pDOS (CST) are given as dashed lines with corresponding color. For Ni and Fe, the LDA+DMFT pDOS for \\(U = 4 \\mathrm{eV}\\) and \\(2 \\mathrm{eV}\\) , respectively are given in green. (b) Experimental VB PES next to calculated spectra using the one-step model with LDA and LDA+DMFT. (c) and (d) Calculated BSF for the CrMnFeCoNi HEA in the LDA and LDA+DMFT approach.",
+    "footnote": [],
+    "bbox": [
+      [
+        70,
+        90,
+        475,
+        460
+      ]
+    ],
+    "page_idx": 4
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Fig. 3 Element specific quasiparticle lifetimes and frequency resolved optical conductivity of the \\(\\mathrm{CrMnFeCoNi}\\) HEA. (a) Element specific lifetimes for Fe and Ni by black lines for calculated pure elements in natural crystal and yellow line pure metals within the fcc structure and \\(\\mathrm{CrMnFeCoNi}\\) lattice constant. Experimental values for are taken from PES (below \\(\\mathrm{E}_{\\mathrm{F}})^{40}\\) and TR-2PPE (above \\(\\mathrm{E}_{\\mathrm{F}})^{41 - 43}\\) data of pure elements. Blue lines in the bottom correspond to LDA+DMFT and green dashed line to the LDA+DMFT case with changed \\(U\\) for Fe \\(= 2\\mathrm{eV}\\) and \\(\\mathrm{Ni} = 4\\mathrm{eV}\\) . (b) Real (left) and imaginary part (right) of complex optical conductivity versus photon energy \\(\\hbar \\omega\\) . LDA calculations are given by the yellow line and LDA+DMFT calculations by the blue line. Experiments are from reflectometry measurements (Exp. 1) and ellipsometry (Exp. 2).",
+    "footnote": [],
+    "bbox": [
+      [
+        72,
+        58,
+        480,
+        421
+      ]
+    ],
+    "page_idx": 5
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Fig. 4 Measurement and calculation of temperature electrical resistivity. (a) Measurements of electrical resistivity versus temperature for several Cantor-Wu alloys, ranging from FeNi to the CrMnFeCoNi HEA. (b) Comparison of experiment and linear response calculation for LDA and LDA+DMFT case for CrMnFeCoNi. (c) and (d) Fermi surface of CrMnFeCoNi for the LDA and LDA+DMFT potential.",
+    "footnote": [],
+    "bbox": [
+      [
+        70,
+        92,
+        460,
+        373
+      ]
+    ],
+    "page_idx": 6
+  }
+]

preprint/preprint__98e188d197f798d498e66156851783607bd993dcbb8ac0d0cb28a16b4678777a/preprint__98e188d197f798d498e66156851783607bd993dcbb8ac0d0cb28a16b4678777a.mmd

@@ -0,0 +1,251 @@
+
+# Interplay between disorder and electronic correlations in compositionally complex alloys  
+
+Jan Minar jminar@ntc.zcu.cz  
+
+University of West Bohemia https://orcid.org/0000- 0001- 9735- 8479  
+
+Munich University of Applied Sciences https://orcid.org/0000- 0002- 7306- 2232  
+
+Saleem Khan University of West Bohemia  
+
+Edoardo Martino École Polytechnique Fédérale de Lausanne  
+
+Xavier Mettan École Polytechnique Fédérale de Lausanne  
+
+Luka Ciric Inst of Physics  
+
+Davor Tolj EPFL  
+
+Trpimir Ivšić École Polytechnique Fédérale de Lausanne  
+
+Andreas Held Ludwig- Maximilians- University Munich  
+
+Marco Caputo Paul Scherrer Institute  
+
+Vladimir Strocov Paul Scherrer Institut https://orcid.org/0000- 0002- 1147- 8486  
+
+Igor Marco Uppsala University  
+
+Hubert Ebert Ludwig- Maximilians- Universität München  
+
+Heinz Huber Munich University of Applied Sciences https://orcid.org/0000- 0003- 2444- 9833  
+
+Hugo Dil École Polytechnique Fédérale de Lausanne https://orcid.org/0000- 0002- 6016- 6120  
+
+László Forró
+
+<--- Page Split --->
+
+# Article  
+
+# Keywords:  
+
+Posted Date: February 29th, 2024  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 3981895/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Additional Declarations: There is NO Competing Interest.  
+
+Version of Record: A version of this preprint was published at Nature Communications on September 12th, 2024. See the published version at https://doi.org/10.1038/s41467- 024- 52349- 8.
+
+<--- Page Split --->
+
+# Interplay between disorder and electronic correlations in compositionally complex alloys  
+
+David Redka \(^{1,2}\) , Saleem Ayaz Khan \(^{1}\) , Edoardo Martino \(^{3}\) , Xavier Mettan \(^{3}\) , Luka Ciric \(^{3}\) , Davor Tolj \(^{3}\) , Trpimir Ivšić \(^{3}\) , Andreas Held \(^{4}\) , Marco Caputo \(^{5}\) , Vladimir N. Strocov \(^{5}\) , Igor Di Marco \(^{6,7}\) , Hubert Ebert \(^{4}\) , Heinz P. Huber \(^{1,2,*}\) , J. Hugo Dil \(^{3,5}\) , László Forró \(^{3,8}\) & Ján Minár \(^{1,*}\) \(^{1}\) New Technologies Research Center, University of West Bohemia, Plzen CZ- 30100, Czech Republic \(^{2}\) Department of Applied Sciences and Mechatronics, Munich University of Applied Sciences HM, Munich DE- 80335, Germany \(^{3}\) Institute of Physics, École Polytechnique Fédérale de Lausanne, Lausanne CH- 1015, Switzerland \(^{4}\) Department of Chemistry, Ludwig- Maximilians- University Munich, Munich DE- 81377, Germany \(^{5}\) Photon Science Division, Paul Scherrer Institut, Villigen CH- 5232, Switzerland \(^{6}\) Institute of Physics, Nicolaus Copernicus University, Toruń PL- 87- 100, Poland \(^{7}\) Department of Physics and Astronomy, Uppsala University, Uppsala SE- 75120, Sweden \(^{8}\) Stavropoulos Center for Complex Quantum Matter, Department of Physics and Astronomy, University of Notre Dame, Notre Dame IN 46556, USA  
+
+Corresponding authors: "heinz.huber@hm.edu; \\* jminar@ntc.zcu.cz;  
+
+Exploring the intricate interplay between disorder and correlations in compositionally complex alloys, our study employs resonant and valence band photoemission, optical conductivity, and electrical resistivity, complemented by density functional theory- based linear response calculations. By applying dynamical mean- field theory, we identify correlation signatures and damping in spectra, emphasizing the significance of many- body effects, especially in states far from the Fermi edge. Electronic transport remains dominated by chemical and magnetic disorder. Our results advance understanding of element specific electronic correlations in CrMnFeCoNi, elucidating the complex physical nature of compositionally complex alloys.  
+
+Compositionally complex alloys (CCAs), which comprise the diverse range of medium- to high- entropy alloys (HEAs), are an exciting class of materials, consisting of randomly distributed multi- principal elements on crystalline lattices \(^{1,2}\) . Aiming on HEAs, the denomination stems from the entropy term overruling the enthalpy of formation of individual phases when mixing a large number of elements, hence escaping phase separation. They exhibit exceptional mechanical attributes including elevated toughness, minimal plastic deformation, and enhanced tensile and yield strengths \(^{3,4}\) , along with intriguing electronic as well phononic transport properties \(^{5 - 8}\) . CCAs, including HEAs, bridge the structural gap between crystalline solids and amorphous materials, exhibiting long- range periodicity but with atom variations on lattice sites inducing site- disorder akin to Anderson localization \(^{9}\) . For years, the question of electron propagation in such an environment has persisted \(^{10}\) , lacking the translational invariance ensuring the validity of the Bloch theorem for propagating electronic waves, yielding electronic localization. Empirical observations report electric resistivity \((\rho)\) approaching the Ioffe- Regel limit with a subdued \(\mathrm{d}\rho /\mathrm{d}T\) dependence \(^{11}\) , demonstrating that the effect of increasing residual resistivity in Cantor- Wu alloys may be linked to magnetic disorder effects \(^{12}\) . Regarding the \(\mathrm{d}\rho /\mathrm{d}T\) dependence various explanations have been suggested, including Anderson localization \(^{13}\) , or quantum interference in accordance with Mooij correlation \(^{14}\) . For the latter, it becomes evident that, in addition to electron- phonon and spin scattering, many- body effects arising from the electron- electron interaction may play a significant role. However, to date, this issue has not been thoroughly investigated.  
+
+In this work, we probe the role of electronic correlation next to disorder effects in CrMnFeCoNi, likely the most studied HEA and prototype CCA, by employing resonant (ResPES) and valence- band (VB) photoemission spectroscopy (PES), as well optical conductivity measurements. All these measurements are supported and explained by electronic structure calculations based on density functional theory (DFT) and dynamic mean- field theory (DMFT). Finally, we also provide an in- depth analysis of the temperature- dependent electrical resistivity, both theoretically and experimentally.   
+
+## Results and discussion  
+
+Element resolved photoemission spectroscopy. In ResPES, photon energies proximate to the X- Ray absorption (XAS) \(L_{3}\) - edge are deployed, engendering interference between the direct photoemission channel \((2\mathrm{p}^{6}3\mathrm{d}^{n} + h\omega \rightarrow 2\mathrm{p}^{6}3\mathrm{d}^{n - 1} + \mathrm{e}_{\mathrm{f}})\) and the dipole transition of a core electron to an unoccupied state \((2\mathrm{p}^{6}3\mathrm{d}^{n} + h\omega \rightarrow 2\mathrm{p}^{5}3\mathrm{d}^{n + 1})\) . The subsequent decay of this intermediate state \((2\mathrm{p}^{5}3\mathrm{d}^{n + 1}\rightarrow 2\mathrm{p}^{6}3\mathrm{d}^{n - 1} + \mathrm{e}_{\mathrm{f}})\) , akin to an Auger process, gives rise to a strong resonant behavior due to the interference of both channels \(^{15,16}\) . ResPES measurements reveal site- or element- specific information on the electronic structure suitable for probing complex alloys where band- overlapping or hybridization is commonplace \(^{17}\) . Element specific \(L_{3}\) XAS spectra are depicted next to ResPES photoelectron energy distribution curves (EDCs) in Fig. 1 on a binding energy \((E_{\mathrm{B}})\) scale relative to the Fermi energy \(E_{\mathrm{F}}\) (calibrated by the Fermi level of gold).  
+
+For Cr, and in contrast to the other elements, the EDCs exhibit a pronounced maximum at constant \(E_{\mathrm{B}}\) of \(1.8\mathrm{eV}\) for an extended range of incident photon energies. This suggests an absent or low contribution of Auger signals in the EDCs, due to the missing transition from the radiationless Raman scattering regime (constant binding energy) to the resonant Raman Auger regime (constant kinetic energy), close to the \(L_{3}\) XAS maximum \(^{17 - 20}\) . For pure Cr, a VB peak in the EDCs of ResPES measurements was observed at \(1.2\mathrm{eV}^{20}\) . For Mn we find an EDC maximum at the \(L_{3}\) edge at \(3.6\mathrm{eV}\) , with a faint shoulder extending up to \(E_{\mathrm{B}} = 7.5\mathrm{eV}\) , hinting to a constant kinetic energy above the resonance. For Fe, Co, and Ni, the main features of the EDCs are clearly attributable to Auger signals \(^{19 - 22}\) . The EDC maxima for the XAS \(L_{3}\) edge are \(4.7\mathrm{eV}\) , \(4.2\mathrm{eV}\) , and \(7.2\mathrm{eV}\) \(E_{\mathrm{B}}\) , respectively. In no case a
+
+<--- Page Split --->
+
+dominant VB contribution at constant \(E_{\mathrm{B}}\) is observed. Comparing these data with those of pure elements provides information on band filling, hybridization, electronic correlations, chemical disorder and crystal field effects. Ni, sharing the fcc crystal structure and comparable lattice constant with the \(\mathrm{CrMnFeCoNi}\) HEA, mainly allows for a focus on the effects induced by chemical disorder. The measured EDCs display a shift of the well- known \(6\mathrm{eV}\) satellite for pure \(\mathrm{Ni}^{23,24}\) towards \(7.2\mathrm{eV}\) in the \(\mathrm{CrMnFeCoNi}\) HEA. Shifts as large as \(1.4\mathrm{eV}\) towards higher \(E_{\mathrm{B}}\) have been observed for Ni based alloys and intermetallic compounds \(^{22,25,26}\) , and explained by \(d\) - band filling through the hybridization of wave functions located on different lattice sites after alloying with more electropositive elements \(^{25,27,28}\) .  
+
+![](images/Figure_1.jpg)
+
+<center>Fig. 1. Element specific XAS \(L_{3}\) curves (right side of graphs as white lines) and corresponding ResPES EDC for the elements of the \(\mathrm{CrMnFeCoNi}\) HEA. In the EDC the superimposed white line corresponds to the EDC intensities at the XAS \(L_{3}\) maximum. Energetic positions of the XAS measurements and EDCs are marked by arrows. </center>  
+
+Fig. 2(a) depicts the calculated \(d\) - band partial density of states (pDOS) of individual elements, for LDA (yellow solid line) and LDA+DMFT (blue solid line), next to the EDCs from Fig. 1 (black solid line). For calculation details see Methods and Supplementary Information (SI). Since we are interested only in the peak position, all graphs are normalized to their maximum. Marked differences emerge between LDA+DMFT and pure LDA, including strong satellites (Mn, Co, and Ni) and a generalized band- narrowing (Ni and Co). Eventually, the spectral weight (excluding satellites) shifts to lower \(E_{\mathrm{B}}\) by including DMFT. For Cr and Fe, \(d\) - band satellites merge with the \(sp\) - pDOS (see detailed Figures on band resolved pDOS in the SI). With increasing \(d\) - band filling, there is a progressive split- off of the formed Hubbard bands towards higher \(E_{\mathrm{B}}\) , although this effect is mitigated between Mn and Fe by the large difference in \(U\) . In Cr and Fe, the value of \(U\) is so small with respect to the bandwidth that no detached satellite peak can form; rather, it reflects a renormalization of the DOS, altering the \(d\) - band shape from rectangular to triangular. The positions of the satellites in the pDOS are compared to the ResPES data. For Mn, the small shoulder in the ResPES data at \(7.5\mathrm{eV}\) coincides with the satellite at \(8\mathrm{eV}\) in the pDOS. A comparable analysis for Fe and Co is not possible. For Ni, the pDOS data reveal a satellite at \(8.2\mathrm{eV}\) BE, elevated by \(1\mathrm{eV}\) compared to that determined experimentally via ResPES.   
+
+Element specific on- site Coulomb interaction. We apply the Cini- Sawatzky Theory (CST) \(^{29 - 31}\) by comparing the measured EDCs (assumed as Auger spectra) with the self- convoluted single- particle pDOS. Within the CST framework, Auger signals may be categorized into distinct regimes, according to the ratio between electronic bandwidth \(W\) and on- site Coulomb interaction \(U\) . For \(U \gg W\) , the spectra correspond to the quasi- atomic limit with split- off satellites at high \(E_{\mathrm{B}}\) , whereas being band- like for \(U \ll W\) . For the latter Lander \(^{32}\) proposes that the Auger signal equals the self- convolution of the single- particle band. At \(U \sim W\) , a complex interplay occurs, resulting in the superposition of both states. According to CST, for systems with nearly filled \(d\) - bands, a discernible shift of the Auger spectra towards lower \(E_{\mathrm{B}}\) is found, displacing the maximum of the self- convolution of the DOS by \(U\) relative to the Auger signals \(^{25,33,34}\) . This is explained by the energy difference of the two- hole state and the two- one- hole states being equal to the Coulomb interaction on the two- particle energy scale. Consequently, the ResPES maxima (black solid lines) are aligned with the self- convolution which are depicted in Fig. 2(a) by dashed colored lines. It is evident, that for Cr, only the VB is measured by ResPES, as the EDCs spectral weight coincidences with that of the SP band. For Mn and Fe, which are approximately situated in the band- like limit with \(U \ll W\) , the self- convolution and ResPES data on the two- electron scale show a congruence in their peaks. For Co, a minor offset of approximately \(1.6\mathrm{eV}\) is identified, which is slightly smaller than our suggested \(U\) of \(2.5\mathrm{eV}\) (see Methods). For Ni, this offset amounts to \(4\mathrm{eV}\) , which is slightly larger than the value of \(3\mathrm{eV}\) used as \(U\) in the LDA+DMFT calculation. This trend is expected, considering the different filling of the \(3d\) - band \(^{25,35 - 37}\) , as well as the limitations of the DMFT solver \(^{38}\) . In order to investigate the sensitivity of the CST we calculate the \(\mathrm{CrMnFeCoNi}\) HEA with the initially given \(U\) values, but increasing them for Fe from \(1.5\mathrm{eV}\) to \(2\mathrm{eV}\) and Ni from \(3\mathrm{eV}\) to \(4\mathrm{eV}\) . The pDOS (solid line) as well self- convoluted signal (dashed line) are given in Fig. 2(a) as green lines. For Fe there is barely a difference in the pDOS visible with a slight increase of the satellite. No shift in the self- convolution signal is found. For Ni, the \(d\) - block shifts also barely recognizably towards the Fermi edge, but the correlation satellite splits off significantly towards higher binding energies of almost \(10\mathrm{eV}\) . Consequently, the main peak of self- convolution hardly changes. The difference between EDC and self- convolution is \(4\mathrm{eV}\) , which corresponds exactly to the Hubbard \(U\) from the LDA+DMFT calculations. The variation of the element specific \(U\) value of Fe and Ni does not change the pDOS of the other elements (see the more detailed plot in the SI). However, since the extreme displacement of the split- off
+
+<--- Page Split --->
+
+satellite is not observed experimentally, we keep the initial chosen pure element \(U\) values in our calculations.  
+
+![](images/Figure_2.jpg)
+
+<center>Fig. 2 Element resolved electronic spectra of CrMnFeCoNi. (a) Calculated pDOS for elements in CrMnFeCoNi with LDA and LDA+DMFT as yellow and blue solid lines, respectively. Measured EDCs in the XAS \(L_{3}\) absorption maximum as black solid lines. The self-convolution of the pDOS (CST) are given as dashed lines with corresponding color. For Ni and Fe, the LDA+DMFT pDOS for \(U = 4 \mathrm{eV}\) and \(2 \mathrm{eV}\) , respectively are given in green. (b) Experimental VB PES next to calculated spectra using the one-step model with LDA and LDA+DMFT. (c) and (d) Calculated BSF for the CrMnFeCoNi HEA in the LDA and LDA+DMFT approach. </center>  
+
+Fig. 2(b) presents VB PES measurements for \(\hbar \omega = 1200 \mathrm{eV}\) , alongside one- step model calculations \(^{39}\) for the LDA as well LDA+DMFT potentials (see SI for details). Experimentally, a satellite feature is visible, attributable to Ni at approximately \(7 \mathrm{eV}\) , as corroborated by ResPES. The LDA+DMFT calculation reveals a peak at approximately \(8 \mathrm{eV}\) , which perfectly overlaps with the Ni satellite in the calculated pDOS. The offset between theory and experiment is attributable to the perturbative nature of the DMFT solver, and aligns well with a \(1 \mathrm{eV}\) offset observed in pure Ni \(^{40}\) , which confirms our choice of \(U\) a posteriori. Despite the satellite, the shoulder spanning roughly from \(3 \mathrm{eV}\) to \(4 \mathrm{eV}\) , paralleling the experimental observations, is also reproduced in the LDA+DMFT calculation. The plateau ranging from \(5 \mathrm{eV}\) \(E_{\mathrm{B}}\) to \(8 \mathrm{eV}\) in the LDA+DMFT calculation resonates well with the experimental data, which extends from \(4 \mathrm{eV}\) \(E_{\mathrm{B}}\) to \(7 \mathrm{eV}\) . The LDA approach fails to adequately capture any of these correlation fingerprints. Despite the strong agreement between experimental and theoretical results, we still observe a discrepancy in the bandwidth, which is slightly narrower in LDA+DMFT. This difference can however have an extrinsic origin, as e.g. arise from subtle contributions from a minor surface oxide layer (oxygen 1s peak in wide scan XAS measurement provided in the SI). Also, the experimental background (intensity at lowest \(E_{\mathrm{B}}\) ), accounts for a systematic  
+
+deviation. Besides, we are able to conclude that the assumption of Hubbard \(U\) parameters for CrMnFeCoNi, aligned with the pure metals, is justified, and electronic correlations play a site- specific role comparable to that of pure 3d transition metals.  
+
+Broadening of the band structure. The influence of the broadening of states due to chemical disorder and quasiparticle lifetimes, can be discerned in the computed Bloch spectral function (BSF) depicted in Fig. 2(c) and 2(d) for LDA and LDA+DMFT, respectively. Near \(E_{\mathrm{F}}\) , both calculations reveal a scarcely dispersive \(d\) - band block with strongly localized electrons, while the parabolic dispersion of the \(sp\) - bands becomes apparent at higher \(E_{\mathrm{B}}\) . The \(d\) - bands around \(2 \mathrm{eV}\) are for the LDA+DMFT case so extensively smeared, especially along the symmetry line X- \(\Gamma\) - L, that subbands cannot be resolved. The strongly localized satellite states are perceptible over a constant smeared background up to \(9 \mathrm{eV}\) . Arguing qualitatively, the spectral width of the \(d\) - states implies that the lifetimes of these electrons must be exceedingly short. Interestingly, this arises mainly from correlation effects at high \(E_{\mathrm{B}}\) , but from the combined action of correlations and chemical/magnetic disorder in the vicinity of \(E_{\mathrm{F}}\) .  
+
+Element resolved quasiparticle lifetimes. Quasiparticle lifetimes \(\tau\) can be obtained from the self- energy function \(\Sigma\) from DMFT via \(\hbar /\tau = 2 \mathrm{Im}(\Sigma)\) . Fig. 3(a) displays element- specific lifetimes for Fe and Ni, obtained by evaluating the Greens function on the real energy axis. Data for other elements are plotted in the SI. For context, black lines show pure element calculations in their natural crystal structures, alongside corresponding experimental data from literature \(^{40 - 43}\) . These calculations align well, although the Ni lifetimes for excited states above \(1 \mathrm{eV}\) slightly exceed experimental observations. For all elements, near \(E_{\mathrm{F}}\) , a Fermi liquid theory- like behavior emerges, with \(\tau \propto (E - E_{\mathrm{F}})^{- 2.43}\) . To assess the effects of altered lattice constants on the element- specific self- energy, lifetimes for Fe and Ni within the fcc lattice but the HEA's lattice constant are illustrated in yellow in Fig. 3(a). While Ni shows a negligible variation, \(\gamma\) - Fe exhibits notably reduced lifetimes, despite having the same Hubbard \(U\) . This is not surprising, considering that \(\gamma\) - Fe is known to have stronger magnetic fluctuations leading to a complex magnetic landscape \(^{44}\) . The bottom row of Fig. 3(a) contrasts the pure elements in the fcc HEA structure against the CPA- derived disordered paramagnetic state. Here, both elements show an increased \(\tau\) , akin to those in their natural crystals. For the LDA+DMFT calculation of CrMnFeCoNi with \(U = 4 \mathrm{eV}\) for Ni and \(2 \mathrm{eV}\) for Fe, the green lines in Fig. 3(a) depict the results. For Ni, lifetimes above \(E_{\mathrm{F}}\) demonstrate a \(3 / 4\) reduction compared to the \(3 \mathrm{eV}\) calculation, reflecting the ratio of \(U\) values. A comparable trend is given for Fe. As observed in the pDOS data, other elements are not affected by the variation of \(U\) (see SI). Our calculations indicate a minor influence of chemical disorder on DMFT derived lifetimes, particularly when contrasted with data for the pure elements and taking the effect of changing crystal structure into account. Furthermore, \(\tau\) exhibit a nearly linear dependency to variations in \(U\) .
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Fig. 3 Element specific quasiparticle lifetimes and frequency resolved optical conductivity of the \(\mathrm{CrMnFeCoNi}\) HEA. (a) Element specific lifetimes for Fe and Ni by black lines for calculated pure elements in natural crystal and yellow line pure metals within the fcc structure and \(\mathrm{CrMnFeCoNi}\) lattice constant. Experimental values for are taken from PES (below \(\mathrm{E}_{\mathrm{F}})^{40}\) and TR-2PPE (above \(\mathrm{E}_{\mathrm{F}})^{41 - 43}\) data of pure elements. Blue lines in the bottom correspond to LDA+DMFT and green dashed line to the LDA+DMFT case with changed \(U\) for Fe \(= 2\mathrm{eV}\) and \(\mathrm{Ni} = 4\mathrm{eV}\) . (b) Real (left) and imaginary part (right) of complex optical conductivity versus photon energy \(\hbar \omega\) . LDA calculations are given by the yellow line and LDA+DMFT calculations by the blue line. Experiments are from reflectometry measurements (Exp. 1) and ellipsometry (Exp. 2). </center>  
+
+Frequency resolved optical conductivity. The dynamic response of electrons in the \(\mathrm{CrMnFeCoNi}\) HEA is further probed by means of complex optical conductivity \(\sigma (\omega)\) measurements. Typically, \(\sigma (\omega)\) reveals for metals a Drude peak in \(\mathrm{Re}(\sigma (\omega))\) , which broadens with increasing scattering rate and eventually may merge with higher- energy interband transitions. Thus separation into inter- and intra- band contributions becomes challenging for correlated 3d transition metals. We therefore compute the \(\sigma (\omega)\) tensor via the Kubo formalism \(^{45}\) by the current- current correlation function \(^{46}\) .  
+
+\[\sigma_{\mu \nu}(\omega) = \frac{i\hbar}{\pi^2}\frac{1}{\Omega}\int_{\Omega}d^3 r\int_{\Omega}d^3 r^{\prime}\int_{E_B}^{\infty}dE^{\prime}\int_{E_B}^{\infty}dE\Theta_T(E - E_F)\Theta_T(E_F\] \[-E^{\prime})\] \[\times \left\{\frac{\mathrm{tr}[j_\mu (\mathbf{r})\mathrm{Im}G^+ (E^{\prime})j_\nu (\mathbf{r}^{\prime})\mathrm{Im}G^+ (E)]}{(E^{\prime} - E - i\Gamma_{\mathrm{ep}})(\hbar\omega + E - E^{\prime} + i\Gamma_{\mathrm{ep}})\] \[+\frac{\mathrm{tr}[j_\nu (\mathbf{r}^{\prime})\mathrm{Im}G^+ (E^{\prime})j_\mu (\mathbf{r})\mathrm{Im}G^+ (E)]}{(E^{\prime} - E - i\Gamma_{\mathrm{ep}})(\hbar\omega + E^{\prime} - E + i\Gamma_{\mathrm{ep}})}\right\}\]  
+
+Due to the fully relativistic formulation of the current density operator \(j(\mathbf{r})\) , both paramagnetic and diamagnetic terms are implicitly included, with the latter yielding a Drude- like contribution \(^{46,47}\) . However, as in our approach electron- phonon collision driven damping mechanisms are not considered a priori, a phenomenological complex photon energy \(\Gamma_{\mathrm{ep}}\) is introduced overruling the infinite small lifetime \(0^{+}\) from adiabatic switching on of the external perturbative field. We adopt a constant \(\Gamma_{\mathrm{ep}}\) value of \(0.340\mathrm{eV}\) , attributed to the substantially diminished electronic mean free path, proximate to the lattice parameter, as well the given Fermi velocity for such disordered alloys \(^{48}\) . Our calculations utilize both LDA and LDA+DMFT, with the latter including energy dependent electron- electron scattering \(\Gamma_{\mathrm{ee}}\) through the imaginary part of the self- energy inherently incorporated in the Greens function. Although the DMFT scheme solely considers \(d\) - band states in \(\Sigma^{49}\) , an empirical choice (e.g. F of \(\Gamma_{\mathrm{ee}}\) for the \(sp\) - band is not obligatory. It has been shown analytically \(^{50}\) as well numerically within the \(GW\) formalism \(^{51}\) that \(d\) - band screening dominates the total self- energy, and thus scattering rates, when considering open \(d\) - band metals.   
+
+Calculation results are depicted in Fig. 3(b) (LDA in yellow, LDA+DMFT in blue) alongside experimental data (see SI). The left subfigure displays the real part of the optical conductivity, \(\mathrm{Re}(\sigma (\omega))\) , corresponding to the absorptive component. The results from LDA and LDA+DMFT show for low \(\hbar \omega\) a constant trend with a minor offset between each other, and dispersive variations in the VIS range where both curves decline. Quasiparticle lifetimes from the DMFT scheme improve the absorptive part of \(\sigma (\omega)\) drastically, especially in the visible and UV range, where a perfect agreement with experiment is found. Here the pure LDA calculations overestimate the absorption. Also for the LDA+DMFT case, at the highest photon energies of \(5\mathrm{eV}\) and above, the fine structure found in the LDA calculation is blurred (see small oscillation between \(6\mathrm{eV}\) and \(8\mathrm{eV}\) for real and imaginary part). This behavior reflects the smearing of sub- bands, induced through the quasi- particle lifetimes, as seen within the BSF in Fig. 3(c) and 3(d). For \(\mathrm{Im}(\sigma (\omega))\) , corresponding to the dispersive or reflective part, the situation is similar. The experiments show low values for small \(\hbar \omega\) (approaching zero), comparable to the Drude- behavior, with a resonance peak in the VIS region. The LDA calculation overestimates this peak threefold and is non- symmetric on a log scale, thus not purely Drude- like. The resonance position in \(\mathrm{Im}(\sigma (\omega))\) is found at \(\hbar \omega = 2.3\mathrm{eV}\) experimentally, and \(2.7\mathrm{eV}\) in the LDA and at \(2.4\mathrm{eV}\) in the LDA+DMFT calculations. Besides improving spectral position, LDA+DMFT calculations reduce the resonance maximum significantly, achieving an improved experimental alignment.  
+
+Temperature dependence of electrical resistivity. Revisiting temperature- dependent electrical resistivity, we focus on correlation effects. Fig. 4(a) depicts our four- point resistivity measurements for different metals, from Ni to \(\mathrm{CrMnFeCoNi}\) HEA, over a broad temperature range (4 K to \(800\mathrm{K}\) ). A dominant increase in residual resistivity is observed when transitioning from FeNi to \(\mathrm{CrFeNi}\) , which is attributed to the coupling of ferromagnetic and antiferromagnetic elements within the alloys \(^{12}\) . This results in smearing across both spin channels in the BSF of \(\mathrm{CrFeNi}\) , whereas FeNi exhibits solely smearing in the minority spin channel \(^{12}\) . Thus, Ni and NiFe show well- defined quasiparticle transport properties within the majority channel with progressive increase in upward curvature up to Curie temperature at \(625\mathrm{K}\) and to \(835\mathrm{K}\) , respectively. For the alloys exhibiting higher residual resistivity, similar trends are
+
+<--- Page Split --->
+
+observed: \(\rho_0\) ranges around \(100\mu \Omega \mathrm{cm}\) , and \(\mathrm{d}\rho /\mathrm{d}T\) remains low (and after subtraction of \(\rho_0\) , \(\mathrm{d}\rho /\mathrm{d}T\) seems to be identical).  
+
+![](images/Figure_4.jpg)
+
+<center>Fig. 4 Measurement and calculation of temperature electrical resistivity. (a) Measurements of electrical resistivity versus temperature for several Cantor-Wu alloys, ranging from FeNi to the CrMnFeCoNi HEA. (b) Comparison of experiment and linear response calculation for LDA and LDA+DMFT case for CrMnFeCoNi. (c) and (d) Fermi surface of CrMnFeCoNi for the LDA and LDA+DMFT potential. </center>  
+
+We compare the CrMnFeCoNi measurement with linear response calculation within the Kubo- Greenwood formalism \(^{52}\) . The alloy analogy model \(^{52}\) is employed to mimic temperature- dependent lattice vibrations, inherently integrating electron- phonon scattering processes \(^{53}\) . Corroborating our BSF observations, both LDA and LDA+DMFT results yield comparable \(\rho_0\) of \(67\mu \Omega \mathrm{cm}\) . By Fermi surface analysis the resistivity may be calculated from the \(k\) - space smearing (electron mean free path), as well the surface area \(^{8}\) . Figs. 4(c) and 4(d) show the Fermi surfaces calculated with the LDA and LDA+DMFT potentials, which exhibit remarking similarities. This supports our linear response results. However, the substantial deviation from the experimental value of \(105\mu \Omega \mathrm{cm}\) could be influenced by potential short- range ordering \(^{54 - 56}\) , which has been experimentally shown to increase resistivity \(^{57}\) . Such mechanisms are not included within the CPA framework and need to be captured with more sophisticated methodologies like non- local CPA \(^{58,59}\) . The localization mechanism must also be considered. While Hubbard localization is addressed within the LDA+DMFT framework, the onset of Anderson localization is inherently not captured by the CPA approach \(^{60,61}\) . The temperature dependence of \(\rho (T)\) is presented in a normalized format \(\rho (T) / \rho_0\) in Fig. 4(b) and demonstrates that while LDA closely aligns with experimental data, even indicating a reduced \(\mathrm{d}\rho /\mathrm{d}T\) , LDA+DMFT yields perfect agreement across all temperatures. This aligns with the BSFs, where LDA+DMFT exhibits increased smearing at higher \(E_{\mathrm{B}}\) , thus impacting electronic transport at elevated temperatures.  
+
+Conclusion. Our findings identify CrMnFeCoNi as a material characterized by pronounced electronic correlations. These correlations exist in addition to the prevailing chemical and magnetic disorder, which smear the band structure in the vicinity of \(E_{\mathrm{F}}\) and thus primarily cause the high residual resistivity. The extent of many body effects by means of on- site Coulomb interaction within the Hubbard model mirrors those of the containing pure elements. Correlation effects gain in significance with increasing distance from the Fermi edge, which was demonstrated both experimentally and theoretically utilizing electronic spectroscopies and temperature dependent electronic transport. Especially in the calculation of the optical response, accounting for quasiparticle lifetimes dramatically improve the absorptive as well dispersive part of the optical conductivity in the VIS- UV range, as energy dependent electron- electron scattering may overrule electron- phonon scattering rates.   
+
+## Methods  
+
+Sample preparation and characterization. All alloys were first synthesized by mixing and pressing powders of elemental metals (total mass \(1.5\mathrm{g}\) ) into pellets to achieve the targeted composition. The pellets were placed in an arc- melting chamber, with glowing elemental zirconium used to remove any residual oxygen. Arc melting was performed three times, and the alloys were subsequently annealed in an evacuated quartz ampoule for one month at \(1030^{\circ}\mathrm{C}\) to enhance sample homogeneity. The specimens were cut into approximately \(0.5\mathrm{mm}\) thick discs for photoemission and optical conductivity measurements and into bars of dimensions \(8 \times 3 \times 0.5\mathrm{mm}^3\) for four- point electrical resistivity measurement. The series of prepared samples included Ni, NiFe, NiFeCr, NiFeCrCo, and NiFeCrCoMn. Characterization of the HEA samples using X- ray diffraction confirmed a well- defined fcc crystalline structure. Scanning electron microscopy and electron- dispersive X- ray spectroscopy analyses revealed the actual composition and satisfactory dispersion of constituent elements.  
+
+Electronic spectroscopies measurements. XAS and ResPES measurements were conducted at the ADRESS beamline of the Paul Scherrer Institut \(^{62}\) on the CrMnFeCoNi alloy at a base temperature of \(20\mathrm{K}\) . The photon energy was scanned along each of the relevant \(L_3\) absorption edges at steps of \(100\mathrm{meV}\) , while at the same time recording a valence band spectrum with a hemispherical electron analyzer at a resolution better than \(100\mathrm{meV}\) .  
+
+Optical conductivity measurements. For the optical conductivity measurements, the CrMnFeCoNi sample was ground with a SiC paper and subsequently polished with diamond paste with decreasing grain sizes to achieve a root mean square surface roughness of \(2\mathrm{nm}\) . The spectra were obtained by ellipsometry (Sentech SE 850) in the visible to infrared range and by reflectance measurements with subsequent Kramers- Kronig transformation in the far infrared part of the electromagnetic spectrum.  
+
+LDA+DMFT calculations. DFT calculations of CrMnFeCoNi were performed within the fully relativistic spin polarized multiple scattering Korringa- Kohn- Rostoker (SPR- KKR) Greens function formalism \(^{63,64}\) . Many- body correlation effects beyond local density approximation (LDA) were added via DMFT, as implemented in SPR- KKR \(^{49}\) . Chemical disorder was accounted for by the coherent potential approximation (CPA) \(^{65,66}\) and the paramagnetic state at \(20\mathrm{K}^{67}\) was mimicked by the disordered local moment scheme \(^{68}\) . An appealing feature of this approach is that it allows to consider local quantum and disorder fluctuations on the same footing. This approach has already been successfully realized by other groups \(^{69}\) . For LDA+DMFT calculations, the Hubbard \(U\) for \(3d\) electrons was set to the values found for pure elements, while the Hund exchange was \(J = 0.94\mathrm{eV}\) for all elements. The \(U\) values for Cr, Mn, Fe, Co, Ni were equal to \(2.0\mathrm{eV}^{70}\) , \(3.0\mathrm{eV}^{38}\) , \(1.5\mathrm{eV}\) , \(2.5\mathrm{eV}\) and \(3.0\mathrm{eV}^{40}\) , respectively. For further details, see SI.  
+
+Application of the Cini- Sawatzky theory. We applied a self- convolution to the calculated partial density of states and compared the results within the CST \(^{29 - 31}\) with the experimental data obtained
+
+<--- Page Split --->
+
+fromResPES. This approach aims to replicate the two- hole interactions using single- particle ground states, necessitating the presentation of the resultant self- convolution on a two- particle energy scale34. Consequently, the new energy axis is multiplied by a factor of two. Prior to convolution, the pDOS data were interpolated on a refined energy grid.  
+
+Calculation of spectroscopic and transport properties. In our study, valence band photoemission39 and electronic transport calculations, including the optical conductivity tensor, were conducted using the SPR- KKR code and a Kubo- framework for linear response, respectively. These calculations incorporated electron- phonon interactions within the alloy analogy model52, considering temperature- dependent atomic displacements and a comprehensive k- point mesh for accuracy. The optical response to external electromagnetic fields was detailed through a current- current correlation function reformulated in Green's function terms, capturing the Drude contribution within a fully relativistic framework46. Our findings primarily focus on the isotropic \(\sigma (\omega)\) in paramagnetic CrMnFeCoNi HEA. For an in- depth explanation of methodologies, including the computational parameters and models employed, the reader is referred to the SI.  
+
+## Data availability  
+
+The data supporting the results of this study are available on reasonable request from the corresponding author.  
+
+## References  
+
+1. George, E. P., Raabe, D. & Ritchie, R. O. High-entropy alloys. Nat. Rev. Mater. 4, 515-534 (2019).  
+2. Miracle, D. B. & Senkov, O. N. A critical review of high entropy alloys and related concepts. Acta Mater. 122, 448-511 (2017).  
+3. Gludovatz, B. et al. A fracture-resistant high-entropy alloy for cryogenic applications. Science 345, 1153-1158 (2014).  
+4. Li, Z., Zhao, S., Ritchie, R. O. & Meyers, M. A. Mechanical properties of high-entropy alloys with emphasis on face-centered cubic alloys. Prog. Mater. Sci. 102, 296-345 (2019).  
+5. Körmann, F., Ikeda, Y., Grabowski, B. & Sluiter, M. H. F. Phonon broadening in high entropy alloys. Npj Comput. Mater. 3, 36 (2017).  
+6. Mu, S. et al. Uncovering electron scattering mechanisms in NiFeCoCrMn derived concentrated solid solution and high entropy alloys. Npj Comput. Mater. 5, 1 (2019).  
+7. Fan, Z., Wang, H., Wu, Y., Liu, X. J. & Lu, Z. P. Thermoelectric high-entropy alloys with low lattice thermal conductivity. RSC Adv. 6, 52164-52170 (2016).  
+8. Robarts, H. C. et al. Extreme Fermi Surface Smearing in a Maximally Disordered Concentrated Solid Solution. Phys. Rev. Lett. 124, 046402 (2020).  
+9. Anderson, P. W. Absence of Diffusion in Certain Random Lattices. Phys. Rev. 109, 1492-1505 (1958).  
+10. Belitz, D. & Schirmacher, W. Theory of phonon-controlled conductivity in high-resistivity conductors. J. Phys. C Solid State Phys. 16, 913-926 (1983).  
+11. Jin, K. et al. Tailoring the physical properties of Ni-based single-phase equiatomic alloys by modifying the chemical complexity. Sci. Rep. 6, 20159 (2016).  
+12. Zhang, Y. et al. Influence of chemical disorder on energy dissipation and defect evolution in advanced alloys. J. Mater. Res. 31, 2363-2375 (2016).  
+13. Kourov, N. I. et al. Peculiar features of physical properties of the rapid quenched AlCrFeCoNiCu high-entropy alloy. J. Alloys Compd. 636, 304-309 (2015).  
+14. Ciuchi, S., Di Sante, D., Dobrosavljevic, V. & Fratini, S. The origin of Mooij correlations in disordered metals. Npj Quantum Mater. 3, 44 (2018).  
+15. Fano, U. Effects of Configuration Interaction on Intensities and Phase Shifts. Phys. Rev. 124, 1866-1878 (1961).  
+
+16. Sakisaka, Y. et al. Photoemission study of the valence-band satellite of Ni(110). Phys. Rev. B 36, 6383-6389 (1987).  
+17. Formoso, V. et al. Resonant valence-band photoemission spectroscopy on the Fe62Ni20Cr18 alloy. Eur. Phys. J. B 43, 463-470 (2005).  
+18. Kay, A. et al. Multi-Atom Resonant Photoemission: A Method for Determining Near-Neighbor Atomic Identities and Bonding. Science 281, 679-683 (1998).  
+19. Weinelt, M. et al. Resonant Photoemission at the 2 p Edges of Ni: Resonant Raman and Interference Effects. Phys. Rev. Lett. 78, 967-970 (1997).  
+20. Hufner, S. et al. Observation of the two-hole satellite in Cr and Fe metal by resonant photoemission at the 2 p absorption energy. Phys. Rev. B 61, 12582-12585 (2000).  
+21. Yablonskikh, M. V., Shkvarin, A. S., Yarmoshenko, Y. M., Skorikov, N. A. & Titov, A. N. Resonant photoemission at the L3 absorption edge of Mn and Ti and the electronic structure of 1T-Mn 0.2 TiSe 2. J. Phys. Condens. Matter 24, 045504 (2012).  
+22. Hsu, L.-S., Tsang, K.-L. & Chung, S.-C. Resonant photoemission and photoabsorption at the Ni L2,3 edges of NiGa and Ni3Ga. J. Magn. Magn. Mater. 177-181, 1031-1032 (1998).  
+23. Guillot, C. et al. Resonant Photoemission in Nickel Metal. Phys. Rev. Lett. 39, 1632-1635 (1977).  
+24. Clauberg, R., Gudat, W., Kisker, E., Kuhlmann, E. & Rothberg, G. M. Nature of the Resonant 6-eV Satellite in Ni: Photoelectron Spin-Polarization Analysis. Phys. Rev. Lett. 47, 1314-1317 (1981).  
+25. Bennett, P. A., Fuggle, J. C., Hillebrecht, F. U., Lenselink, A. & Sawatzky, G. A. Electronic structure of Ni and Pd alloys. III. Correlation effects in the Auger spectra of Ni alloys. Phys. Rev. B 27, 2194-2209 (1983).  
+26. Schneider, W.-D., Reihl, B., Martensson, N. & Arko, A. J. Resonant photoemission study of U Ni 5 - x Cu x. Phys. Rev. B 26, 423-426 (1982).  
+27. Fuggle, J. C. et al. Electronic structure of Ni and Pd alloys. I. X-ray photoelectron spectroscopy of the valence bands. Phys. Rev. B 27, 2145-2178 (1983).  
+28. Hillebrecht, F. U., Fuggle, J. C., Bennett, P. A., Zohnierek, Z. & Freiburg, Ch. Electronic structure of Ni and Pd alloys. II. X-ray photoelectron core-level spectra. Phys. Rev. B 27, 2179-2193 (1983).  
+29. Cini, M. Density of states of two interacting holes in a solid. Solid State Commun. 20, 605-607 (1976).  
+30. Sawatzky, G. A. Quasiatomic Auger Spectra in Narrow-Band Metals. Phys. Rev. Lett. 39, 504-507 (1977).  
+31. Cini, M. Comment on quasiatomic Auger spectra in narrow-band metals. Phys. Rev. B 17, 2788-2789 (1978).  
+32. Lander, J. J. Auger Peaks in the Energy Spectra of Secondary Electrons from Various Materials. Phys. Rev. 91, 1382-1387 (1953).  
+33. Treglia, G., Desjonqueres, M. C., Ducastelle, F. & Spanjaard, D. Correlation effects on Auger spectra in unfilled d band metals. J. Phys. C Solid State Phys. 14, 4347-4355 (1981).  
+34. Chainani, A. et al. Evidence for Weakly Correlated Oxygen Holes in the Highest- T c Cuprate Superconductor HgBa 2 Ca 2 Cu 3 O 8 + δ. Phys. Rev. Lett. 119, 057001 (2017).  
+35. Vilmercati, P. et al. Direct probe of the variability of Coulomb correlation in iron pnictide superconductors. Phys. Rev. B 85, 235133 (2012).  
+36. Cini, M. & Verdozzi, C. Many-body effects in the electron spectroscopies of incompletely filled bands. Il Nuovo Cimento D 9, 1-21 (1987).  
+37. Rausch, R. & Potthoff, M. Multiplons in the two-hole excitation spectra of the one-dimensional Hubbard model. New J. Phys. 18, 023033 (2016).  
+38. Di Marco, I. et al. γ-Mn at the border between weak and strong correlations. Eur. Phys. J. B 72, 473-478 (2009).  
+39. Braun, J., Minár, J., Ebert, H., Katsnelson, M. I. & Lichtenstein, A. I. Spectral Function of Ferromagnetic 3 d Metals: A Self-Consistent LSDA + DMFT Approach Combined with the One-Step Model of Photoemission. Phys. Rev. Lett. 97, 227601 (2006).
+
+<--- Page Split --->
+
+40. Sánchez-Barriga, J. et al. Effects of spin-dependent quasiparticle renormalization in Fe, Co, and Ni photoemission spectra:An experimental and theoretical study. Phys. Rev. B 85, 205109 (2012). 
+41. Knoesel, E., Hotzel, A. & Wolf, M. Temperature dependence of surface state lifetimes, dephasing rates and binding energies on Cu(111) studied with time-resolved photoemission. J. Electron Spectrosc. Relat. Phenom. 88-91, 577-584 (1998). 
+42. Hong, J. & Mills, D. L. Theory of the spin dependence of the inelastic mean free path of electrons in ferromagnetic metals: A model study. Phys. Rev. B 59, 13840-13848 (1999). 
+43. Bauer, M., Marienfeld, A. & Aeschlimann, M. Hot electron lifetimes in metals probed by time-resolved two-photon photoemission. Prog. Surf. Sci. 90, 319-376 (2015). 
+44. Igoshev, P. A., Efremov, A. V., Poteryaev, A. I., Katanin, A. A. & Anisimov, V. I. Magnetic fluctuations and effective magnetic moments in \(\gamma\) -iron due to electronic structure peculiarities. Phys. Rev. B 88, 155120 (2013). 
+45. Kubo, R. Statistical-Mechanical Theory of Irreversible Processes. I. General Theory and Simple Applications to Magnetic and Conduction Problems. J. Phys. Soc. Jpn. 12, 570-586 (1957). 
+46. Huhne, T. & Ebert, H. Fully relativistic description of the magneto-optical properties of arbitrary layered systems. Phys. Rev. B 60, 12982-12989 (1999). 
+47. Banhart, J. Optical Conductivity of Disordered Alloys Calculated from First Principles. Phys. Rev. Lett. 82, 2139-2142 (1999). 
+48. Samolyuk, G. D. et al. Temperature dependent electronic transport in concentrated solid solutions of the 3 d-transition metals Ni, Fe, Co and Cr from first principles. Phys. Rev. B 98, 165141 (2018). 
+49. Minár, J. Correlation effects in transition metals and their alloys studied using the fully self-consistent KKR-based LSDA + DMFT scheme. J. Phys. Condens. Matter 23, 253201 (2011). 
+50. Zarate, E., Apell, P. & Echenique, P. M. Calculation of low-energy-electron lifetimes. Phys. Rev. B 60, 2326-2332 (1999). 
+51. Nechaev, I. A., Chulkov, E. V. & Echenique, P. M. G W lifetimes of quasiparticle excitations in paramagnetic transition metals. Phys. Rev. B 76, 245125 (2007). 
+52. Ebert, H. et al. Calculating linear-response functions for finite temperatures on the basis of the alloy analogy model. Phys. Rev. B 91, 165132 (2015). 
+53. Xu, J. et al. Thermal Conductivity and Electrical Resistivity of Solid Iron at Earth's Core Conditions from First Principles. Phys. Rev. Lett. 121, 096601 (2018). 
+54. Mizuno, M., Sugita, K. & Araki, H. Prediction of short-range order in CrMnFeCoNi high-entropy alloy. Results Phys. 34, 105285 (2022). 
+55. Singh, P., Smirnov, A. V. & Johnson, D. D. Atomic short-range order and incipient long-range order in high-entropy alloys. Phys. Rev. B 91, 224204 (2015). 
+56. Zhang, R. et al. Short-range order and its impact on the CrCoNi medium-entropy alloy. Nature 581, 283-287 (2020). 
+57. Tanimoto, H., Hozumi, R. & Kawamura, M. Electrical resistivity and short-range order in rapid-quenched CrMnFeCoNi high-entropy alloy. J. Alloys Compd. 896, 163059 (2022). 
+58. Woodgate, C. D., Hedlund, D., Lewis, L. H. & Staunton, J. B. Interplay between magnetism and short-range order in medium- and high-entropy alloys: CrCoNi, CrFeCoNi, and CrMnFeCoNi. Phys. Rev. Mater. 7, 053801 (2023). 
+59. Lowitzer, S., Ködderitzsch, D., Ebert, H. & Staunton, J. B. Electronic transport in ferromagnetic alloys and the Slater-Pauling curve. Phys. Rev. B 79, 115109 (2009). 
+60. Ekuma, C. E. et al. Typical medium dynamical cluster approximation for the study of Anderson localization in three dimensions. Phys. Rev. B 89, 081107 (2014). 
+61. Terletska, H., Zhang, Y., Chioncel, L., Vollhardt, D. & Jarrell, M. Typical-medium multiple-scattering theory for disordered systems with Anderson localization. Phys. Rev. B 95, 134204 (2017).  
+
+62. Strocov, V. N. et al. Soft-X-ray ARPES facility at the ADRESS beamline of the SLS: concepts, technical realisation and scientific applications. J. Synchrotron Radiat. 21, 32-44 (2014). 
+63. Ebert, H., Ködderitzsch, D. & Minár, J. Calculating condensed matter properties using the KKR-Green's function method—recent developments and applications. Rep. Prog. Phys. 74, 096501 (2011). 
+64. Ebert, H. The Munich SPR-KKR package. version 8.6 http://olymp.cup.uni-muenchen.de/ak/ebert/SPRKKR (2022). 
+65. Faulkner, J. S. & Stocks, G. M. Calculating properties with the coherent-potential approximation. Phys. Rev. B 21, 3222-3244 (1980). 
+66. Ebert, H., Vernes, A. & Banhart, J. Relativistic bandstructure of disordered magnetic alloys. Solid State Commun. 104, 243-247 (1997). 
+67. Wang, S., Zhang, T., Hou, H. & Zhao, Y. The Magnetic, Electronic, and Thermodynamic Properties of High Entropy Alloy CrMnFeCoNi: A First-Principles Study. Phys. Status Solidi B 255, 1800306 (2018). 
+68. Gyorffy, B. L., Pindor, A. J., Staunton, J., Stocks, G. M. & Winter, H. A first-principles theory of ferromagnetic phase transitions in metals. J. Phys. F Met. Phys. 15, 1337-1386 (1985). 
+69. Östlin, A., Vitos, L. & Chioncel, L. Correlated electronic structure with uncorrelated disorder. Phys. Rev. B 98, 235135 (2018). 
+70. Belozerov, A. S., Katanin, A. A. & Anisimov, V. I. Itinerant magnetism of chromium under pressure: a DFT+DMFT study. J. Phys. Condens. Matter 33, 385601 (2021).  
+
+## Acknowledgments  
+
+This work was supported by Deutsche Forschungsgemeinschaft under Grant 528706678 and the project MEBIOSYS, funded as project No. CZ.02.01.01/00/22_008/0004634 by Programme Johannes Amos Comnenius, call Excellent Research. This research is also part of the project No. 2022/45/P/ST3/04247 co- funded by the National Science Centre and the European Union's Horizon 2020 research and innovation programme under the Marie Skodowska- Curie grant agreement no 945339. For the purpose of Open Access, the author has applied a CC- BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission.  
+
+## Author contributions  
+
+J.M., L.F. and H.D. developed the concept. E.M., X.M., L.C., D.T. and T.I. prepared the samples and performed transport measurements. M.C., V.S. and H.D. were responsible for the photoemission experiments at the Swiss Light Source. D.R. and J.M. conducted the CST analysis. D.R. and S.K. carried out the DFT calculations. I.D.M. interpreted the DMFT results and H.H. discussed the optics. H.E., J.M. and A.H. developed the KKR package. D.R. wrote the manuscript in close collaboration with J.M., L.F. and H.D. All authors contributed to discussions and commented on the manuscript.  
+
+## Additional information  
+
+Supplementary Information accompanies this paper.  
+
+Competing interests: The authors declare no competing interests.
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+20240222Supplement.pdf
+
+<--- Page Split --->

preprint/preprint__98e188d197f798d498e66156851783607bd993dcbb8ac0d0cb28a16b4678777a/preprint__98e188d197f798d498e66156851783607bd993dcbb8ac0d0cb28a16b4678777a_det.mmd

@@ -0,0 +1,317 @@
+<|ref|>title<|/ref|><|det|>[[44, 108, 852, 175]]<|/det|>
+# Interplay between disorder and electronic correlations in compositionally complex alloys  
+
+<|ref|>text<|/ref|><|det|>[[44, 196, 245, 240]]<|/det|>
+Jan Minar jminar@ntc.zcu.cz  
+
+<|ref|>text<|/ref|><|det|>[[50, 268, 750, 312]]<|/det|>
+University of West Bohemia https://orcid.org/0000- 0001- 9735- 8479  
+
+<|ref|>text<|/ref|><|det|>[[50, 315, 750, 334]]<|/det|>
+Munich University of Applied Sciences https://orcid.org/0000- 0002- 7306- 2232  
+
+<|ref|>text<|/ref|><|det|>[[44, 340, 300, 382]]<|/det|>
+Saleem Khan University of West Bohemia  
+
+<|ref|>text<|/ref|><|det|>[[44, 387, 430, 429]]<|/det|>
+Edoardo Martino École Polytechnique Fédérale de Lausanne  
+
+<|ref|>text<|/ref|><|det|>[[44, 434, 430, 476]]<|/det|>
+Xavier Mettan École Polytechnique Fédérale de Lausanne  
+
+<|ref|>text<|/ref|><|det|>[[44, 480, 188, 520]]<|/det|>
+Luka Ciric Inst of Physics  
+
+<|ref|>text<|/ref|><|det|>[[44, 526, 137, 565]]<|/det|>
+Davor Tolj EPFL  
+
+<|ref|>text<|/ref|><|det|>[[44, 572, 430, 612]]<|/det|>
+Trpimir Ivšić École Polytechnique Fédérale de Lausanne  
+
+<|ref|>text<|/ref|><|det|>[[44, 618, 392, 659]]<|/det|>
+Andreas Held Ludwig- Maximilians- University Munich  
+
+<|ref|>text<|/ref|><|det|>[[44, 665, 252, 705]]<|/det|>
+Marco Caputo Paul Scherrer Institute  
+
+<|ref|>text<|/ref|><|det|>[[44, 711, 600, 752]]<|/det|>
+Vladimir Strocov Paul Scherrer Institut https://orcid.org/0000- 0002- 1147- 8486  
+
+<|ref|>text<|/ref|><|det|>[[44, 757, 220, 798]]<|/det|>
+Igor Marco Uppsala University  
+
+<|ref|>text<|/ref|><|det|>[[44, 804, 416, 844]]<|/det|>
+Hubert Ebert Ludwig- Maximilians- Universität München  
+
+<|ref|>text<|/ref|><|det|>[[44, 850, 750, 891]]<|/det|>
+Heinz Huber Munich University of Applied Sciences https://orcid.org/0000- 0003- 2444- 9833  
+
+<|ref|>text<|/ref|><|det|>[[44, 896, 790, 937]]<|/det|>
+Hugo Dil École Polytechnique Fédérale de Lausanne https://orcid.org/0000- 0002- 6016- 6120  
+
+<|ref|>text<|/ref|><|det|>[[44, 941, 155, 959]]<|/det|>
+László Forró
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[44, 106, 103, 124]]<|/det|>
+# Article  
+
+<|ref|>title<|/ref|><|det|>[[44, 144, 135, 163]]<|/det|>
+# Keywords:  
+
+<|ref|>text<|/ref|><|det|>[[44, 181, 336, 201]]<|/det|>
+Posted Date: February 29th, 2024  
+
+<|ref|>text<|/ref|><|det|>[[44, 220, 474, 240]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 3981895/v1  
+
+<|ref|>text<|/ref|><|det|>[[42, 257, 916, 301]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+<|ref|>text<|/ref|><|det|>[[42, 317, 535, 338]]<|/det|>
+Additional Declarations: There is NO Competing Interest.  
+
+<|ref|>text<|/ref|><|det|>[[42, 372, 920, 416]]<|/det|>
+Version of Record: A version of this preprint was published at Nature Communications on September 12th, 2024. See the published version at https://doi.org/10.1038/s41467- 024- 52349- 8.
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[72, 78, 924, 97]]<|/det|>
+# Interplay between disorder and electronic correlations in compositionally complex alloys  
+
+<|ref|>text<|/ref|><|det|>[[80, 108, 920, 272]]<|/det|>
+David Redka \(^{1,2}\) , Saleem Ayaz Khan \(^{1}\) , Edoardo Martino \(^{3}\) , Xavier Mettan \(^{3}\) , Luka Ciric \(^{3}\) , Davor Tolj \(^{3}\) , Trpimir Ivšić \(^{3}\) , Andreas Held \(^{4}\) , Marco Caputo \(^{5}\) , Vladimir N. Strocov \(^{5}\) , Igor Di Marco \(^{6,7}\) , Hubert Ebert \(^{4}\) , Heinz P. Huber \(^{1,2,*}\) , J. Hugo Dil \(^{3,5}\) , László Forró \(^{3,8}\) & Ján Minár \(^{1,*}\) \(^{1}\) New Technologies Research Center, University of West Bohemia, Plzen CZ- 30100, Czech Republic \(^{2}\) Department of Applied Sciences and Mechatronics, Munich University of Applied Sciences HM, Munich DE- 80335, Germany \(^{3}\) Institute of Physics, École Polytechnique Fédérale de Lausanne, Lausanne CH- 1015, Switzerland \(^{4}\) Department of Chemistry, Ludwig- Maximilians- University Munich, Munich DE- 81377, Germany \(^{5}\) Photon Science Division, Paul Scherrer Institut, Villigen CH- 5232, Switzerland \(^{6}\) Institute of Physics, Nicolaus Copernicus University, Toruń PL- 87- 100, Poland \(^{7}\) Department of Physics and Astronomy, Uppsala University, Uppsala SE- 75120, Sweden \(^{8}\) Stavropoulos Center for Complex Quantum Matter, Department of Physics and Astronomy, University of Notre Dame, Notre Dame IN 46556, USA  
+
+<|ref|>text<|/ref|><|det|>[[291, 260, 707, 273]]<|/det|>
+Corresponding authors: "heinz.huber@hm.edu; \\* jminar@ntc.zcu.cz;  
+
+<|ref|>text<|/ref|><|det|>[[117, 290, 881, 378]]<|/det|>
+Exploring the intricate interplay between disorder and correlations in compositionally complex alloys, our study employs resonant and valence band photoemission, optical conductivity, and electrical resistivity, complemented by density functional theory- based linear response calculations. By applying dynamical mean- field theory, we identify correlation signatures and damping in spectra, emphasizing the significance of many- body effects, especially in states far from the Fermi edge. Electronic transport remains dominated by chemical and magnetic disorder. Our results advance understanding of element specific electronic correlations in CrMnFeCoNi, elucidating the complex physical nature of compositionally complex alloys.  
+
+<|ref|>text<|/ref|><|det|>[[70, 403, 481, 827]]<|/det|>
+Compositionally complex alloys (CCAs), which comprise the diverse range of medium- to high- entropy alloys (HEAs), are an exciting class of materials, consisting of randomly distributed multi- principal elements on crystalline lattices \(^{1,2}\) . Aiming on HEAs, the denomination stems from the entropy term overruling the enthalpy of formation of individual phases when mixing a large number of elements, hence escaping phase separation. They exhibit exceptional mechanical attributes including elevated toughness, minimal plastic deformation, and enhanced tensile and yield strengths \(^{3,4}\) , along with intriguing electronic as well phononic transport properties \(^{5 - 8}\) . CCAs, including HEAs, bridge the structural gap between crystalline solids and amorphous materials, exhibiting long- range periodicity but with atom variations on lattice sites inducing site- disorder akin to Anderson localization \(^{9}\) . For years, the question of electron propagation in such an environment has persisted \(^{10}\) , lacking the translational invariance ensuring the validity of the Bloch theorem for propagating electronic waves, yielding electronic localization. Empirical observations report electric resistivity \((\rho)\) approaching the Ioffe- Regel limit with a subdued \(\mathrm{d}\rho /\mathrm{d}T\) dependence \(^{11}\) , demonstrating that the effect of increasing residual resistivity in Cantor- Wu alloys may be linked to magnetic disorder effects \(^{12}\) . Regarding the \(\mathrm{d}\rho /\mathrm{d}T\) dependence various explanations have been suggested, including Anderson localization \(^{13}\) , or quantum interference in accordance with Mooij correlation \(^{14}\) . For the latter, it becomes evident that, in addition to electron- phonon and spin scattering, many- body effects arising from the electron- electron interaction may play a significant role. However, to date, this issue has not been thoroughly investigated.  
+
+<|ref|>text<|/ref|><|det|>[[70, 834, 480, 928], [515, 403, 927, 457]]<|/det|>
+In this work, we probe the role of electronic correlation next to disorder effects in CrMnFeCoNi, likely the most studied HEA and prototype CCA, by employing resonant (ResPES) and valence- band (VB) photoemission spectroscopy (PES), as well optical conductivity measurements. All these measurements are supported and explained by electronic structure calculations based on density functional theory (DFT) and dynamic mean- field theory (DMFT). Finally, we also provide an in- depth analysis of the temperature- dependent electrical resistivity, both theoretically and experimentally.   
+
+<|ref|>sub_title<|/ref|><|det|>[[516, 465, 722, 480]]<|/det|>
+## Results and discussion  
+
+<|ref|>text<|/ref|><|det|>[[515, 488, 927, 707]]<|/det|>
+Element resolved photoemission spectroscopy. In ResPES, photon energies proximate to the X- Ray absorption (XAS) \(L_{3}\) - edge are deployed, engendering interference between the direct photoemission channel \((2\mathrm{p}^{6}3\mathrm{d}^{n} + h\omega \rightarrow 2\mathrm{p}^{6}3\mathrm{d}^{n - 1} + \mathrm{e}_{\mathrm{f}})\) and the dipole transition of a core electron to an unoccupied state \((2\mathrm{p}^{6}3\mathrm{d}^{n} + h\omega \rightarrow 2\mathrm{p}^{5}3\mathrm{d}^{n + 1})\) . The subsequent decay of this intermediate state \((2\mathrm{p}^{5}3\mathrm{d}^{n + 1}\rightarrow 2\mathrm{p}^{6}3\mathrm{d}^{n - 1} + \mathrm{e}_{\mathrm{f}})\) , akin to an Auger process, gives rise to a strong resonant behavior due to the interference of both channels \(^{15,16}\) . ResPES measurements reveal site- or element- specific information on the electronic structure suitable for probing complex alloys where band- overlapping or hybridization is commonplace \(^{17}\) . Element specific \(L_{3}\) XAS spectra are depicted next to ResPES photoelectron energy distribution curves (EDCs) in Fig. 1 on a binding energy \((E_{\mathrm{B}})\) scale relative to the Fermi energy \(E_{\mathrm{F}}\) (calibrated by the Fermi level of gold).  
+
+<|ref|>text<|/ref|><|det|>[[515, 713, 927, 920]]<|/det|>
+For Cr, and in contrast to the other elements, the EDCs exhibit a pronounced maximum at constant \(E_{\mathrm{B}}\) of \(1.8\mathrm{eV}\) for an extended range of incident photon energies. This suggests an absent or low contribution of Auger signals in the EDCs, due to the missing transition from the radiationless Raman scattering regime (constant binding energy) to the resonant Raman Auger regime (constant kinetic energy), close to the \(L_{3}\) XAS maximum \(^{17 - 20}\) . For pure Cr, a VB peak in the EDCs of ResPES measurements was observed at \(1.2\mathrm{eV}^{20}\) . For Mn we find an EDC maximum at the \(L_{3}\) edge at \(3.6\mathrm{eV}\) , with a faint shoulder extending up to \(E_{\mathrm{B}} = 7.5\mathrm{eV}\) , hinting to a constant kinetic energy above the resonance. For Fe, Co, and Ni, the main features of the EDCs are clearly attributable to Auger signals \(^{19 - 22}\) . The EDC maxima for the XAS \(L_{3}\) edge are \(4.7\mathrm{eV}\) , \(4.2\mathrm{eV}\) , and \(7.2\mathrm{eV}\) \(E_{\mathrm{B}}\) , respectively. In no case a
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[68, 58, 481, 248]]<|/det|>
+dominant VB contribution at constant \(E_{\mathrm{B}}\) is observed. Comparing these data with those of pure elements provides information on band filling, hybridization, electronic correlations, chemical disorder and crystal field effects. Ni, sharing the fcc crystal structure and comparable lattice constant with the \(\mathrm{CrMnFeCoNi}\) HEA, mainly allows for a focus on the effects induced by chemical disorder. The measured EDCs display a shift of the well- known \(6\mathrm{eV}\) satellite for pure \(\mathrm{Ni}^{23,24}\) towards \(7.2\mathrm{eV}\) in the \(\mathrm{CrMnFeCoNi}\) HEA. Shifts as large as \(1.4\mathrm{eV}\) towards higher \(E_{\mathrm{B}}\) have been observed for Ni based alloys and intermetallic compounds \(^{22,25,26}\) , and explained by \(d\) - band filling through the hybridization of wave functions located on different lattice sites after alloying with more electropositive elements \(^{25,27,28}\) .  
+
+<|ref|>image<|/ref|><|det|>[[80, 259, 470, 655]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[68, 664, 481, 721]]<|/det|>
+<center>Fig. 1. Element specific XAS \(L_{3}\) curves (right side of graphs as white lines) and corresponding ResPES EDC for the elements of the \(\mathrm{CrMnFeCoNi}\) HEA. In the EDC the superimposed white line corresponds to the EDC intensities at the XAS \(L_{3}\) maximum. Energetic positions of the XAS measurements and EDCs are marked by arrows. </center>  
+
+<|ref|>text<|/ref|><|det|>[[68, 728, 481, 920], [515, 58, 927, 221]]<|/det|>
+Fig. 2(a) depicts the calculated \(d\) - band partial density of states (pDOS) of individual elements, for LDA (yellow solid line) and LDA+DMFT (blue solid line), next to the EDCs from Fig. 1 (black solid line). For calculation details see Methods and Supplementary Information (SI). Since we are interested only in the peak position, all graphs are normalized to their maximum. Marked differences emerge between LDA+DMFT and pure LDA, including strong satellites (Mn, Co, and Ni) and a generalized band- narrowing (Ni and Co). Eventually, the spectral weight (excluding satellites) shifts to lower \(E_{\mathrm{B}}\) by including DMFT. For Cr and Fe, \(d\) - band satellites merge with the \(sp\) - pDOS (see detailed Figures on band resolved pDOS in the SI). With increasing \(d\) - band filling, there is a progressive split- off of the formed Hubbard bands towards higher \(E_{\mathrm{B}}\) , although this effect is mitigated between Mn and Fe by the large difference in \(U\) . In Cr and Fe, the value of \(U\) is so small with respect to the bandwidth that no detached satellite peak can form; rather, it reflects a renormalization of the DOS, altering the \(d\) - band shape from rectangular to triangular. The positions of the satellites in the pDOS are compared to the ResPES data. For Mn, the small shoulder in the ResPES data at \(7.5\mathrm{eV}\) coincides with the satellite at \(8\mathrm{eV}\) in the pDOS. A comparable analysis for Fe and Co is not possible. For Ni, the pDOS data reveal a satellite at \(8.2\mathrm{eV}\) BE, elevated by \(1\mathrm{eV}\) compared to that determined experimentally via ResPES.   
+
+<|ref|>text<|/ref|><|det|>[[515, 226, 928, 925]]<|/det|>
+Element specific on- site Coulomb interaction. We apply the Cini- Sawatzky Theory (CST) \(^{29 - 31}\) by comparing the measured EDCs (assumed as Auger spectra) with the self- convoluted single- particle pDOS. Within the CST framework, Auger signals may be categorized into distinct regimes, according to the ratio between electronic bandwidth \(W\) and on- site Coulomb interaction \(U\) . For \(U \gg W\) , the spectra correspond to the quasi- atomic limit with split- off satellites at high \(E_{\mathrm{B}}\) , whereas being band- like for \(U \ll W\) . For the latter Lander \(^{32}\) proposes that the Auger signal equals the self- convolution of the single- particle band. At \(U \sim W\) , a complex interplay occurs, resulting in the superposition of both states. According to CST, for systems with nearly filled \(d\) - bands, a discernible shift of the Auger spectra towards lower \(E_{\mathrm{B}}\) is found, displacing the maximum of the self- convolution of the DOS by \(U\) relative to the Auger signals \(^{25,33,34}\) . This is explained by the energy difference of the two- hole state and the two- one- hole states being equal to the Coulomb interaction on the two- particle energy scale. Consequently, the ResPES maxima (black solid lines) are aligned with the self- convolution which are depicted in Fig. 2(a) by dashed colored lines. It is evident, that for Cr, only the VB is measured by ResPES, as the EDCs spectral weight coincidences with that of the SP band. For Mn and Fe, which are approximately situated in the band- like limit with \(U \ll W\) , the self- convolution and ResPES data on the two- electron scale show a congruence in their peaks. For Co, a minor offset of approximately \(1.6\mathrm{eV}\) is identified, which is slightly smaller than our suggested \(U\) of \(2.5\mathrm{eV}\) (see Methods). For Ni, this offset amounts to \(4\mathrm{eV}\) , which is slightly larger than the value of \(3\mathrm{eV}\) used as \(U\) in the LDA+DMFT calculation. This trend is expected, considering the different filling of the \(3d\) - band \(^{25,35 - 37}\) , as well as the limitations of the DMFT solver \(^{38}\) . In order to investigate the sensitivity of the CST we calculate the \(\mathrm{CrMnFeCoNi}\) HEA with the initially given \(U\) values, but increasing them for Fe from \(1.5\mathrm{eV}\) to \(2\mathrm{eV}\) and Ni from \(3\mathrm{eV}\) to \(4\mathrm{eV}\) . The pDOS (solid line) as well self- convoluted signal (dashed line) are given in Fig. 2(a) as green lines. For Fe there is barely a difference in the pDOS visible with a slight increase of the satellite. No shift in the self- convolution signal is found. For Ni, the \(d\) - block shifts also barely recognizably towards the Fermi edge, but the correlation satellite splits off significantly towards higher binding energies of almost \(10\mathrm{eV}\) . Consequently, the main peak of self- convolution hardly changes. The difference between EDC and self- convolution is \(4\mathrm{eV}\) , which corresponds exactly to the Hubbard \(U\) from the LDA+DMFT calculations. The variation of the element specific \(U\) value of Fe and Ni does not change the pDOS of the other elements (see the more detailed plot in the SI). However, since the extreme displacement of the split- off
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[68, 58, 480, 85]]<|/det|>
+satellite is not observed experimentally, we keep the initial chosen pure element \(U\) values in our calculations.  
+
+<|ref|>image<|/ref|><|det|>[[70, 90, 475, 460]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[68, 469, 481, 580]]<|/det|>
+<center>Fig. 2 Element resolved electronic spectra of CrMnFeCoNi. (a) Calculated pDOS for elements in CrMnFeCoNi with LDA and LDA+DMFT as yellow and blue solid lines, respectively. Measured EDCs in the XAS \(L_{3}\) absorption maximum as black solid lines. The self-convolution of the pDOS (CST) are given as dashed lines with corresponding color. For Ni and Fe, the LDA+DMFT pDOS for \(U = 4 \mathrm{eV}\) and \(2 \mathrm{eV}\) , respectively are given in green. (b) Experimental VB PES next to calculated spectra using the one-step model with LDA and LDA+DMFT. (c) and (d) Calculated BSF for the CrMnFeCoNi HEA in the LDA and LDA+DMFT approach. </center>  
+
+<|ref|>text<|/ref|><|det|>[[68, 586, 481, 928]]<|/det|>
+Fig. 2(b) presents VB PES measurements for \(\hbar \omega = 1200 \mathrm{eV}\) , alongside one- step model calculations \(^{39}\) for the LDA as well LDA+DMFT potentials (see SI for details). Experimentally, a satellite feature is visible, attributable to Ni at approximately \(7 \mathrm{eV}\) , as corroborated by ResPES. The LDA+DMFT calculation reveals a peak at approximately \(8 \mathrm{eV}\) , which perfectly overlaps with the Ni satellite in the calculated pDOS. The offset between theory and experiment is attributable to the perturbative nature of the DMFT solver, and aligns well with a \(1 \mathrm{eV}\) offset observed in pure Ni \(^{40}\) , which confirms our choice of \(U\) a posteriori. Despite the satellite, the shoulder spanning roughly from \(3 \mathrm{eV}\) to \(4 \mathrm{eV}\) , paralleling the experimental observations, is also reproduced in the LDA+DMFT calculation. The plateau ranging from \(5 \mathrm{eV}\) \(E_{\mathrm{B}}\) to \(8 \mathrm{eV}\) in the LDA+DMFT calculation resonates well with the experimental data, which extends from \(4 \mathrm{eV}\) \(E_{\mathrm{B}}\) to \(7 \mathrm{eV}\) . The LDA approach fails to adequately capture any of these correlation fingerprints. Despite the strong agreement between experimental and theoretical results, we still observe a discrepancy in the bandwidth, which is slightly narrower in LDA+DMFT. This difference can however have an extrinsic origin, as e.g. arise from subtle contributions from a minor surface oxide layer (oxygen 1s peak in wide scan XAS measurement provided in the SI). Also, the experimental background (intensity at lowest \(E_{\mathrm{B}}\) ), accounts for a systematic  
+
+<|ref|>text<|/ref|><|det|>[[515, 58, 927, 127]]<|/det|>
+deviation. Besides, we are able to conclude that the assumption of Hubbard \(U\) parameters for CrMnFeCoNi, aligned with the pure metals, is justified, and electronic correlations play a site- specific role comparable to that of pure 3d transition metals.  
+
+<|ref|>text<|/ref|><|det|>[[515, 134, 927, 379]]<|/det|>
+Broadening of the band structure. The influence of the broadening of states due to chemical disorder and quasiparticle lifetimes, can be discerned in the computed Bloch spectral function (BSF) depicted in Fig. 2(c) and 2(d) for LDA and LDA+DMFT, respectively. Near \(E_{\mathrm{F}}\) , both calculations reveal a scarcely dispersive \(d\) - band block with strongly localized electrons, while the parabolic dispersion of the \(sp\) - bands becomes apparent at higher \(E_{\mathrm{B}}\) . The \(d\) - bands around \(2 \mathrm{eV}\) are for the LDA+DMFT case so extensively smeared, especially along the symmetry line X- \(\Gamma\) - L, that subbands cannot be resolved. The strongly localized satellite states are perceptible over a constant smeared background up to \(9 \mathrm{eV}\) . Arguing qualitatively, the spectral width of the \(d\) - states implies that the lifetimes of these electrons must be exceedingly short. Interestingly, this arises mainly from correlation effects at high \(E_{\mathrm{B}}\) , but from the combined action of correlations and chemical/magnetic disorder in the vicinity of \(E_{\mathrm{F}}\) .  
+
+<|ref|>text<|/ref|><|det|>[[515, 386, 927, 864]]<|/det|>
+Element resolved quasiparticle lifetimes. Quasiparticle lifetimes \(\tau\) can be obtained from the self- energy function \(\Sigma\) from DMFT via \(\hbar /\tau = 2 \mathrm{Im}(\Sigma)\) . Fig. 3(a) displays element- specific lifetimes for Fe and Ni, obtained by evaluating the Greens function on the real energy axis. Data for other elements are plotted in the SI. For context, black lines show pure element calculations in their natural crystal structures, alongside corresponding experimental data from literature \(^{40 - 43}\) . These calculations align well, although the Ni lifetimes for excited states above \(1 \mathrm{eV}\) slightly exceed experimental observations. For all elements, near \(E_{\mathrm{F}}\) , a Fermi liquid theory- like behavior emerges, with \(\tau \propto (E - E_{\mathrm{F}})^{- 2.43}\) . To assess the effects of altered lattice constants on the element- specific self- energy, lifetimes for Fe and Ni within the fcc lattice but the HEA's lattice constant are illustrated in yellow in Fig. 3(a). While Ni shows a negligible variation, \(\gamma\) - Fe exhibits notably reduced lifetimes, despite having the same Hubbard \(U\) . This is not surprising, considering that \(\gamma\) - Fe is known to have stronger magnetic fluctuations leading to a complex magnetic landscape \(^{44}\) . The bottom row of Fig. 3(a) contrasts the pure elements in the fcc HEA structure against the CPA- derived disordered paramagnetic state. Here, both elements show an increased \(\tau\) , akin to those in their natural crystals. For the LDA+DMFT calculation of CrMnFeCoNi with \(U = 4 \mathrm{eV}\) for Ni and \(2 \mathrm{eV}\) for Fe, the green lines in Fig. 3(a) depict the results. For Ni, lifetimes above \(E_{\mathrm{F}}\) demonstrate a \(3 / 4\) reduction compared to the \(3 \mathrm{eV}\) calculation, reflecting the ratio of \(U\) values. A comparable trend is given for Fe. As observed in the pDOS data, other elements are not affected by the variation of \(U\) (see SI). Our calculations indicate a minor influence of chemical disorder on DMFT derived lifetimes, particularly when contrasted with data for the pure elements and taking the effect of changing crystal structure into account. Furthermore, \(\tau\) exhibit a nearly linear dependency to variations in \(U\) .
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[72, 58, 480, 421]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[68, 434, 481, 579]]<|/det|>
+<center>Fig. 3 Element specific quasiparticle lifetimes and frequency resolved optical conductivity of the \(\mathrm{CrMnFeCoNi}\) HEA. (a) Element specific lifetimes for Fe and Ni by black lines for calculated pure elements in natural crystal and yellow line pure metals within the fcc structure and \(\mathrm{CrMnFeCoNi}\) lattice constant. Experimental values for are taken from PES (below \(\mathrm{E}_{\mathrm{F}})^{40}\) and TR-2PPE (above \(\mathrm{E}_{\mathrm{F}})^{41 - 43}\) data of pure elements. Blue lines in the bottom correspond to LDA+DMFT and green dashed line to the LDA+DMFT case with changed \(U\) for Fe \(= 2\mathrm{eV}\) and \(\mathrm{Ni} = 4\mathrm{eV}\) . (b) Real (left) and imaginary part (right) of complex optical conductivity versus photon energy \(\hbar \omega\) . LDA calculations are given by the yellow line and LDA+DMFT calculations by the blue line. Experiments are from reflectometry measurements (Exp. 1) and ellipsometry (Exp. 2). </center>  
+
+<|ref|>text<|/ref|><|det|>[[70, 584, 481, 735]]<|/det|>
+Frequency resolved optical conductivity. The dynamic response of electrons in the \(\mathrm{CrMnFeCoNi}\) HEA is further probed by means of complex optical conductivity \(\sigma (\omega)\) measurements. Typically, \(\sigma (\omega)\) reveals for metals a Drude peak in \(\mathrm{Re}(\sigma (\omega))\) , which broadens with increasing scattering rate and eventually may merge with higher- energy interband transitions. Thus separation into inter- and intra- band contributions becomes challenging for correlated 3d transition metals. We therefore compute the \(\sigma (\omega)\) tensor via the Kubo formalism \(^{45}\) by the current- current correlation function \(^{46}\) .  
+
+<|ref|>equation<|/ref|><|det|>[[70, 740, 454, 850]]<|/det|>
+\[\sigma_{\mu \nu}(\omega) = \frac{i\hbar}{\pi^2}\frac{1}{\Omega}\int_{\Omega}d^3 r\int_{\Omega}d^3 r^{\prime}\int_{E_B}^{\infty}dE^{\prime}\int_{E_B}^{\infty}dE\Theta_T(E - E_F)\Theta_T(E_F\] \[-E^{\prime})\] \[\times \left\{\frac{\mathrm{tr}[j_\mu (\mathbf{r})\mathrm{Im}G^+ (E^{\prime})j_\nu (\mathbf{r}^{\prime})\mathrm{Im}G^+ (E)]}{(E^{\prime} - E - i\Gamma_{\mathrm{ep}})(\hbar\omega + E - E^{\prime} + i\Gamma_{\mathrm{ep}})\] \[+\frac{\mathrm{tr}[j_\nu (\mathbf{r}^{\prime})\mathrm{Im}G^+ (E^{\prime})j_\mu (\mathbf{r})\mathrm{Im}G^+ (E)]}{(E^{\prime} - E - i\Gamma_{\mathrm{ep}})(\hbar\omega + E^{\prime} - E + i\Gamma_{\mathrm{ep}})}\right\}\]  
+
+<|ref|>text<|/ref|><|det|>[[70, 855, 481, 923], [515, 59, 927, 280]]<|/det|>
+Due to the fully relativistic formulation of the current density operator \(j(\mathbf{r})\) , both paramagnetic and diamagnetic terms are implicitly included, with the latter yielding a Drude- like contribution \(^{46,47}\) . However, as in our approach electron- phonon collision driven damping mechanisms are not considered a priori, a phenomenological complex photon energy \(\Gamma_{\mathrm{ep}}\) is introduced overruling the infinite small lifetime \(0^{+}\) from adiabatic switching on of the external perturbative field. We adopt a constant \(\Gamma_{\mathrm{ep}}\) value of \(0.340\mathrm{eV}\) , attributed to the substantially diminished electronic mean free path, proximate to the lattice parameter, as well the given Fermi velocity for such disordered alloys \(^{48}\) . Our calculations utilize both LDA and LDA+DMFT, with the latter including energy dependent electron- electron scattering \(\Gamma_{\mathrm{ee}}\) through the imaginary part of the self- energy inherently incorporated in the Greens function. Although the DMFT scheme solely considers \(d\) - band states in \(\Sigma^{49}\) , an empirical choice (e.g. F of \(\Gamma_{\mathrm{ee}}\) for the \(sp\) - band is not obligatory. It has been shown analytically \(^{50}\) as well numerically within the \(GW\) formalism \(^{51}\) that \(d\) - band screening dominates the total self- energy, and thus scattering rates, when considering open \(d\) - band metals.   
+
+<|ref|>text<|/ref|><|det|>[[515, 286, 927, 680]]<|/det|>
+Calculation results are depicted in Fig. 3(b) (LDA in yellow, LDA+DMFT in blue) alongside experimental data (see SI). The left subfigure displays the real part of the optical conductivity, \(\mathrm{Re}(\sigma (\omega))\) , corresponding to the absorptive component. The results from LDA and LDA+DMFT show for low \(\hbar \omega\) a constant trend with a minor offset between each other, and dispersive variations in the VIS range where both curves decline. Quasiparticle lifetimes from the DMFT scheme improve the absorptive part of \(\sigma (\omega)\) drastically, especially in the visible and UV range, where a perfect agreement with experiment is found. Here the pure LDA calculations overestimate the absorption. Also for the LDA+DMFT case, at the highest photon energies of \(5\mathrm{eV}\) and above, the fine structure found in the LDA calculation is blurred (see small oscillation between \(6\mathrm{eV}\) and \(8\mathrm{eV}\) for real and imaginary part). This behavior reflects the smearing of sub- bands, induced through the quasi- particle lifetimes, as seen within the BSF in Fig. 3(c) and 3(d). For \(\mathrm{Im}(\sigma (\omega))\) , corresponding to the dispersive or reflective part, the situation is similar. The experiments show low values for small \(\hbar \omega\) (approaching zero), comparable to the Drude- behavior, with a resonance peak in the VIS region. The LDA calculation overestimates this peak threefold and is non- symmetric on a log scale, thus not purely Drude- like. The resonance position in \(\mathrm{Im}(\sigma (\omega))\) is found at \(\hbar \omega = 2.3\mathrm{eV}\) experimentally, and \(2.7\mathrm{eV}\) in the LDA and at \(2.4\mathrm{eV}\) in the LDA+DMFT calculations. Besides improving spectral position, LDA+DMFT calculations reduce the resonance maximum significantly, achieving an improved experimental alignment.  
+
+<|ref|>text<|/ref|><|det|>[[515, 688, 927, 907]]<|/det|>
+Temperature dependence of electrical resistivity. Revisiting temperature- dependent electrical resistivity, we focus on correlation effects. Fig. 4(a) depicts our four- point resistivity measurements for different metals, from Ni to \(\mathrm{CrMnFeCoNi}\) HEA, over a broad temperature range (4 K to \(800\mathrm{K}\) ). A dominant increase in residual resistivity is observed when transitioning from FeNi to \(\mathrm{CrFeNi}\) , which is attributed to the coupling of ferromagnetic and antiferromagnetic elements within the alloys \(^{12}\) . This results in smearing across both spin channels in the BSF of \(\mathrm{CrFeNi}\) , whereas FeNi exhibits solely smearing in the minority spin channel \(^{12}\) . Thus, Ni and NiFe show well- defined quasiparticle transport properties within the majority channel with progressive increase in upward curvature up to Curie temperature at \(625\mathrm{K}\) and to \(835\mathrm{K}\) , respectively. For the alloys exhibiting higher residual resistivity, similar trends are
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[70, 58, 480, 86]]<|/det|>
+observed: \(\rho_0\) ranges around \(100\mu \Omega \mathrm{cm}\) , and \(\mathrm{d}\rho /\mathrm{d}T\) remains low (and after subtraction of \(\rho_0\) , \(\mathrm{d}\rho /\mathrm{d}T\) seems to be identical).  
+
+<|ref|>image<|/ref|><|det|>[[70, 92, 460, 373]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[68, 380, 480, 447]]<|/det|>
+<center>Fig. 4 Measurement and calculation of temperature electrical resistivity. (a) Measurements of electrical resistivity versus temperature for several Cantor-Wu alloys, ranging from FeNi to the CrMnFeCoNi HEA. (b) Comparison of experiment and linear response calculation for LDA and LDA+DMFT case for CrMnFeCoNi. (c) and (d) Fermi surface of CrMnFeCoNi for the LDA and LDA+DMFT potential. </center>  
+
+<|ref|>text<|/ref|><|det|>[[68, 454, 480, 862]]<|/det|>
+We compare the CrMnFeCoNi measurement with linear response calculation within the Kubo- Greenwood formalism \(^{52}\) . The alloy analogy model \(^{52}\) is employed to mimic temperature- dependent lattice vibrations, inherently integrating electron- phonon scattering processes \(^{53}\) . Corroborating our BSF observations, both LDA and LDA+DMFT results yield comparable \(\rho_0\) of \(67\mu \Omega \mathrm{cm}\) . By Fermi surface analysis the resistivity may be calculated from the \(k\) - space smearing (electron mean free path), as well the surface area \(^{8}\) . Figs. 4(c) and 4(d) show the Fermi surfaces calculated with the LDA and LDA+DMFT potentials, which exhibit remarking similarities. This supports our linear response results. However, the substantial deviation from the experimental value of \(105\mu \Omega \mathrm{cm}\) could be influenced by potential short- range ordering \(^{54 - 56}\) , which has been experimentally shown to increase resistivity \(^{57}\) . Such mechanisms are not included within the CPA framework and need to be captured with more sophisticated methodologies like non- local CPA \(^{58,59}\) . The localization mechanism must also be considered. While Hubbard localization is addressed within the LDA+DMFT framework, the onset of Anderson localization is inherently not captured by the CPA approach \(^{60,61}\) . The temperature dependence of \(\rho (T)\) is presented in a normalized format \(\rho (T) / \rho_0\) in Fig. 4(b) and demonstrates that while LDA closely aligns with experimental data, even indicating a reduced \(\mathrm{d}\rho /\mathrm{d}T\) , LDA+DMFT yields perfect agreement across all temperatures. This aligns with the BSFs, where LDA+DMFT exhibits increased smearing at higher \(E_{\mathrm{B}}\) , thus impacting electronic transport at elevated temperatures.  
+
+<|ref|>text<|/ref|><|det|>[[68, 870, 480, 925], [515, 58, 926, 236]]<|/det|>
+Conclusion. Our findings identify CrMnFeCoNi as a material characterized by pronounced electronic correlations. These correlations exist in addition to the prevailing chemical and magnetic disorder, which smear the band structure in the vicinity of \(E_{\mathrm{F}}\) and thus primarily cause the high residual resistivity. The extent of many body effects by means of on- site Coulomb interaction within the Hubbard model mirrors those of the containing pure elements. Correlation effects gain in significance with increasing distance from the Fermi edge, which was demonstrated both experimentally and theoretically utilizing electronic spectroscopies and temperature dependent electronic transport. Especially in the calculation of the optical response, accounting for quasiparticle lifetimes dramatically improve the absorptive as well dispersive part of the optical conductivity in the VIS- UV range, as energy dependent electron- electron scattering may overrule electron- phonon scattering rates.   
+
+<|ref|>sub_title<|/ref|><|det|>[[515, 243, 592, 257]]<|/det|>
+## Methods  
+
+<|ref|>text<|/ref|><|det|>[[515, 265, 926, 472]]<|/det|>
+Sample preparation and characterization. All alloys were first synthesized by mixing and pressing powders of elemental metals (total mass \(1.5\mathrm{g}\) ) into pellets to achieve the targeted composition. The pellets were placed in an arc- melting chamber, with glowing elemental zirconium used to remove any residual oxygen. Arc melting was performed three times, and the alloys were subsequently annealed in an evacuated quartz ampoule for one month at \(1030^{\circ}\mathrm{C}\) to enhance sample homogeneity. The specimens were cut into approximately \(0.5\mathrm{mm}\) thick discs for photoemission and optical conductivity measurements and into bars of dimensions \(8 \times 3 \times 0.5\mathrm{mm}^3\) for four- point electrical resistivity measurement. The series of prepared samples included Ni, NiFe, NiFeCr, NiFeCrCo, and NiFeCrCoMn. Characterization of the HEA samples using X- ray diffraction confirmed a well- defined fcc crystalline structure. Scanning electron microscopy and electron- dispersive X- ray spectroscopy analyses revealed the actual composition and satisfactory dispersion of constituent elements.  
+
+<|ref|>text<|/ref|><|det|>[[515, 480, 926, 567]]<|/det|>
+Electronic spectroscopies measurements. XAS and ResPES measurements were conducted at the ADRESS beamline of the Paul Scherrer Institut \(^{62}\) on the CrMnFeCoNi alloy at a base temperature of \(20\mathrm{K}\) . The photon energy was scanned along each of the relevant \(L_3\) absorption edges at steps of \(100\mathrm{meV}\) , while at the same time recording a valence band spectrum with a hemispherical electron analyzer at a resolution better than \(100\mathrm{meV}\) .  
+
+<|ref|>text<|/ref|><|det|>[[515, 574, 926, 671]]<|/det|>
+Optical conductivity measurements. For the optical conductivity measurements, the CrMnFeCoNi sample was ground with a SiC paper and subsequently polished with diamond paste with decreasing grain sizes to achieve a root mean square surface roughness of \(2\mathrm{nm}\) . The spectra were obtained by ellipsometry (Sentech SE 850) in the visible to infrared range and by reflectance measurements with subsequent Kramers- Kronig transformation in the far infrared part of the electromagnetic spectrum.  
+
+<|ref|>text<|/ref|><|det|>[[515, 679, 926, 875]]<|/det|>
+LDA+DMFT calculations. DFT calculations of CrMnFeCoNi were performed within the fully relativistic spin polarized multiple scattering Korringa- Kohn- Rostoker (SPR- KKR) Greens function formalism \(^{63,64}\) . Many- body correlation effects beyond local density approximation (LDA) were added via DMFT, as implemented in SPR- KKR \(^{49}\) . Chemical disorder was accounted for by the coherent potential approximation (CPA) \(^{65,66}\) and the paramagnetic state at \(20\mathrm{K}^{67}\) was mimicked by the disordered local moment scheme \(^{68}\) . An appealing feature of this approach is that it allows to consider local quantum and disorder fluctuations on the same footing. This approach has already been successfully realized by other groups \(^{69}\) . For LDA+DMFT calculations, the Hubbard \(U\) for \(3d\) electrons was set to the values found for pure elements, while the Hund exchange was \(J = 0.94\mathrm{eV}\) for all elements. The \(U\) values for Cr, Mn, Fe, Co, Ni were equal to \(2.0\mathrm{eV}^{70}\) , \(3.0\mathrm{eV}^{38}\) , \(1.5\mathrm{eV}\) , \(2.5\mathrm{eV}\) and \(3.0\mathrm{eV}^{40}\) , respectively. For further details, see SI.  
+
+<|ref|>text<|/ref|><|det|>[[515, 883, 926, 921]]<|/det|>
+Application of the Cini- Sawatzky theory. We applied a self- convolution to the calculated partial density of states and compared the results within the CST \(^{29 - 31}\) with the experimental data obtained
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[69, 58, 480, 131]]<|/det|>
+fromResPES. This approach aims to replicate the two- hole interactions using single- particle ground states, necessitating the presentation of the resultant self- convolution on a two- particle energy scale34. Consequently, the new energy axis is multiplied by a factor of two. Prior to convolution, the pDOS data were interpolated on a refined energy grid.  
+
+<|ref|>text<|/ref|><|det|>[[69, 138, 480, 323]]<|/det|>
+Calculation of spectroscopic and transport properties. In our study, valence band photoemission39 and electronic transport calculations, including the optical conductivity tensor, were conducted using the SPR- KKR code and a Kubo- framework for linear response, respectively. These calculations incorporated electron- phonon interactions within the alloy analogy model52, considering temperature- dependent atomic displacements and a comprehensive k- point mesh for accuracy. The optical response to external electromagnetic fields was detailed through a current- current correlation function reformulated in Green's function terms, capturing the Drude contribution within a fully relativistic framework46. Our findings primarily focus on the isotropic \(\sigma (\omega)\) in paramagnetic CrMnFeCoNi HEA. For an in- depth explanation of methodologies, including the computational parameters and models employed, the reader is referred to the SI.  
+
+<|ref|>sub_title<|/ref|><|det|>[[70, 330, 210, 345]]<|/det|>
+## Data availability  
+
+<|ref|>text<|/ref|><|det|>[[69, 352, 480, 376]]<|/det|>
+The data supporting the results of this study are available on reasonable request from the corresponding author.  
+
+<|ref|>sub_title<|/ref|><|det|>[[70, 384, 170, 398]]<|/det|>
+## References  
+
+<|ref|>text<|/ref|><|det|>[[67, 404, 483, 911]]<|/det|>
+1. George, E. P., Raabe, D. & Ritchie, R. O. High-entropy alloys. Nat. Rev. Mater. 4, 515-534 (2019).  
+2. Miracle, D. B. & Senkov, O. N. A critical review of high entropy alloys and related concepts. Acta Mater. 122, 448-511 (2017).  
+3. Gludovatz, B. et al. A fracture-resistant high-entropy alloy for cryogenic applications. Science 345, 1153-1158 (2014).  
+4. Li, Z., Zhao, S., Ritchie, R. O. & Meyers, M. A. Mechanical properties of high-entropy alloys with emphasis on face-centered cubic alloys. Prog. Mater. Sci. 102, 296-345 (2019).  
+5. Körmann, F., Ikeda, Y., Grabowski, B. & Sluiter, M. H. F. Phonon broadening in high entropy alloys. Npj Comput. Mater. 3, 36 (2017).  
+6. Mu, S. et al. Uncovering electron scattering mechanisms in NiFeCoCrMn derived concentrated solid solution and high entropy alloys. Npj Comput. Mater. 5, 1 (2019).  
+7. Fan, Z., Wang, H., Wu, Y., Liu, X. J. & Lu, Z. P. Thermoelectric high-entropy alloys with low lattice thermal conductivity. RSC Adv. 6, 52164-52170 (2016).  
+8. Robarts, H. C. et al. Extreme Fermi Surface Smearing in a Maximally Disordered Concentrated Solid Solution. Phys. Rev. Lett. 124, 046402 (2020).  
+9. Anderson, P. W. Absence of Diffusion in Certain Random Lattices. Phys. Rev. 109, 1492-1505 (1958).  
+10. Belitz, D. & Schirmacher, W. Theory of phonon-controlled conductivity in high-resistivity conductors. J. Phys. C Solid State Phys. 16, 913-926 (1983).  
+11. Jin, K. et al. Tailoring the physical properties of Ni-based single-phase equiatomic alloys by modifying the chemical complexity. Sci. Rep. 6, 20159 (2016).  
+12. Zhang, Y. et al. Influence of chemical disorder on energy dissipation and defect evolution in advanced alloys. J. Mater. Res. 31, 2363-2375 (2016).  
+13. Kourov, N. I. et al. Peculiar features of physical properties of the rapid quenched AlCrFeCoNiCu high-entropy alloy. J. Alloys Compd. 636, 304-309 (2015).  
+14. Ciuchi, S., Di Sante, D., Dobrosavljevic, V. & Fratini, S. The origin of Mooij correlations in disordered metals. Npj Quantum Mater. 3, 44 (2018).  
+15. Fano, U. Effects of Configuration Interaction on Intensities and Phase Shifts. Phys. Rev. 124, 1866-1878 (1961).  
+
+<|ref|>text<|/ref|><|det|>[[515, 58, 928, 928]]<|/det|>
+16. Sakisaka, Y. et al. Photoemission study of the valence-band satellite of Ni(110). Phys. Rev. B 36, 6383-6389 (1987).  
+17. Formoso, V. et al. Resonant valence-band photoemission spectroscopy on the Fe62Ni20Cr18 alloy. Eur. Phys. J. B 43, 463-470 (2005).  
+18. Kay, A. et al. Multi-Atom Resonant Photoemission: A Method for Determining Near-Neighbor Atomic Identities and Bonding. Science 281, 679-683 (1998).  
+19. Weinelt, M. et al. Resonant Photoemission at the 2 p Edges of Ni: Resonant Raman and Interference Effects. Phys. Rev. Lett. 78, 967-970 (1997).  
+20. Hufner, S. et al. Observation of the two-hole satellite in Cr and Fe metal by resonant photoemission at the 2 p absorption energy. Phys. Rev. B 61, 12582-12585 (2000).  
+21. Yablonskikh, M. V., Shkvarin, A. S., Yarmoshenko, Y. M., Skorikov, N. A. & Titov, A. N. Resonant photoemission at the L3 absorption edge of Mn and Ti and the electronic structure of 1T-Mn 0.2 TiSe 2. J. Phys. Condens. Matter 24, 045504 (2012).  
+22. Hsu, L.-S., Tsang, K.-L. & Chung, S.-C. Resonant photoemission and photoabsorption at the Ni L2,3 edges of NiGa and Ni3Ga. J. Magn. Magn. Mater. 177-181, 1031-1032 (1998).  
+23. Guillot, C. et al. Resonant Photoemission in Nickel Metal. Phys. Rev. Lett. 39, 1632-1635 (1977).  
+24. Clauberg, R., Gudat, W., Kisker, E., Kuhlmann, E. & Rothberg, G. M. Nature of the Resonant 6-eV Satellite in Ni: Photoelectron Spin-Polarization Analysis. Phys. Rev. Lett. 47, 1314-1317 (1981).  
+25. Bennett, P. A., Fuggle, J. C., Hillebrecht, F. U., Lenselink, A. & Sawatzky, G. A. Electronic structure of Ni and Pd alloys. III. Correlation effects in the Auger spectra of Ni alloys. Phys. Rev. B 27, 2194-2209 (1983).  
+26. Schneider, W.-D., Reihl, B., Martensson, N. & Arko, A. J. Resonant photoemission study of U Ni 5 - x Cu x. Phys. Rev. B 26, 423-426 (1982).  
+27. Fuggle, J. C. et al. Electronic structure of Ni and Pd alloys. I. X-ray photoelectron spectroscopy of the valence bands. Phys. Rev. B 27, 2145-2178 (1983).  
+28. Hillebrecht, F. U., Fuggle, J. C., Bennett, P. A., Zohnierek, Z. & Freiburg, Ch. Electronic structure of Ni and Pd alloys. II. X-ray photoelectron core-level spectra. Phys. Rev. B 27, 2179-2193 (1983).  
+29. Cini, M. Density of states of two interacting holes in a solid. Solid State Commun. 20, 605-607 (1976).  
+30. Sawatzky, G. A. Quasiatomic Auger Spectra in Narrow-Band Metals. Phys. Rev. Lett. 39, 504-507 (1977).  
+31. Cini, M. Comment on quasiatomic Auger spectra in narrow-band metals. Phys. Rev. B 17, 2788-2789 (1978).  
+32. Lander, J. J. Auger Peaks in the Energy Spectra of Secondary Electrons from Various Materials. Phys. Rev. 91, 1382-1387 (1953).  
+33. Treglia, G., Desjonqueres, M. C., Ducastelle, F. & Spanjaard, D. Correlation effects on Auger spectra in unfilled d band metals. J. Phys. C Solid State Phys. 14, 4347-4355 (1981).  
+34. Chainani, A. et al. Evidence for Weakly Correlated Oxygen Holes in the Highest- T c Cuprate Superconductor HgBa 2 Ca 2 Cu 3 O 8 + δ. Phys. Rev. Lett. 119, 057001 (2017).  
+35. Vilmercati, P. et al. Direct probe of the variability of Coulomb correlation in iron pnictide superconductors. Phys. Rev. B 85, 235133 (2012).  
+36. Cini, M. & Verdozzi, C. Many-body effects in the electron spectroscopies of incompletely filled bands. Il Nuovo Cimento D 9, 1-21 (1987).  
+37. Rausch, R. & Potthoff, M. Multiplons in the two-hole excitation spectra of the one-dimensional Hubbard model. New J. Phys. 18, 023033 (2016).  
+38. Di Marco, I. et al. γ-Mn at the border between weak and strong correlations. Eur. Phys. J. B 72, 473-478 (2009).  
+39. Braun, J., Minár, J., Ebert, H., Katsnelson, M. I. & Lichtenstein, A. I. Spectral Function of Ferromagnetic 3 d Metals: A Self-Consistent LSDA + DMFT Approach Combined with the One-Step Model of Photoemission. Phys. Rev. Lett. 97, 227601 (2006).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[66, 60, 483, 920]]<|/det|>
+40. Sánchez-Barriga, J. et al. Effects of spin-dependent quasiparticle renormalization in Fe, Co, and Ni photoemission spectra:An experimental and theoretical study. Phys. Rev. B 85, 205109 (2012). 
+41. Knoesel, E., Hotzel, A. & Wolf, M. Temperature dependence of surface state lifetimes, dephasing rates and binding energies on Cu(111) studied with time-resolved photoemission. J. Electron Spectrosc. Relat. Phenom. 88-91, 577-584 (1998). 
+42. Hong, J. & Mills, D. L. Theory of the spin dependence of the inelastic mean free path of electrons in ferromagnetic metals: A model study. Phys. Rev. B 59, 13840-13848 (1999). 
+43. Bauer, M., Marienfeld, A. & Aeschlimann, M. Hot electron lifetimes in metals probed by time-resolved two-photon photoemission. Prog. Surf. Sci. 90, 319-376 (2015). 
+44. Igoshev, P. A., Efremov, A. V., Poteryaev, A. I., Katanin, A. A. & Anisimov, V. I. Magnetic fluctuations and effective magnetic moments in \(\gamma\) -iron due to electronic structure peculiarities. Phys. Rev. B 88, 155120 (2013). 
+45. Kubo, R. Statistical-Mechanical Theory of Irreversible Processes. I. General Theory and Simple Applications to Magnetic and Conduction Problems. J. Phys. Soc. Jpn. 12, 570-586 (1957). 
+46. Huhne, T. & Ebert, H. Fully relativistic description of the magneto-optical properties of arbitrary layered systems. Phys. Rev. B 60, 12982-12989 (1999). 
+47. Banhart, J. Optical Conductivity of Disordered Alloys Calculated from First Principles. Phys. Rev. Lett. 82, 2139-2142 (1999). 
+48. Samolyuk, G. D. et al. Temperature dependent electronic transport in concentrated solid solutions of the 3 d-transition metals Ni, Fe, Co and Cr from first principles. Phys. Rev. B 98, 165141 (2018). 
+49. Minár, J. Correlation effects in transition metals and their alloys studied using the fully self-consistent KKR-based LSDA + DMFT scheme. J. Phys. Condens. Matter 23, 253201 (2011). 
+50. Zarate, E., Apell, P. & Echenique, P. M. Calculation of low-energy-electron lifetimes. Phys. Rev. B 60, 2326-2332 (1999). 
+51. Nechaev, I. A., Chulkov, E. V. & Echenique, P. M. G W lifetimes of quasiparticle excitations in paramagnetic transition metals. Phys. Rev. B 76, 245125 (2007). 
+52. Ebert, H. et al. Calculating linear-response functions for finite temperatures on the basis of the alloy analogy model. Phys. Rev. B 91, 165132 (2015). 
+53. Xu, J. et al. Thermal Conductivity and Electrical Resistivity of Solid Iron at Earth's Core Conditions from First Principles. Phys. Rev. Lett. 121, 096601 (2018). 
+54. Mizuno, M., Sugita, K. & Araki, H. Prediction of short-range order in CrMnFeCoNi high-entropy alloy. Results Phys. 34, 105285 (2022). 
+55. Singh, P., Smirnov, A. V. & Johnson, D. D. Atomic short-range order and incipient long-range order in high-entropy alloys. Phys. Rev. B 91, 224204 (2015). 
+56. Zhang, R. et al. Short-range order and its impact on the CrCoNi medium-entropy alloy. Nature 581, 283-287 (2020). 
+57. Tanimoto, H., Hozumi, R. & Kawamura, M. Electrical resistivity and short-range order in rapid-quenched CrMnFeCoNi high-entropy alloy. J. Alloys Compd. 896, 163059 (2022). 
+58. Woodgate, C. D., Hedlund, D., Lewis, L. H. & Staunton, J. B. Interplay between magnetism and short-range order in medium- and high-entropy alloys: CrCoNi, CrFeCoNi, and CrMnFeCoNi. Phys. Rev. Mater. 7, 053801 (2023). 
+59. Lowitzer, S., Ködderitzsch, D., Ebert, H. & Staunton, J. B. Electronic transport in ferromagnetic alloys and the Slater-Pauling curve. Phys. Rev. B 79, 115109 (2009). 
+60. Ekuma, C. E. et al. Typical medium dynamical cluster approximation for the study of Anderson localization in three dimensions. Phys. Rev. B 89, 081107 (2014). 
+61. Terletska, H., Zhang, Y., Chioncel, L., Vollhardt, D. & Jarrell, M. Typical-medium multiple-scattering theory for disordered systems with Anderson localization. Phys. Rev. B 95, 134204 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[515, 58, 928, 404]]<|/det|>
+62. Strocov, V. N. et al. Soft-X-ray ARPES facility at the ADRESS beamline of the SLS: concepts, technical realisation and scientific applications. J. Synchrotron Radiat. 21, 32-44 (2014). 
+63. Ebert, H., Ködderitzsch, D. & Minár, J. Calculating condensed matter properties using the KKR-Green's function method—recent developments and applications. Rep. Prog. Phys. 74, 096501 (2011). 
+64. Ebert, H. The Munich SPR-KKR package. version 8.6 http://olymp.cup.uni-muenchen.de/ak/ebert/SPRKKR (2022). 
+65. Faulkner, J. S. & Stocks, G. M. Calculating properties with the coherent-potential approximation. Phys. Rev. B 21, 3222-3244 (1980). 
+66. Ebert, H., Vernes, A. & Banhart, J. Relativistic bandstructure of disordered magnetic alloys. Solid State Commun. 104, 243-247 (1997). 
+67. Wang, S., Zhang, T., Hou, H. & Zhao, Y. The Magnetic, Electronic, and Thermodynamic Properties of High Entropy Alloy CrMnFeCoNi: A First-Principles Study. Phys. Status Solidi B 255, 1800306 (2018). 
+68. Gyorffy, B. L., Pindor, A. J., Staunton, J., Stocks, G. M. & Winter, H. A first-principles theory of ferromagnetic phase transitions in metals. J. Phys. F Met. Phys. 15, 1337-1386 (1985). 
+69. Östlin, A., Vitos, L. & Chioncel, L. Correlated electronic structure with uncorrelated disorder. Phys. Rev. B 98, 235135 (2018). 
+70. Belozerov, A. S., Katanin, A. A. & Anisimov, V. I. Itinerant magnetism of chromium under pressure: a DFT+DMFT study. J. Phys. Condens. Matter 33, 385601 (2021).  
+
+<|ref|>sub_title<|/ref|><|det|>[[516, 416, 682, 430]]<|/det|>
+## Acknowledgments  
+
+<|ref|>text<|/ref|><|det|>[[515, 431, 927, 565]]<|/det|>
+This work was supported by Deutsche Forschungsgemeinschaft under Grant 528706678 and the project MEBIOSYS, funded as project No. CZ.02.01.01/00/22_008/0004634 by Programme Johannes Amos Comnenius, call Excellent Research. This research is also part of the project No. 2022/45/P/ST3/04247 co- funded by the National Science Centre and the European Union's Horizon 2020 research and innovation programme under the Marie Skodowska- Curie grant agreement no 945339. For the purpose of Open Access, the author has applied a CC- BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission.  
+
+<|ref|>sub_title<|/ref|><|det|>[[516, 579, 700, 592]]<|/det|>
+## Author contributions  
+
+<|ref|>text<|/ref|><|det|>[[515, 593, 927, 715]]<|/det|>
+J.M., L.F. and H.D. developed the concept. E.M., X.M., L.C., D.T. and T.I. prepared the samples and performed transport measurements. M.C., V.S. and H.D. were responsible for the photoemission experiments at the Swiss Light Source. D.R. and J.M. conducted the CST analysis. D.R. and S.K. carried out the DFT calculations. I.D.M. interpreted the DMFT results and H.H. discussed the optics. H.E., J.M. and A.H. developed the KKR package. D.R. wrote the manuscript in close collaboration with J.M., L.F. and H.D. All authors contributed to discussions and commented on the manuscript.  
+
+<|ref|>sub_title<|/ref|><|det|>[[516, 728, 714, 742]]<|/det|>
+## Additional information  
+
+<|ref|>text<|/ref|><|det|>[[516, 744, 847, 757]]<|/det|>
+Supplementary Information accompanies this paper.  
+
+<|ref|>text<|/ref|><|det|>[[515, 767, 917, 780]]<|/det|>
+Competing interests: The authors declare no competing interests.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[43, 42, 312, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[43, 92, 768, 112]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 129, 313, 149]]<|/det|>
+20240222Supplement.pdf
+
+<--- Page Split --->

preprint/preprint__990b6b7a4019afa38768bea301e6129480c22c61c21e3e2575ba8092b42152e6/images_list.json

@@ -0,0 +1,55 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "Figures",
+    "footnote": [],
+    "bbox": [
+      [
+        98,
+        130,
+        930,
+        750
+      ]
+    ],
+    "page_idx": 18
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2 | The average time for retrieval of ANNs for each query spectrum on one IVF-PQ index, as a function of the number of spectra in the archive, in millions, if one query (blue), 10 queries (orange), 100 queries (green), 1000 queries (gold) and 10000 queries (purple), respectively, were executed in a batch. As shown, the per-query retrieval time scales linearly with archive size.",
+    "footnote": [],
+    "bbox": [
+      [
+        127,
+        88,
+        867,
+        515
+      ]
+    ],
+    "page_idx": 19
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3 | Recall of TNNs by Spectroscape. (a) Histograms of per-query recall of the TNNs, tested with 20,000 randomly selected queries from the archive. TNNs are defined by two parameters: N, the number of closest neighbors to be retained as TNNs (N = 10 and 100), and t, the minimum dot product to be considered a TNN (t = 0.5 to 0.9). On both left (N = 10) and right panels (N = 100), the per-query recall distributions are dominated by the rightmost column, representing the fraction",
+    "footnote": [],
+    "bbox": [
+      [
+        120,
+        140,
+        870,
+        760
+      ]
+    ],
+    "page_idx": 20
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4 | Examples of spectrum clusters visualized by Spectroscape. (a) Identification of unidentified spectra by association with neighbors with a precursor mass shift. The query node (PSM shown in top left panel) is unidentified but its nearest neighbors, with a precursor mass shift of 230.04 Da, are identified as DDPLTNLTFAFVAEK/2. The mass shift equals to the mass of the oligopeptide DD. Therefore, the unidentified query node and all its black neighbors are likely the semi-tryptic peptide PLTNLTFAFVAEK/2, which can also be confirmed by the fact that a library spectrum of that peptide ion is located in the middle of that cluster. The bottom left panel shows a very good PSM when we re-annotate the query node with this semi-tryptic peptide. (b) A query spectrum as a bridge of two clusters of PSMs corresponding to two peptides, LGPAIATGNVVVMK/2 and HVNPVQALSEFK/2. It is clearly a chimeric spectrum as the two PSMs almost do not share any matched peaks, as shown by the same spectrum annotated with respect to the two peptide ions (left panel). (c) The same peptide ion NQPGNTLTEILETPATAQQEVDHATDM[147]VSR/3 yields two slightly different fragmentation patterns on two mass spectrometers, as can be visualized clearly as two subclusters. The NIST HCD library spectrum (top left), represented as the node with a dotted outline that lies in the middle of the two subclusters, does not capture either of the fragmentation patterns very well.",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 22
+  }
+]

preprint/preprint__990b6b7a4019afa38768bea301e6129480c22c61c21e3e2575ba8092b42152e6/preprint__990b6b7a4019afa38768bea301e6129480c22c61c21e3e2575ba8092b42152e6.mmd

@@ -0,0 +1,253 @@
+
+# Spectroscape: Real-time Query and Visualization of a Spectral Archive of Millions of Tandem Mass Spectra  
+
+Henry Lam ( \(\boxed{\bullet}\) kehlam@ust.hk) The Hong Kong University of Science and Technology https://orcid.org/0000- 0001- 7928- 0364 Long Wu The Hong Kong University of Science and Technology https://orcid.org/0000- 0002- 3566- 9818 Ayman Hoque The Hong Kong University of Science and Technology  
+
+## Brief Communication  
+
+Keywords: Proteomics, mass spectrometry, spectral archive, big data, data visualization  
+
+Posted Date: October 25th, 2022  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 2166780/v1  
+
+License: \(\circledcirc\) This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+
+# Spectroscape: Real-time Query and Visualization of a Spectral Archive of Millions of Tandem Mass Spectra  
+
+Long \(\mathrm{Wu}^{1,2}\) , Ayman Hoque<sup>1</sup>, Henry Lam<sup>1,\*</sup>  
+
+<sup>1</sup> Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology  <sup>2</sup> Department of Electrical and Computer Engineering, The Hong Kong University of Science and Technology  
+
+\* Corresponding author  Henry Lam  Department of Chemical and Biological Engineering  The Hong Kong University of Science and Technology  Clear Water Bay  Hong Kong Special Administrative Region  China  Email: kehlam@ust.hk  
+
+Keywords: Proteomics, mass spectrometry, spectral archive, big data, data visualization
+
+<--- Page Split --->
+
+## Abstract  
+
+In proteomics, spectral archives organize the enormous amounts of publicly available peptide tandem mass spectra by similarity, offering opportunities for error correction and novel discoveries. Here we adapted an indexing algorithm developed by Facebook for organizing online multimedia resources to tandem mass spectra, and achieved practically instantaneous retrieval and spectrum clustering of approximate nearest neighbors in a large spectral archive. An interactive web- based graphical user interface enables the user to view a query spectrum in its clustered "neighborhood," which facilitate contextual validation of peptide identifications and exploration of the dark proteome.  
+
+## Main  
+
+The tandem mass spectrum is the basic unit of proteomic data, each representing a direct experimental observation of a peptide, and by inference, the parent protein. A modern mass spectrometer can easily produce tens of thousands of MS/MS spectra every hour. Such rapid data generation, when combined with the increasing prevalence of these machines, has led to an astronomical amount of data. In ProteomeXchange, the leading data repository in proteomics, over 5,300 datasets are deposited in 2021 alone, with the number of datasets deposited steadily increasing every year. \(^{1 - 3}\) Within these datasets, the number of MS/MS spectra currently available in the public domain is on the order of tens of billions. \(^{2,3}\) Moreover, this data volume probably represents only a small fraction of all proteomic data generated, as typically only data- sets directly associated with publications are deposited. Currently, proteomics data are largely stored in data repositories, which organize the data by data- sets associated with individual publications or projects. \(^{4 - 7}\) In effect, the spectra are stored in the same unprocessed data files and organized in the same structure as when they were submitted, usually as supporting data of a publication. As many have argued, while this is an important first step, such organization is not conducive to meaningful data reuse. \(^{8,9}\)  
+
+The emerging paradigm explored by several proteomics databases is to organize data by spectral similarity, in what is called a spectral archive. \(^{10,11}\) A spectral archive preserves all individual spectra, identified or not, and spectra similar to each other are grouped in a process
+
+<--- Page Split --->
+
+called spectrum clustering. In theory, this organization should maximize the potential for data reuse. For instance, by analyzing clusters of similar spectra, the unidentified ones among them can become identified through association, errors in identifications can be corrected, outliers can be detected, and cross- references across data- sets can be made. Therefore, once in a spectral archive, the identification of a given spectrum can evolve and hopefully self- correct over time. \(^{11 - 13}\) Another application of spectral archives, which was demonstrated previously, was to discover unexpected post- translational modifications or amino acid substitutions by detecting mass shifts between connected members in a cluster. \(^{14}\) However, it is computationally challenging to maintain large spectral archives, since in principle each spectrum needs to be compared to every other spectrum, though in practice the problem is made more tractable by various shortcuts. Some common strategies are: (i) progressively reducing the search space by merging spectra immediately in newly found clusters, (ii) limit comparisons to spectra with similar precursor m/z values, and (iii) avoiding comparisons of dissimilar spectra by grouping similar spectra first using various dimensionality reduction schemes. \(^{10,13,15 - 21}\)  
+
+Despite the obvious promise of spectral archives, the adoption of the idea by the wider proteomics community has been slow. In our view, the main reason is that spectrum clustering is computationally intensive, and does not seem accessible to the typical proteomics researcher. Moreover, the outcome of the spectrum clustering is not visible to the human user, and must be further analyzed by other computational tools, as existing tools lack a graphical user interface that allow the user to interact with the spectral archive. To break this inertia, we developed a platform called Spectroscape, that enables the real- time query and visualization of a spectral archive (Figure 1). Unlike other spectrum clustering tools that focus on pre- building the spectral archive off- line, our method enables query- driven real- time clustering, providing the detailed cluster structures with connectivity information. This is made possible by an indexing scheme based on the inverted file and product quantization encoding (IVF- PQ) algorithm in the Facebook AI Similarity Search (FAISS) \(^{22}\) library that groups spectra in “neighborhoods” in high- dimensional space, defined by approximate spectral similarity. Given any query spectrum that the user supplies, the method efficiently retrieves all its approximate nearest neighbors in the entire repository and performs real- time spectrum clustering by computing accurate pairwise distances among the query and the neighbors to reveal any cluster(s) in the neighborhood. The user can then visualize the
+
+<--- Page Split --->
+
+result in an interactive web- based user interface. In effect, the platform allows the user to search the entire data repository by spectral similarity, and instantaneously receive not only a list of best- matching spectra to the query, but also the detailed structure of any cluster in its neighborhood. Unlike other clustering algorithms, Spectroscape is fast enough even without limiting comparisons by using the precursor m/z value, making it suitable for discovering unexpected post- translational modifications or sequence variants<sup>23,24</sup>.  
+
+In Spectroscape, a spectrum is represented by a vector in 4096- dimensional space. The chosen variant of the IVF- PQ algorithm converts such a vector into a 17- byte string that serves as an approximate address of its location. In brief, the 4096- dimensional space is first divided into 256 regions (called "buckets") in a training step that takes in a random sample of MS/MS spectra and uses the familiar k- means clustering algorithm<sup>25</sup> to partition them into 256 groups. In doing so, the bucket boundaries are chosen such that buckets should contain similar number of spectra, assuming that the training data is representative of experimental MS/MS spectra. Then, the residual vector, the difference between the spectral vector and the centroid of the bucket it belongs to, is sub- divided into 16 vectors, and for each of them, a similar process is carried out to map it to one of 256 possible "sub- buckets." Thus, for every spectrum, its address consists of two parts: a bucket number, which denotes which of the 256 buckets it belongs to, and the 16 sub- bucket numbers of its residual vector. The beauty of this indexing approach lies in the efficient computation of distances between any two addresses (Supplementary Method). Although the address is only an approximation of the actual location of the spectral vector, this method enables the efficient retrieval of top approximate nearest neighbors (ANNs) of any query spectrum. With suitable parameters, which were optimized in this study (Supplementary Method, Supplementary Figure 1), the probability of recovering the true nearest neighbors among the ANNs is very high. It then remains to perform detailed spectrum clustering among the approximate nearest neighbors to reveal the full cluster structure in the neighborhood of the query spectrum, but since the number of retrieved ANNs is capped, the time taken for detailed clustering would not scale with the archive size.  
+
+We have tested the method on about 25 million spectra of the Human Proteome Project dataset,<sup>25</sup> and verified that the query time scales linearly with archive size (Figure 2). One individual query can be executed in less than 5 ms. The per- query search time can be further
+
+<--- Page Split --->
+
+reduced to less than 1 ms if many queries are processed in a batch, thanks to amortization of the computational overhead and further parallelization. This is achieved on our modest server featuring a single graphical processing unit (GPU). If we extrapolate the running time to 1 billion spectra in the archive, one individual query would take less than 0.1 s, still fast enough to provide a real- time interactive experience.  
+
+The success of the indexing scheme is best assessed by the recall, defined as the probability of finding the true nearest neighbors (TNNs) among the retrieved approximate nearest neighbors. By testing a random sample of 20,000 query spectra, we found the overall recall is over \(97\%\) . For \(93\%\) of the queries, perfect \((100\%)\) recall is achieved (Figure 3). Here, we define the TNNs as the N most similar spectra \((N \leq 10)\) to the query in the entire repository, with the condition that the spectral dot product is at least 0.7, a typical cutoff for a good spectral match. The overall recall would be lower at \(96\%\) (with \(80\%\) of queries achieving perfect recall) if we require the retrieval of up to the 100 most similar spectra with dot product at least 0.7. If we relax the dot product threshold to include fainter matches as TNNs, the recall is slightly reduced, while the opposite is true if we raise the threshold, indicating that the retrieval accuracy of the algorithm is higher if the neighbors are more similar. Fortunately, this limitation, shared by many ANN- based algorithms, is irrelevant for our intended application, as there is no information gained by finding such far- away neighbors, essentially random spectral matches.  
+
+Under our optimized default settings, about 2,000 distinct ANNs are retrieved for a given query by the index. For the purpose of visualization, Spectroscape re- ranks the ANNs by accurate dot products to obtain the list of TNNs. Then it performs true pairwise similarity calculations of the retained TNNs to determine the cluster structure to be displayed in a graph. This occurs in real time without any noticeable lag time. Each node of the graph is a spectrum, with the query positioned in the center, and two nodes are joined if their true similarity is above a certain user- defined threshold. Hence, groups of highly similar spectra will tend to form tightly- connected clusters. The node is color- coded based on the peptide identification of the spectrum, or colored black if it is unidentified. The edge is colored black if it connects spectra of similar precursor m/z values, and orange if the two spectra are not similar in precursor m/z. This enables the user to quickly detect discordant identifications among cluster members, as well as mass- shifted spectral
+
+<--- Page Split --->
+
+matches indicative of PTMs or sequence variants. The edges connecting two nodes with dissimilar m/z or mass value can be removed by toggle an option by the user on the web interface.  
+
+To provide a visual impression of the tightness of the cluster, a force- directed graph drawing scheme was employed (Supplementary Method), such that the edge lengths are not a simple function of the spectral similarity, but also depend on the cluster structure. (An option exists to turn off this behavior and set the edge length based solely on the similarity between the query and the neighbor.) The force- directed graph drawing is especially conducive for detecting loosely connected sub- clusters and highlighting outlying and bridge nodes. This feature can have useful applications. For one, chimeric spectra can be readily detected, as they tend to bridge two clusters of different identifications. For another, the sub- clusters can reveal subtle differences in fragmentation patterns of the same peptide ion, perhaps due to instrument differences. Some examples of how the visual cues provided by Spectroscape can help confirm identifications or lead to new discoveries are shown in Figure 4 and Supplementary Figure 2.  
+
+Another interesting feature of this platform is that consensus spectra in spectral libraries can also be loaded into the archive and be retrieved along with the experimental spectra. In effect, a conventional spectral library search is carried out as part of the query. In most cases, the consensus spectrum tends to be located near the cluster center comprising its experimental replicates, which is expected since the consensus spectrum is a weighted average of the replicates. However, a given query is often more similar to some experimental replicates than it is to the consensus. By its nature, the consensus construction process distills only the most reproducible features of the peptide fragmentation pattern, and may throw away partially reproducible features that can also be used to support some spectral matches. \(^{26}\) Occasionally, it can be seen that the consensus spectrum resides at the outer fringe of the cluster, suggesting that it may not be truly representative of the experimental replicates. Figure 4c offers such an example, in contrast to the ideal scenario shown in Supplementary Figure 2a. Therefore, with the ability to query the entire spectral repository, we envisage a more accurate and sensitive method for spectral identification that bypasses the step of model generation (i.e., construction of the consensus spectrum) from observations, but relies on matching experimental observations directly and in a contextual manner.  
+
+The web- based interface also enables the user to interact with the spectral archive in real time. For example, the user can search the archive with any displayed node as a new query by
+
+<--- Page Split --->
+
+double- clicking it, thereby hopping from node to node as he/she navigates the "neighborhood" for more insights. It also provides many options for the user to customize how the spectrum clusters are displayed, by changing the sizes and colors of the nodes and edges, and tuning the parameters of the force- directed graph. The interface also shows the full spectrum of any node in an interactive spectrum viewer, or a "butterfly" plot of the spectra of two selected nodes shown head- to- tail for comparison. A full table of nodes and edges containing their information is available in a separate tab. Some screenshots of the Spectroscape interface are shown in Supplementary Figure 3.  
+
+In terms of applications, Spectroscape is the first platform in proteomics to allow the user to assess the plausibility of any peptide- spectrum match (PSM) in the context of prior data, and do so in a visual and interactive manner. A user will be visually prompted to question a PSM that contradicts the identifications of vast majority of its nearest neighbors. This is in contrast to the current practice of relying on search engines and statistical validation tools that operate on the data at hand, which do not take advantage of the huge amount of publicly available data nor provide alternative explanations. With this platform, sequence variants and chimeric spectra can also be readily discovered, which sometimes offer better interpretations than the answers found by the search engine. Although such manual validation is impractical as a general procedure, often important biological discoveries in proteomics (e.g., novel proteoforms) hinge on a small number of PSMs. Journal editors and reviewers are tasked to assess the scientific merits of such discoveries, often with limited means to view the key spectra in question, let alone seeing them in context. Spectroscape can supplement recent efforts to facilitate and standardize data reporting at the spectrum level<sup>27</sup> and help address this unmet need. Finally, Spectroscape can eventually be deployed as an integrated platform for both data analysis and data sharing, as the query process and the spectral archive building process are algorithmically the same. Any new data submitted can be quickly indexed, and become part of the spectral archive and connected to prior knowledge. Once the indexing is complete, the data submitter can browse his/her data through Spectroscape. This type of workflow is nothing new in genomics, as sequencing data is almost always analyzed in the context of prior data thanks to convenient tools like BLAST, and through that process, any new data deposited into public databases are quickly integrated and become useful to everyone. With the concerted efforts of the proteomics community, we believe that this platform can grow and evolve to play a similar role for proteomics.
+
+<--- Page Split --->
+
+## Data availability  
+
+The data used for building the spectral archive is available from ProteomeXchange repository with identifier PXD000561. The web- based interface to explore the spectral archive built on this dataset can be accessed from http://omics.ust.hk:8709/.  
+
+## Code availability  
+
+Spectroscape is implemented in C++ and JavaScript and is freely available as open- source software under MIT license at https://github.com/wulongict/SpectralArchive/ for individual groups to build their own spectral archive. Detailed installation and running instructions are provided in the README file.
+
+<--- Page Split --->
+
+## Methods  
+
+## Datasets  
+
+The Human Proteome Project dataset, consisting of 2,210 LC- MS runs, was downloaded from ProteomeXchange (identifier PXD00056128). The raw data files were converted into mzXML format using MSConvert29. The resulting 24,988,311 MS2 spectra were searched with MSFragger (v20190628)30 with search parameter as follows. Precursor and fragment mass tolerance are 0.05Da and up to 2 missed cleavages are allowed. One fixed modification (cysteine carbamidomethylation), and 7 variable modifications (methionine oxidation, protein n- terminus acetylation, asparagine deamidation, peptide n- terminus carbamidomethylation, threonine carbamidomethylation, and tryptophan dioxidation) were considered. The sequence database of human proteome is downloaded from Uniprot31 with accession id UP000005640. Decoy database is created with in- house Perl script, where the amino acid sequence within a tryptic peptide is shuffled while keeping the K and R fixed. The search results were processed by PeptideProphet and iProphet in the Trans Proteomic Pipeline32- 34 (TPP v6.0.0 OmegaBlock). All spectra regardless of identification confidence were loaded into the spectral archive; the iProphet probability was used only to determine if the corresponding node is colored when displayed. Three consensus spectral libraries containing over 900,000 HCD spectra total from human samples compiled by the National Institute of Standards and Technology (NIST), USA, was downloaded from https://chemdata.nist.gov/dokuwiki/doku.php?id=peptidew:lib:humanhcd20160503 (dated 2020- 05- 19), converted to sptxt format by SpectraST26,35, and loaded into the archive.  
+
+## Spectrum preprocessing  
+
+All spectra are preprocessed as follows. First, spectra with number of peaks lower than a user- defined threshold (6 by default), are recorded as an empty spectrum. Next, only the top 50 most intense peaks for each spectrum are retained, and the intensity is rank- transformed to a scale of 50 (most intense), 49, 48, ..., 2, 1. To maximize the information content, isotopic peaks and peaks near the precursor m/z were excluded, and no more than 6 peaks within a sliding window of 50 m/z are kept.36 Then the peak list is vectorized by binning, with a bin width of 0.5 m/z, resulting
+
+<--- Page Split --->
+
+in a vector of 4096 dimensions, the required input to the IVF- PQ algorithm. In the binning step, the two flanking bins of each peak are filled with half of the rank- transform intensity.  
+
+## Training and optimizing the index  
+
+The FAISS IVF- PQ index needs to be initialized in a training step. A randomly selected set of 100,000 spectra was used to find an optimal partition of the high- dimensional space that distribute the spectra among "buckets" and "sub- buckets." The details of the algorithm were described in the article "Billion- scale similarity search with GPUs,"<sup>22</sup> and supplemented in Supplementary Method. The process takes about 15 minutes on our server. This process can be repeated using a different set of training spectra and different product quantization schemes<sup>37</sup> to produce distinct indices. At the expense of running time, combined use of multiple indices can improve recall, because pairs of similar spectra that straddle a bucket boundary of one index are unlikely to do the same in another index. Thus the number of indices used (nindices) is a parameter that can be chosen to strike a balance between recall and running time. Moreover, at the time of query, one can limit the number of closest buckets (nprobes) to search for approximate nearest neighbors, which is another parameter that can be chosen. After optimization, we chose the parameter settings of nindices = 2 and nprobes = 8 for the rest of the study (Supplementary Method, Supplementary Figure 1), though the latter can be changed at query time by the user in the web interface of Spectroscape.  
+
+## Adding data to the spectral archive  
+
+Tandem mass spectra in MS data files (in mzXML, mzML, mgf or sptxt formats) are first preprocessed as described above, and the 17- byte address of each spectrum is computed by the IVF- PQ algorithm for each index. Meanwhile, to enable the subsequent clustering and visualization steps, the preprocessed spectrum is stored in a memory- efficient format in a separate file (called the "MZ" file) whereby the m/z values of the retained 50 peaks were converted to a two- byte integer and listed in descending order of intensity, resulting in a 100- byte representation. Other metadata of the spectrum such as the source data file name and scan number, its precursor m/z value, and its peptide identification (if any, loaded separately from the corresponding pepXML<sup>38</sup> or comma- delimited file containing search engine results) is stored in an SQLite
+
+<--- Page Split --->
+
+database<sup>39</sup> in a bookkeeping step (Figure 1a). The corresponding entries in the indices, the MZ file and the SQLite database are linked by a unique spectrum ID.  
+
+## Retrieval by spectral similarity  
+
+A query request can be sent through a web browser to the server, which executes the query in a cgi script written in \(\mathrm{C + + }\) . The user can either use one spectrum already loaded into the spectral archive as the query (by providing the unique spectrum ID, or by searching by the source data file name and the scan number, or the peptide identification), or upload a new query spectrum to the server through the web service. Searching a batch of queries is also possible on the server side via command- line scripts.  
+
+Given a new query spectrum, the same preprocessing steps are taken and its address (one per index) is computed. For each index, the IVF- PQ algorithm then retrieves all the existing addresses in the closest nprobes buckets, and computes the approximate dot product between each address to the query's address (Supplementary Method). The top 1024 most similar addresses are returned as the "approximate nearest neighbors" (ANNs) of the query. The ANNs returned by all indices are combined, and the corresponding preprocessed spectra were read from the MZ files. Then, the accurate dot product is calculated between the full query spectrum and the full spectra of the ANNs to determine the "true nearest neighbors" (TNNs).  
+
+## Clustering and visualization  
+
+For visualization, the N most similar TNNs (N defaults to 20 but can be chosen by the user) can be shown in a graph in Spectroscape, with each spectrum shown as a node. For these N spectra, pairwise accurate dot products are computed to get a fully connected graph. Edges were considered for visualization on web browser if their Euclidian distance exceed a user- defined threshold. The graph is then plotted by the web browser using a force- directed graph drawing algorithm implemented in a free JavaScript library D3.js (https://d3js.org/). In brief, this graph visualization simulates the physical behavior of a network of electrically charged spheres (the nodes) connected by springs (the edges), with several parameters controllable by the user (Supplementary Method). The query spectrum is displayed in the center of the graph as a larger circle to distinguish it from its neighbors. The color of the node is calculated based on the peptide sequence using a hash
+
+<--- Page Split --->
+
+function to map a string to a RGB color. The color black was reserved to indicate an unidentified spectrum. Optionally, edges between spectra of different precursor \(\mathrm{m / z}\) values can be shown as a different color, to aid the discovery of potential PTMs and variants. Information about the spectra and their computed similarities are shown if the mouse pointer hovers over nodes and edges. Single clicking on any node displays the full spectrum by the Lorikeet spectrum viewer. If the spectrum is identified by search engine with FDR \(< 1\%\) , it will be annotated with expected ions. To help the user evaluate alternative explanations for the query node, the user can also enter another peptide identification, upon which the annotation will be updated. Double clicking on any node will start a new query with the clicked node as the query spectrum.  
+
+## Running time evaluation  
+
+Spectroscape was tested on a server equipped with an AMD Ryzen Threadripper PRO 3975WX CPU with 32 cores at 3.5 GHz, one NVIDIA RTX3080Ti 12 GB GPU, and 512 GB of DDR4 3200 MHz RAM. Training 6 IVF- PQ indices using 100,000 spectra took about 15 minutes in total. Adding one spectrum to the archive, including all the preprocessing, indexing and bookkeeping steps, took less than 0.3 ms, averaged over 6 indices and the nearly 25 million spectra. To estimate retrieval time, different query batch sizes (1, 10, 100, 1000 and 10000) were tested; a random sample of spectra already in the spectral archive are submitted for query to evaluate the effect of parallelization. The running times for various steps done on the server were recorded by the commands in the \(\mathrm{C + + }\) standard library chrono, and averaged over many executions. The time taken for sending query requests over the internet, or for plotting the graph in the web browser is not estimated given the inherent variability of such tasks.  
+
+## Recall evaluation  
+
+To evaluate the retrieval performance of the IVF- PQ algorithm, we measure the recall \(R_{i}\) of a given query \(\pmb{q}_{i}\) by:  
+
+\[R_{i}(t) = \frac{|A(\pmb{q}_{i})\cap T(\pmb{q}_{i},N,t)|}{|T(\pmb{q}_{i},N,t)|}\]  
+
+where \(A(\pmb{q}_{i})\) and \(T(\pmb{q}_{i},N,t)\) is the sets of retrieved ANNs of \(\pmb{q}_{i}\) and its TNNs, respectively, and \(t\) is the minimum dot product threshold to retain as a TNN. Here, the set of a TNNs of a given query
+
+<--- Page Split --->
+
+\(T(\pmb {q}_i,N,t)\) is obtained by comparing the query spectrum to every spectrum in the archive without any index, and retaining up to \(N\) nearest neighbors whose accurate dot products with the query are above a certain cut- off \(t\) . The recall \(R_{i}\) of 20,000 randomly selected queries were plotted in two histograms corresponding to top 10 and top 100 TNNs (Figure 3a). The dot product cut- off was varied between 0.5 and 0.9, to evaluate the impact of true spectral similarity on the recall, as it is expected that the ANN- based similarity retrieval should be more effective as the neighbors are closer in reality. The approximate dot products and the true dot products of all retrieved ANNs of one typical query are plotted in a scatterplot (Figure 3b). The overall recall is defined over all the 20,000 queries as follows.  
+
+\[R_{overall} = \frac{\sum_{i = 1}^{20000}|A(\pmb{q}_i)\cap T(\pmb{q}_i,N,t)|}{\sum_{i = 1}^{20000}|T(\pmb{q}_i,N,t)|}\]
+
+<--- Page Split --->
+
+## Reference
+
+<--- Page Split --->
+
+13. Griss, J. et al. Recognizing millions of consistently unidentified spectra across hundreds of shotgun proteomics datasets. Nat Methods 13, 651–656 (2016).  
+
+14. Guthals, A., Watrous, J. D., Dorrestein, P. C. & Bandeira, N. The spectral networks paradigm in high throughput mass spectrometry. Mol Biosyst 8, 2535 (2012).  
+
+15. The, M. & Käll, L. MaRaCluster: A Fragment Rarity Metric for Clustering Fragment Spectra in Shotgun Proteomics. J Proteome Res 15, 713–720 (2016).  
+
+16. To, P. K. P., Wu, L., Chan, C. M., Hoque, A. & Lam, H. ClusterSheep: A Graphics Processing Unit-Accelerated Software Tool for Large-Scale Clustering of Tandem Mass Spectra from Shotgun Proteomics. J Proteome Res 20, 5359–5367 (2021).  
+
+17. Bittremieux, W., Meysman, P., Noble, W. S. & Laukens, K. Fast Open Modification Spectral Library Searching through Approximate Nearest Neighbor Indexing. J Proteome Res 17, 3463–3474 (2018).  
+
+18. Bittremieux, W., Laukens, K. & Noble, W. S. Extremely Fast and Accurate Open Modification Spectral Library Searching of High-Resolution Mass Spectra Using Feature Hashing and Graphics Processing Units. J Proteome Res 18, 3792–3799 (2019).  
+
+19. Bittremieux, W., May, D. H., Bilmes, J. & Noble, W. S. A learned embedding for efficient joint analysis of millions of mass spectra. Nat Methods 19, 675–678 (2022).  
+
+20. Wang, L., Li, S. & Tang, H. msCRUSH: Fast Tandem Mass Spectral Clustering Using Locality Sensitive Hashing. J Proteome Res acs.jproteome.8b00448 (2018) doi:10.1021/acs.jproteome.8b00448.  
+
+21. Bittremieux, W., Laukens, K., Noble, W. S. & Dorrestein, P. C. Large-scale tandem mass spectrum clustering using fast nearest neighbor searching. Rapid Communications in Mass Spectrometry (2021) doi:10.1002/rcm.9153.  
+
+22. Johnson, J., Douze, M. & Jégou, H. Billion-scale similarity search with GPUs. CoRR abs/1702.08734, (2017).  
+
+23. Ma, C. W. M. & Lam, H. Hunting for Unexpected Post-Translational Modifications by Spectral Library Searching with Tier-Wise Scoring. J Proteome Res 13, 2262–2271 (2014).
+
+<--- Page Split --->
+
+24. Chick, J. M. et al. A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides. Nat Biotechnol 33, 743–749 (2015).  
+
+25. MacQueen, J. Some methods for classification and analysis of multivariate observations. in Berkeley Symposium on Mathematical Statistics and Probability Vol. 5.1, 281-297 (1967) 281–297 (1967).  
+
+26. Lam, H. et al. Development and validation of a spectral library searching method for peptide identification from MS/MS. Proteomics 7, 655–667 (2007).  
+
+27. Deutsch, E. W. et al. Universal Spectrum Identifier for mass spectra. Nat Methods 18, 768–770 (2021).  
+
+28. Kim, M.-S. et al. A draft map of the human proteome. Nature 509, 575–581 (2014).  
+
+29. Adusumilli, R. & Mallick, P. Data Conversion with ProteoWizard msConvert. in 339–368 (2017). doi:10.1007/978-1-4939-6747-6_23.  
+
+30. Kong, A. T., Leprevost, F. v, Avtonomov, D. M., Mellacheruvu, D. & Nesvizhskii, A. I. MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry–based proteomics. Nat Methods 14, 513–520 (2017).  
+
+31. Bateman, A. et al. UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res 49, D480–D489 (2021).  
+
+32. Ma, K., Vitek, O. & Nesvizhskii, A. I. A statistical model-building perspective to identification of MS/MS spectra with PeptideProphet. BMC Bioinformatics 13, S1 (2012).  
+
+33. Shteynberg, D. et al. iProphet: Multi-level Integrative Analysis of Shotgun Proteomic Data Improves Peptide and Protein Identification Rates and Error Estimates. Molecular & Cellular Proteomics 10, M111.007690 (2011).  
+
+34. Deutsch, E. W. et al. Trans-Proteomic Pipeline, a standardized data processing pipeline for large-scale reproducible proteomics informatics. Proteomics Clin Appl 9, 745–754 (2015).  
+
+35. Lam, H. et al. Building consensus spectral libraries for peptide identification in proteomics. Nat Methods 5, 873–875 (2008).
+
+<--- Page Split --->
+
+36. Shao, W., Zhu, K. & Lam, H. Refining similarity scoring to enable decoy-free validation in spectral library searching. \*Proteomics\* 13, 3273–3283 (2013).  
+
+37. Ge, T., He, K., Ke, Q. & Sun, J. Optimized Product Quantization. \*IEEE Trans Pattern Anal Mach Intell\* 36, 744–755 (2014).  
+
+38. Hoopmann, M. R., Mendoza, L., Deutsch, E. W., Shteynberg, D. & Moritz, R. L. An Open Data Format for Visualization and Analysis of Cross-Linked Mass Spectrometry Results. \*J Am Soc Mass Spectrom\* 27, 1728–1734 (2016).  
+
+39. Hipp, R. D. SQLite. Preprint at https://www.sqlite.org/index.html (2020).
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+<center>Figures </center>  
+
+<center>Figure 1 | Building and searching of spectral archive by Spectroscape. In the building step, the indices are built by training on 100,000 spectra. The new data files can be added to the indices after preprocessing. The preprocessed spectra are also stored in an MZ file. An SQL database is created to keep all the meta information of each spectrum, including identifications (if any). In the searching step, ANNs are retrieved from indices for each preprocessed query spectrum. Those ANNs are re-ranked by the accurate dot product to retain up to 100 true nearest neighbors (TNNs) for clustering, which is visualized as a force-directed graph. </center>
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2 | The average time for retrieval of ANNs for each query spectrum on one IVF-PQ index, as a function of the number of spectra in the archive, in millions, if one query (blue), 10 queries (orange), 100 queries (green), 1000 queries (gold) and 10000 queries (purple), respectively, were executed in a batch. As shown, the per-query retrieval time scales linearly with archive size. </center>
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3 | Recall of TNNs by Spectroscape. (a) Histograms of per-query recall of the TNNs, tested with 20,000 randomly selected queries from the archive. TNNs are defined by two parameters: N, the number of closest neighbors to be retained as TNNs (N = 10 and 100), and t, the minimum dot product to be considered a TNN (t = 0.5 to 0.9). On both left (N = 10) and right panels (N = 100), the per-query recall distributions are dominated by the rightmost column, representing the fraction </center>
+
+<--- Page Split --->
+
+of queries achieving perfect recall. In general, Spectroscape achieves higher per- query recall if one requires higher minimum dot product similarity to be counted as TNNs, or if one considers fewer TNNs. (b) A scatterplot of true dot product versus the approximate dot product as computed by the IVF- PQ algorithm; each data point represents one retrieved ANN of one of 20,000 random queries. About 600,000 retrieved ANNs in total were plotted. The color indicates density of data points by kernel density estimation. The overall correlation coefficient between the approximated dot product and the true dot product is 0.857.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+  
+
+![PLACEHOLDER_22_1]
+
+
+<--- Page Split --->
+![PLACEHOLDER_23_0]
+
+<center>Figure 4 | Examples of spectrum clusters visualized by Spectroscape. (a) Identification of unidentified spectra by association with neighbors with a precursor mass shift. The query node (PSM shown in top left panel) is unidentified but its nearest neighbors, with a precursor mass shift of 230.04 Da, are identified as DDPLTNLTFAFVAEK/2. The mass shift equals to the mass of the oligopeptide DD. Therefore, the unidentified query node and all its black neighbors are likely the semi-tryptic peptide PLTNLTFAFVAEK/2, which can also be confirmed by the fact that a library spectrum of that peptide ion is located in the middle of that cluster. The bottom left panel shows a very good PSM when we re-annotate the query node with this semi-tryptic peptide. (b) A query spectrum as a bridge of two clusters of PSMs corresponding to two peptides, LGPAIATGNVVVMK/2 and HVNPVQALSEFK/2. It is clearly a chimeric spectrum as the two PSMs almost do not share any matched peaks, as shown by the same spectrum annotated with respect to the two peptide ions (left panel). (c) The same peptide ion NQPGNTLTEILETPATAQQEVDHATDM[147]VSR/3 yields two slightly different fragmentation patterns on two mass spectrometers, as can be visualized clearly as two subclusters. The NIST HCD library spectrum (top left), represented as the node with a dotted outline that lies in the middle of the two subclusters, does not capture either of the fragmentation patterns very well. </center>
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- SpectralArchiveVisualizationSUPPINFOv5.1.docx
+
+<--- Page Split --->

preprint/preprint__990b6b7a4019afa38768bea301e6129480c22c61c21e3e2575ba8092b42152e6/preprint__990b6b7a4019afa38768bea301e6129480c22c61c21e3e2575ba8092b42152e6_det.mmd

@@ -0,0 +1,323 @@
+<|ref|>title<|/ref|><|det|>[[44, 107, 940, 208]]<|/det|>
+# Spectroscape: Real-time Query and Visualization of a Spectral Archive of Millions of Tandem Mass Spectra  
+
+<|ref|>text<|/ref|><|det|>[[44, 228, 883, 365]]<|/det|>
+Henry Lam ( \(\boxed{\bullet}\) kehlam@ust.hk) The Hong Kong University of Science and Technology https://orcid.org/0000- 0001- 7928- 0364 Long Wu The Hong Kong University of Science and Technology https://orcid.org/0000- 0002- 3566- 9818 Ayman Hoque The Hong Kong University of Science and Technology  
+
+<|ref|>sub_title<|/ref|><|det|>[[45, 404, 230, 423]]<|/det|>
+## Brief Communication  
+
+<|ref|>text<|/ref|><|det|>[[45, 443, 800, 463]]<|/det|>
+Keywords: Proteomics, mass spectrometry, spectral archive, big data, data visualization  
+
+<|ref|>text<|/ref|><|det|>[[45, 481, 327, 500]]<|/det|>
+Posted Date: October 25th, 2022  
+
+<|ref|>text<|/ref|><|det|>[[44, 519, 474, 538]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 2166780/v1  
+
+<|ref|>text<|/ref|><|det|>[[44, 556, 910, 599]]<|/det|>
+License: \(\circledcirc\) This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+<|ref|>title<|/ref|><|det|>[[195, 90, 837, 157]]<|/det|>
+# Spectroscape: Real-time Query and Visualization of a Spectral Archive of Millions of Tandem Mass Spectra  
+
+<|ref|>text<|/ref|><|det|>[[325, 205, 669, 225]]<|/det|>
+Long \(\mathrm{Wu}^{1,2}\) , Ayman Hoque<sup>1</sup>, Henry Lam<sup>1,\*</sup>  
+
+<|ref|>text<|/ref|><|det|>[[113, 270, 880, 370]]<|/det|>
+<sup>1</sup> Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology  <sup>2</sup> Department of Electrical and Computer Engineering, The Hong Kong University of Science and Technology  
+
+<|ref|>text<|/ref|><|det|>[[113, 480, 552, 682]]<|/det|>
+\* Corresponding author  Henry Lam  Department of Chemical and Biological Engineering  The Hong Kong University of Science and Technology  Clear Water Bay  Hong Kong Special Administrative Region  China  Email: kehlam@ust.hk  
+
+<|ref|>text<|/ref|><|det|>[[113, 716, 806, 736]]<|/det|>
+Keywords: Proteomics, mass spectrometry, spectral archive, big data, data visualization
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 196, 110]]<|/det|>
+## Abstract  
+
+<|ref|>text<|/ref|><|det|>[[113, 126, 886, 330]]<|/det|>
+In proteomics, spectral archives organize the enormous amounts of publicly available peptide tandem mass spectra by similarity, offering opportunities for error correction and novel discoveries. Here we adapted an indexing algorithm developed by Facebook for organizing online multimedia resources to tandem mass spectra, and achieved practically instantaneous retrieval and spectrum clustering of approximate nearest neighbors in a large spectral archive. An interactive web- based graphical user interface enables the user to view a query spectrum in its clustered "neighborhood," which facilitate contextual validation of peptide identifications and exploration of the dark proteome.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 378, 170, 397]]<|/det|>
+## Main  
+
+<|ref|>text<|/ref|><|det|>[[113, 408, 885, 794]]<|/det|>
+The tandem mass spectrum is the basic unit of proteomic data, each representing a direct experimental observation of a peptide, and by inference, the parent protein. A modern mass spectrometer can easily produce tens of thousands of MS/MS spectra every hour. Such rapid data generation, when combined with the increasing prevalence of these machines, has led to an astronomical amount of data. In ProteomeXchange, the leading data repository in proteomics, over 5,300 datasets are deposited in 2021 alone, with the number of datasets deposited steadily increasing every year. \(^{1 - 3}\) Within these datasets, the number of MS/MS spectra currently available in the public domain is on the order of tens of billions. \(^{2,3}\) Moreover, this data volume probably represents only a small fraction of all proteomic data generated, as typically only data- sets directly associated with publications are deposited. Currently, proteomics data are largely stored in data repositories, which organize the data by data- sets associated with individual publications or projects. \(^{4 - 7}\) In effect, the spectra are stored in the same unprocessed data files and organized in the same structure as when they were submitted, usually as supporting data of a publication. As many have argued, while this is an important first step, such organization is not conducive to meaningful data reuse. \(^{8,9}\)  
+
+<|ref|>text<|/ref|><|det|>[[114, 813, 883, 886]]<|/det|>
+The emerging paradigm explored by several proteomics databases is to organize data by spectral similarity, in what is called a spectral archive. \(^{10,11}\) A spectral archive preserves all individual spectra, identified or not, and spectra similar to each other are grouped in a process
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 87, 884, 449]]<|/det|>
+called spectrum clustering. In theory, this organization should maximize the potential for data reuse. For instance, by analyzing clusters of similar spectra, the unidentified ones among them can become identified through association, errors in identifications can be corrected, outliers can be detected, and cross- references across data- sets can be made. Therefore, once in a spectral archive, the identification of a given spectrum can evolve and hopefully self- correct over time. \(^{11 - 13}\) Another application of spectral archives, which was demonstrated previously, was to discover unexpected post- translational modifications or amino acid substitutions by detecting mass shifts between connected members in a cluster. \(^{14}\) However, it is computationally challenging to maintain large spectral archives, since in principle each spectrum needs to be compared to every other spectrum, though in practice the problem is made more tractable by various shortcuts. Some common strategies are: (i) progressively reducing the search space by merging spectra immediately in newly found clusters, (ii) limit comparisons to spectra with similar precursor m/z values, and (iii) avoiding comparisons of dissimilar spectra by grouping similar spectra first using various dimensionality reduction schemes. \(^{10,13,15 - 21}\)  
+
+<|ref|>text<|/ref|><|det|>[[113, 467, 884, 880]]<|/det|>
+Despite the obvious promise of spectral archives, the adoption of the idea by the wider proteomics community has been slow. In our view, the main reason is that spectrum clustering is computationally intensive, and does not seem accessible to the typical proteomics researcher. Moreover, the outcome of the spectrum clustering is not visible to the human user, and must be further analyzed by other computational tools, as existing tools lack a graphical user interface that allow the user to interact with the spectral archive. To break this inertia, we developed a platform called Spectroscape, that enables the real- time query and visualization of a spectral archive (Figure 1). Unlike other spectrum clustering tools that focus on pre- building the spectral archive off- line, our method enables query- driven real- time clustering, providing the detailed cluster structures with connectivity information. This is made possible by an indexing scheme based on the inverted file and product quantization encoding (IVF- PQ) algorithm in the Facebook AI Similarity Search (FAISS) \(^{22}\) library that groups spectra in “neighborhoods” in high- dimensional space, defined by approximate spectral similarity. Given any query spectrum that the user supplies, the method efficiently retrieves all its approximate nearest neighbors in the entire repository and performs real- time spectrum clustering by computing accurate pairwise distances among the query and the neighbors to reveal any cluster(s) in the neighborhood. The user can then visualize the
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 88, 884, 240]]<|/det|>
+result in an interactive web- based user interface. In effect, the platform allows the user to search the entire data repository by spectral similarity, and instantaneously receive not only a list of best- matching spectra to the query, but also the detailed structure of any cluster in its neighborhood. Unlike other clustering algorithms, Spectroscape is fast enough even without limiting comparisons by using the precursor m/z value, making it suitable for discovering unexpected post- translational modifications or sequence variants<sup>23,24</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[113, 256, 885, 802]]<|/det|>
+In Spectroscape, a spectrum is represented by a vector in 4096- dimensional space. The chosen variant of the IVF- PQ algorithm converts such a vector into a 17- byte string that serves as an approximate address of its location. In brief, the 4096- dimensional space is first divided into 256 regions (called "buckets") in a training step that takes in a random sample of MS/MS spectra and uses the familiar k- means clustering algorithm<sup>25</sup> to partition them into 256 groups. In doing so, the bucket boundaries are chosen such that buckets should contain similar number of spectra, assuming that the training data is representative of experimental MS/MS spectra. Then, the residual vector, the difference between the spectral vector and the centroid of the bucket it belongs to, is sub- divided into 16 vectors, and for each of them, a similar process is carried out to map it to one of 256 possible "sub- buckets." Thus, for every spectrum, its address consists of two parts: a bucket number, which denotes which of the 256 buckets it belongs to, and the 16 sub- bucket numbers of its residual vector. The beauty of this indexing approach lies in the efficient computation of distances between any two addresses (Supplementary Method). Although the address is only an approximation of the actual location of the spectral vector, this method enables the efficient retrieval of top approximate nearest neighbors (ANNs) of any query spectrum. With suitable parameters, which were optimized in this study (Supplementary Method, Supplementary Figure 1), the probability of recovering the true nearest neighbors among the ANNs is very high. It then remains to perform detailed spectrum clustering among the approximate nearest neighbors to reveal the full cluster structure in the neighborhood of the query spectrum, but since the number of retrieved ANNs is capped, the time taken for detailed clustering would not scale with the archive size.  
+
+<|ref|>text<|/ref|><|det|>[[114, 819, 883, 891]]<|/det|>
+We have tested the method on about 25 million spectra of the Human Proteome Project dataset,<sup>25</sup> and verified that the query time scales linearly with archive size (Figure 2). One individual query can be executed in less than 5 ms. The per- query search time can be further
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 88, 884, 213]]<|/det|>
+reduced to less than 1 ms if many queries are processed in a batch, thanks to amortization of the computational overhead and further parallelization. This is achieved on our modest server featuring a single graphical processing unit (GPU). If we extrapolate the running time to 1 billion spectra in the archive, one individual query would take less than 0.1 s, still fast enough to provide a real- time interactive experience.  
+
+<|ref|>text<|/ref|><|det|>[[113, 231, 885, 566]]<|/det|>
+The success of the indexing scheme is best assessed by the recall, defined as the probability of finding the true nearest neighbors (TNNs) among the retrieved approximate nearest neighbors. By testing a random sample of 20,000 query spectra, we found the overall recall is over \(97\%\) . For \(93\%\) of the queries, perfect \((100\%)\) recall is achieved (Figure 3). Here, we define the TNNs as the N most similar spectra \((N \leq 10)\) to the query in the entire repository, with the condition that the spectral dot product is at least 0.7, a typical cutoff for a good spectral match. The overall recall would be lower at \(96\%\) (with \(80\%\) of queries achieving perfect recall) if we require the retrieval of up to the 100 most similar spectra with dot product at least 0.7. If we relax the dot product threshold to include fainter matches as TNNs, the recall is slightly reduced, while the opposite is true if we raise the threshold, indicating that the retrieval accuracy of the algorithm is higher if the neighbors are more similar. Fortunately, this limitation, shared by many ANN- based algorithms, is irrelevant for our intended application, as there is no information gained by finding such far- away neighbors, essentially random spectral matches.  
+
+<|ref|>text<|/ref|><|det|>[[113, 584, 885, 865]]<|/det|>
+Under our optimized default settings, about 2,000 distinct ANNs are retrieved for a given query by the index. For the purpose of visualization, Spectroscape re- ranks the ANNs by accurate dot products to obtain the list of TNNs. Then it performs true pairwise similarity calculations of the retained TNNs to determine the cluster structure to be displayed in a graph. This occurs in real time without any noticeable lag time. Each node of the graph is a spectrum, with the query positioned in the center, and two nodes are joined if their true similarity is above a certain user- defined threshold. Hence, groups of highly similar spectra will tend to form tightly- connected clusters. The node is color- coded based on the peptide identification of the spectrum, or colored black if it is unidentified. The edge is colored black if it connects spectra of similar precursor m/z values, and orange if the two spectra are not similar in precursor m/z. This enables the user to quickly detect discordant identifications among cluster members, as well as mass- shifted spectral
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 88, 883, 135]]<|/det|>
+matches indicative of PTMs or sequence variants. The edges connecting two nodes with dissimilar m/z or mass value can be removed by toggle an option by the user on the web interface.  
+
+<|ref|>text<|/ref|><|det|>[[113, 153, 884, 435]]<|/det|>
+To provide a visual impression of the tightness of the cluster, a force- directed graph drawing scheme was employed (Supplementary Method), such that the edge lengths are not a simple function of the spectral similarity, but also depend on the cluster structure. (An option exists to turn off this behavior and set the edge length based solely on the similarity between the query and the neighbor.) The force- directed graph drawing is especially conducive for detecting loosely connected sub- clusters and highlighting outlying and bridge nodes. This feature can have useful applications. For one, chimeric spectra can be readily detected, as they tend to bridge two clusters of different identifications. For another, the sub- clusters can reveal subtle differences in fragmentation patterns of the same peptide ion, perhaps due to instrument differences. Some examples of how the visual cues provided by Spectroscape can help confirm identifications or lead to new discoveries are shown in Figure 4 and Supplementary Figure 2.  
+
+<|ref|>text<|/ref|><|det|>[[113, 454, 886, 840]]<|/det|>
+Another interesting feature of this platform is that consensus spectra in spectral libraries can also be loaded into the archive and be retrieved along with the experimental spectra. In effect, a conventional spectral library search is carried out as part of the query. In most cases, the consensus spectrum tends to be located near the cluster center comprising its experimental replicates, which is expected since the consensus spectrum is a weighted average of the replicates. However, a given query is often more similar to some experimental replicates than it is to the consensus. By its nature, the consensus construction process distills only the most reproducible features of the peptide fragmentation pattern, and may throw away partially reproducible features that can also be used to support some spectral matches. \(^{26}\) Occasionally, it can be seen that the consensus spectrum resides at the outer fringe of the cluster, suggesting that it may not be truly representative of the experimental replicates. Figure 4c offers such an example, in contrast to the ideal scenario shown in Supplementary Figure 2a. Therefore, with the ability to query the entire spectral repository, we envisage a more accurate and sensitive method for spectral identification that bypasses the step of model generation (i.e., construction of the consensus spectrum) from observations, but relies on matching experimental observations directly and in a contextual manner.  
+
+<|ref|>text<|/ref|><|det|>[[114, 858, 883, 905]]<|/det|>
+The web- based interface also enables the user to interact with the spectral archive in real time. For example, the user can search the archive with any displayed node as a new query by
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 88, 884, 266]]<|/det|>
+double- clicking it, thereby hopping from node to node as he/she navigates the "neighborhood" for more insights. It also provides many options for the user to customize how the spectrum clusters are displayed, by changing the sizes and colors of the nodes and edges, and tuning the parameters of the force- directed graph. The interface also shows the full spectrum of any node in an interactive spectrum viewer, or a "butterfly" plot of the spectra of two selected nodes shown head- to- tail for comparison. A full table of nodes and edges containing their information is available in a separate tab. Some screenshots of the Spectroscape interface are shown in Supplementary Figure 3.  
+
+<|ref|>text<|/ref|><|det|>[[112, 281, 886, 881]]<|/det|>
+In terms of applications, Spectroscape is the first platform in proteomics to allow the user to assess the plausibility of any peptide- spectrum match (PSM) in the context of prior data, and do so in a visual and interactive manner. A user will be visually prompted to question a PSM that contradicts the identifications of vast majority of its nearest neighbors. This is in contrast to the current practice of relying on search engines and statistical validation tools that operate on the data at hand, which do not take advantage of the huge amount of publicly available data nor provide alternative explanations. With this platform, sequence variants and chimeric spectra can also be readily discovered, which sometimes offer better interpretations than the answers found by the search engine. Although such manual validation is impractical as a general procedure, often important biological discoveries in proteomics (e.g., novel proteoforms) hinge on a small number of PSMs. Journal editors and reviewers are tasked to assess the scientific merits of such discoveries, often with limited means to view the key spectra in question, let alone seeing them in context. Spectroscape can supplement recent efforts to facilitate and standardize data reporting at the spectrum level<sup>27</sup> and help address this unmet need. Finally, Spectroscape can eventually be deployed as an integrated platform for both data analysis and data sharing, as the query process and the spectral archive building process are algorithmically the same. Any new data submitted can be quickly indexed, and become part of the spectral archive and connected to prior knowledge. Once the indexing is complete, the data submitter can browse his/her data through Spectroscape. This type of workflow is nothing new in genomics, as sequencing data is almost always analyzed in the context of prior data thanks to convenient tools like BLAST, and through that process, any new data deposited into public databases are quickly integrated and become useful to everyone. With the concerted efforts of the proteomics community, we believe that this platform can grow and evolve to play a similar role for proteomics.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 168, 248, 186]]<|/det|>
+## Data availability  
+
+<|ref|>text<|/ref|><|det|>[[115, 205, 883, 277]]<|/det|>
+The data used for building the spectral archive is available from ProteomeXchange repository with identifier PXD000561. The web- based interface to explore the spectral archive built on this dataset can be accessed from http://omics.ust.hk:8709/.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 300, 252, 318]]<|/det|>
+## Code availability  
+
+<|ref|>text<|/ref|><|det|>[[114, 339, 884, 437]]<|/det|>
+Spectroscape is implemented in C++ and JavaScript and is freely available as open- source software under MIT license at https://github.com/wulongict/SpectralArchive/ for individual groups to build their own spectral archive. Detailed installation and running instructions are provided in the README file.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 203, 110]]<|/det|>
+## Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 155, 184, 172]]<|/det|>
+## Datasets  
+
+<|ref|>text<|/ref|><|det|>[[113, 194, 885, 660]]<|/det|>
+The Human Proteome Project dataset, consisting of 2,210 LC- MS runs, was downloaded from ProteomeXchange (identifier PXD00056128). The raw data files were converted into mzXML format using MSConvert29. The resulting 24,988,311 MS2 spectra were searched with MSFragger (v20190628)30 with search parameter as follows. Precursor and fragment mass tolerance are 0.05Da and up to 2 missed cleavages are allowed. One fixed modification (cysteine carbamidomethylation), and 7 variable modifications (methionine oxidation, protein n- terminus acetylation, asparagine deamidation, peptide n- terminus carbamidomethylation, threonine carbamidomethylation, and tryptophan dioxidation) were considered. The sequence database of human proteome is downloaded from Uniprot31 with accession id UP000005640. Decoy database is created with in- house Perl script, where the amino acid sequence within a tryptic peptide is shuffled while keeping the K and R fixed. The search results were processed by PeptideProphet and iProphet in the Trans Proteomic Pipeline32- 34 (TPP v6.0.0 OmegaBlock). All spectra regardless of identification confidence were loaded into the spectral archive; the iProphet probability was used only to determine if the corresponding node is colored when displayed. Three consensus spectral libraries containing over 900,000 HCD spectra total from human samples compiled by the National Institute of Standards and Technology (NIST), USA, was downloaded from https://chemdata.nist.gov/dokuwiki/doku.php?id=peptidew:lib:humanhcd20160503 (dated 2020- 05- 19), converted to sptxt format by SpectraST26,35, and loaded into the archive.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 681, 307, 700]]<|/det|>
+## Spectrum preprocessing  
+
+<|ref|>text<|/ref|><|det|>[[114, 720, 884, 871]]<|/det|>
+All spectra are preprocessed as follows. First, spectra with number of peaks lower than a user- defined threshold (6 by default), are recorded as an empty spectrum. Next, only the top 50 most intense peaks for each spectrum are retained, and the intensity is rank- transformed to a scale of 50 (most intense), 49, 48, ..., 2, 1. To maximize the information content, isotopic peaks and peaks near the precursor m/z were excluded, and no more than 6 peaks within a sliding window of 50 m/z are kept.36 Then the peak list is vectorized by binning, with a bin width of 0.5 m/z, resulting
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 88, 883, 135]]<|/det|>
+in a vector of 4096 dimensions, the required input to the IVF- PQ algorithm. In the binning step, the two flanking bins of each peak are filled with half of the rank- transform intensity.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 156, 386, 176]]<|/det|>
+## Training and optimizing the index  
+
+<|ref|>text<|/ref|><|det|>[[113, 197, 885, 584]]<|/det|>
+The FAISS IVF- PQ index needs to be initialized in a training step. A randomly selected set of 100,000 spectra was used to find an optimal partition of the high- dimensional space that distribute the spectra among "buckets" and "sub- buckets." The details of the algorithm were described in the article "Billion- scale similarity search with GPUs,"<sup>22</sup> and supplemented in Supplementary Method. The process takes about 15 minutes on our server. This process can be repeated using a different set of training spectra and different product quantization schemes<sup>37</sup> to produce distinct indices. At the expense of running time, combined use of multiple indices can improve recall, because pairs of similar spectra that straddle a bucket boundary of one index are unlikely to do the same in another index. Thus the number of indices used (nindices) is a parameter that can be chosen to strike a balance between recall and running time. Moreover, at the time of query, one can limit the number of closest buckets (nprobes) to search for approximate nearest neighbors, which is another parameter that can be chosen. After optimization, we chose the parameter settings of nindices = 2 and nprobes = 8 for the rest of the study (Supplementary Method, Supplementary Figure 1), though the latter can be changed at query time by the user in the web interface of Spectroscape.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 605, 392, 623]]<|/det|>
+## Adding data to the spectral archive  
+
+<|ref|>text<|/ref|><|det|>[[113, 645, 885, 875]]<|/det|>
+Tandem mass spectra in MS data files (in mzXML, mzML, mgf or sptxt formats) are first preprocessed as described above, and the 17- byte address of each spectrum is computed by the IVF- PQ algorithm for each index. Meanwhile, to enable the subsequent clustering and visualization steps, the preprocessed spectrum is stored in a memory- efficient format in a separate file (called the "MZ" file) whereby the m/z values of the retained 50 peaks were converted to a two- byte integer and listed in descending order of intensity, resulting in a 100- byte representation. Other metadata of the spectrum such as the source data file name and scan number, its precursor m/z value, and its peptide identification (if any, loaded separately from the corresponding pepXML<sup>38</sup> or comma- delimited file containing search engine results) is stored in an SQLite
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[115, 88, 883, 135]]<|/det|>
+database<sup>39</sup> in a bookkeeping step (Figure 1a). The corresponding entries in the indices, the MZ file and the SQLite database are linked by a unique spectrum ID.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 157, 361, 176]]<|/det|>
+## Retrieval by spectral similarity  
+
+<|ref|>text<|/ref|><|det|>[[114, 198, 884, 348]]<|/det|>
+A query request can be sent through a web browser to the server, which executes the query in a cgi script written in \(\mathrm{C + + }\) . The user can either use one spectrum already loaded into the spectral archive as the query (by providing the unique spectrum ID, or by searching by the source data file name and the scan number, or the peptide identification), or upload a new query spectrum to the server through the web service. Searching a batch of queries is also possible on the server side via command- line scripts.  
+
+<|ref|>text<|/ref|><|det|>[[113, 369, 884, 573]]<|/det|>
+Given a new query spectrum, the same preprocessing steps are taken and its address (one per index) is computed. For each index, the IVF- PQ algorithm then retrieves all the existing addresses in the closest nprobes buckets, and computes the approximate dot product between each address to the query's address (Supplementary Method). The top 1024 most similar addresses are returned as the "approximate nearest neighbors" (ANNs) of the query. The ANNs returned by all indices are combined, and the corresponding preprocessed spectra were read from the MZ files. Then, the accurate dot product is calculated between the full query spectrum and the full spectra of the ANNs to determine the "true nearest neighbors" (TNNs).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 594, 338, 612]]<|/det|>
+## Clustering and visualization  
+
+<|ref|>text<|/ref|><|det|>[[113, 634, 885, 890]]<|/det|>
+For visualization, the N most similar TNNs (N defaults to 20 but can be chosen by the user) can be shown in a graph in Spectroscape, with each spectrum shown as a node. For these N spectra, pairwise accurate dot products are computed to get a fully connected graph. Edges were considered for visualization on web browser if their Euclidian distance exceed a user- defined threshold. The graph is then plotted by the web browser using a force- directed graph drawing algorithm implemented in a free JavaScript library D3.js (https://d3js.org/). In brief, this graph visualization simulates the physical behavior of a network of electrically charged spheres (the nodes) connected by springs (the edges), with several parameters controllable by the user (Supplementary Method). The query spectrum is displayed in the center of the graph as a larger circle to distinguish it from its neighbors. The color of the node is calculated based on the peptide sequence using a hash
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 87, 885, 319]]<|/det|>
+function to map a string to a RGB color. The color black was reserved to indicate an unidentified spectrum. Optionally, edges between spectra of different precursor \(\mathrm{m / z}\) values can be shown as a different color, to aid the discovery of potential PTMs and variants. Information about the spectra and their computed similarities are shown if the mouse pointer hovers over nodes and edges. Single clicking on any node displays the full spectrum by the Lorikeet spectrum viewer. If the spectrum is identified by search engine with FDR \(< 1\%\) , it will be annotated with expected ions. To help the user evaluate alternative explanations for the query node, the user can also enter another peptide identification, upon which the annotation will be updated. Double clicking on any node will start a new query with the clicked node as the query spectrum.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 340, 312, 359]]<|/det|>
+## Running time evaluation  
+
+<|ref|>text<|/ref|><|det|>[[113, 379, 886, 662]]<|/det|>
+Spectroscape was tested on a server equipped with an AMD Ryzen Threadripper PRO 3975WX CPU with 32 cores at 3.5 GHz, one NVIDIA RTX3080Ti 12 GB GPU, and 512 GB of DDR4 3200 MHz RAM. Training 6 IVF- PQ indices using 100,000 spectra took about 15 minutes in total. Adding one spectrum to the archive, including all the preprocessing, indexing and bookkeeping steps, took less than 0.3 ms, averaged over 6 indices and the nearly 25 million spectra. To estimate retrieval time, different query batch sizes (1, 10, 100, 1000 and 10000) were tested; a random sample of spectra already in the spectral archive are submitted for query to evaluate the effect of parallelization. The running times for various steps done on the server were recorded by the commands in the \(\mathrm{C + + }\) standard library chrono, and averaged over many executions. The time taken for sending query requests over the internet, or for plotting the graph in the web browser is not estimated given the inherent variability of such tasks.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 684, 255, 702]]<|/det|>
+## Recall evaluation  
+
+<|ref|>text<|/ref|><|det|>[[115, 723, 883, 772]]<|/det|>
+To evaluate the retrieval performance of the IVF- PQ algorithm, we measure the recall \(R_{i}\) of a given query \(\pmb{q}_{i}\) by:  
+
+<|ref|>equation<|/ref|><|det|>[[378, 791, 618, 833]]<|/det|>
+\[R_{i}(t) = \frac{|A(\pmb{q}_{i})\cap T(\pmb{q}_{i},N,t)|}{|T(\pmb{q}_{i},N,t)|}\]  
+
+<|ref|>text<|/ref|><|det|>[[115, 853, 883, 901]]<|/det|>
+where \(A(\pmb{q}_{i})\) and \(T(\pmb{q}_{i},N,t)\) is the sets of retrieved ANNs of \(\pmb{q}_{i}\) and its TNNs, respectively, and \(t\) is the minimum dot product threshold to retain as a TNN. Here, the set of a TNNs of a given query
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 88, 886, 319]]<|/det|>
+\(T(\pmb {q}_i,N,t)\) is obtained by comparing the query spectrum to every spectrum in the archive without any index, and retaining up to \(N\) nearest neighbors whose accurate dot products with the query are above a certain cut- off \(t\) . The recall \(R_{i}\) of 20,000 randomly selected queries were plotted in two histograms corresponding to top 10 and top 100 TNNs (Figure 3a). The dot product cut- off was varied between 0.5 and 0.9, to evaluate the impact of true spectral similarity on the recall, as it is expected that the ANN- based similarity retrieval should be more effective as the neighbors are closer in reality. The approximate dot products and the true dot products of all retrieved ANNs of one typical query are plotted in a scatterplot (Figure 3b). The overall recall is defined over all the 20,000 queries as follows.  
+
+<|ref|>equation<|/ref|><|det|>[[339, 339, 657, 384]]<|/det|>
+\[R_{overall} = \frac{\sum_{i = 1}^{20000}|A(\pmb{q}_i)\cap T(\pmb{q}_i,N,t)|}{\sum_{i = 1}^{20000}|T(\pmb{q}_i,N,t)|}\]
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[115, 91, 216, 110]]<|/det|>
+## Reference
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 886, 138]]<|/det|>
+13. Griss, J. et al. Recognizing millions of consistently unidentified spectra across hundreds of shotgun proteomics datasets. Nat Methods 13, 651–656 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[113, 148, 884, 197]]<|/det|>
+14. Guthals, A., Watrous, J. D., Dorrestein, P. C. & Bandeira, N. The spectral networks paradigm in high throughput mass spectrometry. Mol Biosyst 8, 2535 (2012).  
+
+<|ref|>text<|/ref|><|det|>[[113, 208, 884, 255]]<|/det|>
+15. The, M. & Käll, L. MaRaCluster: A Fragment Rarity Metric for Clustering Fragment Spectra in Shotgun Proteomics. J Proteome Res 15, 713–720 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[113, 267, 884, 342]]<|/det|>
+16. To, P. K. P., Wu, L., Chan, C. M., Hoque, A. & Lam, H. ClusterSheep: A Graphics Processing Unit-Accelerated Software Tool for Large-Scale Clustering of Tandem Mass Spectra from Shotgun Proteomics. J Proteome Res 20, 5359–5367 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[113, 353, 884, 426]]<|/det|>
+17. Bittremieux, W., Meysman, P., Noble, W. S. & Laukens, K. Fast Open Modification Spectral Library Searching through Approximate Nearest Neighbor Indexing. J Proteome Res 17, 3463–3474 (2018).  
+
+<|ref|>text<|/ref|><|det|>[[113, 438, 884, 514]]<|/det|>
+18. Bittremieux, W., Laukens, K. & Noble, W. S. Extremely Fast and Accurate Open Modification Spectral Library Searching of High-Resolution Mass Spectra Using Feature Hashing and Graphics Processing Units. J Proteome Res 18, 3792–3799 (2019).  
+
+<|ref|>text<|/ref|><|det|>[[113, 525, 884, 574]]<|/det|>
+19. Bittremieux, W., May, D. H., Bilmes, J. & Noble, W. S. A learned embedding for efficient joint analysis of millions of mass spectra. Nat Methods 19, 675–678 (2022).  
+
+<|ref|>text<|/ref|><|det|>[[113, 585, 884, 658]]<|/det|>
+20. Wang, L., Li, S. & Tang, H. msCRUSH: Fast Tandem Mass Spectral Clustering Using Locality Sensitive Hashing. J Proteome Res acs.jproteome.8b00448 (2018) doi:10.1021/acs.jproteome.8b00448.  
+
+<|ref|>text<|/ref|><|det|>[[113, 670, 884, 745]]<|/det|>
+21. Bittremieux, W., Laukens, K., Noble, W. S. & Dorrestein, P. C. Large-scale tandem mass spectrum clustering using fast nearest neighbor searching. Rapid Communications in Mass Spectrometry (2021) doi:10.1002/rcm.9153.  
+
+<|ref|>text<|/ref|><|det|>[[113, 757, 884, 805]]<|/det|>
+22. Johnson, J., Douze, M. & Jégou, H. Billion-scale similarity search with GPUs. CoRR abs/1702.08734, (2017).  
+
+<|ref|>text<|/ref|><|det|>[[113, 816, 884, 864]]<|/det|>
+23. Ma, C. W. M. & Lam, H. Hunting for Unexpected Post-Translational Modifications by Spectral Library Searching with Tier-Wise Scoring. J Proteome Res 13, 2262–2271 (2014).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 88, 886, 152]]<|/det|>
+24. Chick, J. M. et al. A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides. Nat Biotechnol 33, 743–749 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[111, 174, 884, 247]]<|/det|>
+25. MacQueen, J. Some methods for classification and analysis of multivariate observations. in Berkeley Symposium on Mathematical Statistics and Probability Vol. 5.1, 281-297 (1967) 281–297 (1967).  
+
+<|ref|>text<|/ref|><|det|>[[111, 260, 884, 306]]<|/det|>
+26. Lam, H. et al. Development and validation of a spectral library searching method for peptide identification from MS/MS. Proteomics 7, 655–667 (2007).  
+
+<|ref|>text<|/ref|><|det|>[[111, 320, 884, 365]]<|/det|>
+27. Deutsch, E. W. et al. Universal Spectrum Identifier for mass spectra. Nat Methods 18, 768–770 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[111, 379, 830, 400]]<|/det|>
+28. Kim, M.-S. et al. A draft map of the human proteome. Nature 509, 575–581 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[111, 413, 884, 460]]<|/det|>
+29. Adusumilli, R. & Mallick, P. Data Conversion with ProteoWizard msConvert. in 339–368 (2017). doi:10.1007/978-1-4939-6747-6_23.  
+
+<|ref|>text<|/ref|><|det|>[[111, 473, 884, 546]]<|/det|>
+30. Kong, A. T., Leprevost, F. v, Avtonomov, D. M., Mellacheruvu, D. & Nesvizhskii, A. I. MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry–based proteomics. Nat Methods 14, 513–520 (2017).  
+
+<|ref|>text<|/ref|><|det|>[[111, 560, 884, 605]]<|/det|>
+31. Bateman, A. et al. UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res 49, D480–D489 (2021).  
+
+<|ref|>text<|/ref|><|det|>[[111, 619, 884, 665]]<|/det|>
+32. Ma, K., Vitek, O. & Nesvizhskii, A. I. A statistical model-building perspective to identification of MS/MS spectra with PeptideProphet. BMC Bioinformatics 13, S1 (2012).  
+
+<|ref|>text<|/ref|><|det|>[[111, 679, 884, 750]]<|/det|>
+33. Shteynberg, D. et al. iProphet: Multi-level Integrative Analysis of Shotgun Proteomic Data Improves Peptide and Protein Identification Rates and Error Estimates. Molecular & Cellular Proteomics 10, M111.007690 (2011).  
+
+<|ref|>text<|/ref|><|det|>[[111, 764, 884, 810]]<|/det|>
+34. Deutsch, E. W. et al. Trans-Proteomic Pipeline, a standardized data processing pipeline for large-scale reproducible proteomics informatics. Proteomics Clin Appl 9, 745–754 (2015).  
+
+<|ref|>text<|/ref|><|det|>[[111, 824, 884, 870]]<|/det|>
+35. Lam, H. et al. Building consensus spectral libraries for peptide identification in proteomics. Nat Methods 5, 873–875 (2008).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 88, 884, 130]]<|/det|>
+36. Shao, W., Zhu, K. & Lam, H. Refining similarity scoring to enable decoy-free validation in spectral library searching. \*Proteomics\* 13, 3273–3283 (2013).  
+
+<|ref|>text<|/ref|><|det|>[[113, 147, 884, 194]]<|/det|>
+37. Ge, T., He, K., Ke, Q. & Sun, J. Optimized Product Quantization. \*IEEE Trans Pattern Anal Mach Intell\* 36, 744–755 (2014).  
+
+<|ref|>text<|/ref|><|det|>[[113, 209, 884, 281]]<|/det|>
+38. Hoopmann, M. R., Mendoza, L., Deutsch, E. W., Shteynberg, D. & Moritz, R. L. An Open Data Format for Visualization and Analysis of Cross-Linked Mass Spectrometry Results. \*J Am Soc Mass Spectrom\* 27, 1728–1734 (2016).  
+
+<|ref|>text<|/ref|><|det|>[[113, 295, 746, 316]]<|/det|>
+39. Hipp, R. D. SQLite. Preprint at https://www.sqlite.org/index.html (2020).
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[98, 130, 930, 750]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[115, 92, 192, 111]]<|/det|>
+<center>Figures </center>  
+<|ref|>image_caption<|/ref|><|det|>[[113, 755, 884, 879]]<|/det|>
+<center>Figure 1 | Building and searching of spectral archive by Spectroscape. In the building step, the indices are built by training on 100,000 spectra. The new data files can be added to the indices after preprocessing. The preprocessed spectra are also stored in an MZ file. An SQL database is created to keep all the meta information of each spectrum, including identifications (if any). In the searching step, ANNs are retrieved from indices for each preprocessed query spectrum. Those ANNs are re-ranked by the accurate dot product to retain up to 100 true nearest neighbors (TNNs) for clustering, which is visualized as a force-directed graph. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[127, 88, 867, 515]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 536, 883, 608]]<|/det|>
+<center>Figure 2 | The average time for retrieval of ANNs for each query spectrum on one IVF-PQ index, as a function of the number of spectra in the archive, in millions, if one query (blue), 10 queries (orange), 100 queries (green), 1000 queries (gold) and 10000 queries (purple), respectively, were executed in a batch. As shown, the per-query retrieval time scales linearly with archive size. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 140, 870, 760]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 787, 884, 888]]<|/det|>
+<center>Figure 3 | Recall of TNNs by Spectroscape. (a) Histograms of per-query recall of the TNNs, tested with 20,000 randomly selected queries from the archive. TNNs are defined by two parameters: N, the number of closest neighbors to be retained as TNNs (N = 10 and 100), and t, the minimum dot product to be considered a TNN (t = 0.5 to 0.9). On both left (N = 10) and right panels (N = 100), the per-query recall distributions are dominated by the rightmost column, representing the fraction </center>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[114, 88, 884, 229]]<|/det|>
+of queries achieving perfect recall. In general, Spectroscape achieves higher per- query recall if one requires higher minimum dot product similarity to be counted as TNNs, or if one considers fewer TNNs. (b) A scatterplot of true dot product versus the approximate dot product as computed by the IVF- PQ algorithm; each data point represents one retrieved ANN of one of 20,000 random queries. About 600,000 retrieved ANNs in total were plotted. The color indicates density of data points by kernel density estimation. The overall correlation coefficient between the approximated dot product and the true dot product is 0.857.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[120, 131, 835, 470]]<|/det|>  
+
+<|ref|>image<|/ref|><|det|>[[120, 530, 880, 870]]<|/det|>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[115, 140, 886, 490]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[113, 506, 884, 848]]<|/det|>
+<center>Figure 4 | Examples of spectrum clusters visualized by Spectroscape. (a) Identification of unidentified spectra by association with neighbors with a precursor mass shift. The query node (PSM shown in top left panel) is unidentified but its nearest neighbors, with a precursor mass shift of 230.04 Da, are identified as DDPLTNLTFAFVAEK/2. The mass shift equals to the mass of the oligopeptide DD. Therefore, the unidentified query node and all its black neighbors are likely the semi-tryptic peptide PLTNLTFAFVAEK/2, which can also be confirmed by the fact that a library spectrum of that peptide ion is located in the middle of that cluster. The bottom left panel shows a very good PSM when we re-annotate the query node with this semi-tryptic peptide. (b) A query spectrum as a bridge of two clusters of PSMs corresponding to two peptides, LGPAIATGNVVVMK/2 and HVNPVQALSEFK/2. It is clearly a chimeric spectrum as the two PSMs almost do not share any matched peaks, as shown by the same spectrum annotated with respect to the two peptide ions (left panel). (c) The same peptide ion NQPGNTLTEILETPATAQQEVDHATDM[147]VSR/3 yields two slightly different fragmentation patterns on two mass spectrometers, as can be visualized clearly as two subclusters. The NIST HCD library spectrum (top left), represented as the node with a dotted outline that lies in the middle of the two subclusters, does not capture either of the fragmentation patterns very well. </center>
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 311, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 92, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 130, 510, 150]]<|/det|>
+- SpectralArchiveVisualizationSUPPINFOv5.1.docx
+
+<--- Page Split --->

preprint/preprint__9937ad734a3858e02f648619b966cc973f874dbf719528959aedd62faea1712d/images_list.json

@@ -0,0 +1,242 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Fig. 1: NPH3 binds to polyacidic phospholipids via its C-terminal domain.",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        103,
+        875,
+        404
+      ]
+    ],
+    "page_idx": 25
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Fig. 2: An amphipathic helix within the C-terminal domain is required for NPH3 phospholipid binding, membrane association and plasma membrane localization.",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        80,
+        880,
+        328
+      ]
+    ],
+    "page_idx": 26
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Fig. 3: Interaction of NPH3 and 14-3-3 proteins is triggered by blue light irradiation and abolished by mutation of the antepenultimate NPH3 residue.",
+    "footnote": [],
+    "bbox": [
+      [
+        115,
+        80,
+        653,
+        435
+      ]
+    ],
+    "page_idx": 27
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Fig. 4: 14-3-3 binding is required for proper NPH3 function in phototropic hypocotyl bending and its light-triggered detachment from the plasma membrane.",
+    "footnote": [],
+    "bbox": [
+      [
+        110,
+        78,
+        660,
+        680
+      ]
+    ],
+    "page_idx": 28
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Fig. 5: The phosphorylation status of the NPH3 14-3-3 binding site is dynamically modulated by the light regime.",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        80,
+        880,
+        410
+      ]
+    ],
+    "page_idx": 30
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Fig. 6: Functional relevance of the subcellular localization of NPH3.",
+    "footnote": [],
+    "bbox": [
+      [
+        110,
+        80,
+        884,
+        660
+      ]
+    ],
+    "page_idx": 31
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "Fig. S1:",
+    "footnote": [],
+    "bbox": [
+      [
+        113,
+        125,
+        864,
+        390
+      ]
+    ],
+    "page_idx": 33
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_1.jpg",
+    "caption": "Fig. S2:",
+    "footnote": [],
+    "bbox": [
+      [
+        100,
+        75,
+        718,
+        810
+      ]
+    ],
+    "page_idx": 34
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_2.jpg",
+    "caption": "Fig. S3:",
+    "footnote": [],
+    "bbox": [
+      [
+        112,
+        80,
+        764,
+        522
+      ]
+    ],
+    "page_idx": 36
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_3.jpg",
+    "caption": "Fig. S4:",
+    "footnote": [],
+    "bbox": [
+      [
+        115,
+        80,
+        450,
+        305
+      ]
+    ],
+    "page_idx": 37
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1",
+    "footnote": [],
+    "bbox": [
+      [
+        44,
+        95,
+        940,
+        479
+      ]
+    ],
+    "page_idx": 39
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2",
+    "footnote": [],
+    "bbox": [
+      [
+        52,
+        50,
+        940,
+        368
+      ]
+    ],
+    "page_idx": 40
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3",
+    "footnote": [],
+    "bbox": [
+      [
+        50,
+        54,
+        920,
+        690
+      ]
+    ],
+    "page_idx": 41
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_4.jpg",
+    "caption": "Figure 4",
+    "footnote": [],
+    "bbox": [
+      [
+        52,
+        50,
+        660,
+        777
+      ]
+    ],
+    "page_idx": 43
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Figure 5",
+    "footnote": [],
+    "bbox": [
+      [
+        50,
+        282,
+        944,
+        710
+      ]
+    ],
+    "page_idx": 44
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_6.jpg",
+    "caption": "Figure 6",
+    "footnote": [],
+    "bbox": [
+      [
+        55,
+        170,
+        915,
+        878
+      ]
+    ],
+    "page_idx": 45
+  }
+]

preprint/preprint__9937ad734a3858e02f648619b966cc973f874dbf719528959aedd62faea1712d/preprint__9937ad734a3858e02f648619b966cc973f874dbf719528959aedd62faea1712d.mmd

@@ -0,0 +1,599 @@
+
+# Light-triggered and phosphorylation-dependent 14-3-3 association with NON-PHOTOTROPIC HYPOCOTYL 3 is required for hypocotyl phototropism  
+
+Lea Reuter  ZMBP- University of Tuebingen  Tanja Schmidt  ZMBP- University of Tuebingen  https://orcid.org/0000- 0001- 9203- 8847  Prabha Manishankar  ZMBP- University of Tuebingen  Christian Throm  ZMBP- University of Tuebingen  https://orcid.org/0000- 0003- 2914- 7025  Jutta Keicher  ZMBP- University of Tuebingen  Andrea Bock  ZMBP- University of Tuebingen  Claudia Oecking ( \(\square\) claudia.oecking@zmbp.uni- tuebingen.de)  ZMBP- University of Tuebingen  
+
+## Article  
+
+Keywords: Auxin- dependent Phototrophic Growth Response, Polyacidic Phospholipids, Plasma Membrane Association, NPH3 Dephosphorylation  
+
+Posted Date: May 4th, 2021  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 467981/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License  
+
+Version of Record: A version of this preprint was published at Nature Communications on October 21st, 2021. See the published version at https://doi.org/10.1038/s41467- 021- 26332- 6.
+
+<--- Page Split --->
+
+1 Light- triggered and phosphorylation- dependent 14- 3- 3 association with 2 NON- PHOTOTROPIC HYPOcOTYL 3 is required for hypocotyl phototropism 3 4 Lea Reuter#, Tanja Schmidt#, Prabha Manishankar, Christian Throm, Jutta Keicher, Andrea 5 Bock, Claudia Oecking\* 6 7 Center for Plant Molecular Biology (ZMBP), Plant Physiology, University of Tübingen, 8 Germany 9 10 # Authors contributed equally 11 \* Corresponding author. Email: claudia.oecking@zmbp.uni- tuebingen.de
+
+<--- Page Split --->
+
+## ABSTRACT  
+
+NON- PHOTOTROPIC HYPOcOTYL 3 (NPH3) is a key component of the auxin- dependent plant phototropic growth response. We show that NPH3 directly binds polyacidic phospholipids, required for plasma membrane association in darkness. We further demonstrate that blue light induces an immediate phosphorylation of a C- terminal 14- 3- 3 binding motif in NPH3. Subsequent association of 14- 3- 3 proteins is causal for the light- induced release of NPH3 from the membrane and required for NPH3 dephosphorylation. In the cytosol, NPH3 dynamically transitions into membrane- less condensate- like structures. The dephosphorylated state of the 14- 3- 3 binding site and NPH3 membrane recruitment are recoverable in darkness. NPH3 variants that constitutively localize either to the membrane or to condensates are non- functional, revealing a fundamental role of the 14- 3- 3 mediated dynamic change in NPH3 localization for auxin- dependent phototropism. This novel mechanism of regulation might be of general nature, given that several members of the NPH3- like family interact with 14- 3- 3 via a C- terminal motif.  
+
+## Introduction  
+
+Developmental plasticity of plants is impressively demonstrated by the phototropic response, through which plants align their growth with incoming blue light (BL) \(^{1}\) . Shoots typically grow towards the light by generating a lateral gradient of the growth promoting phytohormone auxin. Here, the hormone concentration is higher on the shaded side as compared with the lit side, resulting in differential growth. It is well established that the phototropins phot1 and phot2 function as primary photoreceptors controlling phototropism in Arabidopsis \(^{2,3,4}\) . Phototropins are plasma membrane (PM)- associated, light- activated protein kinases and indeed, BL- induced autophosphorylation turned out to be a primary and essential step for the asymmetric growth response \(^{5}\) . In this context, members of the 14- 3- 3 family were identified as phot1 interactors in Arabidopsis. Eukaryotic 14- 3- 3 proteins are known to interact with a multitude of polypeptides in a phosphorylation- dependent manner, thereby regulating distinct cellular processes \(^{6}\) . Plant 14- 3- 3 are crucial components regulating auxin transport- related development and polarity of PIN- FORMED (PIN) auxin efflux carriers \(^{7}\) . As yet, however, a functional role of phot1/14- 3- 3 association could not be proven \(^{5,8}\) . Furthermore, evidence for trans- phosphorylation activity of phototropins is surprisingly limited. The polar localization of PIN proteins within the PM made them likely candidates promoting formation of the auxin gradient that precedes phototropic growth \(^{9}\) . Indeed, a mutant lacking the three major PINs expressed in aerial plant parts (PIN3, PIN4, PIN7) is severely compromised in phototropism \(^{10}\) . Notably, unilateral illumination polarizes PIN3 specifically to the inner lateral side of hypocotyl endodermis cells, aligning PIN3 polarity with the light direction and presumably redirecting auxin flow towards the shaded side \(^{11}\) . Moreover, the
+
+<--- Page Split --->
+
+activity of PINs is positively regulated by two protein kinase families from the AGCVIII class, namely PINOID and D6 PROTEIN KINASES \(^{12}\) . Though phototropins belong to the same kinase class, direct PIN phosphorylation could not be demonstrated \(^{11}\) . Taken together, signaling events that couple photoreceptor activation to changes in PIN polarization and consequently auxin relocation remain mainly elusive.  
+
+In this regard, the PM- associated NON- PHOTOTROPIC HYPOCOTYL 3 (NPH3) might represent a promising component of early phototropic signaling events. It acts downstream of the photoreceptors and appears to be instrumental for auxin redistribution \(^{3,4,13,14}\) . NPH3 possesses – in addition to the central NPH3 domain – two putative protein- protein interaction domains, a C- terminal coiled- coil (CC) domain and a N- terminal BTB/POZ (broad- complex, tramtrack, bric a brac/Pox virus and zinc finger) domain \(^{1,15}\) (Fig. S1). Indeed, NPH3 physically interacts not only with the photoreceptor phot1 but also with further early signaling elements, such as ROOT PHOTOTROPISM (RPT2) \(^{16}\) – another member of the plant- specific NPH3/RPT2- like family (NRL) – and defined members of the PHYTOCROME KINASE (PKS) family \(^{17,18}\) . Interestingly, NPH3 exists in a phosphorylated form in dark- grown seedlings and becomes rapidly dephosphorylated upon phot1 activation \(^{19,20}\) . Later on, the alteration in phosphorylation status was shown to correlate closely with light- driven changes in the subcellular localization of NPH3 which detaches from the PM upon irradiation, forming aggregated particles in the cytosol \(^{21}\) . As found for light- triggered dephosphorylation \(^{19}\) , formation of the NPH3 particles is reversible upon darkness or prolonged irradiation \(^{21}\) . One factor required for the recovery of phosphorylated NPH3 at the PM over periods of prolonged irradiation is its interaction partner RPT2 \(^{21}\) . Altogether, this has led to the current model that the phosphorylation status of NPH3 determines its subcellular localization and function: phosphorylation of NPH3 promotes its action in mediating phototropic signaling from the PM, whereas NPH3 dephosphorylation reduces it by internalizing NPH3 into aggregates \(^{4,13,21,22}\) . As yet, however, the functional significance of NPH3 (de)phosphorylation remains poorly understood \(^{20,23}\) .  
+
+Here, we identified members of the 14- 3- 3 family as novel interactors and major regulators of NPH3. Our analyses revealed that BL induces phosphorylation of the antepenultimate NPH3 residue which in turn enables 14- 3- 3 association. Complex formation interferes with the ability of NPH3 to bind to polyacidic phospholipids, resulting in its displacement from the PM. Accumulation of NPH3 in the cytosol causes formation of membrane- less condensates. Intriguingly, both PM association and 14- 3- 3 triggered PM dissociation are required for NPH3 function. Taking the reversibility of the light- induced processes into account, the phototropin- triggered and 14- 3- 3- mediated dynamic change in the subcellular localization of NPH3 seems to be crucial for its proper function in the phototropic response.
+
+<--- Page Split --->
+
+## Results  
+
+## PM association of NPH3 is phospholipid-dependent and requires its C-terminal domain  
+
+Association of the hydrophilic NPH3 with the PM is known since its discovery in 1999 1. As yet, the molecular mechanism of NPH3 membrane recruitment in darkness remains elusive. MACCHI- BOU 4 (MAB4)/ ENHANCER OF PINOID (ENP), another member of the NRL family, was recently shown to associate with the PM in a PIN- dependent manner 24. Besides protein- protein interactions, hydrophobic as well as protein- lipid interactions can cause membrane anchoring of proteins. Several members of the AGCVIII kinase family - though not phot1- contain a basic and hydrophobic (BH) motif in the middle domain of the kinase. This polybasic motif interacts directly with phospholipids and is required for PM binding 25. When we applied the BH score prediction 26 to NPH3, two putative BH motifs were identified in its C- terminal domain (Fig. S2B). To examine the importance of electronegativity for NPH3 PM association in the dark, we made use of a genetic system that depletes the polyacidic phosphoinositide (PI) phosphatidylinositol- 4- phosphate (PI4P) at the PM via lipid anchoring of the catalytic domain of the yeast SAC1 PI4P phosphatase 27, 28. Transient co- expression of NPH3 together with SAC1, but not the catalytically inactive version SAC1DEAD, displaced NPH3 from the PM into discrete cytosolic bodies in darkness (Fig. 1A), reminiscent of the aggregated particles that have been observed upon BL treatment 21, 22. The strong and unique electrostatic signature of the plant PM is powered by the additive effect of PI4P and the phospholipids phosphatidic acid (PA) and phosphatidylserine (PS) 28, 29, 30, 31. In lipid overlay assays, NPH3 bound to several phospholipids characterized by polyacidic headgroups, namely PA as well as the PIs PI3P, PI4P, PI5P, PI(3,4)P2, PI(3,5)P2, PI(4,5)P2 and PI(3,4,5)P3 (Fig. 1B). NPH3 did neither bind to phospholipids with monoacidic headgroups, such as phosphatidylinositol or PS, nor to phospholipids with neutral headgroups, namely phosphatidylcholine (PC) and phosphatidylethanolamine (PE). Deletion of the C- terminal 51 residues of NPH3 (NPH3ΔC51, still comprising the CC domain, Fig. S1) abolished lipid binding, while the bacterially expressed C- terminal 51 residues of NPH3 (NPH3- C51) turned out to be sufficient to bind to polyacidic phospholipids (Fig. 1B). Moreover, NPH3- C51 bound to large unilamellar liposomes containing the polyacidic phospholipids PI4P or PA, but not to liposomes composed of only neutral phospholipids such as PC and PE (Fig. 1C). Apparently, the C- terminal 51 residues of NPH3 enable electrostatic association with membrane bilayers irrespective of posttranslational protein modifications or association with other proteins. As expected, transient expression of GFP:NPH3ΔC51 in N. benthamiana (native or 35S promoter) revealed loss of PM recruitment in the dark, as evident by the presence of discrete bodies in the cytosol (Fig. 1D, Fig. S2A). This resembles the scenario observed upon co- expression of NPH3 and SAC1 (Fig. 1A) as well as upon
+
+<--- Page Split --->
+
+transient expression of NPH3ΔC65:GFP in guard cells of Vicia faba \(^{32}\) . By contrast, deletion of the N-terminal domain (NPH3ΔN54, still comprising the BTB domain, Fig. S1) did not affect PM association of NPH3 in darkness (Fig. 1D, Fig. S2A).  
+
+## An amphipathic helix is essential for phospholipid binding and PM association of NPH3 in vivo  
+
+As already mentioned, two polybasic motifs with a BH score above the critical threshold value of 0.6 (window size 11 as recommended for the detection of motifs closer to the termini, \(^{26}\) ) were identified in the C- terminal domain of NPH3: (i) a R- rich motif (R736- R742) close to the C- terminal tail and (ii) a K- rich motif further upstream (W700- M713) (Fig. 2A, Fig. S2B). The latter is predicted to form an amphipathic helix, organized with clearly distinct positively charged and hydrophobic faces. The hydrophobic moment – a measure of the amphiphilicity - was calculated to be 0.58 (Fig. S2C), similar to the PM anchor of Remorin \(^{33}\) . In order to test the requirement of the two motifs for membrane association, all five basic amino acids within the R- rich motif were replaced by alanine (NPH3- 5KR/A). Furthermore, both hydrophobicity and positive charge of the amphipathic helix were decreased by exchange of four hydrophobic residues (NPH3- 4WLM/A) and of four lysine residues (NPH3- 4K/A), respectively (Fig. 2A; Fig. S2B). The ability of any of the three NPH3 replacement variants to bind polyacidic phospholipids in vitro was significantly impaired (Fig. 2B, C). Nonetheless, the GFP:NPH3- 5KR/A mutant remained PM- associated in the dark when transiently expressed in N. benthamiana (Fig. 2D). To verify that the terminal R- rich motif is dispensable for PM recruitment in vivo, NPH3 was truncated by the C- terminal 28 residues (NPH3ΔC28). Indeed, PM anchoring was unaffected (Fig. 2D; Fig. S2D). By contrast, modification of either the amphiphilicity or the hydrophobicity of the amphipathic helix gave rise to cytosolic particle- like structures in darkness (Fig. 2D, Fig. S2D). Though these particles differ in shape and size, strict co- localization of the respective NPH3 variants was observed upon co- expression (Fig. S2E). Taken together, these experiments revealed the necessity of the amphipathic helix for PM anchoring in vivo and indicate hydrophobic interactions to also contribute to PM- association of NPH3. Thus, one attractive hypothesis is that the positively charged residues interact electrostatically with polyacidic phospholipids of the PM followed by partial membrane penetration. By this means, interactions with both the polar headgroups and the hydrocarbon region of the bilayer would be established in darkness, causing anchor properties of NPH3 similar to intrinsic proteins.
+
+<--- Page Split --->
+
+## 14-3-3 proteins interact with NPH3 via a C-terminal binding motif in a BL-dependent manner  
+
+A yeast two hybrid screen performed in our lab (see \(^{34}\) ) identified NPH3 as putative interactor of several Arabidopsis 14- 3- 3 isoforms, among those epsilon and omega (Fig. 3A). 14- 3- 3 mediated regulation of NPH3 might thus represent an early event in phototropic signaling. Complex formation of NPH3 and 14- 3- 3 was confirmed in planta by co- immunoprecipitation (ColP) of fluorophore- tagged proteins transiently co- expressed in N. benthamiana leaves (Fig. 3B). To elucidate the impact of light on 14- 3- 3/NPH3 complex assembly, transgenic Arabidopsis lines expressing 14- 3- 3 epsilon:GFP under control of the native promoter \(^{7}\) and, as control, UBQ10::GFP were employed. Three- days old etiolated seedlings were either maintained in complete darkness or irradiated with BL (1 \(\mu\) mol m \(^{- 2}\) sec \(^{- 1}\) ) for 30 minutes. Potential targets of 14- 3- 3 epsilon:GFP were identified by stringent ColP- experiments coupled with mass spectrometry (MS)- based protein identification. As expected, several known 14- 3- 3 clients \(^{7}\) were detected by MS, and remarkably, NPH3 emerged as a major 14- 3- 3 interactor (Table S1). Binding capability of characterized 14- 3- 3 targets, such as the H \(^+\) - ATPase (AHA1) and cytosolic invertase 1 (CINV1), was not modified by BL treatment. By contrast, NPH3 turned out to be a BL- dependent 14- 3- 3 interactor in planta (Fig. 3C, Table S1). ColP of fluorophore- tagged proteins transiently co- expressed in N. benthamiana leaves confirmed that physical association of NPH3 and 14- 3- 3 is not detectable in darkness while BL irradiation triggers complex formation (Fig. 3D). Assuming 14- 3- 3 association to depend on phosphorylation of the target protein, this observation is in apparent contrast to the light- induced dephosphorylation of NPH3 \(^{19}\) .  
+
+The specific phosphorylatable 14- 3- 3 binding sequences of numerous target proteins are most flexible and disordered \(^{35}\) . Since both the N- and C- terminal domain of NPH3 are predicted to be intrinsically disordered (Fig. S1, \(^{36}\) ), the corresponding truncated versions were analyzed by yeast two hybrid assays. While NPH3ΔN54 was capable of 14- 3- 3 binding, deletion of the C- terminal 51 residues (NPH3ΔC51) abolished 14- 3- 3 association, suggesting that the 14- 3- 3 binding site – in addition to the membrane targeting motif- localizes downstream of the CC domain (Fig. 3A). We therefore exchanged amino acid residues, phosphorylation of which has recently been demonstrated in planta (S722, S723, S744, S746, \(^{37,38}\) ), for a non- phosphorylatable alanine. Strikingly, 14- 3- 3 binding was not affected in all but one NPH3 mutant: replacement of S744 – the antepenultimate residue of NPH3 – prevented 14- 3- 3 association both in yeast (Fig. 3A) and in planta (Fig. 3B), suggesting a phosphorylation- dependent C- terminal 14- 3- 3 binding motif (pS/pTX \(_{1 - 2}\) - COOH) \(^{39}\) in NPH3. Phosphomimic variants (NPH3- S744D/S744E), however, do not allow for 14- 3- 3 binding (Fig. 3A), a characteristic of almost all 14- 3- 3 clients \(^{40}\) .
+
+<--- Page Split --->
+
+14- 3- 3 association is required for NPH3 function and its BL- induced PM dissociation To address the issue of functional significance of 14- 3- 3 association in vivo, GFP- tagged NPH3 variants were expressed in a T- DNA induced loss of function allele of NPH3, nph3- 7 \(^{41}\) . GFP:NPH3 was fully functional in restoring the severe impairment of hypocotyl phototropism in nph3- 7, regardless of whether expression was driven by the native or the 35S CaMV promoter (Fig. 4A, Fig. S3A), thus confirming previous data \(^{21,22}\) . By contrast, phototropic hypocotyl bending was still significantly reduced when NPH3 incapable of 14- 3- 3 association (GFP:NPH3- S744A) was expressed (Fig. 4A, Fig. S3A), indicating that BL- induced interaction with 14- 3- 3 is required for proper NPH3 function. Though NPH3 is hydrophilic in nature, both GFP:NPH3 and GFP:NPH3- S744A localized to the cell periphery in the hypocotyl of etiolated transgenic seedlings (Fig. 4B, Fig. S3B), suggesting PM association as described previously for NPH3 \(^{1,21,22}\) . Within minutes, however, the BL laser used to excite GFP (488 nm, activates phototropins), induced detachment of NPH3 from the PM into discrete bodies/particle- like structures in the cytoplasm (Video S1). This BL- induced shift in subcellular localization is mediated by phot1 activity \(^{21}\) and again, could be observed independent of whether expression of NPH3 was under control of the endogenous (Fig. S3B; \(^{22}\) ) or the 35S promoter (Fig. 4B, \(^{21}\) ). By contrast, GFP:NPH3- S744A remained mainly PM- associated upon irradiation (Fig. 4B, Video S2, Fig. S3B, C). Mutation of the 14- 3- 3 binding site does thus not affect PM association of NPH3 in darkness but prevents BL- triggered PM dissociation, suggesting that light- induced binding of 14- 3- 3 proteins to the antepenultimate, presumably phosphorylated residue S744 is required to internalize NPH3 from the PM into cytosolic particles. Nonetheless, the suspected phosphorylation of S744 might per se decrease the interaction of NPH3 with polyacidic phospholipids, hence triggering PM dissociation. Yet, the appropriate phosphomimic version of NPH3 (NPH3- S744D) was neither impaired in phospholipid- interaction in vitro (Fig. 1B) nor PM recruitment in vivo (Fig. 1D). Altogether, the C- terminal domain plays a dual role in determining the subcellular localization of NPH3: it mediates phospholipid- dependent PM association and allows for PM dissociation as a result of 14- 3- 3 association. We confirmed our findings in transiently transformed N. benthamiana leaves (Fig. 4C, D; Videos S3, S4). Here, primarily RFP- tagged proteins were employed since excitation of RFP (558 nm) – unlike GFP (488 nm) – does not activate phototropins. This enabled us to conditionally activate phot1 by means of the GFP laser. It became evident that NPH3 – instead of being directly internalized into discrete bodies – initially detaches from the PM and moves along cytoplasmic strands comparable to soluble polypeptides (Video S3). Body formation in the cytosol is initiated after a lag time of approximately 4 to 5 minutes. Generation of particle- like structures might thus depend on soluble NPH3 exceeding a critical
+
+<--- Page Split --->
+
+concentration in the cytosol. Upon co- expression of GFP- tagged 14- 3- 3s, colocalization with NPH3 was observed in such particles (Fig. S3D).  
+
+## NPH3 forms membrane-less condensates in the cytosol  
+
+BL- induced PM dissociation and particle assembly of NPH3 in the cytosol seem to be separate and consecutive processes (Video S3). As yet, the identity of these particles has not been determined. NPH3ΔC51 is devoid of the amphipathic helix and localized to cytosolic particles in darkness (Fig. 1D). Subcellular fractionation clearly illustrated that the lack of the C- terminal region shifts NPH3 from a membrane- associated state to the soluble fraction (Fig. 1E). This reveals a non- membrane- attached state of NPH3 in discrete bodies as has been suggested for NPH3 aggregates generated upon BL irradiation \(^{21}\) . Apparently, the mechanisms of NPH3 targeting towards and away from the PM are distinct from vesicle- mediated transport of transmembrane proteins. This is in line with the observation that NPH3 is insensitive to an inhibitor of endosomal trafficking \(^{21}\) . Considering the lack of the 14- 3- 3 binding motif in NPH3ΔC51, 14- 3- 3 association seems dispensable for NPH3 body formation in the cytosol. To confirm this assumption, we examined NPH3 variants incapable of 14- 3- 3 binding, namely (i) NPH3- 4K/A- S744A and (ii) NPH3- S744A, the latter upon co- expression with SAC1. Indeed, prevention of 14- 3- 3 association did not affect assembly of NPH3- 4K/A- S744A particles in darkness (Fig. S2D). Similar to NPH3, NPH3- S744A localized to cytosolic particles in the dark upon co- expression of SAC1 but not SAC1DEAD (Fig. 1A). Generation of NPH3 particles is hence feasible in the absence of 14- 3- 3s and might be due to intrinsic properties of NPH3 when exceeding a critical concentration in the cytosol. Taking constitutive PM association of NPH3- S744A in the absence of SAC1 into account, 14- 3- 3 association seems to be crucial for initial PM detachment while formation of discrete bodies in the cytosol occurs as an autonomous process.  
+
+The dynamic generation and morphology of NPH3 bodies is reminiscent of membrane- less biomolecular condensates which are micron- scale compartments in cells lacking surrounding membranes. An important organizing principle is liquid- liquid phase separation driven by multivalent macromolecular interactions – either mediated by modular interaction domains or disordered regions \(^{42}\) . NPH3 is characterized by both intrinsically disordered regions and interaction domains such as the BTB and the CC domain (Fig. S1). We performed single- cell time- lapse imaging of RFP:NPH3 body formation to investigate whether NPH3 undergoes transition from a solute to a condensed state in N. benthamiana. Indeed, formation of particle- like structures in the cytosol is initiated after approx. 4 min and the fluorescence intensity per body gradually increased over time as a result of the growth in size (Fig. 4E, F). In contrast to the signal intensity, the number of bodies reached a maximum after approx. 10
+
+<--- Page Split --->
+
+to 15 min and afterwards started to decrease as a result of body fusion (Fig. 4E, G). Worth mentioning, these features are characteristic criteria of biomolecular condensates \(^{42,43}\) .  
+
+Phosphorylation of the 14- 3- 3 binding site in NPH3 is light- dependent and reversible In dark- grown seedlings, NPH3 exists as a phosphorylated protein irrespective of phot1 activity. Light- induced dephosphorylation of NPH3 is almost a dogma in the literature. It has been recognized as a slight shift in electrophoretic mobility of NPH3 upon SDS- PAGE \(^{19}\) and requires – in accordance with the light- induced formation of particle- like structures in the cytosol \(^{21}\) – the photoreceptor phot1. In the following, (de)phosphorylation of NPH3, represented by a modification of its electrophoretic mobility, will be referred to as ‘general’ (de)phosphorylation of NPH3. Nonetheless, the data presented so far suggest that light- triggered and presumably S744 phosphorylation- dependent 14- 3- 3 association contributes to NPH3 function – an obvious antagonism to the ‘dogma of dephosphorylation’. A phosphosite- specific peptide antibody (α- pS744) was therefore established (antigen: \(^{734}\) PPRKPRRWRNS(P)- IS \(^{746}\) ) and an antibody against the unmodified peptide (α- NPH3) served as control. Examination of GFP:NPH3 in either N. benthamiana leaves or transgenic Arabidopsis lines revealed the typical enhanced electrophoretic mobility upon BL excitation (Fig. 5), indicative of a ‘general’ dephosphorylation \(^{19,20,21}\) . Intriguingly, the α- pS744 antibody recognized GFP:NPH3, but not GFP:NPH3- S744A, exclusively upon BL irradiation (Fig. 5). BL hence triggers two different posttranslational modifications of NPH3: (i) the phosphorylation of the 14- 3- 3 binding site (S744) and (ii) a ‘general’ dephosphorylation. Yet, neither of the modifications could be observed for GFP:NPH3- S744A (Fig. 5A). To uncover light- induced 14- 3- 3 association at the molecular level, an IP of GFP:NPH3 was conducted and combined with 14- 3- 3 Far Western analysis. Phosphorylation of S744 indeed enabled binding of purified recombinant 14- 3- 3 proteins to NPH3 upon SDS PAGE (Fig. 5A, B). Prolonged irradiation or transfer of BL- irradiated seedlings to darkness is known to confer PM re- association of NPH3 \(^{21}\) , correlating with a reduced electrophoretic mobility, indicative of a ‘general’ re- phosphorylation \(^{19,21}\) . Remarkably, we observed simultaneous dephosphorylation of S744 (Fig. 5B, C), effectively preventing binding of 14- 3- 3 to NPH3 (Fig. 5B). Taken together, the dark/light- dependent phosphorylation status of S744 determines 14- 3- 3 association with NPH3. In addition, the phosphorylation status of the 14- 3- 3 binding site and of NPH3 ‘in general’ is modulated by the light regime in an opposite manner, giving rise to a coinciding, but inverse pattern. Time course analyses, however, proved S744 phosphorylation of NPH3 to precede ‘general’ dephosphorylation upon BL treatment (Fig. 5C). ‘General’ dephosphorylation of NPH3 has been assumed to determine PM release of NPH3 coupled to particle assembly in the cytosol \(^{4,13,21,22}\) . Our data now clearly indicate S744 phosphorylation- dependent 14- 3- 3 association to be the cause of PM dissociation, but
+
+<--- Page Split --->
+
+not of condensate assembly in the cytosol. 'General' dephosphorylation might thus be coupled to PM dissociation and/or condensate formation. We examined the 'general' phosphorylation status of both NPH3 and NPH3- S744A when co- expressed with SAC1. Despite the fact that either NPH3 variant constitutively localized to cytosolic condensates (Fig. 1A), NPH3 was phosphorylated in darkness and shifted to the dephosphorylated status upon BL treatment, while NPH3- S744A exhibited a permanent phosphorylated state (Fig. 5D). 'General' dephosphorylation of NPH3 is thus not coupled to PM dissociation. Moreover, it is neither a prerequisite nor a consequence of condensate assembly, rather it seems to require prior light- triggered and S744 phosphorylation- dependent 14- 3- 3 association (Fig. 5A, D). Taken together, we suggest (Fig. 6E) that BL- induced and phosphorylation- dependent 14- 3- 3 association releases NPH3 from the PM into the cytosol and very likely provokes 'general' dephosphorylation of NPH3. Formation of NPH3 condensates is, however, determined by the biological properties of PM- detached NPH3.  
+
+## Cycling of NPH3 might be key to function  
+
+The light- triggered and reversible shift in subcellular localization of NPH3 has led to the hypothesis that PM localization of NPH3 promotes its action in mediating phototropic signaling. In turn, NPH3 present in soluble condensates is considered to be inactive \(^{13,21,22}\) . The functional relevance of the transient changes in subcellular NPH3 localization is, however, still not known. To assess the functionality of NPH3 variants constitutively localizing to condensates, GFP:NPH3- 4K/A (Fig. 2D) as well as GFP:NPH3ΔC51 (Fig. 1D) were expressed in the loss of function Arabidopsis mutant nph3- 7. Worth mentioning, the electrophoretic mobility of GFP:NPH3- 4K/A corresponded to the dephosphorylated version of NPH3 and was not modified by light treatment (Fig. 6C). In line with the hypothesis mentioned above, NPH3 mutants constitutively present in condensates did not restore hypocotyl phototropism (Fig. 6A, B, Videos S5, S6). Contrary to the hypothesis, however, GFP:NPH3- S744A - despite exhibiting constitutive PM localization (Fig. 4B) - is also largely incapable of mediating phototropic hypocotyl bending in nph3- 7 (Fig. 4A). To verify significantly impaired activity of permanently PM- attached NPH3, we examined NPH3ΔC28 in addition. Comparable to the results obtained in N. benthamiana (Fig. 2D, Fig. S2D), NPH3ΔC28 remained PM- associated upon activation of phot1 in stable transgenic Arabidopsis lines (Fig. 6B, Video S7) and its electrophoretic mobility was not modified by BL treatment (Fig. 6C). Noteworthy, both NPH3- S744A and NPH3ΔC28 still interacted with phot 1 (Fig. 6D), indicating that complex formation at the PM is not compromised. Nevertheless, permanent attachment of NPH3 to the PM turned out to be insufficient for triggering the phototropic response in nph3- 7 (Fig. 6A).
+
+<--- Page Split --->
+
+Taken together, neither NPH3 mutants permanently detached from the PM nor NPH3 versions permanently attached to the PM seem to be fully functional (Fig. 6A, E). So, what is the underlying mechanism of NPH3 function? We examined NPH3ΔN54 (Fig. 1D, Fig. S2A, Video S8) in more detail. Similar to NPH3, NPH3ΔN54 associated to the PM in etiolated seedlings (Fig. 6B). Upon irradiation it (i) became phosphorylated at S744 (Fig. 6C), (ii) exhibited an increased electrophoretic mobility, indicative of a 'general' dephosphorylation (Fig. 6C) and (iii) detached from the PM followed by condensate formation in the cytosol (Fig. 6B, Video S9). Furthermore, all these processes were reverted when seedlings were re-transferred to darkness (Fig. 6B, C). Intriguingly, expression of NPH3ΔN54 completely restored phototropic hypocotyl bending in nph3- 7 (Fig. 6A) as did NPH3 (Fig. 4A). Thus, 14- 3- 3 mediated cycling of NPH3 between the PM and the cytosol might be of utmost importance for functionality (Fig. 6E).  
+
+## Discussion  
+
+Our data provide novel insight into the molecular mechanisms defining NPH3 function in BL- induced phototropic hypocotyl bending. We applied a combination of genetic, biochemical, physiological and live cell imaging approaches to uncover the impact of 14- 3- 3 proteins on NPH3, in particular its BL- triggered, phosphorylation- dependent and functionally essential release from the PM. Association of NPH3 with the PM is known since decades, but how it is recruited to this compartment is unknown. We demonstrated that NPH3 attaches to the PM in a phospholipid- dependent manner in darkness (Fig. 1A). The electrostatic interaction with polyacidic phospholipids (Fig. 1B, C) is mediated by four basic residues of an amphipathic helix, the hydrophobic face of which further contributes to PM association (Fig. 2D). We therefore suggest the amphipathic helix to be embedded in the PM inner- leaflet with its hydrophobic interface inserted in the hydrophobic core of the bilayer while the positively charged interface is arranged on the PM surface, interacting with the lipid polar heads. The molecular mechanism underlying PM association of NPH3 is thus different from the NRL protein MAB4/ENP which is recruited to the PM by interaction with PIN proteins 24. The amphipathic helix of NPH3 (amino acids 700- 713) localizes downstream of the CC domain of NPH3 in its C- terminal region which also encompasses the 14- 3- 3 binding site (S744) (Fig. 2A).  
+
+We discovered that BL induces two distinct posttranslational modifications in NPH3 (Fig. 5): (i) the immediate phosphorylation of S744 which in turn enables association of 14- 3- 3 proteins with NPH3, followed by (ii) the well- described dephosphorylation, represented by an enhanced electrophoretic mobility of NPH3 ('general' dephosphorylation) 19, 20, 21. The - as yet unrecognized - BL- induced NPH3 phosphorylation event linked to 14- 3- 3 association is of utmost importance since it is essential for (i) the BL- triggered internalization of NPH3 from
+
+<--- Page Split --->
+
+the PM (Fig. 4B) and (ii) the function of NPH3 in phototropic hypocotyl bending (Fig. 4A). However, expression of NPH3- S744A which is incapable of 14- 3- 3 interaction, partially restored the severe impairment of hypocotyl phototropism in \(nph3 - 7\) (Fig. 4A). This might be due to functional redundancy among certain members of the NRL protein family. Indeed, RPT2 is required for hypocotyl phototropism at light intensities utilized in our assays \(^{21}\) and its expression is induced and stabilized by BL treatment \(^{44}\) . RPT2 might thus partially substitute for NPH3. The same applies to DEFECTIVELY ORGANIZED TRIBUTARIES 3 (DOT3), the, as yet, functionally uncharacterized closest homolog of NPH3 \(^{13}\) . Worth mentioning, RPT2, DOT3 and also MAB4/ENP are capable of interacting with 14- 3- 3 proteins in yeast (Fig. S4). In each case, exchange of the antepenultimate residue (serine) abolished 14- 3- 3 association (Fig. S4), suggesting that phosphorylation- dependent 14- 3- 3 binding is not limited to NPH3 but rather represents a more widespread mechanism of NRL regulation. However, residual activity of NPH3- S744A in phototropic hypocotyl bending might alternatively be caused by its permanent association with the PM. Light treatment could induce a reorganization of NPH3- S744A within/along the PM which might allow for phototropic responsiveness to a certain level. Addressing these alternatives represents a formidable challenge for future research.  
+
+NPH3 has been described to re- localize directly from the PM into discrete bodies in the cytosol upon light treatment \(^{21,22}\) . It became, however, evident that it initially detaches from the PM into the cytosol (Video S3). Here, NPH3 undergoes a dynamic transition from a dilute to a condensed state, resulting in the formation of membrane- less biomolecular compartments (Fig. 1E; Fig. 4E). Biomolecular condensates are emerging as an important concept in signaling, also in plants \(^{45}\) . Their formation can be driven by multivalent interactions with other macromolecules, by intrinsically disordered regions within a single molecule or both \(^{42,46}\) . Interestingly, 14- 3- 3 proteins are dispensable for condensate assembly in the cytosol, as demonstrated by 14- 3- 3 binding- deficient NPH3 variants (Fig. 1A, D; Fig. S2D). Further studies will reveal whether condensate formation of the PM- detached NPH3 is essential for its action.  
+
+As described above, the light- triggered modifications of the phosphorylation pattern of NPH3 are highly complex. Our observations disproved the view that BL- triggered 'general' dephosphorylation events determine PM dissociation of NPH3 \(^{13,21,22}\) . First of all, dephosphorylation of NPH3 – i.e. a decrease in negative charge – is entirely inappropriate to interfere with membrane association relying on electrostatic interactions with polyacidic phospholipids. Furthermore, investigation of the seven NPH3 phosphorylation sites that were recently identified in etiolated Arabidopsis seedlings revealed that the phosphorylation status of these NPH3 residues was neither required for PM association in darkness nor BL- induced release of NPH3 into the cytosol \(^{23}\) . By contrast, single site mutation of the 14- 3- 3 binding
+
+<--- Page Split --->
+
+site in NPH3 (S744A) abolished PM dissociation upon BL treatment (Fig. 4B- D), indicating light- induced and phosphorylation- dependent 14- 3- 3 association to mediate PM release of NPH3. Given that the amphipathic helix localizes approximately 30 – 45 residues upstream of the 14- 3- 3 binding site (Fig. 2A), 14- 3- 3 binding to NPH3 is expected to induce a substantial conformational change that liberates the amphipathic helix from the PM. The molecular mechanism of NPH3 internalization is hence different from the - likewise PM- associated - photoreceptor phot1, trafficking of which occurs via vesicles through the endosomal recycling pathway \(^{47}\) . Now, what about the BL- triggered 'general' dephosphorylation of NPH3? Based on our findings, this posttranslational modification temporally succeeded light- induced S744 phosphorylation (Fig. 5C). Furthermore, 'general' dephosphorylation was coupled to BL- triggered S744 phosphorylation, irrespective of the subcellular localization of NPH3 (Fig. 5A, D). We therefore assume phosphorylation- dependent 14- 3- 3 binding to be required for BL- induced 'general' dephosphorylation of NPH3 as well - a hypothesis that will be examined by future research. Re- transfer of BL- irradiated seedlings to darkness triggers (i) dephosphorylation of S744 linked to 14- 3- 3 dissociation. 14- 3- 3 release is expected to result in a (re)exposure of the amphipathic helix, which subsequently enables (ii) re- association with the PM and presumably (iii) re- phosphorylation of NPH3, represented by a reduced electrophoretic mobility ('general' re- phosphorylation) (Fig. 5B, C). Intriguingly, neither NPH3 variants that constitutively localize to the PM nor mutant versions constitutively detached from the PM are capable of restoring the severe defect in hypocotyl phototropism in nph3- 7. Complementation of the nph3- 7 phenotype exclusively could be observed upon expression of NPH3 variants that exhibit a light regime- driven dynamic change in subcellular localization (Fig. 6A, B, C). In summary, we propose a model where S744 phosphorylation- dependent and 14- 3- 3 driven cycling of NPH3 between the PM and the cytosol critically determine NPH3 function in mediating phototropic signaling in Arabidopsis (Fig. 6E). In the past, it has been hypothesized that the light- induced internalization of phot1 – first described in 2002 \(^{48}\) - may be coupled to light- triggered re- localization of auxin transporters. Functionality of phot1, however, was unaffected when internalization of the photoreceptor was effectively prevented by PM tethering via lipid anchoring \(^{49}\) . Altogether, the change in subcellular localization does not seem to be essential for signaling of phot1, but of its downstream signaling component NPH3 (Fig. 6E). Light- induced and 14- 3- 3- mediated detachment of NPH3 from the PM might hence account for BL- driven changes in PIN polarity required for hypocotyl phototropism. Plant 14- 3- 3 proteins have been shown to contribute to the subcellular polar localization of PIN auxin efflux carrier and consequently auxin transport- dependent growth \(^{7}\) . NRL proteins in turn act as signal transducers in processes involving auxin (re)distribution in response to developmental or environmental signals \(^{13}\) , hence
+
+<--- Page Split --->
+
+providing a likely link between 14- 3- 3 and PIN polarity. One subfamily of the NRL protein family consists of MAB4/ ENP- like (MEL) polypeptides, playing a critical role in auxin- regulated organogenesis in Arabidopsis \(^{50,51,52}\) . MEL proteins exhibited a polar localization at the cell periphery which was almost identical to that of PIN proteins \(^{53,54}\) and were recently shown to maintain PIN polarity by limiting lateral diffusion \(^{24}\) . Thus, one attractive hypothesis is that certain NRL proteins contribute either to the maintenance or to a dynamic change of the subcellular polarity of PIN auxin carriers, thereby regulating auxin (re)distribution. Given that several NRL proteins are able to interact with 14- 3- 3 via a C- terminal binding motif (Fig. S4), phosphorylation- dependent 14- 3- 3 association might constitute a crucial mechanism of regulation for NRL proteins and consequently polarity of PIN proteins.  
+
+## Material and Methods  
+
+## Plant materials, transformation and growth conditions  
+
+Arabidopsis thaliana (ecotype Columbia- 0 (Col- 0)) expressing 14- 3- 3 epsilon:GFP under control of the native promoter has been described recently \(^{7}\) . Seeds of A. thaliana nph3- 7 (SALK_110039, Col- 0 background) were obtained from the Nottingham Arabidopsis Stock Centre. T- DNA insertion was confirmed by genomic PCR analysis and homozygous lines were identified. Stable transformation of nph3- 7 followed standard procedures. Seeds were surface sterilized and planted on solid half- strength Murashige and Skoog (MS) medium (pH 5.8). Following stratification in the dark for 48- 72 h at 4°C, seeds were exposed to fluorescent white light for 4 h. Subsequently, seedlings were grown at 20°C in darkness for 68 h. Light treatment of etiolated seedlings was done as specified in the Figure legends. Independent experiments were carried out at least in triplicates with the same significant results. Representative images are presented. Statistics were evaluated with Excel (Microsoft).  
+
+Transient transformation of 3- 4 weeks old Nicotiana benthamiana plants was performed exactly as described \(^{55}\) . Freshly transformed tobacco plants were kept under constant light for 24 h, subsequently transferred to darkness for 17 h (dark adaptation) and finally irradiated or kept in darkness as specified in the Figure legends.  
+
+## Cloning procedures  
+
+A 2.1 kb NPH3 promoter fragment was PCR- amplified from Col- 0 genomic DNA and the cDNA of NPH3 was amplified from Col- 0 cDNA. The respective primers were characterized by Bsal restriction sites allowing for the usage of the Golden Gate based modular assembly of synthetic genes for transgene expression in plants \(^{56}\) . Following A- tailing, the individual PCR products were directly ligated into the pGEM- T Easy (Promega) vector yielding level I vectors LI A- B pNPH3 and LI C- D NPH3, respectively. Golden Gate level II assembly was
+
+<--- Page Split --->
+
+performed by Bsal cut ligation and by using the modules LI A- B pNPH3, LI B- C GFP or LI B- C mCherry, LI C- D NPH3, LI dy D- E, LI E- F nos- T and LI F- G Hygro exactly as described \(^{56}\) . For ColP of fluorophore- tagged NPH3 and 14- 3- 3 transiently expressed in N. benthamiana, the corresponding cDNA was cloned into the 2in1 GATEWAY™ compatible vector pFRETcg- 2in1- NC \(^{57}\) via GATEWAY™ technology. Cloning of N- terminally fluorophore- tagged NPH3 variants (GFP and/or RFP) into the destination vectors pB7WGR2 and/or pH7WGF2 \(^{58}\) for stable or transient overexpression followed standard GATEWAY™ procedures. Transgenic plants were selected based on the hygromycin resistance conferred by pH7WGF2 and homozygous lines were established. The 35S- driven PHOT1:GFP \(^{47}\) and the 35S::MAP:mCherry:SAC1/SAC1DEAD transformation vectors \(^{28}\) as well as the utilized Golden Gate level I vectors \(^{56}\) have been described before, respectively.  
+
+Site- directed mutagenesis was performed by PCR. PCR products and products of mutagenesis were verified by sequencing.  
+
+A complete list of oligonucleotides used for PCR is provided below.  
+
+## Expression and purification of proteins  
+
+For expression of the Arabidopsis 14- 3- 3 isoform omega as RGS(His)6- tagged protein in Escherichia coli M15, the corresponding cDNA was amplified by PCR and cloned into the expression vector pQE- 30 (Qiagen). Purification was done by using Ni \(^{2 + }\) - NTA agarose (Qiagen) according to the manufacturer's protocol. For expression of the Arabidopsis NPH3 C- terminal 51 residues fused to GST in E. coli BL21(DE3), the corresponding cDNA fragment was amplified by PCR and cloned into the GST expression vector pGEX- 4T- 1. GST fusion proteins were purified from transformed bacteria using GSH- Sepharose according to the manufacturer's protocol (Cytiva). Free GST protein was expressed and purified as a negative control.  
+
+## Cell-free protein expression  
+
+Reactions were performed using the TNT® T7 Quick Coupled Transcription/Translation System (Promega) with 1 μg of vector (NPH3 or variants in pGADT7) for a 50 μl reaction. Protein expression was carried out at 30°C for 90 min. Immunodetection was performed by using an anti- HA antibody (HA- tag encoded by pGADT7).  
+
+## Preparation of microsomal membranes  
+
+Microsomal membrane fractions were prepared from transiently transformed N. benthamiana leaves. Tissue was homogenized with 3 mL homogenization buffer per g fresh weight (50 mM Hepes (pH 7.8), 500 mM sucrose, 1 % (w/v) PVP- 40, 3 mM DTT, 3 mM EDTA,
+
+<--- Page Split --->
+
+supplemented with Complete Protease Inhibitor Mixture (Roche) and Phosphatase Inhibitor Mix 1 (Serva)). The homogenate was centrifuged at 10,000 g for 20 min at 4 °C. The supernatant was filtered through MiraCloth and subsequently centrifuged at 100,000 g for 45 min at 4 °C. The microsomal pellet was resuspended in 5 mM Tris/MES (pH 6.5), 330 mM sucrose, 2 mM DTT, supplemented with Complete Protease Inhibitor Mixture (Roche) and Phosphatase Inhibitor Mix 1 (Serva).  
+
+## Phospholipid binding assays  
+
+For lipid binding assays, either NPH3 variants expressed in a cell free system or purified recombinant GST fusion proteins were applied. Lipid overlay assays using PIP- strips were performed following the manufacturer's instructions (Echelon). In brief, membranes were blocked overnight at 4°C in a blocking buffer with 4% fatty acid- free BSA in PBS- T (0.1% Tween). Purified proteins (0.1 μg/ml blocking buffer) or 10- 50 μl of the cell free expression reaction (volume adjusted according to prior immunodetection of individual reactions) were incubated with PIP- strip membranes for 1 h at room temperature and washed three times for 10 min with PBS- T. Subsequently, detection of bound proteins was done by immunodetection of either GST (GST fusion proteins) or the HA- tag (cell free expression). Liposome binding assays were conducted essentially as described by \(^{59}\) with slight modifications. All lipids were obtained from Avanti Polar Lipids. Liposomes were prepared from 400 nmol of total lipids at the following molar ratios: PC:PE, 1:1; PC: PE:P14P, 2:2:1; PC:PE:PA, 2:2:1. The binding buffer (150 mM KCl, 25 mM Tris–HCl pH 7.5, 1 mM DTT, 0.5 mM EDTA) was supplemented with Complete Protease Inhibitor Mixture (Roche). Purified GST- NPH3- C51 variants in binding buffer were centrifuged at 50,000 g to get rid of any possible precipitates. Following incubation of liposomes and proteins, the liposome pellet was washed twice with binding buffer. Liposome- bound GST- NPH3- C51 variants were detected by immunoblotting with anti- GST antibodies.  
+
+## Y2H, SDS-PAGE and Western Blotting  
+
+For yeast two- hybrid analyses, the individual constructs were cloned into the vectors pGADT7 and pGBKT7 and co- transformed into the yeast strain PJ69- 4A. Activity of the ADE2 reporter was analyzed by growth of co- transformed yeast on SD medium lacking adenine. SDS- PAGE, Western blotting and immunodetection followed standard procedures. Total proteins were extracted from 3- day- old etiolated Arabidopsis seedlings (50 seedlings) or transiently transformed N. benthamiana leaves (2 leaf disks) by directly grinding in 100 μl 2 x SDS sample buffer under red safe light illumination. Chemiluminescence detection was performed with an Amersham Image Quant800 (Cytiva) system.
+
+<--- Page Split --->
+
+The rabbit anti- NPH3- S744P antibody was generated with the phosphorylated synthetic peptide \(\mathsf{NH}_2\) - PPRKPRRWRN- S( \(\mathsf{PO}_3\mathsf{H}_2\) )- IS- COOH followed by affinity- purifications against the non- phosphorylated and phosphorylated peptide at Eurogentec (Liege, Belgium).  
+
+## ColP and mass spectrometry analysis  
+
+Arabidopsis seedlings expressing 14- 3- 3 epsilon- GFP (endogenous promoter) and, as control, GFP (UBQ10 promoter) were grown in the dark on half- strength MS plates for 3 days. Subsequently, the etiolated seedlings were either kept in darkness or treated with overhead BL (1 \(\mu \mathrm{mol} \mathrm{m}^{- 2} \mathrm{sec}^{- 1}\) ) for 30 min. Three grams of plant tissue were used under red safe light illumination for immunoprecipitation as described \(^{60}\) . The final precipitate in Laemmli buffer was analyzed by mass spectrometry (MS) at the University of Tübingen Proteome Center. Following a tryptic in gel digestion, LC- MS/MS analysis was performed on a Proxeon Easy- nLC coupled to an QExactiveHF mass spectrometer (method: 60 min, Top7, HCD). Processing of the data was conducted using MaxQuant software (vs 1.5.2.8). The spectra were searched against an Arabidopsis thaliana UniProt database. Raw data processing was done with 1% false discovery rate setting.  
+
+intensities of well- known 14- 3- 3 client proteins (Fig. 3C) were converted to normalized abundance of the bait protein. Fold changes in relative abundance of BL treatment versus darkness (BL vs. D) were calculated (Table S1).  
+
+Arabidopsis nph3- 7 ectopically expressing GFP: NPH3 and N. benthamiana leaves transiently overexpressing fluorophore- tagged proteins were immunoprecipitated under red safe light illumination according to \(^{61}\) . Growth and light irradiation of the plants is specified elsewhere.  
+
+In vivo interaction of phot1: GFP and N- terminally RFP- tagged NPH3 variants was tested by using solubilized microsomal proteins obtained from dark adapted N. benthamiana plants ectopically co- expressing the proteins of interest. Solubilization was achieved by adding \(0.5\%\) Triton X- 100 to resuspended microsomal proteins followed by centrifugation at 50,000 g for 30 min at \(4^{\circ} \mathrm{C}\) . The supernatant was added to GFP- Trap Beads (ChromoTek) and incubated at \(4^{\circ} \mathrm{C}\) for 1 h. Precipitated beads were washed six times with 50mM HEPES pH 7.8, 150mM NaCl, \(0.2\%\) Triton X- 100. Finally, proteins were eluted by SDS sample buffer and separated by SDS- PAGE.  
+
+## 14-3-3 Far-Western  
+
+Anti- GFP immunoprecipitates obtained from Arabidopsis nph3- 7 stably overexpressing GFP: NPH3 were separated by SDS- PAGE and transferred to nitrocellulose. Nonspecific
+
+<--- Page Split --->
+
+sites were blocked by incubation with \(4\%\) (w/v) milk powder in TBS at room temperature for at least 1 h. Subsequently, the membrane was incubated overnight at \(4^{\circ}C\) (followed by 1 h at room temperature) with purified recombinant RGS(His) \(_6\) - tagged 14- 3- 3 isoform omega of Arabidopsis diluted to \(20\mu \mathrm{gml}^{- 1}\) in \(50~\mathrm{mM}\) MOPS/NaOH, pH 6.5, \(20\%\) (w/v) glycerol, 5mM \(\mathrm{MgCl}_2\) , and 2mM DTT. After washing with TBS, immunodetection of RGS(His) \(_6\) - tagged 14- 3- 3 was performed by applying the anti- RGS(His) \(_6\) antibody (Qiagen) in combination with a secondary anti- mouse HRP antibody.  
+
+## Hypocotyl Phototropism analysis  
+
+A. thaliana seedlings were grown in the dark on vertically oriented half-strength MS plates for 48 h. Etiolated seedlings were then transferred to a LED chamber and illuminated with unilateral BL (1 \(\mu \mathrm{mol} \mathrm{m}^{-2} \mathrm{sec}^{-1}\) ) for 24 h. Plates were scanned and the inner hypocotyl angle was measured for each seedling using Fiji. The curvature angle was calculated as the difference between \(180^{\circ}\) and the measured value.  
+
+## Confocal microscopy  
+
+Live- cell imaging was performed using the Leica TCS SP8 (upright) confocal laser scanning microscope. For excitation and emission of fluorophores, the following laser settings were used: GFP, excitation 488 nm, emission 505- 530 nm; RFP, excitation 558 nm, emission 600- 630 nm. All CLSM images in a single experiment were captured with the same settings using the Leica Confocal Software. All the experiments were repeated at least three times. Images were processed using LAS X light. Single- cell time- lapse imaging was carried out on live leaf tissue samples from \(N\) . benthamiana transiently expressing RFP:NPH3. PM- detachment was induced by means of the GFP- laser (488 nM) and image acquisition (RFP- laser) was done for the duration of 32 min by scanning 30 consecutive planes along the Z axis covering the entire thickness of an epidermal cell. Z- projection was done for each 3,5 min interval. For all image quantifications, randomly sampled unsaturated confocal images ( \(512 \times 512\) pixels, \(225 \times 225 \mu \mathrm{m}\) ) were used with an image analysis protocol implemented in the ImageJ software \(^{62}\) as previously described \(^{63}\) . A random image was selected from the dataset and parameters such as local threshold, background noise, object size and shape were determined. The obtained parameters were used for image analysis of the whole dataset following exactly the published step by step protocol \(^{63}\) .
+
+<--- Page Split --->
+
+<table><tr><td>pGEM-T Easy/<br/>yeast &amp; bacterial<br/>expression vectors/<br/>mutagenesis</td><td></td></tr><tr><td>NPH3_Smal_F</td><td>TATcccgggCATGTGGGAATCTGAGAGCGAC</td></tr><tr><td>NPH3ΔN53_Sma_F</td><td>TATcccgggCATCTTCTGGTTAAGATCGGC</td></tr><tr><td>NPH3-C51_EcoRI_F</td><td>TATgaattcTCTTCTTCGGCTTGACACG</td></tr><tr><td>NPH3_Sall_R</td><td>TATgtcgacTCATGAAATTGAGTTCCT</td></tr><tr><td>NPH3ΔC51_Sall_R</td><td>TATgtcgacCTATGGCGTGTTCTTCACTTTCCC</td></tr><tr><td>NPH3_S743A_Sall_R</td><td>TATgtcgacTCATGAAATTGcGTTCTCCTCACTCGTCT</td></tr><tr><td>NPH3_S743D_Sall_R</td><td>TATgtcgacTCATGAAATgtcGTTCCCTCACTCGTCTTGTTTC</td></tr><tr><td>NPH3_S743E_Sall_R</td><td>TATgtcgacTCATGAAATTtcGTTCCCTCACTCGTCTTGTTTC</td></tr><tr><td>NPH3_S745A_Sall_R</td><td>TATgtcgacTCATGCAATTGAGTTCCCTCACTCGTCT</td></tr><tr><td>NPH3_3KR/A_Sall_R</td><td>TATgtcgacTCATGAAATTGAGTTGcGCATgtcTTCTGGTTTCgcGGGGGGTGGATGATC</td></tr><tr><td>NPH3_5KR/A_Sall_R</td><td>TATgtcgacTCATGAAATTGAGTTGcGCATgtcTgTGGTgcCgCGGGGGGTGGATGATC</td></tr><tr><td>NPH3_4K/A_F</td><td>GCTTGGACCAGCGGTTGGgcGGCGCTAAGTGcACTGACTGcGATGAGTGGCAGAGGAGAG</td></tr><tr><td>NPH3_4K/A_R</td><td>CTCTCCTGTCCACTCATCgcGTCAGTgcACTTAGCgcCgcCCAACCGCTGGTCCAAGC</td></tr><tr><td>NPH3_4WLM/A_F</td><td>TCGGCTTGACCAGCGGTgcGAGAAGAGCAGGAAGTAAGcGACTAAGcGAGGTGGACAGGAGAGCCAT</td></tr><tr><td>NPH3_4WLM/A_R</td><td>ATGGCTCTCCTGTCCACTCgcCTATGCgCTTTACTTgcCTCTTTCgcACCGCTGGTCCAAGCCGA</td></tr><tr><td>NPH3_S721A_F</td><td>CAGGAGAGCCATGACATAGCCTCTGGAGGAGAACAAGCT</td></tr><tr><td>NPH3_S721A_R</td><td>AGCTTGTTCTCCTCCAGAGGCTATGTCATGGCTCTCCT</td></tr><tr><td>NPH3_S722A_F</td><td>GAGAGCCATGACATATCCGCTGGAGGAGAACAAGCTGGT</td></tr><tr><td>NPH3_S722A_R</td><td>ACCAGCTTGTTCTCCTCCAGCGATATGTCATGGCTCTC</td></tr><tr><td>14-3-3omega_BamHI_F</td><td>TATggatccATGGCGCTGGCGGTGAAGAG</td></tr><tr><td>14-3-3omega_EcoRI_F</td><td>TATgaattcATGGCGCTGGCGGCTGAAGAG</td></tr><tr><td>14-3-3omega_Sall_R</td><td>TATgtcgacTCACTGCTGTTCCTCGGT</td></tr><tr><td>GATEWAY</td><td></td></tr><tr><td>NPH3_attB1_F</td><td>AAAAAGCAGGCTTAATGTGGGAATCTGAGAGCGAC</td></tr><tr><td>NPH3ΔN53_attB1_F</td><td>AAAAAGCAGGCTTAATGGATCTCTGGTTAAGATCGGC</td></tr><tr><td>NPH3_attB2_R</td><td>AGAAAGCTGGGTGTCATGAAATTGAGTTTCCTCCA</td></tr><tr><td>NPH3_S743A_attB2_R</td><td>AGAAAGCTGGGTGTCATGAAATTGCTTCCTCCAATCGTC</td></tr><tr><td>NPH3_S743D_attB2_R</td><td>AGAAAGCTGGGTGTCATGAAATTGCTTCCTCCAATCGTC</td></tr><tr><td>NPH3_5KR/A_attB2_R</td><td>AGAAAGCTGGGTGTCATGAAATTGCTTCCTCCAATCGTC</td></tr><tr><td>NPH3ΔC28_attB2_R</td><td>AGAAAGCTGGGTGTCATGCCTCTCCTGCTCCATCATCTT</td></tr><tr><td>NPH3ΔC51_attB2_R</td><td>AGAAAGCTGGGTGTCATGCCTCTCCTGCTCCATCATCTT</td></tr><tr><td>NPH3_attB4_R</td><td>GAAAAGTTGGGTGTCATGAAATTGAGTTTCCTCCA</td></tr><tr><td>NPH3_S743A_attB4_R</td><td>GAAAAGTTGGGTGTCATGAAATTGCTTCCTCCAATCGTC</td></tr><tr><td>14-3-3omega_attB3_F</td><td>ataataaagttgtaATGGCGCTGGCGGCT</td></tr></table>
+
+<--- Page Split --->
+
+<table><tr><td>14-3-3omega_attB2_R</td><td>agaaagctgggtgCTGCTGTTCCTCGGT</td></tr><tr><td>attB1 adapter</td><td>GGGGACAAGTTTGTACAAAAAAGCAGGCT</td></tr><tr><td>attB2 adapter</td><td>GGGGACCACTTTGTACAAGAAAGCTGGGT</td></tr><tr><td>attB3 adapter</td><td>GGGGACAACTTTGTATAATAAAGTTG</td></tr><tr><td>attB4 adapter</td><td>GGGGACAACTTTGTATAGAAAAGTTGGGT</td></tr><tr><td colspan="2">GOLDEN GATE</td></tr><tr><td>NPH3prom_A-B_F</td><td>AACAggtctcAGCGGAAACCCCACATTAATCAGACAGAATC</td></tr><tr><td>NPH3prom_A-B_R</td><td>AACAggtctcACAGAACACAAGTTAAACACTCTCTGTAGTTG</td></tr><tr><td>NPH3_C-D_F</td><td>AACAggtctcACCACTGTGGGAATCTGAGAGCGAC</td></tr><tr><td>NPH3ΔN53_C-D_F</td><td>AACAggtctcACCACTGGATCTTGTTGAATACGGC</td></tr><tr><td>NPH3_C-D_R</td><td>AACAggtctcACCTTTCATGAAATTGAGTTTCCTCCA</td></tr><tr><td>NPH3_S743A_C-D_R</td><td>AACAggtctcACCTTTCATGAAATTGGCGTTCTCCTCCATCGTC</td></tr><tr><td>NPH3ΔC28_C-D_R</td><td>AACAggtctcACCTTTCATGCTCTCGTCACTCATCTT</td></tr><tr><td>NPH3ΔC51_C-D_R</td><td>AACAggtctcACCTTTCATGGCGTGTTCTTCACTTTCCC</td></tr></table>
+
+## Acknowledgements 
+
+We are grateful to Yvon Jaillais and John M. Christie for providing the constructs 
+
+35S::MAP:SAC1/SAC1DEAD and 35S::PHOT1:GFP, respectively. We furthermore thank 
+
+Sandra Richter for SP8 support and John M. Christie for sharing data and stimulating 
+
+discussions. MS analysis was done at the Proteome Centre, University of Tübingen, and we 
+
+thank Irina Droste-Borel for help in data assessment. Research in our laboratory was 
+
+supported by the German Research Foundation (DFG) with a grant to C.O. (CRC 1101- 
+
+B09). 
+
+## References 
+
+1. Motchoulski A, Liscum E. Arabidopsis NPH3: A NPH1 photoreceptor-interacting protein essential for phototropism. *Science* 286, 961-964 (1999). 
+
+2. Christie JM, Murphy AS. Shoot phototropism in higher plants: new light through old concepts. *Am J Bot* 100, 35-46 (2013). 
+
+3. Fankhauser C, Christie JM. Plant phototropic growth. *Current biology: CB* 25, R384-389 (2015). 
+
+4. Legris M, Boccaccini A. Stem phototropism toward blue and ultraviolet light. *Physiol Plant* 169, 357-368 (2020). 
+
+5. Inoue S, Kinoshita T, Matsumoto M, Nakayama KI, Doi M, Shimazaki K. Blue light-induced autophosphorylation of phototropin is a primary step for signaling. *Proc Natl Acad Sci U S A* 105, 5626-5631 (2008). 
+
+6. Mackintosh C. Dynamic interactions between 14-3-3 proteins and phosphoproteins regulate diverse cellular processes. *Biochem J* 381, 329-342 (2004).
+
+<--- Page Split --->
+
+7. Keicher J, et al. Arabidopsis 14-3-3 epsilon members contribute to polarity of PIN auxin carrier and auxin transport-related development. Elife 6, (2017). 
+
+8. Sullivan S, Thomson CE, Kaiserli E, Christie JM. Interaction specificity of Arabidopsis 14-3-3 proteins with phototropin receptor kinases. FEBS letters 583, 2187-2193 (2009). 
+
+9. Rakusová H, Fendrych M, Friml J. Intracellular trafficking and PIN-mediated cell polarity during tropic responses in plants. Curr Opin Plant Biol 23, 116-123 (2015). 
+
+10. Willige BC, et al. D6PK AGCVIII kinases are required for auxin transport and phototropic hypocotyl bending in Arabidopsis. Plant Cell 25, 1674-1688 (2013). 
+
+11. Ding Z, et al. Light-mediated polarization of the PIN3 auxin transporter for the phototropic response in Arabidopsis. Nat Cell Biol 13, 447-452 (2011). 
+
+12. Zoureldou M, et al. Auxin efflux by PIN-FORMED proteins is activated by two different protein kinases, D6 PROTEIN KINASE and PINOID. Elife 3, (2014). 
+
+13. Christie JM, Suetsugu N, Sullivan S, Wada M. Shining Light on the Function of NPH3/RPT2-Like Proteins in Phototropin Signaling. Plant Physiol 176, 1015-1024 (2018). 
+
+14. Hohm T, Preuten T, Fankhauser C. Phototropism: translating light into directional growth. Am J Bot 100, 47-59 (2013). 
+
+15. Pedmale UV, Celaya RB, Liscum E. Phototropism: mechanism and outcomes. Arabidopsis Book 8, e0125 (2010). 
+
+16. Inada S, Ohgishi M, Mayama T, Okada K, Sakai T. RPT2 is a signal transducer involved in phototropic response and stomatal opening by association with phototropin 1 in Arabidopsis thaliana. Plant Cell 16, 887-896 (2004). 
+
+17. de Carbonnel M, et al. The Arabidopsis PHYTOCHROME KINASE SUBSTRATE2 protein is a phototropin signaling element that regulates leaf flattening and leaf positioning. Plant Physiol 152, 1391-1405 (2010). 
+
+18. Lariquet P, et al. PHYTOCHROME KINASE SUBSTRATE 1 is a phototropin 1 binding protein required for phototropism. Proc Natl Acad Sci U S A 103, 10134-10139 (2006). 
+
+19. Pedmale UV, Liscum E. Regulation of phototropic signaling in Arabidopsis via phosphorylation state changes in the phototropin 1-interacting protein NPH3. J Biol Chem 282, 19992-20001 (2007). 
+
+20. Tsuchida-Mayama T, et al. Mapping of the phosphorylation sites on the phototropic signal transducer, NPH3. Plant Science 174, 626-633 (2008). 
+
+21. Haga K, Tsuchida-Mayama T, Yamada M, Sakai T. Arabidopsis ROOT PHOTOTROPISM2 Contributes to the Adaptation to High-Intensity Light in Phototropic Responses. Plant Cell 27, 1098-1112 (2015). 
+
+22. Sullivan S, Kharshiing E, Laird J, Sakai T, Christie JM. Deetiolation Enhances Phototropism by Modulating NON-PHOTOTROPIC HYPOCOTYL3 Phosphorylation Status. Plant Physiol 180, 1119-1131 (2019).
+
+<--- Page Split --->
+
+23. Kimura T, Haga K, Nomura Y, Higaki T, Nakagami H, Sakai T. The Phosphorylation Status of NPH3 Affects Photosensory Adaptation During the Phototropic Response. bioRxiv https://doi.org/10.1101/2020.12.01.407205, (2020).
+24. Glanc M, et al. AGC kinases and MAB4/MEL proteins maintain PIN polarity by limiting lateral diffusion in plant cells. Current biology, doi: 10.1016/j.cub.2021.02.028 (2021).
+25. Barbosa IC, Shikata H, Zoureldou M, Heilmann M, Heilmann I, Schwechheimer C. Phospholipid composition and a polybasic motif determine D6 PROTEIN KINASE polar association with the plasma membrane and tropic responses. Development 143, 4687-4700 (2016).
+26. Brzeska H, Guag J, Remmert K, Chacko S, Korn ED. An experimentally based computer search identifies unstructured membrane-binding sites in proteins: application to class I myosins, PAKS, and CARMIL. J Biol Chem 285, 5738-5747 (2010).
+27. Gronnier J, et al. Structural basis for plant plasma membrane protein dynamics and organization into functional nanodomains. Elife 6, (2017).
+28. Simon ML, et al. A PtdIns(4)P-driven electrostatic field controls cell membrane identity and signalling in plants. Nat Plants 2, 16089 (2016).
+29. Heilmann I. Phosphoinositide signaling in plant development. Development 143, 2044-2055 (2016).
+30. Platre MP, et al. A Combinatorial Lipid Code Shapes the Electrostatic Landscape of Plant Endomembranes. Dev Cell 45, 465-480 e411 (2018).
+31. Tejos R, et al. Bipolar Plasma Membrane Distribution of Phosphoinositides and Their Requirement for Auxin-Mediated Cell Polarity and Patterning in Arabidopsis. Plant Cell 26, 2114-2128 (2014).
+32. Inoue S, Kinoshita T, Takemiya A, Doi M, Shimazaki K. Leaf positioning of Arabidopsis in response to blue light. Mol Plant 1, 15-26 (2008).
+33. Perraki A, et al. Plasma membrane localization of Solanum tuberosum remorin from group 1, homolog 3 is mediated by conformational changes in a novel C-terminal anchor and required for the restriction of potato virus X movement]. Plant Physiol 160, 624-637 (2012).
+34. Jaspert N, Throm C, Oecking C. Arabidopsis 14-3-3 proteins: fascinating and less fascinating aspects. Frontiers in Plant Science 2, 96 (2011).
+35. Bustos DM, Iglesias AA. Intrinsic disorder is a key characteristic in partners that bind 14-3-3 proteins. Proteins 63, 35-42 (2006).
+36. Piovesan D, et al. MobiDB: intrinsically disordered proteins in 2021. Nucleic acids research 49, D361-d367 (2021).
+37. Mergner J, et al. Mass-spectrometry-based draft of the Arabidopsis proteome. Nature 579, 409-414 (2020).
+38. Wang P, et al. Mapping proteome-wide targets of protein kinases in plant stress responses. Proc Natl Acad Sci U S A 117, 3270-3280 (2020).
+
+<--- Page Split --->
+
+39. Coblitz B, Wu M, Shikano S, Li M. C-terminal binding: an expanded repertoire and function of 14-3-3 proteins. FEBS letters 580, 1531-1535 (2006).  
+
+40. Johnson C, Crowther S, Stafford MJ, Campbell DG, Toth R, MacKintosh C. Bioinformatic and experimental survey of 14-3-3-binding sites. Biochem J 427, 69-78 (2010).  
+
+41. Kansup J, Tsugama D, Liu S, Takano T. Arabidopsis G-protein beta subunit AGB1 interacts with NPH3 and is involved in phototroprism. Biochem Biophys Res Commun 445, 54-57 (2014).  
+
+42. Banani SF, Lee HO, Hyman AA, Rosen MK. Biomolecular condensates: organizers of cellular biochemistry. Nat Rev Mol Cell Biol 18, 285-298 (2017).  
+
+43. Zavaliev R, Mohan R, Chen T, Dong X. Formation of NPR1 Condensates Promotes Cell Survival during the Plant Immune Response. Cell 182, 1093-1108 e1018 (2020).  
+
+44. Kimura T, Tsuchida-Mayama T, Imai H, Okajima K, Ito K, Sakai T. Arabidopsis ROOT PHOTOTROPISM2 Is a Light-Dependent Dynamic Modulator of Phototropin1. Plant Cell 32, 2004-2019 (2020).  
+
+45. Emenecker RJ, Holehouse AS, Strader LC. Biological Phase Separation and Biomolecular Condensates in Plants. Annual review of plant biology, DOI: 10.1146/annurev-arplant-081720-015238 (2021).  
+
+46. Ruff KM, Roberts S, Chilkoti A, Pappu RV. Advances in Understanding Stimulus-Responsive Phase Behavior of Intrinsically Disordered Protein Polymers. Journal of molecular biology 430, 4619-4635 (2018).  
+
+47. Kaiserli E, Sullivan S, Jones MA, Feeney KA, Christie JM. Domain swapping to assess the mechanistic basis of Arabidopsis phototropin 1 receptor kinase activation and endocytosis by blue light. Plant Cell 21, 3226-3244 (2009).  
+
+48. Sakamoto K, Briggs WR. Cellular and subcellular localization of phototropin 1. Plant Cell 14, 1723-1735 (2002).  
+
+49. Preuten T, Blackwood L, Christie JM, Fankhauser C. Lipid anchoring of Arabidopsis phototropin 1 to assess the functional significance of receptor internalization: should I stay or should I go? New Phytol 206, 1038-1050 (2015).  
+
+50. Cheng Y, Qin G, Dai X, Zhao Y. NPY1, a BTB-NPH3-like protein, plays a critical role in auxin-regulated organogenesis in Arabidopsis. Proc Natl Acad Sci U S A 104, 18825-18829 (2007).  
+
+51. Furutani M, et al. The gene MACCHI-BOU 4/ENHANCER OF PINOID encodes a NPH3-like protein and reveals similarities between organogenesis and phototropism at the molecular level. Development 134, 3849-3859 (2007).  
+
+52. Treml BS, et al. The gene ENHANCER OF PINOID controls cotyledon development in the Arabidopsis embryo. Development 132, 4063-4074 (2005).  
+
+53. Furutani M, Nakano Y, Tasaka M. MAB4-induced auxin sink generates local auxin gradients in Arabidopsis organ formation. Proc Natl Acad Sci U S A 111, 1198-1203 (2014).
+
+<--- Page Split --->
+
+54. Furutani M, et al. Polar-localized NPH3-like proteins regulate polarity and endocytosis of PIN-FORMED auxin efflux carriers. Development 138, 2069-2078 (2011).55. Blatt MR, Grefen C. Applications of fluorescent marker proteins in plant cell biology. Methods Mol Biol 1062, 487-507 (2014).56. Binder A, et al. A modular plasmid assembly kit for multigene expression, gene silencing and silencing rescue in plants. PLoS One 9, e88218 (2014).57. Hecker A, Wallmeroth N, Peter S, Blatt MR, Harter K, Grefen C. Binary 2in1 Vectors Improve in Planta (Co)localization and Dynamic Protein Interaction Studies. Plant Physiol 168, 776-787 (2015).58. Karimi M, Depicker A, Hilson P. Recombinational cloning with plant gateway vectors. Plant Physiol 145, 1144-1154 (2007).59. Julkowska MM, Rankenberg JM, Testerink C. Liposome-binding assays to assess specificity and affinity of phospholipid-protein interactions. Methods Mol Biol 1009, 261-271 (2013).60. Park M, Touihri S, Muller I, Mayer U, Jurgens G. Sec1/Munc18 protein stabilizes fusion-competent syntaxin for membrane fusion in Arabidopsis cytokinesis. Dev Cell 22, 989-1000 (2012).61. Albert I, et al. An RLP23-SOBIR1-BAK1 complex mediates NLP-triggered immunity. Nature Plants 1, 15140 (2015).62. Schindelin J, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 9, 676-682 (2012).63. Zavaliev R, Epel BL. Imaging callose at plasmodesmata using aniline blue: quantitative confocal microscopy. Methods Mol Biol 1217, 105-119 (2015).
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Fig. 1: NPH3 binds to polyacidic phospholipids via its C-terminal domain. </center>  
+
+(A), (D) Representative confocal microscopy images of leaf epidermal cells from transiently transformed N. benthamiana adapted to darkness (Z-stack projections of NPH3ΔC51 (D) as well as NPH3 variants (NPH3\*) co-expressed with SAC1 variants (SAC1\*) (A) are shown). Scale bars, \(25 \mu \mathrm{m}\). (B) Lipid overlay assay performed with either in vitro transcribed and translated HA:NPH3 and HA:NPH3ΔC51 or purified GST and GST:NPH3-C51 variants. Immunodetection was performed by using anti-HA or anti-GST antibodies, respectively. See main text for abbreviations. (C) Liposome binding assay using large unilamellar liposomes containing the neutral phospholipids PE and PC mixed with either the polyacidic PI4P or PA as specified. Anti-GST immunoblot of GST:NPH3-C51 is shown. (E) Representative immunoblots with anti-GFP after subcellular fractionation of protein extracts prepared from N. benthamiana leaves transiently expressing 35S::GFP:NPH3 variants and adapted to darkness. Proteins in each fraction (7.5 \(\mu \mathrm{g}\)) were separated on 7.5% SDS-PAGE gels. Note that the total amount of soluble proteins (S) is approximately 15 times higher as compared to the total amount of microsomal proteins (M) after 100,000 g centrifugation.
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Fig. 2: An amphipathic helix within the C-terminal domain is required for NPH3 phospholipid binding, membrane association and plasma membrane localization. </center>  
+
+(A) Domain structure and primary sequence of NPH3 showing the two putative BH domains (amphipathic helix and R-rich motif) within the C-terminal region. Stars depict residues of either the R-rich motif or the amphipathic helix substituted by alanine (A) in the NPH3 variants, blue circle depicts the 14-3-3 binding site (see Fig. 3).  
+
+(B) Lipid overlay assay performed with purified GST:NPH3-C51 variants (C51\*).  
+
+(C) Liposome binding assay using large unilamellar liposomes containing the neutral PE and PC mixed with the polyacidic PA. Anti-GST immunoblot of GST:NPH3-C51 variants is shown.  
+
+(D) Representative confocal microscopy images of leaf epidermal cells from transiently transformed N. benthamiana adapted to darkness (Z-stack projections of NPH3-4K/A and NPH3-4WLM/A are shown). Scale bars, 25 μm.
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Fig. 3: Interaction of NPH3 and 14-3-3 proteins is triggered by blue light irradiation and abolished by mutation of the antepenultimate NPH3 residue. </center>  
+
+(A) Yeast two-hybrid interaction analysis of the Arabidopsis 14-3-3 isoform omega with NPH3 wild type and mutant variants (upper panel). Expression of the diverse NPH3 fusion proteins in yeast was confirmed by anti-HA-immunodetection (lower panel). AD, activating domain; BD, binding domain.  
+
+(B, D) In vivo interaction of mCherry:NPH3 variants and 14-3-3 omega:mEGFP in transiently transformed N. benthamiana leaves. Expression of transgenes was driven by the 35S promoter. Freshly transformed tobacco plants were either kept under constant light for 42 h  
+
+(B) or kept under constant light for 24 h and subsequently transferred to darkness for 17h with (BL) or without (D) blue light treatment (5 μmol m⁻² sec⁻¹) for the last 40 minutes (D). The crude extract was immunoprecipitated using GFP beads and separated on 11% SDS-PAGE gels, followed by immunoblotting with anti-GFP and anti-RFP antibodies, respectively.  
+
+(C) Arabidopsis 14-3-3 epsilon interactors were identified by mass spectrometry analysis of anti-GFP immunoprecipitations (two biological replicates) from etiolated seedlings expressing 14-3-3 epsilon:GFP either maintained in darkness or irradiated with blue light (1 μmol m⁻² sec⁻¹) for 30 min. Protein intensities of 14-3-3 client proteins were normalized to relative abundance of the bait protein (Table S1). Fold changes in relative abundance (mean ± SD, logarithmic scale) of blue light treatment versus darkness are given. AHA1, AHA2, Arabidopsis H⁺-ATPase; CINV1, cytosolic invertase 1; EIN2, ethylene insensitive 2; PhyA, phytochrome A; SPS1, sucrose phosphate synthase 1.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Fig. 4: 14-3-3 binding is required for proper NPH3 function in phototropic hypocotyl bending and its light-triggered detachment from the plasma membrane. </center>  
+
+(A) Quantification of the hypocotyl phototropism response (mean \(\pm\) SEM) in 3-days old etiolated seedlings exposed for 12h to unilateral blue light (1 \(\mu\) mol m\(^{-2}\) sec\(^{-1}\)) (n>30 seedlings per experiment, one representative experiment of two replicates is shown). Expression of transgenes in \(nph3-7\) was driven by the 35S promoter. Student's t-test, different letters mark statistically significant differences (P<0.05), same letters mark statistically non-significant differences.  
+
+(B, C, D) Representative confocal microscopy images of hypocotyl cells from transgenic etiolated Arabidopsis \(nph3- 7\) seedlings (B) or of leaf epidermal cells from transiently transformed N. benthamiana (Z-stack projections of BL-treated NPH3 are shown) (C, D). The plants were either kept in darkness (D) or treated with blue light (BL) (N. benthamiana: approx. 11 min and \(nph3- 7\) : approx. 6 min by means of the GFP-laser). Expression of transgenes was driven by the 35S promoter (B, C) or the native NPH3 promoter (D). Scale bars, 25 \(\mu\) m.
+
+<--- Page Split --->
+
+941 (E, F, G) Single-cell time-lapse imaging of RFP:NPH3 condensation induced by GFP-laser 942 treatment. The image of time point 0 image was taken in the absence of the GFP-laser. Z- 943 stack projections from selected time points (E), fluorescence intensity per body (mean 944 ± SEM) (F) and number of bodies (G) are shown. Scale bars, 25 μm. 945
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Fig. 5: The phosphorylation status of the NPH3 14-3-3 binding site is dynamically modulated by the light regime. </center>  
+
+(A, B, C) Immunoblot analysis of total protein extracts (C) or anti- GFP immunoprecipitates and 14- 3- 3 Far- Western (A, B) from Arabidopsis \(nph3 - 7\) ectopically expressing GFP: NPH3 or GFP: NPH3- S744A. 3- days old etiolated seedlings were treated with cycloheximide (100 \(\mu M\) ) for 1 h (B) and either maintained in darkness (D), treated with blue light (BL) (1 \(\mu mol m^{- 2}\) sec \(^{- 1}\) ) for the indicated time (A: 30 min), or re- transferred to darkness (1 h) after 30 min of irradiation (R- D). Proteins were separated on 7.5% SDS- PAGE gels. The upper panel in (B) shows representative confocal microscopy images of hypocotyl cells from transgenic etiolated Arabidopsis seedlings under the specified conditions. Scale bars, 25 \(\mu m\) . (D) Immunoblot analysis of transiently transformed N. benthamiana leaves co- expressing SAC1: RFP with either GFP: NPH3 or GFP: NPH3- S744A and adapted to darkness (see Fig. 1A). Expression of transgenes was driven by the 35S promoter. Total protein extracts were separated on 7.5% SDS- PAGE gels. The closed and open arrowheads indicate the positions of 'generally' phosphorylated and dephosphorylated NPH3 proteins, respectively.
+
+<--- Page Split --->
+![](images/Figure_6.jpg)
+
+<center>Fig. 6: Functional relevance of the subcellular localization of NPH3. </center>  
+
+(A) Quantification of the hypocotyl phototropism response (mean ± SEM) in 3-days old etiolated seedlings exposed for 12 h to unilateral blue light (1 μmol m⁻² sec⁻¹) (n>30 seedlings per experiment, one representative experiment of two replicates is shown). Expression of wild-type and mutant variants of GFP:NPH3 in nph3-7 was driven by the 35S promoter. Student's t-test, different letters mark statistically significant differences (P<0.05), same letters mark statistically non-significant differences. 
+(B) Representative confocal microscopy images of hypocotyl cells from transgenic Arabidopsis nph3-7 seedlings ectopically expressing mutant variants of GFP:NPH3. 3-days old etiolated seedlings were either maintained in darkness (D), treated with blue light (BL) (approx. 11 min by means of the GFP-laser) or re-transferred to darkness (1 h) (R-D) after 30 min of irradiation (1 μmol m⁻² sec⁻¹). Scale bars, 25 μm. 
+(C) Immunoblot analysis of etiolated Arabidopsis nph3-7 seedlings ectopically expressing mutant variants of GFP:NPH3 and treated as described in (B). Total protein extracts were separated on 7.5% SDS-PAGE gels. All samples shown in one panel are from the same blot, the dashed line was inserted to indicate an expected modification of the molecular weight of
+
+<--- Page Split --->
+
+982 NPH3 due to truncations. The closed and open arrowheads indicate the positions of 'generally' phosphorylated and dephosphorylated NPH3 proteins, respectively. 983 (D) In vivo interaction of RFP:NPH3 and phot1:GFP in transiently transformed N. 984 benthamiana leaves adapted to darkness. Expression of transgenes was driven by the 35S promoter. Microsomal proteins were immunoprecipitated using GFP beads and separated on 987 11% SDS-PAGE gels, followed by immunoblotting with anti-GFP and anti-RFP antibodies, respectively. 989 (E) Model depicting the light-regime triggered changes in the phosphorylation status, subcellular localization and phototropic responsiveness of NPH3. BL-induced and 991 phosphorylation-dependent (S744, blue) binding of 14-3-3 proteins releases NPH3 from the 992 PM into the cytosol followed by condensate formation. Residues that are phosphorylated in 993 darkness (yellow) and become dephosphorylated upon light treatment give rise to a shift in 994 electrophoretic mobility ('general' phosphorylation status). Re-transfer to darkness reverts all 995 BL-triggered processes, finally resulting in PM re-association. Cycling of NPH3 between the 996 PM and the cytosol seems to be essential for proper function. Vice versa, NPH3 variants 997 either constitutively attached to (red flash) or constitutively detached (red arrowhead) from 998 the PM are non-functional.
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+<center>Fig. S1: </center>  
+
+Domain structure of NPH3 and MobiDB plot (https://mobidb.org/) of intrinsically disordered regions in NPH3. BTB domain, broad- complex, tramtrack, bric a brac domain.
+
+<--- Page Split --->
+![](images/Figure_unknown_1.jpg)
+
+<center>Fig. S2: </center>  
+
+(A), (D), (E) Representative confocal microscopy images of leaf epidermal cells from transiently transformed N. benthamiana. (A) The plants were either kept in darkness (D) or treated with BL (approx.11 min by means of the GFP-laser, Z-stack projections are shown). (D, E) The plants were adapted to darkness (D: Z-stack projections). Scale bars, \(25 \mu \mathrm{m}\). (B) BH score profiles (window size 11) of NPH3 and mutant variants. Putative BH-domains are indicated by red arrows.
+
+<--- Page Split --->
+
+1016 (C) Helical wheel projection showing amphipathy of the predicted helix (residues 700- 713) 1017 within the C-terminal domain of NPH3. Overall helix hydrophobicity (H) and the hydrophobic 1018 moment (μH) are given for NPH3 and mutant variants. 1019
+
+<--- Page Split --->
+![](images/Figure_unknown_2.jpg)
+
+<center>Fig. S3: </center>  
+
+(A) Quantification of the hypocotyl phototropism response (mean ± SEM) in 3-days old etiolated seedlings exposed for 12h to unilateral blue light (1 \(\mu\) mol m\(^{-2}\) sec\(^{-1}\)) (n>30 seedlings per experiment, one representative experiment of two replicates is shown). Expression of transgenes in \(nph3-7\) was driven by the \(NPH3\) promoter. Student's t-test, different letters mark statistically significant differences (P<0.05), same letters mark statistically non-significant differences.  
+
+(B, C, D) Representative confocal microscopy images of hypocotyl cells from transgenic etiolated Arabidopsis \(nph3- 7\) seedlings (B, C) or of leaf epidermal cells from transiently transformed N. benthamiana (here, Z-stack projections are shown) (D). The plants were either kept in darkness (D) or treated with blue light (BL) (nph3- 7: 1 \(\mu\) mol m\(^{-2}\) sec\(^{-1}\) and N. benthamiana: 10 \(\mu\) mol m\(^{-2}\) sec\(^{-1}\)) for 40 min. Expression of transgenes was driven by the \(NPH3\) promoter (B) or the 35S promoter (C, D). Scale bars, 25 \(\mu\) m.
+
+<--- Page Split --->
+![](images/Figure_unknown_3.jpg)
+
+<center>Fig. S4: </center>  
+
+Yeast two- hybrid interaction analysis of the Arabidopsis 14- 3- 3 isoform omega with various NRL wild type and mutant variants (exchange of the antepenultimate residue (serine), respectively). Yeast growth was recorded after 3 days (upper panel) or 5 days (lower panel).
+
+<--- Page Split --->
+
+1041 Table S1: 
+
+1042 Analysis of 14-3-3 epsilon-GFP immunoprecipitates via mass spectrometry (MS) based on
+1043 two biological replicates. This table lists only known 14-3-3 clients in addition to NPH3. 
+
+1044 
+
+<table><tr><td rowspan="2">AGI code</td><td rowspan="2">gene name</td><td rowspan="2">description</td><td>Mol. weight (kDa)</td><td colspan="2">peptides R1</td><td colspan="2">Sequence coverage R1 (%)</td><td colspan="2">Intensity R1</td><td colspan="2">Normalized intensity R1</td><td rowspan="2">Intensity (BL/dark) R1</td></tr><tr><td>28,9</td><td>dark</td><td>BL</td><td>dark</td><td>BL</td><td>dark</td><td>BL</td><td>dark</td><td>BL</td></tr><tr><td>AT1G22300</td><td>GRF10</td><td>14-3-3-like protein GF14 epsilon</td><td>28,9</td><td>32</td><td>30</td><td>79,5</td><td>79,5</td><td>3,230E+11</td><td>3,740E+11</td><td>3E+11</td><td>3E+11</td><td>1</td></tr><tr><td>AT1G35580</td><td>CINV1</td><td>Alkaline/neutral invertase CINV1</td><td>62,834</td><td>35</td><td>32</td><td>64,2</td><td>64,2</td><td>3,4953E+10</td><td>2,9811E+10</td><td>3,245E+10</td><td>2,391E+10</td><td>0,735</td></tr><tr><td>AT2G18960</td><td>AHA1</td><td>ATPase 1, plasma membrane-type</td><td>104,22</td><td>31</td><td>36</td><td>36,4</td><td>43,4</td><td>2970100000</td><td>4414600000</td><td>2760000000</td><td>3540000000</td><td>1,28</td></tr><tr><td>AT4G30190</td><td>AHA2</td><td>ATPase 2, plasma membrane-type</td><td>104,4</td><td>33</td><td>37</td><td>39,9</td><td>42,6</td><td>401660000</td><td>380890000</td><td>372900000</td><td>305450000</td><td>0,819</td></tr><tr><td>AT5G84330</td><td>NPH3</td><td>Non-phototropic hypocaly 3</td><td>81,872</td><td>5</td><td>22</td><td>7,4</td><td>39</td><td>53537000</td><td>1145400000</td><td>49700000</td><td>919000000</td><td>18,5</td></tr><tr><td>AT1G09570</td><td>PhyA</td><td>Phytochrome A</td><td>125,02</td><td>1</td><td>1</td><td>0,9</td><td>1,2</td><td>4661700</td><td>5509900</td><td>4328000</td><td>4419000</td><td>1,02</td></tr><tr><td>AT5G11110</td><td>SPS1</td><td>Sucrose-phosphate synthase 1</td><td>117,32</td><td>13</td><td>18</td><td>15,6</td><td>22,7</td><td>246270000</td><td>519340000</td><td>228700000</td><td>416500000</td><td>1,819</td></tr><tr><td>AT5G03280</td><td>EN2</td><td>Ethylene-insensitive protein 2</td><td>140,95</td><td>6</td><td>9</td><td>6,6</td><td>9</td><td>81033000</td><td>136130000</td><td>75240000</td><td>109200000</td><td>1,45</td></tr><tr><td>AT3G45780</td><td>Phot1</td><td>Phototropin-1</td><td>111,69</td><td>10</td><td>14</td><td>8,8</td><td>14,7</td><td>169850000</td><td>316360000</td><td>157700000</td><td>253700000</td><td>1,61</td></tr></table>
+
+<table><tr><td>AT1G22300</td><td>GRF10</td><td>14-3-3-like protein GF14 epsilon</td><td>28,9</td><td>33</td><td>33</td><td>76,4</td><td>76,4</td><td>3,3401E+11</td><td>3,9971E+11</td><td>3E+11</td><td>3E+11</td><td>1</td></tr></table>
+
+1046 
+
+1047 
+
+1048 
+
+1049 
+
+1050 
+
+1051 
+
+1052 
+
+1053 
+
+1054 
+
+1055 
+
+1056 
+
+1057 
+
+1058 
+
+1059 
+
+1060 
+
+1061 
+
+1062 
+
+1063 
+
+1064 
+
+1065 
+
+1066 
+
+1067 
+
+1068 
+
+1069 
+
+1070 
+
+1071 
+
+1072 
+
+1073 
+
+1074 
+
+1075 
+
+1076 
+
+1077 
+
+1078 
+
+1079 
+
+1080 
+
+1081 
+
+1082 
+
+1083 
+
+1084 
+
+1085 
+
+1086 
+
+1087 
+
+1088 
+
+1089 
+
+1090 
+
+1101 
+
+1111 
+
+
+
+<--- Page Split --->
+
+## Figures  
+
+![](images/Figure_1.jpg)
+
+<center>Figure 1 </center>  
+
+NPH3 binds to polyacidic phospholipids via its C- terminal domain. (A), (D) Representative confocal microscopy images of leaf epidermal cells from transiently transformed N. benthamiana adapted to darkness (Z- stack projections of NPH3DC51 (D) as well as NPH3 variants (NPH3\*) co- expressed with SAC1 variants (SAC1\*) (A) are shown). Scale bars, \(25 \mu \mathrm{m}\) . (B) Lipid overlay assay performed with either in vitro transcribed and translated HA:NPH3 and HA:NPH3DC51 or purified GST and GST:NPH3- C51 variants. Immunodetection was performed by using anti- HA or anti- GST antibodies, respectively. See main text for abbreviations. (C) Liposome binding assay using large unilamellar liposomes containing the neutral phospholipids PE and PC mixed with either the polyacidic PI4P or PA as specified. Anti- GST immunoblot of GST:NPH3- C51 is shown. (E) Representative immunoblots with anti- GFP after subcellular fractionation of protein extracts prepared from N. benthamiana leaves transiently expressing 35S::GFP:NPH3 variants and adapted to darkness. Proteins in each fraction (7.5 \(\mu \mathrm{g}\) ) were separated on 7.5% SDS- PAGE gels. Note that the total amount of soluble proteins (S) is approximately 15 times higher as compared to the total amount of microsomal proteins (M) after 100,000 g centrifugation.
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2 </center>  
+
+An amphipathic helix within the C- terminal domain is required for NPH3 phospholipid binding, membrane association and plasma membrane localization. (A) Domain structure and primary sequence of NPH3 showing the two putative BH domains (amphipathic helix and R- rich motif) within the C- terminal region. Stars depict residues of either the R- rich motif or the amphipathic helix substituted by alanine (A) in the NPH3 variants, blue circle depicts the 14- 3- 3 binding site (see Fig. 3). (B) Lipid overlay assay performed with purified GST: NPH3- C51 variants (C51\*). (C) Liposome binding assay using large unilamellar liposomes containing the neutral PE and PC mixed with the polyacidic PA. Anti- GST immunoblot of GST: NPH3- C51 variants is shown. (D) Representative confocal microscopy images of leaf epidermal cells from transiently transformed N. benthamiana adapted to darkness (Z- stack projections of NPH3- 4K/A and NPH3- 4WLM/A are shown). Scale bars, 25 μm.
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+<center>Figure 3 </center>  
+
+Interaction of NPH3 and 14- 3- 3 proteins is triggered by blue light irradiation and abolished by mutation of the antepenultimate NPH3 residue. (A) Yeast two- hybrid interaction analysis of the Arabidopsis 14- 3- 3 isoform omega with NPH3 wild type and mutant variants (upper panel). Expression of the diverse NPH3 fusion proteins in yeast was confirmed by anti- HA- immunodetection (lower panel). AD, activating domain; BD, binding domain. (B, D) In vivo interaction of mCherry: NPH3 variants and 14- 3- 3 omega:mEGFP in transiently transformed N. benthamiana leaves. Expression of transgenes was driven by the 35S promoter. Freshly transformed tobacco plants were either kept under constant light for 42 h (B) or kept under constant light for 24 h and subsequently transferred to darkness for 17h with (BL) or without (D)
+
+<--- Page Split --->
+
+blue light treatment (5 \(\mu\) mol m- 2 sec- 1) for the last 40 minutes (D). The crude extract was immunoprecipitated using GFP beads and separated on 11% SDS- PAGE gels, followed by immunoblotting with anti- GFP and anti- RFP antibodies, respectively. (C) Arabidopsis 14- 3- 3 epsilon interactors were identified by mass spectrometry analysis of anti- GFP immunoprecipitations (two biological replicates) from etiolated seedlings expressing 14- 3- 3 epsilon:GFP either maintained in darkness or irradiated with blue light (1 \(\mu\) mol m- 2 sec- 1) for 30 min. Protein intensities of 14- 3- 3 client proteins were normalized to relative abundance of the bait protein (Table S1). Fold changes in relative abundance (mean \(\pm\) SD, logarithmic scale) of blue light treatment versus darkness are given. AHA1, AHA2, Arabidopsis H+-ATPase; CINV1, cytosolic invertase 1; EIN2, ethylene insensitive 2; PhyA, phytochrome A; SPS1, sucrose phosphate synthase 1.
+
+<--- Page Split --->
+![](images/Figure_4.jpg)
+
+<center>Figure 4 </center>  
+
+14- 3- 3 binding is required for proper NPH3 function in phototropic hypocotyl bending and its light- triggered detachment from the plasma membrane. (A) Quantification of the hypocotyl phototropism response (mean ± SEM) in 3- days old etiolated seedlings exposed for 12h to unilateral blue light (1 μmol m- 2 sec- 1) (n>30 seedlings per experiment, one representative experiment of two replicates is shown). Expression of transgenes in nph3- 7 was driven by the 35S promoter. Student's t- test, different letters mark
+
+<--- Page Split --->
+
+statistically significant differences (P<0.05), same letters mark statistically non- significant differences. (B, C, D) Representative confocal microscopy images of hypocotyl cells from transgenic etiolated Arabidopsis nph3- 7 seedlings (B) or of leaf epidermal cells from transiently transformed N. benthamiana (Z- stack projections of BL- treated NPH3 are shown) (C, D). The plants were either kept in darkness (D) or treated with blue light (BL) (N. benthamiana: approx. 11 min and nph3- 7: approx. 6 min by means of the GFP- laser). Expression of transgenes was driven by the 35S promoter (B, C) or the native NPH3 promoter (D). Scale bars, 25 μm. (E, F, G) Single- cell time- lapse imaging of RFP: NPH3 condensation induced by GFP- laser treatment. The image of time point 0 image was taken in the absence of the GFP- laser. Z stack projections from selected time points (E), fluorescence intensity per body (mean ± SEM) (F) and number of bodies (G) are shown. Scale bars, 25 μm. 
+
+![](images/Figure_5.jpg)
+
+ 
+
+<center>Figure 5</center> 
+
+The phosphorylation status of the NPH3 14-3-3 binding site is dynamically modulated by the light regime. (A, B, C) Immunoblot analysis of total protein extracts (C) or anti-GFP immunoprecipitates and 14-3-3 Far-Western (A, B) from Arabidopsis nph3-7 ectopically expressing GFP:NPH3 or GFP:NPH3-S744A. 3-days old etiolated seedlings were treated with cycloheximide (100 μM) for 1 h (B) and either maintained in darkness (D), treated with blue light (BL) (1 μmol m-2 sec-1) for the indicated time (A: 30 min), or re-transferred to darkness (1 h) after 30 min of irradiation (R-D). Proteins were separated on 7.5% SDS-PAGE gels. The upper panel in (B) shows representative confocal microscopy images of hypocotyl cells from transgenic etiolated Arabidopsis seedlings under the specified conditions. Scale bars, 25 μm. (D)
+
+<--- Page Split --->
+
+Immunoblot analysis of transiently transformed N. benthamiana leaves co-expressing SAC1:RFP with either GFP:NPH3 or GFP:NPH3-S744A and adapted to darkness (see Fig. 1A). Expression of transgenes was driven by the 35S promoter. Total protein extracts were separated on 7.5% SDS-PAGE gels. The closed and open arrowheads indicate the positions of 'generally' phosphorylated and dephosphorylated NPH3 proteins, respectively. 
+
+![](images/Figure_6.jpg)
+
+ 
+
+<center>Figure 6</center>
+
+<--- Page Split --->
+
+Functional relevance of the subcellular localization of NPH3. (A) Quantification of the hypocotyl phototropism response (mean \(\pm\) SEM) in 3- days old etiolated seedlings exposed for 12 h to unilateral blue light (1 \(\mu\) mol m- 2 sec- 1) (n>30 seedlings per experiment, one representative experiment of two replicates is shown). Expression of wild- type and mutant variants of GFP:NPH3 in nph3- 7 was driven by the 35S promoter. Student's t- test, different letters mark statistically significant differences (P<0.05), same letters mark statistically non- significant differences. (B) Representative confocal microscopy images of hypocotyl cells from transgenic Arabidopsis nph3- 7 seedlings ectopically expressing mutant variants of GFP:NPH3. 3- days old etiolated seedlings were either maintained in darkness (D), treated with blue light (BL) (approx. 11 min by means of the GFP- laser) or re- transferred to darkness (1 h) (R- D) after 30 min of irradiation (1 \(\mu\) mol m- 2 sec- 1). Scale bars, 25 \(\mu\) m. (C) Immunolobt analysis of etiolated Arabidopsis nph3- 7 seedlings ectopically expressing mutant variants of GFP:NPH3 and treated as described in (B). Total protein extracts were separated on 7.5% SDS- PAGE gels. All samples shown in one panel are from the same blot, the dashed line was inserted to indicate an expected modification of the molecular weight of NPH3 due to truncations. The closed and open arrowheads indicate the positions of 'generally' phosphorylated and dephosphorylated NPH3 proteins, respectively. (D) In vivo interaction of RFP:NPH3 and phot1:GFP in transiently transformed N. benthamiana leaves adapted to darkness. Expression of transgenes was driven by the 35S promoter. Microsomal proteins were immunoprecipitated using GFP beads and separated on 11% SDS- PAGE gels, followed by immunoblotting with anti- GFP and anti- RFP antibodies, respectively. (E) Model depicting the light- regime triggered changes in the phosphorylation status, subcellular localization and phototropic responsiveness of NPH3. BL- induced and phosphorylation- dependent (S744, blue) binding of 14- 3- 3 proteins releases NPH3 from the PM into the cytosol followed by condensate formation. Residues that are phosphorylated in darkness (yellow) and become dephosphorylated upon light treatment give rise to a shift in electrophoretic mobility ('general' phosphorylation status). Re- transfer to darkness reverts all BL- triggered processes, finally resulting in PM re- association. Cycling of NPH3 between the PM and the cytosol seems to be essential for proper function. Vice versa, NPH3 variants either constitutively attached to (red flash) or constitutively detached (red arrowhead) from the PM are non- functional.  
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+- VideoS1.mov- VideoS2.mov- VideoS3.mov- VideoS4.mov- VideoS5.mov- VideoS6.mov- VideoS7.mov
+
+<--- Page Split --->
+
+- VideoS8.mov- VideoS9.mov
+
+<--- Page Split --->

preprint/preprint__996236fe7bcffb4d384731573a4c082ea4a44831dd06c020baf33b3c539e363c/images_list.json

@@ -0,0 +1,93 @@
+[
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Figure 1 | Quantitative temporal proteomic analysis of MVA infection. a Schematic indicating the experimental workflow. b Hierarchical clustering of all proteins quantified in the two biological repeats. An enlargement is shown indicating groups of proteins that were significantly down- or upregulated during the course of the experiment. c Scatterplot of all proteins quantified at 18 h of infection, showing average fold change. p-values were estimated using Significance B values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>56,57</sup>.",
+    "footnote": [],
+    "bbox": [
+      [
+        125,
+        120,
+        538,
+        610
+      ]
+    ],
+    "page_idx": 5
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Figure 2 | MVA regulates multiple mediators of intrinsic, innate and adaptive immunity. a Functional enrichment analysis of proteins downregulated >2-fold at ≥1 time point during infection. b Example components from cell surface-related pathways identified in (a). Error bars = range. c Functional enrichment analysis of groups of proteins regulated as indicated in the left-hand column, and examples of components of enriched pathways. Full data is shown in Supplementary Table 2. 2-fold is used throughout the manuscript as a cutoff for downregulation, with proteins unchanged defined as downregulated <1.25 fold at all time points measured. Error bars = range. d Plot of interactions among proteins decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples (red text = proteins connected by red edges). For context, an additional 74 neighbouring proteins are shown (grey text). Relative proximity to the proteins that were decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples is indicated by red shading, and was quantified via random walk with restart (Methods).",
+    "footnote": [],
+    "bbox": [
+      [
+        152,
+        117,
+        571,
+        650
+      ]
+    ],
+    "page_idx": 6
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_1.jpg",
+    "caption": "Extended Data Figure 1. Percentage of cells expressing viral proteins at different MOIs. HFFT-TERTs were infected at the indicated MOI for 18h, then stained with a polyclonal rabbit anti-VACV antibody.",
+    "footnote": [],
+    "bbox": [
+      [
+        118,
+        112,
+        933,
+        616
+      ]
+    ],
+    "page_idx": 7
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_unknown_0.jpg",
+    "caption": "b",
+    "footnote": [],
+    "bbox": [
+      [
+        275,
+        120,
+        800,
+        500
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_2.jpg",
+    "caption": "Extended Data Figure 2. Dot plot of all proteins quantified. Average fold change was calculated from each of two biological replicates. a. Human proteins. Significance B was used to estimate p value, and was adjusted for multiple hypothesis testing using the method of Benjamini-Hochberg<sup>60,57</sup>. b viral proteins, p values not displayed.",
+    "footnote": [],
+    "bbox": [
+      [
+        275,
+        545,
+        762,
+        956
+      ]
+    ],
+    "page_idx": 8
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_3.jpg",
+    "caption": "Figure 3 | Differential protein changes during infection of human fibroblasts with MVA or VACV-WR. a Scatterplot of all proteins quantified at 18 h of infection with both viruses. p-values were estimated using Significance A values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>65,67</sup>. Example profiles for NF-κB pathway proteins are shown in the lower panels. Error bars = range. b Functional enrichment analysis of proteins (i) downregulated by MVA but not VACV Western Reserve (VACV-WR), (ii) downregulated by VACV-WR but not MVA, or (iii) downregulated by both MVA and VACV-WR. c Example components from pathways identified in (b). Error bars = range.",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 9
+  },
+  {
+    "type": "image",
+    "img_path": "images/Figure_5.jpg",
+    "caption": "Extended Data Figure 5. Number of temporal classes of MVA gene expression.",
+    "footnote": [],
+    "bbox": [],
+    "page_idx": 12
+  }
+]

preprint/preprint__996236fe7bcffb4d384731573a4c082ea4a44831dd06c020baf33b3c539e363c/preprint__996236fe7bcffb4d384731573a4c082ea4a44831dd06c020baf33b3c539e363c.mmd

@@ -0,0 +1,332 @@
+
+# Quantitative temporal analysis of modified vaccinia Ankara, the monkeypox and smallpox vaccine  
+
+Jonas Dutra Albaraz University of Cambridge  
+
+Marisa Oliveira University of Cambridge  
+
+Joanne Kite University of Cambridge  
+
+Joao Paulo Harvard Medical School https://orcid.org/0000- 0002- 4291- 413X  
+
+Robin Antrobus University of Cambridge  
+
+Steven Gygi Harvard Medical School https://orcid.org/0000- 0001- 7626- 0034  
+
+Edward Huttlin Harvard Medical School https://orcid.org/0000- 0002- 1822- 1173  
+
+Geoffrey Smith University of Cambridge https://orcid.org/0000- 0002- 3730- 9955  
+
+Michael Weekes (mpw1001@cam.ac.uk) University of Cambridge https://orcid.org/0000- 0003- 3196- 5545  
+
+## Biological Sciences - Article  
+
+Keywords: modified vaccinia Ankara, vaccinia virus, monkeypox, innate immunity, immune evasion, host- pathogen interaction, proteomics  
+
+Posted Date: July 13th, 2022  
+
+DOI: https://doi.org/10.21203/rs.3.rs- 1850393/v1  
+
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+
+Quantitative temporal analysis of modified vaccinia Ankara, the monkeypox and smallpox vaccine  
+
+3 Jonas D. Albaranza<sup>a,b,#</sup>, Marisa Oliveira<sup>a,b,#</sup>, Joanne Kite<sup>a,b,#</sup>, Joao A. Paulo<sup>c</sup>, Robin Antrobus<sup>a,b</sup>, 4 Steven P. Gygi<sup>c</sup>, Edward L. Huttlin<sup>c</sup>, Geoffrey L. Smith<sup>d</sup>, Michael P. Weekes<sup>a,b</sup>.  
+
+5  
+
+## Affiliations:  
+
+a Cambridge Institute for Medical Research, University of Cambridge, Cambridge CB2 0XY, UK  b Department of Medicine, University of Cambridge, Cambridge, CB2 0XY, UK  c Department of Cell Biology, Harvard Medical School, 240 Longwood Avenue, Boston, MA 02115  USA  
+
+d Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK  
+
+# Contributed equally  
+
+Corresponding author  
+
+Correspondence: Michael P. Weekes  
+
+Cambridge Institute for Medical Research, University of Cambridge, Keith Peters Building, Hills  
+
+Road, Cambridge CB2 0XY, UK  
+
+Email: mpw1001@cam.ac.uk  
+
+Telephone: +44 1223 767811  
+
+21  
+
+Keywords: modified vaccinia Ankara; vaccinia virus; monkeypox; innate immunity; immune  
+
+evasion; host- pathogen interaction; proteomics
+
+<--- Page Split --->
+
+## 24 Summary  
+
+Modified vaccinia Ankara (MVA) immunisation is being deployed to curb the current outbreak of monkeypox in multiple countries<sup>1</sup>. Originally authorized for vaccination against smallpox, MVA is a vaccinia virus (VACV) strain that does not replicate in human cells or cause serious adverse events. Here, we conducted a highly multiplexed proteomic analysis<sup>2</sup> to quantify \(\sim 7,500\) cellular proteins and \(\sim 80\%\) of viral proteins at five time points throughout MVA infection of human cells<sup>3</sup>. \(>380\) human proteins were down- regulated \(>2\) - fold by MVA, revealing a profound remodelling of the host proteome. \(>25\%\) of these MVA targets, including multiple components of the nuclear pore complex (NPC), were not shared with VACV- Western Reserve<sup>4</sup>, which is derived from a first generation smallpox vaccine associated with serious adverse events. Using pharmacological inhibition of viral DNA replication and killed virions, we discovered that post- replicative gene expression is necessary for the downregulation of NPC proteins and for elements of MVA antagonism of innate immune sensing. Our approach thus provides the first global view of the impact of MVA infection on the host proteome, offers insights into how MVA interacts with the antiviral defences and identifies cellular mechanisms that may underpin the abortive infection of human cells. These discoveries will prove vital to the rational design of future generations of vaccines.
+
+<--- Page Split --->
+
+## 41 Introduction  
+
+Monkeypox virus (MPXV) is a zoonotic orthopoxvirus endemic in Central and West Africa<sup>5,6</sup>. Since May 2022, thousands of cases of MPXV have been reported in 50 non- endemic countries worldwide, with sustained human- to- human transmission<sup>7</sup>. MPXV causes a smallpox- like illness, with severe disease seen in immunocompromised individuals, children and pregnant women. The ongoing outbreak has been caused by MPXV clade 3 (within the formerly designated "West African" clade)<sup>1</sup>. Mortality is commoner with MPXV clade 1 (formerly designated "Central African" or "Congo Basin" clade), however up to \(3.6\%\) of infections with clade 3 MPXV have previously resulted in death, albeit in some cases in the context of HIV co- infection<sup>8</sup>.  
+
+Although the World Health Organisation (WHO) has not yet advised mass vaccination, post- exposure prophylaxis is recommended for higher- risk MPXV contacts, and pre- exposure prophylaxis for at- risk healthcare workers<sup>7</sup>. Highly effective protection can be provided by live vaccines originally developed against smallpox, which was itself eradicated through vaccination with a related orthopoxvirus, VACV<sup>9</sup>. The most recent generation of smallpox vaccines derive from modified vaccinia Ankara (MVA), itself derived by \(>570\) passages of the VACV strain chorioallantois vaccinia Ankara (CVA) in chicken embryo fibroblasts<sup>10- 12</sup>. Serial passage resulted in loss of \(\sim 30\) kb of the genome, and loss of replicative capacity in human cells<sup>3</sup>. Because MVA has been shown to be safe and immunogenic in both healthy and immunocompromised individuals<sup>13,14</sup>, it has also been investigated extensively as vaccine vector for viruses including Ebola, respiratory syncytial virus, HIV, and SARS- CoV- 2<sup>15</sup>.  
+
+However, the biological effects of MVA at the cellular level are poorly understood, without prior systematic proteomic investigation. Viral infection induces dynamic changes in the host proteome with diverse functional consequences for virus- host interaction<sup>2</sup>, but little information is available about non- productive, abortive infections. We therefore conducted the first systematic proteomic analysis of MVA and host throughout the whole course of infection, including inactivated controls to understand the contribution of the viral particle with no- , or limited viral gene expression.
+
+<--- Page Split --->
+
+## Results  
+
+## Quantitative temporal analysis of MVA infection  
+
+To build a global picture of changes in host and viral proteins throughout the course of MVA infection, we infected telomerase reverse transcriptase (TERT)- immortalized primary human fetal foreskin fibroblasts (HFFF- TERTs) with MVA at multiplicity of infection (MOI) 5 in biological duplicate. HFFF- TERTs have been well established as a model for a variety of different viral infections \(^{2,4,16,17}\) , and MOI 5 infected \(>96\%\) of cells (Extended data Fig. 1). We used an MVA stock derived from the original MVA strain that had undergone 575 passages in vitro, since this is the direct precursor to vaccines in current use or development, including MVA- Bavarian Nordic (MVA- BN) \(^{11}\) . 16- plex TMT and triple- stage mass spectrometry (MS3) with real- time search (RTS) \(^{18,19}\) quantified protein expression over the full course of infection (Fig. 1a). In the same experiment, we included identical analyses with heat- inactivated MVA (MVA- HI), or the viral DNA replication inhibitor cytosine arabinoside (AraC), to understand the contribution of the viral particle with no gene expression, or expression of early viral genes only, respectively (Fig. 1a).  
+
+7545 human and 144 viral proteins were quantified, providing a global view of changes in protein expression during infection. The greatest changes occurred late during infection, and heat inactivation ablated the majority of host protein downregulation (Fig. 1b, Extended data Fig. 2). All data are shown in Supplementary Table 1, in which the worksheet "Plotter" enables interactive generation of temporal graphs of the expression of each of the human or viral proteins quantified.  
+
+## MVA regulates multiple mediators of intrinsic, innate and adaptive immunity  
+
+Over 18 h of infection, 383 human proteins were downregulated \(>2\) - fold, and 66 human proteins upregulated \(>2\) - fold. Interestingly, Database for Annotation, Visualization and Integrated Discovery (DAVID) software \(^{20}\) revealed that diverse groups of cell surface proteins were downregulated early during MVA infection including multiple NK and T- cell ligands suggesting that a key focus of the first phase of infection may be evasion of cellular immunity (Fig. 2a- b). Upregulated proteins included inflammatory mediators such as complement C4- A and lactotransferrin (LTF) \(^{21}\) (Extended data Fig. 3A). To distinguish with high confidence host factors
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Figure 1 | Quantitative temporal proteomic analysis of MVA infection. a Schematic indicating the experimental workflow. b Hierarchical clustering of all proteins quantified in the two biological repeats. An enlargement is shown indicating groups of proteins that were significantly down- or upregulated during the course of the experiment. c Scatterplot of all proteins quantified at 18 h of infection, showing average fold change. p-values were estimated using Significance B values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>56,57</sup>. </center>
+
+<--- Page Split --->
+![](images/Figure_2.jpg)
+
+<center>Figure 2 | MVA regulates multiple mediators of intrinsic, innate and adaptive immunity. a Functional enrichment analysis of proteins downregulated >2-fold at ≥1 time point during infection. b Example components from cell surface-related pathways identified in (a). Error bars = range. c Functional enrichment analysis of groups of proteins regulated as indicated in the left-hand column, and examples of components of enriched pathways. Full data is shown in Supplementary Table 2. 2-fold is used throughout the manuscript as a cutoff for downregulation, with proteins unchanged defined as downregulated <1.25 fold at all time points measured. Error bars = range. d Plot of interactions among proteins decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples (red text = proteins connected by red edges). For context, an additional 74 neighbouring proteins are shown (grey text). Relative proximity to the proteins that were decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples is indicated by red shading, and was quantified via random walk with restart (Methods). </center>
+
+<--- Page Split --->
+![](images/Figure_1.jpg)
+
+<center>Extended Data Figure 1. Percentage of cells expressing viral proteins at different MOIs. HFFT-TERTs were infected at the indicated MOI for 18h, then stained with a polyclonal rabbit anti-VACV antibody. </center>
+
+<--- Page Split --->
+![](images/Figure_unknown_0.jpg)
+
+<center>b </center>  
+
+![](images/Figure_2.jpg)
+
+<center>Extended Data Figure 2. Dot plot of all proteins quantified. Average fold change was calculated from each of two biological replicates. a. Human proteins. Significance B was used to estimate p value, and was adjusted for multiple hypothesis testing using the method of Benjamini-Hochberg<sup>60,57</sup>. b viral proteins, p values not displayed. </center>
+
+<--- Page Split --->
+![](images/Figure_3.jpg)
+
+
+<--- Page Split --->
+
+that were downregulated by viral gene expression at distinct stages of infection, we applied a series of filters that identified differential modulation of host protein expression in the presence of unmodified MVA, MVA+ArA C and MVA- HI (Fig. 2c, Extended data Fig. 3, Supplementary Table 2). Of particular interest, viral genes expressed later during infection exquisitely regulated multiple components of the nuclear pore complex. This was unexpected, since a recent study suggested that nuclear pore proteins may be essential for replication of poxviruses<sup>22</sup>. Proteins downregulated by early- expressed genes included secreted inflammatory mediators such as metallopeptidase inhibitor 2 (TIMP2) and transforming growth factor \(\beta - 1\) (TGFB1). Infection with heat- inactivated virus offered the opportunity to identify proteins that were upregulated upon sensing of the viral particle, but whose expression was limited by viral genes (Fig. 2c, lower panels). These included four poly ADP- ribose polymerase (PARP) proteins including Zinc Finger Antiviral Protein (ZC3HAV1, ZAP), a variety of proteins with key roles in immunity including the viral DNA sensor IFI16, interferon regulator TRIM26 and interferon stimulated genes including IFIT2 and OASL. Certain factors were downregulated in all three conditions, including extracellular matrix components and multiple collagens (Extended data Fig. 3B), suggesting that these changes may represent a cellular response to infection, as opposed to being induced by viral gene expression.  
+
+To understand better how proteins altered during various stages of MVA infection relate to one another and to the larger proteome, we superimposed groups of proteins displaying each characteristic abundance signature (Fig. 2c) onto the BioPlex network of human protein- protein interactions<sup>23</sup>. Graph assortativity was then calculated, measuring the tendency of proteins in a particular group to interact preferentially with each other compared to proteins that are not part of the selected group (Extended data Fig. 3c). Proteins downregulated during later stages of MVA infection showed a strong tendency to self- associate, including nuclear pore components, and a selection of nucleolar proteins associated with ribosome biogenesis (e.g. SURF6, RBM28, RPL7L1, ZC3HAV1, CDK105) (Fig. 2e).  
+
+## Differential regulation of the host proteome by MVA and virulent VACV  
+
+To discover what may underpin the differences in pathogenesis and replication between MVA vaccine, which is avirulent and replication- incompetent in human cells, and the mouse- adapted, virulent, and replication- competent VACV strain Western Reserve (VACV- WR), we next
+
+<--- Page Split --->
+
+characterised similarities and difference in host regulation. We compared results from the present study with our prior proteomic analysis of VACV- WR<sup>4</sup>, which is the reference for studies with orthopoxviruses, is highly neurotropic in mice, and was derived from the New York City Board of Health first- generation smallpox vaccine. Whereas at least \(30\%\) of proteins were co- regulated by both viruses, there were a number of key differences (Fig. 3a- b, Extended data Fig. 4a, Supplemental Table 3). Proteins substantially downregulated during MVA infection but minimally modulated by VACV- WR were enriched in interacting components of the nuclear pore complex and certain glycoproteins. The latter included multiple immune modulators such as semaphorins, which modulate intercellular communication between immune cells<sup>24</sup>, transforming growth factors TGFB1 and TGFB2, and the endoplasmic reticulum stress sensor EIF2AK3 (also known as PERK) (Fig. 3a- c, Extended data Fig. 4b- d). MVA, but not VACV- WR, also induced the downregulation of negative (NFKBIA and NFKBIB, also known as IκBα and IκBβ, respectively) and positive (BIRC2, also known as cIAP1) regulators of NF- κB immune signalling (Fig. 3a, Extended data Fig. 4c- d). IκBα sequester inactive NF- κB in the cytoplasm and their proteasomal degradation releases NF- κB components to accumulate in the nucleus where they activate the expression of immunity- related genes including interferons<sup>25</sup>. Therefore, the down- regulation of NFKBIA and NFKBIB indicate activation of NF- κB during MVA, but not VACV- WR, infection. The presence of NFKBIA re- synthesis during infection with MVA+ArαC but not MVA infection also indicate an MVA blockade of the NF- κB pathway at or downstream of IκBα degradation. Conversely, proteins down- regulated during VACV- WR infection but minimally modulated by MVA were enriched in factors involved in tyrosine phosphorylation and innate immunity. The former pathway included members of the Ephrin family of receptor tyrosine kinases (EPHA4, EPHB2, and EPHB3) and the proto- oncogene FER whereas the latter included multiple interferon- stimulated genes and MAP4K2, an upstream activator of mitogen- activated protein kinases, key regulators of immune signaling downstream of pathogen sensing<sup>26</sup> (Fig. 3b, c, Extended data Fig. 3C).  
+
+## Temporal analysis of MVA protein expression to inform viral-host interactions  
+
+High- definition temporal analysis of viral protein expression can facilitate direct correlation between viral and cellular protein profiles to give insights into viral- host protein interaction<sup>4</sup>. We quantified \(77\%\) of all predicted MVA proteins. To define different classes of protein expression,
+
+<--- Page Split --->
+![](images/Figure_5.jpg)
+
+<center>Figure 3 | Differential protein changes during infection of human fibroblasts with MVA or VACV-WR. a Scatterplot of all proteins quantified at 18 h of infection with both viruses. p-values were estimated using Significance A values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>65,67</sup>. Example profiles for NF-κB pathway proteins are shown in the lower panels. Error bars = range. b Functional enrichment analysis of proteins (i) downregulated by MVA but not VACV Western Reserve (VACV-WR), (ii) downregulated by VACV-WR but not MVA, or (iii) downregulated by both MVA and VACV-WR. c Example components from pathways identified in (b). Error bars = range. </center>
+
+<--- Page Split --->
+![PLACEHOLDER_13_0]
+
+
+<--- Page Split --->
+
+we employed k- means clustering and included data from samples treated with AraC. This suggested that there are at least four distinct temporal protein profiles of viral protein expression (Fig. 4A, Extended data Fig. 5).  
+
+Some of the earliest Tp1 (temporal profile 1) and Tp2 classes of viral protein exhibited increased expression in the presence of AraC, likely explained by the absence of negative regulation from post- replicative proteins. In contrast, the Tp3- class proteins were partly inhibited by AraC, with expression of Tp4 class proteins almost completely inhibited (Fig. 4A- B). Tp4 class proteins included multiple VACV envelope proteins, including B5 and A27 homologues PS/HR and MVA138 whose late expression may relate to roles in virus- specific T- cell memory<sup>27</sup> (Supplementary Table 4). Tp1- class proteins included MVA102 (VACV- WR D5), which uncoats viral genomes early during infection<sup>28</sup>, and MVA019 (VACV- WR C6). We previously demonstrated that C6 targets host restriction factors HDAC4 and HDAC5 for proteasomal degradation<sup>4,29</sup>. This classification thus serves as a particularly useful resource to predict interactions between host factors downregulated at early, intermediate or late times after infection and their viral counterparts and may help to form a rational basis for prediction of the effects of mutations introduced in individual genes during vaccine production by additional passage, plaque purification or insertion of foreign genes derived from other pathogens.
+
+<--- Page Split --->
+![PLACEHOLDER_15_0]
+
+
+<--- Page Split --->
+![PLACEHOLDER_16_0]
+
+<center>Extended Data Figure 5. Number of temporal classes of MVA gene expression. </center>  
+
+The k- means approach was used with 1- 15 classes to cluster viral proteins, and the summed distance of each protein from its cluster centroid was calculated. Although this summed distance necessarily becomes smaller as more clusters are added, the rate of decline decreases with each added group, eventually settling at a fairly constant rate of decline that reflects overfitting; clusters added prior to this point reflect underlying structure in the temporal protein data, whereas clusters subsequently added through overfitting are not informative. The point of inflexion fell between four and six classes, suggesting that there are at least four distinct temporal protein profiles of viral protein expression.
+
+<--- Page Split --->
+
+## 186 Discussion  
+
+Upon infection of human cells, MVA does not produce infectious new virions and is therefore classed as a replication- incompetent vaccine<sup>12</sup>. In the spectrum of vaccines available for use against human diseases, replication- incompetent vaccines lie between attenuated vaccines and inactivated vaccines, without the safety concerns of the former or the reduced immunogenicity of the latter<sup>30</sup>. Despite an ultimately abortive MVA infection of human cells, the viral life- cycle is typical of other poxviruses until early morphogenesis, including early gene expression, genome uncoating, DNA replication, and post- replicative gene expression<sup>31</sup>. Our global analysis of MVA infection highlighted a pervasive modulation of human host proteins and processes. Such an understanding, particularly including regulation of functions related to immunity, is critical for a complete assessment of MVA's applicability and safety as a vaccine against monkeypox and other orthopoxviruses, and as a vaccine vector against other pathogens.  
+
+Interferons and other cytokines induced upon viral sensing are key regulatory molecules in innate immunity and inflammation and play important roles in fine- tuning antigen- specific adaptive immunity during viral infection<sup>32,33</sup>. MVA- induced regulation of the interferon response and NF- \(\kappa\) B signaling, indicated both by limited upregulation of interferon- stimulated gene (ISG) products and direct regulation of NF- \(\kappa\) B, will therefore be likely to impact the immunological memory elicited by vaccination. Interferon regulation is likely to occur at all stages of infection, indicated by the increased response seen upon blockage of post- replicative viral gene expression and in cells exposed to heat- inactivated virions in comparison to cells infected with unmodified MVA. This is in contrast with the antagonism of the interferon response by VACV- WR. For instance, VACV early protein C9 targets ISG products IFIT1, IFIT2, and IFIT3 for proteosomal degradation<sup>34</sup>. Through serial passage in chicken embryo fibroblasts, the MVA genome has accumulated six major deletions and many small insertion and deletions (indels), which, in combination, result in the loss of 25 open reading frames (ORFs) when compared to wildtype VACV strains<sup>3,35</sup>. In addition, 12 ORFs have been split due to mutations, including the ORF encoding a C9 orthologue, explaining the absence of IFIT degradation during MVA infection<sup>35</sup>. Analysis both of viral protein temporal profiles and inactivating mutations are thus likely to yield candidates antagonists for host factors downregulated during VACV- WR, but not MVA infection including Ephrins and MAP4K2.
+
+<--- Page Split --->
+
+Host proteins with concordant modulation by MVA and VACV- WR might indicate that viral mechanisms of host modulation are still functional in MVA. Examples include common downregulation of the antiviral restriction factor HDAC5 by VACV- WR and MVA, mediated by VACV- WR protein C6 (MVA019)4. Proteasomal degradation represents the main mechanism responsible for downregulation of host protein during VACV infection4 and therefore, most host proteins downregulated during MVA infection are likely also to be degraded by the proteasome. Alternatively, common modulation of protein expression might reflect common cellular responses to infection, such as the downregulation of multiple collagens.  
+
+101 proteins were downregulated during MVA, but not VACV- WR infection. These included multiple components of the nuclear pore complex, whose downregulation is likely to disrupt nucleocytoplasmic transport36. The innate immune response to virus infection requires an intact NPC to mediate nucleocytoplasmic transport of transcription factors and mRNAs and therefore, downregulation/degradation of NPC proteins may be a viral immune evasion strategy. Although some viral proteases inactivate nucleoporins (NUPs) by cleavage37, NUP abundance did not vary substantially during infection by multiple human pathogenic viruses, such as human cytomegalovirus2, herpes simplex virus type 116, influenza A virus38, Epstein- Barr virus39, SARSCoV- 240 and HIV41. Therefore, to our knowledge, MVA is the first virus reported to downregulate/degrade NUPs. Three of the NUPs down- regulated by MVA (NUP54, NUP62, NUP88) were identified as host factors necessary for VACV- WR morphogenesis22, raising the possibility that NUP downregulation contributes to arrested virion maturation during MVA infection of human cells. It remains to be determined whether this unique MVA- induced modulation is shared with the parental strain (CVA) or was acquired during the serial passage in chicken embryo fibroblasts.  
+
+Proteins upregulated by heat- inactivated MVA included five members of the poly- ADP- ribosyl polymerase (PARP) family, some of which have well- characterized roles in antiviral and inflammatory responses, such as ZC3HAV1 (also known as PARP13 or ZAP)42. ZAP is an antiviral factor that restricts MVA infection in human cells43. VACV- WR induces the proteasomal degradation of ZAP and some other VACV strains encode proteins that interact with and overcome its antiviral function44,45. Blockade of MVA DNA replication and post- replicative gene expression upregulated ZAP, whilst PARPs 9, 10, 12 and 14 were only upregulated by heat- inactivated virus. This suggests that these latter PARPs might be targeted by early gene product(s), may indicate that they serve as- yet uncharacterized antiviral function, and highlights
+
+<--- Page Split --->
+
+the usefulness of use of our multi- level method to reveal regulation that would otherwise not be observed.  
+
+In sum, our quantitative temporal analysis of the infection of human cells by the MVA vaccine has revealed an extensive remodelling of the host proteome. This highlighted that, at different stages of infection, MVA retains a limited capacity to antagonise innate immune sensing and exhibits features not present in the VACV- WR strain related to the first- generation vaccine. Our data offer an opportunity to dissect how multiple, seemingly redundant MVA and VACV immunomodulatory effectors function and to discover novel viral strategies to escape antiviral immunity. A comprehensive understanding of all these issues may be essential to generate future, fourth- generation smallpox/monkeypox vaccines in addition to highly potent recombinant vaccines against other infections.
+
+<--- Page Split --->
+
+## Methods  
+
+## Cells and viruses  
+
+Primary human foetal foreskin fibroblast cells immortalised with human telomerase (HFFF- TERTs, male) and primary chicken embryo fibroblasts (CEFs) were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with foetal bovine serum (FBS: \(10\%\) v/v), and penicillin/streptomycin at \(37^{\circ}C\) in \(5\%\) CO2. HFFF- TERTs have been tested at regular intervals since isolation to confirm both that the HLA and MICA genotypes, and the morphology and antibiotic resistances are consistent with the original description<sup>44</sup>. In addition, human fibroblasts (dermal or foreskin) are the only permissive cells for human cytomegalovirus (HCMV) in cell culture, and HFFF- TERTs are routinely infected with the HCMV Merlin strain, further limiting the chances that the cells have been contaminated with another cell type<sup>17</sup>. Fresh CEFs were obtained from the Pirbright Institute (Woking, UK) and directly seeded for virus production without further passages. All cell lines used regularly tested negative for mycoplasma.  
+
+Modified vaccinia Ankara (MVA) was obtained as a seed stock prepared from an original MVA stock at passage \(575^{45}\) . MVA was propagated in CEFs, purified by ultracentrifugation through two \(36\%\) (w/v) sucrose cushions and suspended in \(10 \text{mM Tris - HCl pH 9.0}\) . MVA infectivity was determined by immunocytochemistry on HFFF- TERTs cells, by using a polyclonal rabbit anti- VACV antibody<sup>46</sup>. For heat inactivation, MVA stock was diluted ten- fold in DMEM supplemented with \(2\%\) FBS and heated at \(56^{\circ}C\) for \(1 \text{h}\) in a water bath<sup>47</sup>.  
+
+## Viral infections and inhibitors  
+
+For proteomic experiments, \(7.5 \times 10^{5}\) HFFF- TERTs were seeded in a \(25 - \text{cm}^2\) flask two days prior to infection. Cells were infected with MVA, or the equivalent amount of heat- inactivated MVA, at MOI 5 for 2, 4, 8, 12, or \(18 \text{h}\) , with time zero defined as the time the virus was added to the cells. A mock- infected control was harvested at \(12 \text{h}\) post- infection. Where indicated, cells were incubated with cytosine arabinoside (AraC) at \(40 \mu \text{g/ml}\) from the time zero. The time course experiments were performed in biological duplicate.  
+
+## Flow cytometry  
+
+MVA- infected HFFF- TERTs were detached with trypsin- EDTA (Gibco) \(24 \text{h}\) post- infection, passed through a \(70 - \mu \text{m}\) cell strainer, washed in PBS and stained with Zombie Violet viability dye (BioLegend) for \(30 \text{min}\) at \(4^{\circ}C\) in the dark. Cells were collected by centrifugation, washed once
+
+<--- Page Split --->
+
+with PBS and fixed in Cytofix/Cytoperm fixation and permeabilization solution (BD Biosciences) for 30 min at \(4^{\circ}C\) in the dark. Cell were washed twice with Perm/Wash buffer (BD Biosciences) and stained with a polyclonal rabbit anti- VACV antibody<sup>46</sup> diluted 1:500 in Perm/Wash buffer, followed by AlexaFluor 568-conjugated goat anti-rabbit IgG (Invitrogen) diluted 1:100. Stained cells were suspended in PBS before data acquisition with an Attune NxT flow cytometer. Data were analysed with FlowJo software.  
+
+## Whole cell lysate protein digestion  
+
+Two independent biological replicates were performed, 'WCL1' and 'WCL2'. For each replicate, cells were washed twice with PBS, and 250 μl lysis buffer added (6 M Guanidine/50 mM HEPES pH 8.5). Cell lifters (Corning) were used to scrape cells in lysis buffer, which was removed to an eppendorf tube, vortexed extensively then sonicated. Cell debris was removed by centrifuging at 21,000 g for 10 min twice. For half of each sample, dithiothreitol (DTT) was added to a final concentration of 5 mM and samples were incubated for 20 min. Cysteines were alkylated with 14 mM iodoacetamide and incubated 20 min at room temperature in the dark. Excess iodoacetamide was quenched with DTT for 15 min. Samples were diluted with 200 mM HEPES pH 8.5 to 1.5 M Guanidine followed by digestion at room temperature for 3 h with LysC protease at a 1:100 protease- to- protein ratio. Samples were further diluted with 200 mM HEPES pH 8.5 to 0.5 M Guanidine. Trypsin was then added at a 1:100 protease- to- protein ratio followed by overnight incubation at \(37^{\circ}C\) . The reaction was quenched with 5% formic acid, then centrifuged at 21,000 g for 10 min to remove undigested protein. Peptides were subjected to C18 solid- phase extraction (SPE, Sep- Pak, Waters) and vacuum- centrifuged to near- dryness.  
+
+## Peptide labeling with tandem mass tags  
+
+In preparation for TMT labeling, desalted peptides were dissolved in 200 mM HEPES pH 8.5. Peptide concentration was measured by microBCA (Pierce), and 25 μg of peptide labeled with TMT reagent. TMT reagents (0.8 mg) were dissolved in 43 μl anhydrous acetonitrile and 5 μl added to peptide at a final acetonitrile concentration of 30% (v/v). Sample labelling was: 126, MVA (2h infection); 127N, MVA+ArA (2h); 127C, MVA-HI (2h); 128N, MVA (4h); 128C, MVA+ArA (4h); 129N, MVA-HI (4h); 129C, MVA (8h); 130N, MVA+ArA (8h); 130C, MVA-HI (8h); 131N, MVA (12h); 131C, MVA+ArA (12h); 132N, MVA-HI (12h); 132C, MVA (18h); 133N, MVA+ArA (18h); 133C, MVA-HI (18h); 134N, mock infection. Following incubation at room temperature for 1 h, the reaction was quenched with hydroxylamine to a final concentration of
+
+<--- Page Split --->
+
+0.3% (v/v). TMT- labeled samples were combined at a 1:1:1:1:1:1:1:1:1:1:1;1:1:1:1:1:1:1 ratio. The sample was vacuum- centrifuged to near dryness and subjected to C18 SPE (Sep- Pak, Waters). An unfractionated singleshot was analysed initially to ensure similar peptide loading across each TMT channel, thus avoiding the need for excessive electronic normalization. As all normalisation factors were \(>0.5\) and \(<2\) , data for each singleshot experiment was analysed with data for the corresponding fractions to increase the overall number of peptides quantified. Normalisation is discussed in 'Data Analysis', and high pH reversed- phase (HpRP) fractionation is discussed below.  
+
+## Offline HpRP fractionation  
+
+TMT- labelled tryptic peptides were subjected to HpRP fractionation using an Ultimate 3000 RSLC UHPLC system (Thermo Fisher Scientific) equipped with a 2.1 mm internal diameter (ID) \(\times 25\) cm long, \(1.7 \mu \mathrm{m}\) particle Kinetix Evo C18 column (Phenomenex). Mobile phase consisted of A: \(3\%\) acetonitrile (MeCN), B: MeCN and C: \(200 \mathrm{mM}\) ammonium formate pH 10. Isocratic conditions were \(90\%\) A / \(10\%\) C, and C was maintained at \(10\%\) throughout the gradient elution. Separations were conducted at \(45^{\circ}\mathrm{C}\) . Samples were loaded at \(200 \mu \mathrm{l} / \mathrm{min}\) for 5 min. The flow rate was then increased to \(400 \mu \mathrm{l} / \mathrm{min}\) over 5 min, after which the gradient elution proceed as follows: \(0 - 19\%\) B over 10 min, \(19 - 34\%\) B over 14.25 min, \(34 - 50\%\) B over 8.75 min, followed by a 10 min wash at \(90\%\) B. UV absorbance was monitored at \(280 \mathrm{nm}\) and 15 s fractions were collected into 96- well microplates using the integrated fraction collector. Fractions were recombined orthogonally in a checkerboard fashion, combining alternate wells from each column of the plate into a single fraction, and commencing combination of adjacent fractions in alternating rows. Wells were excluded prior to the start or after the cessation of elution of peptide- rich fractions, as identified from the UV trace. This yielded two sets of 12 combined fractions, A and B, which were dried in a vacuum centrifuge and resuspended in \(10 \mu \mathrm{l}\) MS solvent ( \(4\%\) MeCN / \(5\%\) formic acid) prior to LC- MS3. 12 set 'A' fractions were used for MS3 analysis of all experiments, and half of each sample was subjected to MS3 analysis, with the other half subjected to real- time search (RTS) MS3 analysis<sup>48</sup> (described below).  
+
+## LC-MS3 for TMT and TMT/SILAC experiments  
+
+For 'singleshot' analysis of unfractionated peptides from each experiment, mass spectrometry data was acquired using an Orbitrap Lumos. Subsequently, fractions were acquired using an Orbitrap Eclipse.
+
+<--- Page Split --->
+
+For Orbitrap Lumos analyses: An Ultimate 3000 RSLC nano UHPLC equipped with a \(300\mu \mathrm{m}\) ID \(\times 5 \mathrm{mm}\) Acclaim PepMap \(\mu\) - Precolumn (Thermo Fisher Scientific) and a \(75\mu \mathrm{m}\) ID \(\times 50 \mathrm{cm} 2.1 \mu \mathrm{m}\) particle Acclaim PepMap RSLC analytical column was used. Loading solvent was \(0.1\%\) FA, analytical solvent A: \(0.1\%\) FA and B: \(80\%\) MeCN \(+0.1\%\) FA. All separations were carried out at \(40^{\circ}\mathrm{C}\) . Samples were loaded at \(5 \mu \mathrm{L} / \mathrm{min}\) for \(5 \mathrm{min}\) in loading solvent before beginning the analytical gradient. The following gradient was used: \(3 - 7\%\) B over \(2 \mathrm{min}\) , \(7 - 37\%\) B over \(173 \mathrm{min}\) , followed by a \(4 \mathrm{min}\) wash at \(95\%\) B and equilibration at \(3\%\) B for \(15 \mathrm{min}\) . Each analysis used a MultiNotch MS3- based TMT method \(^{49}\) . The following settings were used: MS1: 380- 1500 Th, 120,000 resolution, \(2 \times 10^{5}\) automatic gain control (AGC) target, \(50 \mathrm{ms}\) maximum injection time. MS2: Quadrupole isolation at an isolation width of \(m / z 0.7\) , CID fragmentation (normalised collision energy (NCE) 34) with ion trap scanning in turbo mode, with \(1.5 \times 10^{4}\) AGC target and \(120 \mathrm{ms}\) maximum injection time. MS3: In Synchronous Precursor Selection mode the top 10 MS2 ions were selected for HCD fragmentation (NCE 45) and scanned in the Orbitrap at 60,000 resolution with an AGC target of \(1 \times 10^{5}\) and a maximum accumulation time of \(150 \mathrm{ms}\) . Ions were not accumulated for all parallelisable time. The entire MS/MS/MS cycle had a target time of \(3 \mathrm{s}\) . Data analysis is discussed below.  
+
+For Orbitrap Eclipse analyses: The FAIMS Pro interface (ThermoFisher Scientific, San Jose, CA) was coupled to a Proxeon EASY- nLC 1200 liquid chromatograph (LC) (ThermoFisher Scientific, San Jose, CA). Peptides were separated on a \(100 \mu \mathrm{m}\) inner diameter micropapillary column packed with \(\sim 35 \mathrm{cm}\) of Accucore150 resin (2.6 \(\mu \mathrm{m}\) , \(150 \mathrm{\AA}\) , ThermoFisher Scientific, San Jose, CA). For each analysis, \(1 - 2 \mu \mathrm{g}\) of peptide was loaded onto the column and fractionated over a \(90 \mathrm{min}\) gradient of \(7\) to \(27\%\) acetonitrile in \(0.125\%\) formic acid at a flow rate of \(\sim 600 \mathrm{nL} / \mathrm{min}\) . Mass spectrometric data for each sample were collected using two distinct data acquisition modes (SPS- MS3 and RTS- MS3), all with FAIMS. SPS- MS3 data were collected with a FAIMS compensation voltage (CV) set of - 40 V, - 60 V, and - 80 V, while RTS- MS3 data were collected with a FAIMS CV set of - 30 V, - 50 V, and - 70 V, with each segment as a 1 s TopSpeed method. For both acquisition methods, the scan sequence began with an MS1 spectrum (Orbitrap analysis; resolution, 60,000; mass range, 350- 1350 Th; automatic gain control (AGC) target \(100\%\) ; maximum injection time, auto). Precursors (charge states 2- 5; precursor fit \(50\%\) at \(0.7 \mathrm{Th}\) ; minimum intensity of \(15 \mathrm{K}\) ) were then selected for MS2/MS3 analysis \(^{50}\) . MS2 analysis consisted of collision- induced dissociation (CID) with quadrupole ion trap analysis, using the following
+
+<--- Page Split --->
+
+parameters: scan speed, turbo; AGC target, \(100\%\) ; NCE, 35; q- value, 0.25; maximum injection time, 35 ms; and isolation window, 0.5 Th. MS3 precursors were fragmented by HCD and analyzed using the Orbitrap with the following parameters: resolution, 50,000; NCE, 55; AGC, \(250\%\) ; maximum injection time, 86 ms; maximum synchronous precursor selection (SPS) ions, 10; and isolation window, 1.2 Th. RTS- MS3 data were collected with the "close- out" parameter set to two. Dynamic exclusion was set at 120 sec with \(+ / - 10\) ppm error tolerance.  
+
+## Data analysis  
+
+In the following description, the first report in the literature for each relevant algorithm is listed. Mass spectra were processed using Sequest- based in house software suite, a software pipeline for quantitative proteomics, developed by Professor Steven Gygi's laboratory at Harvard Medical School. MS spectra were converted to mzXML using an extractor built upon Thermo Fisher's RAW File Reader library (version 4.0.26).  
+
+After converting RAW files to mzxml format, MS data were analyzed to correct errors in monoisotopic peak assignment using a published algorithm<sup>51</sup> and then searched against a database of protein sequences. A combined database was constructed from (a) the human Uniprot database (accessed \(20^{\text{th}}\) June 2022, UP000005640), (b) the MVA proteome (accessed \(20^{\text{th}}\) June 2022, UP000172909), (c) common contaminants such as porcine trypsin and endoproteinase LysC. The combined database was concatenated with a reverse database composed of all protein sequences in reversed order. Sequest searches were performed using a 20 ppm precursor ion tolerance<sup>52,53</sup>. Product ion tolerance was set to 1 Da. Oxidation of methionine residues (15.99492 Da) was set as a variable modification. TMT tags on lysine residues and peptide N termini (304.207145 Da) and carbamidomethylation of cysteine residues (57.02146 Da) were included as static modifications. Peptides were assumed to be fully tryptic with up to two missed cleavages.  
+
+To control the fraction of erroneous protein identifications, a target- decoy strategy was employed<sup>54</sup>. Peptide spectral matches (PSMs) were filtered to an initial peptide- level false discovery rate (FDR) of \(1\%\) with subsequent filtering to attain a final protein- level FDR of \(1\%\) . PSM filtering was performed using a linear discriminant analysis, as described previously<sup>54</sup>. This distinguishes correct from incorrect peptide IDs in a manner analogous to the widely used Percolator algorithm (https://noble.gs.washington.edu/proj/percolator/), though employing a
+
+<--- Page Split --->
+
+distinct machine learning algorithm. The following parameters were considered: XCorr, \(\Delta \mathsf{Cn}\) , missed cleavages, peptide length, charge state, and precursor mass accuracy. Peptides shorter than seven amino acids in length or with XCorr less than 1.0 were excluded prior to LDA filtering. Protein assembly was guided by principles of parsimony to produce the smallest set of proteins necessary to account for all observed peptides (algorithm described previously54).  
+
+Proteins were quantified by summing TMT reporter ion counts across all matching peptide- spectral matches, as described previously49. Briefly, a 0.003 Th window around the theoretical m/z of each reporter ion (126, 127n, 127c, 128n, 128c, 129n, 129c, 130n, 130c, 131n, 131c, 132n, 132c, 133n, 133c, 134n) was scanned for ions, and the maximum intensity nearest to the theoretical m/z was used. The primary determinant of quantitation quality is the number of TMT reporter ions detected in each MS3 spectrum, which is directly proportional to the signal- to- noise (S:N) ratio observed for each ion. Conservatively, every individual peptide used for quantitation was required to contribute sufficient TMT reporter ions (minimum of \(\sim 1250\) per spectrum) so that each on its own could be expected to provide a representative picture of relative protein abundance49. An isolation specificity filter with a cutoff of \(50\%\) was employed to minimise peptide co- isolation49. Peptide- spectral matches with poor quality MS3 spectra (more than 15 TMT channels missing and/or a combined S:N ratio of less than 250 across all TMT reporter ions) or no MS3 spectra at all were excluded from quantitation. Peptides meeting the stated criteria for reliable quantitation were then summed by parent protein, in effect weighting the contributions of individual peptides to the total protein signal based on their individual TMT reporter ion yields. Protein quantitation values were exported for further analysis in Excel.  
+
+For protein quantitation, reverse and contaminant proteins were removed, then each reporter ion channel was summed across all quantified proteins and normalised assuming equal protein loading across all channels. For further analysis and display, fractional TMT signals were used (i.e. reporting the fraction of maximal signal observed for each protein in each TMT channel, rather than the absolute normalized signal intensity). This effectively corrected for differences in the numbers of peptides observed per protein. In three instances (for MVA proteins MVA045 and MVA131 and human protein DZIP1L), the S:N in the mock sample was zero. For the purposes of calculating fold change of each infected sample compared to mock, these mock values were set to \(2\%\) of the maximum S:N for any sample in the same WCL experiment.
+
+<--- Page Split --->
+
+For analysis of the viral proteome, the MVA proteome (UP000172909) was matched with the VACV- WR proteome (UP000000344) using the PathoSystems Resource Integration Center (PATRIC)'s proteome comparison tool with default settings<sup>55</sup>.  
+
+Hierarchical centroid clustering based on uncentered Pearson correlation, and k- means clustering were performed using Cluster 3.0 (Stanford University) and visualised using Java Treeview (http://jtreeview.sourceforge.net) unless otherwise noted.  
+
+## Statistical analysis  
+
+Two biological replicates were employed throughout. P- values were estimated using the methods of Significance A or B as indicated in the figure legends, which consider two sides of a normal distribution with non- equal standard deviations, with multiple hypothesis correction using the method of Benjamini- Hochberg in Perseus version 1.5.1.6<sup>56,57</sup>. Blinding or sample- size estimation was not appropriate for this study. There were no inclusion criteria and no data was excluded.  
+
+## Pathway analysis  
+
+The Database for Annotation, Visualisation and Integrated Discovery (DAVID) was used to determine pathway enrichment<sup>20</sup>. A given cluster was always searched against a background of all proteins quantified within the relevant experiment. Downregulated proteins were defined as those factors that decreased \(>2\) - fold in abundance compared to the mock sample at \(\geq 1\) time point during infection. Proteins that were not downregulated were defined as factors that decreased \(< 1.25\) - fold in abundance compared to the mock sample at all times of infection. Upregulated proteins were defined in a similar manner.  
+
+## Interaction analysis  
+
+Interaction analysis was performed on groups of proteins defined by changes in abundance during MVA, MVA+ArAC, MVA- HI and VACV infection indicated in Figures 2c and 3b. Network analyses were performed using interactions from BioPlex 3.0<sup>23</sup>. For this analysis, both 293T and HCT116 networks were merged. All analyses were performed in Mathematica 13.1 (Wolfram Research).  
+
+To assess the tendency for sets of differentially expressed proteins to interact with each other, each was mapped onto the combined BioPlex network after converting all protein identifiers to Entrez GeneIDs. Graph assortativity was then calculated and the process repeated using 1000
+
+<--- Page Split --->
+
+randomized protein sets of equal size. Scores from randomized protein sets were then fit to a Gaussian distribution and Z- scores and p- values were estimated.  
+
+To produce interaction networks shown in Figures 2d and Extended data Fig.4d, proteins were grouped according to their differential expression under each experimental condition. Each group was then mapped onto the BioPlex network and network propagation was used to identify additional nodes that closely associate with proteins in each group. Network propagation was calculated via random walk with restart as described previously<sup>58</sup> with the restart probability set to 0.5. After 40 iterations, proteins were sorted by descending weights. In Figure 2e the top 100 highest ranking proteins were selected for display. In Extended data Fig.4d the top 150 highest ranking proteins were selected for display, though for clarity those proteins that did not interact with any other proteins on this list were excluded.  
+
+## Data Availability  
+
+The mass spectrometry proteomics data will be deposited to the ProteomeXchange Consortium (http://www.proteomexchange.org/) via the PRIDE<sup>59</sup> partner repository. All materials described in this manuscript, and any further details of protocols employed can be obtained on request from the corresponding author by email to mpw1001@cam.ac.uk.  
+
+## Code Availability  
+
+The data analysis pipeline as a whole may be licensed from Harvard Medical School, and constituent components have been published and referenced above. These include a RAW file to mzXML conversion tool, built upon Thermo Fisher's RAW File Reader library (version 4.0.26); a monoisotopic mass assignment tool<sup>51</sup>; the Sequest algorithm<sup>53</sup>; strategies for target- decoy analysis, peptide spectral match filtering and protein assembly<sup>54</sup>, and methods for protein quantification and isolation specificity filtering<sup>49</sup> are recapitulated in Proteome Discoverer Software (Thermo Fisher, UK).  
+
+## Acknowledgments  
+
+This research was funded in part by Medical Research Council Project Grant MR/W025647/1 to M.P.W., R01 GM132129 to J.A.P., R021 GM67945 to S.P.G., National Institute of Health grant U24HG006673 to E.L.H and S.P.G. This study was additionally supported by the Cambridge Biomedical Research Centre, UK. For the purpose of open access, the author has applied a CC
+
+<--- Page Split --->
+
+BY public copyright licence to any Author Accepted Manuscript version arising from this submission.  
+
+## Author Contributions  
+
+Conceptualization: M.P.W. Investigation: J.D.A., M.O., J.K., J.A.P., R.A. Data analysis: J.D.A., M.O., J.K., E.H., M.P.W. Funding acquisition: S.P.G, E.L.H, G.L.S, M.P.W. Supervision: S.P.G, E.L.H, G.L.S, M.P.W. Writing: J.D.A., M.O., J.K., J.A.P., R.A., S.P.G, E.L.H, G.L.S, M.P.W.  
+
+## Competing Interest Statement  
+
+The authors declare no competing interests.
+
+<--- Page Split --->
+
+510 1 Isidro, J. et al. Phylogenomic characterization and signs of microevolution in the 2022 multi- 511 country outbreak of monkeypox virus. Nat Med, doi:10.1038/s41591- 022- 01907- y (2022). 512 2 Weekes, M. P. et al. Quantitative temporal viromics: an approach to investigate host- pathogen 513 interaction. Cell 157, 1460- 1472 (2014). 514 3 Volz, A. & Sutter, G. Modified Vaccinia Virus Ankara: History, Value in Basic Research, and Current 515 Perspectives for Vaccine Development. Adv Virus Res 97, 187- 243, 516 doi:10.1016/bs.avir.2016.07.001 (2017). 517 4 Soday, L. et al. Quantitative temporal proteomic analysis of vaccinia virus infection reveals 518 regulation of histone deacetylases by an interferon antagonist. Cell Rep 27, 1920- 1933 e1927, 519 doi:10.1016/j.celrep.2019.04.042 (2019). 520 5 Yinka- Ogunleye, A. et al. Outbreak of human monkeypox in Nigeria in 2017- 18: a clinical and 521 epidemiological report. Lancet Infect Dis 19, 872- 879, doi:10.1016/S1473- 3099(19)30294- 4 522 (2019). 523 6 Babkin, I. V., Babkina, I. N. & Tikunova, N. V. An Update of Orthopoxvirus Molecular Evolution. 524 Viruses 14, doi:10.3390/v14020388 (2022). 525 7 World Health Organization (WHO). Multi- country monkeypox outbreak: situation update (2022). 526 8 Bunge, E. M. et al. The changing epidemiology of human monkeypox- A potential threat? A 527 systematic review. PLoS Negl Trop Dis 16, e0010141, doi:10.1371/journal.pntd.0010141 (2022). 528 9 Fenner, F., Henderson, D. A., Arita, I., Jezek, Z. & Ladnyi, I. D. Smallpox and its eradication. World 529 Health Organisation, Geneva (1988). 530 10 Sanchez- Sampedro, L. et al. The evolution of poxvirus vaccines. Viruses 7, 1726- 1803, 531 doi:10.3390/v7041726 (2015). 532 11 Stickl, H. & Hochstein- Mintzel, V. [Intracutaneous smallpox vaccination with a weak pathogenic 533 vaccinia virus ("MVA virus")]. Munch Med Wochenschr 113, 1149- 1153 (1971). 534 12 Jacobs, B. L. et al. Vaccinia virus vaccines: past, present and future. Antiviral Res 84, 1- 13, 535 doi:10.1016/j.antiviral.2009.06.006 (2009). 536 13 Greenberg, R. N. et al. Safety, immunogenicity, and surrogate markers of clinical efficacy for 537 modified vaccinia Ankara as a smallpox vaccine in HIV- infected subjects. The Journal of infectious 538 diseases 207, 749- 758, doi:10.1093/infdis/jis753 (2013). 539 14 Overton, E. T. et al. Immunogenicity and safety of three consecutive production lots of the non 540 replicating smallpox vaccine MVA: A randomised, double blind, placebo controlled phase III trial. 541 PLoS One 13, e0195897, doi:10.1371/journal.pone.0195897 (2018). 542 15 Volkmann, A. et al. The Brighton Collaboration standardized template for collection of key 543 information for risk/benefit assessment of a Modified Vaccinia Ankara (MVA) vaccine platform. 544 Vaccine 39, 3067- 3080, doi:10.1016/j.vaccine.2020.08.050 (2021). 545 16 Soh, T. K. et al. Herpes simplex virus- 1 pUL56 degrades GOPC to alter the plasma membrane 546 proteome. Cell Reports 33, 108235 (2020). 547 17 Nightingale, K. et al. High- definition analysis of host protein stability during Human 548 Cytomegalovirus infection reveals antiviral factors and viral evasion mechanisms. Cell Host 549 Microbe 12, 447- 460, doi:10.1016/j.chom.2018.07.011 (2018).
+
+<--- Page Split --->
+
+550 18 Li, J. et al. TMTpro reagents: a set of isobaric labeling mass tags enables simultaneous proteome- 551 wide measurements across 16 samples. Nat Methods 17, 399- 404, doi:10.1038/s41592- 020- 552 0781- 4 (2020). 553 19 Navarrete- Perea, J., Yu, Q., Gygi, S. P. & Paulo, J. A. Streamlined Tandem Mass Tag (SL- TMT) 554 Protocol: An Efficient Strategy for Quantitative (Phospho)proteome Profiling Using Tandem Mass 555 Tag- Synchronous Precursor Selection- MS3. J Proteome Res 17, 2226- 2236, 556 doi:10.1021/acs.jproteome.8b00217 (2018). 557 20 Huang da, W., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene 558 lists using DAVID bioinformatics resources. Nature protocols 4, 44- 57, 559 doi:10.1038/nprot.2008.211 (2009). 560 21 Sortino, O. et al. The Effects of Recombinant Human Lactoferrin on Immune Activation and the 561 Intestinal Microbiome Among Persons Living with Human Immunodeficiency Virus and Receiving 562 Antiretroviral Therapy. The Journal of infectious diseases 219, 1963- 1968, 563 doi:10.1093/infdis/jiz042 (2019). 564 22 Sivan, G. et al. Human genome- wide RNAi screen reveals a role for nuclear pore proteins in 565 poxvirus morphogenesis. Proc Natl Acad Sci U S A 110, 3519- 3524, doi:10.1073/pnas.1300708110 566 (2013). 567 23 Huttlin, E. L. et al. Dual proteome- scale networks reveal cell- specific remodeling of the human 568 interactome. Cell 184, 3022- 3040 e3028, doi:10.1016/j.cell.2021.04.011 (2021). 569 24 Kanth, S. M., Gairhe, S. & Torabi- Parizi, P. The Role of Semaphorins and Their Receptors in Innate 570 Immune Responses and Clinical Diseases of Acute Inflammation. Frontiers in immunology 12, 571 672441, doi:10.3389/fimmu.2021.672441 (2021). 572 25 Vallabhapurapu, S. & Karin, M. Regulation and function of NF- kappaB transcription factors in the 573 immune system. Annu Rev Immunol 27, 693- 733, doi:10.1146/annurev.immunol.021908.132641 574 (2009). 575 26 Arthur, J. S. & Ley, S. C. Mitogen- activated protein kinases in innate immunity. Nat Rev Immunol 576 13, 679- 692, doi:10.1038/nri3495 (2013). 577 27 Tang, J. et al. Human T- cell responses to vaccinia virus envelope proteins. J Virol 80, 10010- 10020, 578 doi:10.1128/JVI.00601- 06 (2006). 579 28 Kilcher, S. et al. siRNA screen of early poxvirus genes identifies the AAA+ ATPase D5 as the virus 580 genome- uncoating factor. Cell Host Microbe 15, 103- 112, doi:10.1016/j.chom.2013.12.008 581 (2014). 582 29 Lu, Y. et al. Histone deacetylase 4 promotes type I interferon signaling, restricts DNA viruses, and 583 is degraded via vaccinia virus protein C6. Proc Natl Acad Sci U S A 116, 11997- 12006, 584 doi:10.1073/pnas.1816399116 (2019). 585 30 Hajj Hussein, I. et al. Vaccines Through Centuries: Major Cornerstones of Global Health. Front 586 Public Health 3, 269, doi:10.3389/fpubh.2015.00269 (2015). 587 31 Sutter, G. & Moss, B. Nonreplicating vaccinia vector efficiently expresses recombinant genes. Proc 588 Natl Acad Sci U S A 89, 10847- 10851, doi:10.1073/pnas.89.22.10847 (1992). 589 32 Pulendran, B. & Ahmed, R. Translating innate immunity into immunological memory: implications 590 for vaccine development. Cell 124, 849- 863, doi:10.1016/j.cell.2006.02.019 (2006). 591 33 Richer, M. J., Nolz, J. C. & Harty, J. T. Pathogen- specific inflammatory milieux tune the antigen 592 sensitivity of CD8(+) T cells by enhancing T cell receptor signaling. Immunity 38, 140- 152, 593 doi:10.1016/j.immuni.2012.09.017 (2013).
+
+<--- Page Split --->
+
+594 34 Liu, R., Olano, L. R., Mirzakhanyan, Y., Gershon, P. D. & Moss, B. Vaccinia Virus Ankyrin-Repeat/F- 595 Box Protein Targets Interferon-Induced IFITs for Proteasomal Degradation. Cell Rep 29, 816- 828 596 e816, doi:10.1016/j.celrep.2019.09.039 (2019). 597 35 Antoine, G., Scheiflinger, F., Dorner, F. & Falkner, F. G. The complete genomic sequence of the 598 modified vaccinia Ankara strain: comparison with other orthopoxviruses. Virology 244, 365- 396, 599 doi:10.1006/viro.1998.9123 (1998). 600 36 Strambio- De- Castillia, C., Niepel, M. & Rout, M. P. The nuclear pore complex: bridging nuclear 601 transport and gene regulation. Nat Rev Mol Cell Biol 11, 490- 501, doi:10.1038/nrm2928 (2010). 602 37 De Jesus- Gonzalez, L. A. et al. The Nuclear Pore Complex Is a Key Target of Viral Proteases to 603 Promote Viral Replication. Viruses 13, doi:10.3390/v13040706 (2021). 604 38 Turnbull, M. L. et al. Role of the B Allele of Influenza A Virus Segment 8 in Setting Mammalian Host 605 Range and Pathogenicity. J Virol 90, 9263- 9284, doi:10.1128/JVI.01205- 16 (2016). 606 39 Ersing, I. et al. A Temporal Proteomic Map of Epstein- Barr Virus Lytic Replication in B Cells. Cell 607 Rep 19, 1479- 1493, doi:10.1016/j.celrep.2017.04.062 (2017). 608 40 Bojkova, D. et al. Proteomics of SARS- CoV- 2- infected host cells reveals therapy targets. Nature 609 583, 469- 472, doi:10.1038/s41586- 020- 2332- 7 (2020). 610 41 Naamati, A. et al. Functional proteomic atlas of HIV infection in primary human CD4+ T cells. eLife 611 8, doi:10.7554/eLife.41431 (2019). 612 42 Ficarelli, M., Neil, S. J. D. & Swanson, C. M. Targeted Restriction of Viral Gene Expression and 613 Replication by the ZAP Antiviral System. Annu Rev Virol 8, 265- 283, doi:10.1146/annurev- virology- 0191919- 104213 (2021). 614 43 Peng, C. et al. Zinc- finger antiviral protein (ZAP) is a restriction factor for replication of modified 615 vaccinia virus Ankara (MVA) in human cells. PLoS Pathog 16, e1008845, 616 doi:10.1371/journal.ppat.1008845 (2020). 617 44 McSharry, B. P., Jones, C. J., Skinner, J. W., Kipling, D. & Wilkinson, G. W. Human telomerase 618 reverse transcriptase- immortalized MRC- 5 and HCA2 human fibroblasts are fully permissive for 619 human cytomegalovirus. J Gen Virol 82, 855- 863, doi:10.1099/0022- 1317- 82- 4- 855 (2001). 620 45 Blanchard, T. J., Alcami, A., Andrea, P. & Smith, G. L. Modified vaccinia virus Ankara undergoes 621 limited replication in human cells and lacks several immunomodulatory proteins: implications for 622 use as a human vaccine. J Gen Virol 79 ( Pt 5), 1159- 1167, doi:10.1099/0022- 1317- 79- 5- 1159 623 (1998). 624 46 Law, M. & Smith, G. L. Antibody neutralization of the extracellular enveloped form of vaccinia 625 virus. Virology 280, 132- 142, doi:10.1006/viro.2000.0750 (2001). 626 47 Cao, H. et al. Innate immune response of human plasmacytoid dendritic cells to poxvirus infection 627 is subverted by vaccinia E3 via its Z- DNA/RNA binding domain. PLoS One 7, e36823, 628 doi:10.1371/journal.pone.0036823 (2012). 629 48 Schweppe, D. K. et al. Full- Featured, Real- Time Database Searching Platform Enables Fast and 630 Accurate Multiplexed Quantitative Proteomics. J Proteome Res 19, 2026- 2034, 631 doi:10.1021/acs.jproteome.9b00860 (2020). 632 49 McAlister, G. C. et al. MultiNotch MS3 enables accurate, sensitive, and multiplexed detection of 633 differential expression across cancer cell line proteomes. Anal Chem 86, 7150- 7158, 634 doi:10.1021/ac502040v (2014). 635 50 Ting, L., Rad, R., Gygi, S. P. & Haas, W. MS3 eliminates ratio distortion in isobaric multiplexed 636 quantitative proteomics. Nat Methods 8, 937- 940, doi:10.1038/nmeth.1714 (2011).
+
+<--- Page Split --->
+
+638 51 Rad, R. et al. Improved Monoisotopic Mass Estimation for Deeper Proteome Coverage. J Proteome Res 20, 591- 598, doi:10.1021/acs.jproteome.0c00563 (2021).639 52 Haas, W. et al. Optimization and use of peptide mass measurement accuracy in shotgun proteomics. Mol Cell Proteomics 5, 1326- 1337, doi:10.1074/mcp.M500339- MCP200 (2006).640 53 Yates, J. R., 3rd. Pivotal role of computers and software in mass spectrometry - SEQUEST and 20 years of tandem MS database searching. J Am Soc Mass Spectrom 26, 1804- 1813, doi:10.1007/s13361- 015- 1220- 0 (2015).641 54 Huttlin, E. L. et al. A tissue- specific atlas of mouse protein phosphorylation and expression. Cell 143, 1174- 1189, doi:10.1016/j.cell.2010.12.001 (2010).642 55 Davis, J. J. et al. The PATRIC Bioinformatics Resource Center: expanding data and analysis capabilities. Nucleic acids research 48, D606- D612, doi:10.1093/nar/gkz943 (2020).643 56 Cox, J. & Mann, M. MaxQuant enables high peptide identification rates, individualized ppb- range mass accuracies and proteome- wide protein quantification. Nature Biotechnology 26, 1367- 1372, doi:10.1038/nbt.1511 (2008).644 57 Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B- Methodol. 57, 289- 300 (1995).645 58 Cowen, L., Ideker, T., Raphael, B. J. & Sharan, R. Network propagation: a universal amplifier of genetic associations. Nat Rev Genet 18, 551- 562, doi:10.1038/nrg.2017.38 (2017).646 59 Vizcaino, J. A. et al. 2016 update of the PRIDE database and its related tools. Nucleic acids research 647 44, D447- 456, doi:10.1093/nar/gkv1145 (2016).
+
+<--- Page Split --->
+
+## Supplementary Files  
+
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+Supplementarytable1. xlsx Supplementarytable2. xlsx Supplementarytable3. xlsx Supplementarytable4. xlsx
+
+<--- Page Split --->

preprint/preprint__996236fe7bcffb4d384731573a4c082ea4a44831dd06c020baf33b3c539e363c/preprint__996236fe7bcffb4d384731573a4c082ea4a44831dd06c020baf33b3c539e363c_det.mmd

@@ -0,0 +1,422 @@
+<|ref|>title<|/ref|><|det|>[[44, 106, 937, 175]]<|/det|>
+# Quantitative temporal analysis of modified vaccinia Ankara, the monkeypox and smallpox vaccine  
+
+<|ref|>text<|/ref|><|det|>[[44, 195, 268, 236]]<|/det|>
+Jonas Dutra Albaraz University of Cambridge  
+
+<|ref|>text<|/ref|><|det|>[[44, 242, 268, 283]]<|/det|>
+Marisa Oliveira University of Cambridge  
+
+<|ref|>text<|/ref|><|det|>[[44, 290, 268, 330]]<|/det|>
+Joanne Kite University of Cambridge  
+
+<|ref|>text<|/ref|><|det|>[[44, 336, 625, 377]]<|/det|>
+Joao Paulo Harvard Medical School https://orcid.org/0000- 0002- 4291- 413X  
+
+<|ref|>text<|/ref|><|det|>[[44, 382, 268, 423]]<|/det|>
+Robin Antrobus University of Cambridge  
+
+<|ref|>text<|/ref|><|det|>[[44, 429, 625, 470]]<|/det|>
+Steven Gygi Harvard Medical School https://orcid.org/0000- 0001- 7626- 0034  
+
+<|ref|>text<|/ref|><|det|>[[44, 475, 625, 516]]<|/det|>
+Edward Huttlin Harvard Medical School https://orcid.org/0000- 0002- 1822- 1173  
+
+<|ref|>text<|/ref|><|det|>[[44, 521, 625, 562]]<|/det|>
+Geoffrey Smith University of Cambridge https://orcid.org/0000- 0002- 3730- 9955  
+
+<|ref|>text<|/ref|><|det|>[[44, 567, 625, 608]]<|/det|>
+Michael Weekes (mpw1001@cam.ac.uk) University of Cambridge https://orcid.org/0000- 0003- 3196- 5545  
+
+<|ref|>sub_title<|/ref|><|det|>[[44, 650, 285, 669]]<|/det|>
+## Biological Sciences - Article  
+
+<|ref|>text<|/ref|><|det|>[[42, 687, 950, 730]]<|/det|>
+Keywords: modified vaccinia Ankara, vaccinia virus, monkeypox, innate immunity, immune evasion, host- pathogen interaction, proteomics  
+
+<|ref|>text<|/ref|><|det|>[[44, 748, 296, 767]]<|/det|>
+Posted Date: July 13th, 2022  
+
+<|ref|>text<|/ref|><|det|>[[42, 786, 474, 805]]<|/det|>
+DOI: https://doi.org/10.21203/rs.3.rs- 1850393/v1  
+
+<|ref|>text<|/ref|><|det|>[[42, 823, 910, 865]]<|/det|>
+License: © This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[71, 112, 884, 160]]<|/det|>
+Quantitative temporal analysis of modified vaccinia Ankara, the monkeypox and smallpox vaccine  
+
+<|ref|>text<|/ref|><|det|>[[71, 172, 884, 219]]<|/det|>
+3 Jonas D. Albaranza<sup>a,b,#</sup>, Marisa Oliveira<sup>a,b,#</sup>, Joanne Kite<sup>a,b,#</sup>, Joao A. Paulo<sup>c</sup>, Robin Antrobus<sup>a,b</sup>, 4 Steven P. Gygi<sup>c</sup>, Edward L. Huttlin<sup>c</sup>, Geoffrey L. Smith<sup>d</sup>, Michael P. Weekes<sup>a,b</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[71, 240, 88, 255]]<|/det|>
+5  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 260, 216, 277]]<|/det|>
+## Affiliations:  
+
+<|ref|>text<|/ref|><|det|>[[70, 282, 885, 380]]<|/det|>
+a Cambridge Institute for Medical Research, University of Cambridge, Cambridge CB2 0XY, UK  b Department of Medicine, University of Cambridge, Cambridge, CB2 0XY, UK  c Department of Cell Biology, Harvard Medical School, 240 Longwood Avenue, Boston, MA 02115  USA  
+
+<|ref|>text<|/ref|><|det|>[[70, 376, 884, 422]]<|/det|>
+d Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK  
+
+<|ref|>text<|/ref|><|det|>[[70, 425, 285, 444]]<|/det|>
+# Contributed equally  
+
+<|ref|>text<|/ref|><|det|>[[70, 450, 301, 469]]<|/det|>
+Corresponding author  
+
+<|ref|>text<|/ref|><|det|>[[70, 512, 425, 530]]<|/det|>
+Correspondence: Michael P. Weekes  
+
+<|ref|>text<|/ref|><|det|>[[70, 535, 884, 554]]<|/det|>
+Cambridge Institute for Medical Research, University of Cambridge, Keith Peters Building, Hills  
+
+<|ref|>text<|/ref|><|det|>[[70, 559, 373, 577]]<|/det|>
+Road, Cambridge CB2 0XY, UK  
+
+<|ref|>text<|/ref|><|det|>[[70, 583, 352, 601]]<|/det|>
+Email: mpw1001@cam.ac.uk  
+
+<|ref|>text<|/ref|><|det|>[[70, 607, 351, 625]]<|/det|>
+Telephone: +44 1223 767811  
+
+<|ref|>text<|/ref|><|det|>[[70, 632, 88, 647]]<|/det|>
+21  
+
+<|ref|>text<|/ref|><|det|>[[70, 667, 884, 687]]<|/det|>
+Keywords: modified vaccinia Ankara; vaccinia virus; monkeypox; innate immunity; immune  
+
+<|ref|>text<|/ref|><|det|>[[70, 692, 490, 710]]<|/det|>
+evasion; host- pathogen interaction; proteomics
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[65, 114, 201, 133]]<|/det|>
+## 24 Summary  
+
+<|ref|>text<|/ref|><|det|>[[63, 148, 886, 530]]<|/det|>
+Modified vaccinia Ankara (MVA) immunisation is being deployed to curb the current outbreak of monkeypox in multiple countries<sup>1</sup>. Originally authorized for vaccination against smallpox, MVA is a vaccinia virus (VACV) strain that does not replicate in human cells or cause serious adverse events. Here, we conducted a highly multiplexed proteomic analysis<sup>2</sup> to quantify \(\sim 7,500\) cellular proteins and \(\sim 80\%\) of viral proteins at five time points throughout MVA infection of human cells<sup>3</sup>. \(>380\) human proteins were down- regulated \(>2\) - fold by MVA, revealing a profound remodelling of the host proteome. \(>25\%\) of these MVA targets, including multiple components of the nuclear pore complex (NPC), were not shared with VACV- Western Reserve<sup>4</sup>, which is derived from a first generation smallpox vaccine associated with serious adverse events. Using pharmacological inhibition of viral DNA replication and killed virions, we discovered that post- replicative gene expression is necessary for the downregulation of NPC proteins and for elements of MVA antagonism of innate immune sensing. Our approach thus provides the first global view of the impact of MVA infection on the host proteome, offers insights into how MVA interacts with the antiviral defences and identifies cellular mechanisms that may underpin the abortive infection of human cells. These discoveries will prove vital to the rational design of future generations of vaccines.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[66, 114, 223, 132]]<|/det|>
+## 41 Introduction  
+
+<|ref|>text<|/ref|><|det|>[[112, 149, 884, 338]]<|/det|>
+Monkeypox virus (MPXV) is a zoonotic orthopoxvirus endemic in Central and West Africa<sup>5,6</sup>. Since May 2022, thousands of cases of MPXV have been reported in 50 non- endemic countries worldwide, with sustained human- to- human transmission<sup>7</sup>. MPXV causes a smallpox- like illness, with severe disease seen in immunocompromised individuals, children and pregnant women. The ongoing outbreak has been caused by MPXV clade 3 (within the formerly designated "West African" clade)<sup>1</sup>. Mortality is commoner with MPXV clade 1 (formerly designated "Central African" or "Congo Basin" clade), however up to \(3.6\%\) of infections with clade 3 MPXV have previously resulted in death, albeit in some cases in the context of HIV co- infection<sup>8</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[112, 353, 884, 612]]<|/det|>
+Although the World Health Organisation (WHO) has not yet advised mass vaccination, post- exposure prophylaxis is recommended for higher- risk MPXV contacts, and pre- exposure prophylaxis for at- risk healthcare workers<sup>7</sup>. Highly effective protection can be provided by live vaccines originally developed against smallpox, which was itself eradicated through vaccination with a related orthopoxvirus, VACV<sup>9</sup>. The most recent generation of smallpox vaccines derive from modified vaccinia Ankara (MVA), itself derived by \(>570\) passages of the VACV strain chorioallantois vaccinia Ankara (CVA) in chicken embryo fibroblasts<sup>10- 12</sup>. Serial passage resulted in loss of \(\sim 30\) kb of the genome, and loss of replicative capacity in human cells<sup>3</sup>. Because MVA has been shown to be safe and immunogenic in both healthy and immunocompromised individuals<sup>13,14</sup>, it has also been investigated extensively as vaccine vector for viruses including Ebola, respiratory syncytial virus, HIV, and SARS- CoV- 2<sup>15</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[112, 629, 884, 770]]<|/det|>
+However, the biological effects of MVA at the cellular level are poorly understood, without prior systematic proteomic investigation. Viral infection induces dynamic changes in the host proteome with diverse functional consequences for virus- host interaction<sup>2</sup>, but little information is available about non- productive, abortive infections. We therefore conducted the first systematic proteomic analysis of MVA and host throughout the whole course of infection, including inactivated controls to understand the contribution of the viral particle with no- , or limited viral gene expression.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[113, 115, 183, 132]]<|/det|>
+## Results  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 163, 525, 181]]<|/det|>
+## Quantitative temporal analysis of MVA infection  
+
+<|ref|>text<|/ref|><|det|>[[111, 186, 884, 469]]<|/det|>
+To build a global picture of changes in host and viral proteins throughout the course of MVA infection, we infected telomerase reverse transcriptase (TERT)- immortalized primary human fetal foreskin fibroblasts (HFFF- TERTs) with MVA at multiplicity of infection (MOI) 5 in biological duplicate. HFFF- TERTs have been well established as a model for a variety of different viral infections \(^{2,4,16,17}\) , and MOI 5 infected \(>96\%\) of cells (Extended data Fig. 1). We used an MVA stock derived from the original MVA strain that had undergone 575 passages in vitro, since this is the direct precursor to vaccines in current use or development, including MVA- Bavarian Nordic (MVA- BN) \(^{11}\) . 16- plex TMT and triple- stage mass spectrometry (MS3) with real- time search (RTS) \(^{18,19}\) quantified protein expression over the full course of infection (Fig. 1a). In the same experiment, we included identical analyses with heat- inactivated MVA (MVA- HI), or the viral DNA replication inhibitor cytosine arabinoside (AraC), to understand the contribution of the viral particle with no gene expression, or expression of early viral genes only, respectively (Fig. 1a).  
+
+<|ref|>text<|/ref|><|det|>[[112, 484, 884, 625]]<|/det|>
+7545 human and 144 viral proteins were quantified, providing a global view of changes in protein expression during infection. The greatest changes occurred late during infection, and heat inactivation ablated the majority of host protein downregulation (Fig. 1b, Extended data Fig. 2). All data are shown in Supplementary Table 1, in which the worksheet "Plotter" enables interactive generation of temporal graphs of the expression of each of the human or viral proteins quantified.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 642, 767, 662]]<|/det|>
+## MVA regulates multiple mediators of intrinsic, innate and adaptive immunity  
+
+<|ref|>text<|/ref|><|det|>[[112, 665, 884, 830]]<|/det|>
+Over 18 h of infection, 383 human proteins were downregulated \(>2\) - fold, and 66 human proteins upregulated \(>2\) - fold. Interestingly, Database for Annotation, Visualization and Integrated Discovery (DAVID) software \(^{20}\) revealed that diverse groups of cell surface proteins were downregulated early during MVA infection including multiple NK and T- cell ligands suggesting that a key focus of the first phase of infection may be evasion of cellular immunity (Fig. 2a- b). Upregulated proteins included inflammatory mediators such as complement C4- A and lactotransferrin (LTF) \(^{21}\) (Extended data Fig. 3A). To distinguish with high confidence host factors
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[125, 120, 538, 610]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[125, 633, 537, 720]]<|/det|>
+<center>Figure 1 | Quantitative temporal proteomic analysis of MVA infection. a Schematic indicating the experimental workflow. b Hierarchical clustering of all proteins quantified in the two biological repeats. An enlargement is shown indicating groups of proteins that were significantly down- or upregulated during the course of the experiment. c Scatterplot of all proteins quantified at 18 h of infection, showing average fold change. p-values were estimated using Significance B values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>56,57</sup>. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[152, 117, 571, 650]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[570, 395, 970, 570]]<|/det|>
+<center>Figure 2 | MVA regulates multiple mediators of intrinsic, innate and adaptive immunity. a Functional enrichment analysis of proteins downregulated >2-fold at ≥1 time point during infection. b Example components from cell surface-related pathways identified in (a). Error bars = range. c Functional enrichment analysis of groups of proteins regulated as indicated in the left-hand column, and examples of components of enriched pathways. Full data is shown in Supplementary Table 2. 2-fold is used throughout the manuscript as a cutoff for downregulation, with proteins unchanged defined as downregulated <1.25 fold at all time points measured. Error bars = range. d Plot of interactions among proteins decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples (red text = proteins connected by red edges). For context, an additional 74 neighbouring proteins are shown (grey text). Relative proximity to the proteins that were decreased upon infection with MVA but unchanged in MVA+ArA and MVA-HI samples is indicated by red shading, and was quantified via random walk with restart (Methods). </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[118, 112, 933, 616]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[140, 618, 936, 644]]<|/det|>
+<center>Extended Data Figure 1. Percentage of cells expressing viral proteins at different MOIs. HFFT-TERTs were infected at the indicated MOI for 18h, then stained with a polyclonal rabbit anti-VACV antibody. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[275, 120, 800, 500]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[251, 543, 262, 554]]<|/det|>
+<center>b </center>  
+
+<|ref|>image<|/ref|><|det|>[[275, 545, 762, 956]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[125, 963, 962, 996]]<|/det|>
+<center>Extended Data Figure 2. Dot plot of all proteins quantified. Average fold change was calculated from each of two biological replicates. a. Human proteins. Significance B was used to estimate p value, and was adjusted for multiple hypothesis testing using the method of Benjamini-Hochberg<sup>60,57</sup>. b viral proteins, p values not displayed. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[125, 115, 960, 940]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 886, 494]]<|/det|>
+that were downregulated by viral gene expression at distinct stages of infection, we applied a series of filters that identified differential modulation of host protein expression in the presence of unmodified MVA, MVA+ArA C and MVA- HI (Fig. 2c, Extended data Fig. 3, Supplementary Table 2). Of particular interest, viral genes expressed later during infection exquisitely regulated multiple components of the nuclear pore complex. This was unexpected, since a recent study suggested that nuclear pore proteins may be essential for replication of poxviruses<sup>22</sup>. Proteins downregulated by early- expressed genes included secreted inflammatory mediators such as metallopeptidase inhibitor 2 (TIMP2) and transforming growth factor \(\beta - 1\) (TGFB1). Infection with heat- inactivated virus offered the opportunity to identify proteins that were upregulated upon sensing of the viral particle, but whose expression was limited by viral genes (Fig. 2c, lower panels). These included four poly ADP- ribose polymerase (PARP) proteins including Zinc Finger Antiviral Protein (ZC3HAV1, ZAP), a variety of proteins with key roles in immunity including the viral DNA sensor IFI16, interferon regulator TRIM26 and interferon stimulated genes including IFIT2 and OASL. Certain factors were downregulated in all three conditions, including extracellular matrix components and multiple collagens (Extended data Fig. 3B), suggesting that these changes may represent a cellular response to infection, as opposed to being induced by viral gene expression.  
+
+<|ref|>text<|/ref|><|det|>[[111, 508, 886, 722]]<|/det|>
+To understand better how proteins altered during various stages of MVA infection relate to one another and to the larger proteome, we superimposed groups of proteins displaying each characteristic abundance signature (Fig. 2c) onto the BioPlex network of human protein- protein interactions<sup>23</sup>. Graph assortativity was then calculated, measuring the tendency of proteins in a particular group to interact preferentially with each other compared to proteins that are not part of the selected group (Extended data Fig. 3c). Proteins downregulated during later stages of MVA infection showed a strong tendency to self- associate, including nuclear pore components, and a selection of nucleolar proteins associated with ribosome biogenesis (e.g. SURF6, RBM28, RPL7L1, ZC3HAV1, CDK105) (Fig. 2e).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 774, 720, 794]]<|/det|>
+## Differential regulation of the host proteome by MVA and virulent VACV  
+
+<|ref|>text<|/ref|><|det|>[[112, 799, 884, 866]]<|/det|>
+To discover what may underpin the differences in pathogenesis and replication between MVA vaccine, which is avirulent and replication- incompetent in human cells, and the mouse- adapted, virulent, and replication- competent VACV strain Western Reserve (VACV- WR), we next
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 110, 886, 737]]<|/det|>
+characterised similarities and difference in host regulation. We compared results from the present study with our prior proteomic analysis of VACV- WR<sup>4</sup>, which is the reference for studies with orthopoxviruses, is highly neurotropic in mice, and was derived from the New York City Board of Health first- generation smallpox vaccine. Whereas at least \(30\%\) of proteins were co- regulated by both viruses, there were a number of key differences (Fig. 3a- b, Extended data Fig. 4a, Supplemental Table 3). Proteins substantially downregulated during MVA infection but minimally modulated by VACV- WR were enriched in interacting components of the nuclear pore complex and certain glycoproteins. The latter included multiple immune modulators such as semaphorins, which modulate intercellular communication between immune cells<sup>24</sup>, transforming growth factors TGFB1 and TGFB2, and the endoplasmic reticulum stress sensor EIF2AK3 (also known as PERK) (Fig. 3a- c, Extended data Fig. 4b- d). MVA, but not VACV- WR, also induced the downregulation of negative (NFKBIA and NFKBIB, also known as IκBα and IκBβ, respectively) and positive (BIRC2, also known as cIAP1) regulators of NF- κB immune signalling (Fig. 3a, Extended data Fig. 4c- d). IκBα sequester inactive NF- κB in the cytoplasm and their proteasomal degradation releases NF- κB components to accumulate in the nucleus where they activate the expression of immunity- related genes including interferons<sup>25</sup>. Therefore, the down- regulation of NFKBIA and NFKBIB indicate activation of NF- κB during MVA, but not VACV- WR, infection. The presence of NFKBIA re- synthesis during infection with MVA+ArαC but not MVA infection also indicate an MVA blockade of the NF- κB pathway at or downstream of IκBα degradation. Conversely, proteins down- regulated during VACV- WR infection but minimally modulated by MVA were enriched in factors involved in tyrosine phosphorylation and innate immunity. The former pathway included members of the Ephrin family of receptor tyrosine kinases (EPHA4, EPHB2, and EPHB3) and the proto- oncogene FER whereas the latter included multiple interferon- stimulated genes and MAP4K2, an upstream activator of mitogen- activated protein kinases, key regulators of immune signaling downstream of pathogen sensing<sup>26</sup> (Fig. 3b, c, Extended data Fig. 3C).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 748, 784, 768]]<|/det|>
+## Temporal analysis of MVA protein expression to inform viral-host interactions  
+
+<|ref|>text<|/ref|><|det|>[[115, 772, 884, 840]]<|/det|>
+High- definition temporal analysis of viral protein expression can facilitate direct correlation between viral and cellular protein profiles to give insights into viral- host protein interaction<sup>4</sup>. We quantified \(77\%\) of all predicted MVA proteins. To define different classes of protein expression,
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[137, 120, 963, 400]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[137, 402, 960, 460]]<|/det|>
+<center>Figure 3 | Differential protein changes during infection of human fibroblasts with MVA or VACV-WR. a Scatterplot of all proteins quantified at 18 h of infection with both viruses. p-values were estimated using Significance A values corrected for multiple hypothesis testing using the Benjamini-Hochberg method<sup>65,67</sup>. Example profiles for NF-κB pathway proteins are shown in the lower panels. Error bars = range. b Functional enrichment analysis of proteins (i) downregulated by MVA but not VACV Western Reserve (VACV-WR), (ii) downregulated by VACV-WR but not MVA, or (iii) downregulated by both MVA and VACV-WR. c Example components from pathways identified in (b). Error bars = range. </center>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[130, 120, 940, 950]]<|/det|>
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 112, 884, 183]]<|/det|>
+we employed k- means clustering and included data from samples treated with AraC. This suggested that there are at least four distinct temporal protein profiles of viral protein expression (Fig. 4A, Extended data Fig. 5).  
+
+<|ref|>text<|/ref|><|det|>[[111, 197, 886, 530]]<|/det|>
+Some of the earliest Tp1 (temporal profile 1) and Tp2 classes of viral protein exhibited increased expression in the presence of AraC, likely explained by the absence of negative regulation from post- replicative proteins. In contrast, the Tp3- class proteins were partly inhibited by AraC, with expression of Tp4 class proteins almost completely inhibited (Fig. 4A- B). Tp4 class proteins included multiple VACV envelope proteins, including B5 and A27 homologues PS/HR and MVA138 whose late expression may relate to roles in virus- specific T- cell memory<sup>27</sup> (Supplementary Table 4). Tp1- class proteins included MVA102 (VACV- WR D5), which uncoats viral genomes early during infection<sup>28</sup>, and MVA019 (VACV- WR C6). We previously demonstrated that C6 targets host restriction factors HDAC4 and HDAC5 for proteasomal degradation<sup>4,29</sup>. This classification thus serves as a particularly useful resource to predict interactions between host factors downregulated at early, intermediate or late times after infection and their viral counterparts and may help to form a rational basis for prediction of the effects of mutations introduced in individual genes during vaccine production by additional passage, plaque purification or insertion of foreign genes derived from other pathogens.
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[118, 120, 364, 870]]<|/det|>
+
+<--- Page Split --->
+<|ref|>image<|/ref|><|det|>[[125, 120, 416, 272]]<|/det|>
+<|ref|>image_caption<|/ref|><|det|>[[130, 291, 564, 302]]<|/det|>
+<center>Extended Data Figure 5. Number of temporal classes of MVA gene expression. </center>  
+
+<|ref|>text<|/ref|><|det|>[[130, 303, 565, 397]]<|/det|>
+The k- means approach was used with 1- 15 classes to cluster viral proteins, and the summed distance of each protein from its cluster centroid was calculated. Although this summed distance necessarily becomes smaller as more clusters are added, the rate of decline decreases with each added group, eventually settling at a fairly constant rate of decline that reflects overfitting; clusters added prior to this point reflect underlying structure in the temporal protein data, whereas clusters subsequently added through overfitting are not informative. The point of inflexion fell between four and six classes, suggesting that there are at least four distinct temporal protein profiles of viral protein expression.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[65, 114, 214, 132]]<|/det|>
+## 186 Discussion  
+
+<|ref|>text<|/ref|><|det|>[[111, 150, 885, 410]]<|/det|>
+Upon infection of human cells, MVA does not produce infectious new virions and is therefore classed as a replication- incompetent vaccine<sup>12</sup>. In the spectrum of vaccines available for use against human diseases, replication- incompetent vaccines lie between attenuated vaccines and inactivated vaccines, without the safety concerns of the former or the reduced immunogenicity of the latter<sup>30</sup>. Despite an ultimately abortive MVA infection of human cells, the viral life- cycle is typical of other poxviruses until early morphogenesis, including early gene expression, genome uncoating, DNA replication, and post- replicative gene expression<sup>31</sup>. Our global analysis of MVA infection highlighted a pervasive modulation of human host proteins and processes. Such an understanding, particularly including regulation of functions related to immunity, is critical for a complete assessment of MVA's applicability and safety as a vaccine against monkeypox and other orthopoxviruses, and as a vaccine vector against other pathogens.  
+
+<|ref|>text<|/ref|><|det|>[[110, 425, 885, 830]]<|/det|>
+Interferons and other cytokines induced upon viral sensing are key regulatory molecules in innate immunity and inflammation and play important roles in fine- tuning antigen- specific adaptive immunity during viral infection<sup>32,33</sup>. MVA- induced regulation of the interferon response and NF- \(\kappa\) B signaling, indicated both by limited upregulation of interferon- stimulated gene (ISG) products and direct regulation of NF- \(\kappa\) B, will therefore be likely to impact the immunological memory elicited by vaccination. Interferon regulation is likely to occur at all stages of infection, indicated by the increased response seen upon blockage of post- replicative viral gene expression and in cells exposed to heat- inactivated virions in comparison to cells infected with unmodified MVA. This is in contrast with the antagonism of the interferon response by VACV- WR. For instance, VACV early protein C9 targets ISG products IFIT1, IFIT2, and IFIT3 for proteosomal degradation<sup>34</sup>. Through serial passage in chicken embryo fibroblasts, the MVA genome has accumulated six major deletions and many small insertion and deletions (indels), which, in combination, result in the loss of 25 open reading frames (ORFs) when compared to wildtype VACV strains<sup>3,35</sup>. In addition, 12 ORFs have been split due to mutations, including the ORF encoding a C9 orthologue, explaining the absence of IFIT degradation during MVA infection<sup>35</sup>. Analysis both of viral protein temporal profiles and inactivating mutations are thus likely to yield candidates antagonists for host factors downregulated during VACV- WR, but not MVA infection including Ephrins and MAP4K2.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 111, 884, 301]]<|/det|>
+Host proteins with concordant modulation by MVA and VACV- WR might indicate that viral mechanisms of host modulation are still functional in MVA. Examples include common downregulation of the antiviral restriction factor HDAC5 by VACV- WR and MVA, mediated by VACV- WR protein C6 (MVA019)4. Proteasomal degradation represents the main mechanism responsible for downregulation of host protein during VACV infection4 and therefore, most host proteins downregulated during MVA infection are likely also to be degraded by the proteasome. Alternatively, common modulation of protein expression might reflect common cellular responses to infection, such as the downregulation of multiple collagens.  
+
+<|ref|>text<|/ref|><|det|>[[111, 317, 884, 650]]<|/det|>
+101 proteins were downregulated during MVA, but not VACV- WR infection. These included multiple components of the nuclear pore complex, whose downregulation is likely to disrupt nucleocytoplasmic transport36. The innate immune response to virus infection requires an intact NPC to mediate nucleocytoplasmic transport of transcription factors and mRNAs and therefore, downregulation/degradation of NPC proteins may be a viral immune evasion strategy. Although some viral proteases inactivate nucleoporins (NUPs) by cleavage37, NUP abundance did not vary substantially during infection by multiple human pathogenic viruses, such as human cytomegalovirus2, herpes simplex virus type 116, influenza A virus38, Epstein- Barr virus39, SARSCoV- 240 and HIV41. Therefore, to our knowledge, MVA is the first virus reported to downregulate/degrade NUPs. Three of the NUPs down- regulated by MVA (NUP54, NUP62, NUP88) were identified as host factors necessary for VACV- WR morphogenesis22, raising the possibility that NUP downregulation contributes to arrested virion maturation during MVA infection of human cells. It remains to be determined whether this unique MVA- induced modulation is shared with the parental strain (CVA) or was acquired during the serial passage in chicken embryo fibroblasts.  
+
+<|ref|>text<|/ref|><|det|>[[111, 665, 884, 880]]<|/det|>
+Proteins upregulated by heat- inactivated MVA included five members of the poly- ADP- ribosyl polymerase (PARP) family, some of which have well- characterized roles in antiviral and inflammatory responses, such as ZC3HAV1 (also known as PARP13 or ZAP)42. ZAP is an antiviral factor that restricts MVA infection in human cells43. VACV- WR induces the proteasomal degradation of ZAP and some other VACV strains encode proteins that interact with and overcome its antiviral function44,45. Blockade of MVA DNA replication and post- replicative gene expression upregulated ZAP, whilst PARPs 9, 10, 12 and 14 were only upregulated by heat- inactivated virus. This suggests that these latter PARPs might be targeted by early gene product(s), may indicate that they serve as- yet uncharacterized antiviral function, and highlights
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 884, 155]]<|/det|>
+the usefulness of use of our multi- level method to reveal regulation that would otherwise not be observed.  
+
+<|ref|>text<|/ref|><|det|>[[111, 173, 885, 386]]<|/det|>
+In sum, our quantitative temporal analysis of the infection of human cells by the MVA vaccine has revealed an extensive remodelling of the host proteome. This highlighted that, at different stages of infection, MVA retains a limited capacity to antagonise innate immune sensing and exhibits features not present in the VACV- WR strain related to the first- generation vaccine. Our data offer an opportunity to dissect how multiple, seemingly redundant MVA and VACV immunomodulatory effectors function and to discover novel viral strategies to escape antiviral immunity. A comprehensive understanding of all these issues may be essential to generate future, fourth- generation smallpox/monkeypox vaccines in addition to highly potent recombinant vaccines against other infections.
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[113, 115, 193, 132]]<|/det|>
+## Methods  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 152, 266, 169]]<|/det|>
+## Cells and viruses  
+
+<|ref|>text<|/ref|><|det|>[[112, 175, 884, 433]]<|/det|>
+Primary human foetal foreskin fibroblast cells immortalised with human telomerase (HFFF- TERTs, male) and primary chicken embryo fibroblasts (CEFs) were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with foetal bovine serum (FBS: \(10\%\) v/v), and penicillin/streptomycin at \(37^{\circ}C\) in \(5\%\) CO2. HFFF- TERTs have been tested at regular intervals since isolation to confirm both that the HLA and MICA genotypes, and the morphology and antibiotic resistances are consistent with the original description<sup>44</sup>. In addition, human fibroblasts (dermal or foreskin) are the only permissive cells for human cytomegalovirus (HCMV) in cell culture, and HFFF- TERTs are routinely infected with the HCMV Merlin strain, further limiting the chances that the cells have been contaminated with another cell type<sup>17</sup>. Fresh CEFs were obtained from the Pirbright Institute (Woking, UK) and directly seeded for virus production without further passages. All cell lines used regularly tested negative for mycoplasma.  
+
+<|ref|>text<|/ref|><|det|>[[112, 448, 884, 589]]<|/det|>
+Modified vaccinia Ankara (MVA) was obtained as a seed stock prepared from an original MVA stock at passage \(575^{45}\) . MVA was propagated in CEFs, purified by ultracentrifugation through two \(36\%\) (w/v) sucrose cushions and suspended in \(10 \text{mM Tris - HCl pH 9.0}\) . MVA infectivity was determined by immunocytochemistry on HFFF- TERTs cells, by using a polyclonal rabbit anti- VACV antibody<sup>46</sup>. For heat inactivation, MVA stock was diluted ten- fold in DMEM supplemented with \(2\%\) FBS and heated at \(56^{\circ}C\) for \(1 \text{h}\) in a water bath<sup>47</sup>.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 607, 370, 625]]<|/det|>
+## Viral infections and inhibitors  
+
+<|ref|>text<|/ref|><|det|>[[112, 630, 884, 770]]<|/det|>
+For proteomic experiments, \(7.5 \times 10^{5}\) HFFF- TERTs were seeded in a \(25 - \text{cm}^2\) flask two days prior to infection. Cells were infected with MVA, or the equivalent amount of heat- inactivated MVA, at MOI 5 for 2, 4, 8, 12, or \(18 \text{h}\) , with time zero defined as the time the virus was added to the cells. A mock- infected control was harvested at \(12 \text{h}\) post- infection. Where indicated, cells were incubated with cytosine arabinoside (AraC) at \(40 \mu \text{g/ml}\) from the time zero. The time course experiments were performed in biological duplicate.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 788, 250, 805]]<|/det|>
+## Flow cytometry  
+
+<|ref|>text<|/ref|><|det|>[[112, 811, 884, 875]]<|/det|>
+MVA- infected HFFF- TERTs were detached with trypsin- EDTA (Gibco) \(24 \text{h}\) post- infection, passed through a \(70 - \mu \text{m}\) cell strainer, washed in PBS and stained with Zombie Violet viability dye (BioLegend) for \(30 \text{min}\) at \(4^{\circ}C\) in the dark. Cells were collected by centrifugation, washed once
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[112, 112, 884, 255]]<|/det|>
+with PBS and fixed in Cytofix/Cytoperm fixation and permeabilization solution (BD Biosciences) for 30 min at \(4^{\circ}C\) in the dark. Cell were washed twice with Perm/Wash buffer (BD Biosciences) and stained with a polyclonal rabbit anti- VACV antibody<sup>46</sup> diluted 1:500 in Perm/Wash buffer, followed by AlexaFluor 568-conjugated goat anti-rabbit IgG (Invitrogen) diluted 1:100. Stained cells were suspended in PBS before data acquisition with an Attune NxT flow cytometer. Data were analysed with FlowJo software.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 270, 414, 290]]<|/det|>
+## Whole cell lysate protein digestion  
+
+<|ref|>text<|/ref|><|det|>[[112, 293, 884, 625]]<|/det|>
+Two independent biological replicates were performed, 'WCL1' and 'WCL2'. For each replicate, cells were washed twice with PBS, and 250 μl lysis buffer added (6 M Guanidine/50 mM HEPES pH 8.5). Cell lifters (Corning) were used to scrape cells in lysis buffer, which was removed to an eppendorf tube, vortexed extensively then sonicated. Cell debris was removed by centrifuging at 21,000 g for 10 min twice. For half of each sample, dithiothreitol (DTT) was added to a final concentration of 5 mM and samples were incubated for 20 min. Cysteines were alkylated with 14 mM iodoacetamide and incubated 20 min at room temperature in the dark. Excess iodoacetamide was quenched with DTT for 15 min. Samples were diluted with 200 mM HEPES pH 8.5 to 1.5 M Guanidine followed by digestion at room temperature for 3 h with LysC protease at a 1:100 protease- to- protein ratio. Samples were further diluted with 200 mM HEPES pH 8.5 to 0.5 M Guanidine. Trypsin was then added at a 1:100 protease- to- protein ratio followed by overnight incubation at \(37^{\circ}C\) . The reaction was quenched with 5% formic acid, then centrifuged at 21,000 g for 10 min to remove undigested protein. Peptides were subjected to C18 solid- phase extraction (SPE, Sep- Pak, Waters) and vacuum- centrifuged to near- dryness.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 641, 459, 660]]<|/det|>
+## Peptide labeling with tandem mass tags  
+
+<|ref|>text<|/ref|><|det|>[[112, 664, 884, 880]]<|/det|>
+In preparation for TMT labeling, desalted peptides were dissolved in 200 mM HEPES pH 8.5. Peptide concentration was measured by microBCA (Pierce), and 25 μg of peptide labeled with TMT reagent. TMT reagents (0.8 mg) were dissolved in 43 μl anhydrous acetonitrile and 5 μl added to peptide at a final acetonitrile concentration of 30% (v/v). Sample labelling was: 126, MVA (2h infection); 127N, MVA+ArA (2h); 127C, MVA-HI (2h); 128N, MVA (4h); 128C, MVA+ArA (4h); 129N, MVA-HI (4h); 129C, MVA (8h); 130N, MVA+ArA (8h); 130C, MVA-HI (8h); 131N, MVA (12h); 131C, MVA+ArA (12h); 132N, MVA-HI (12h); 132C, MVA (18h); 133N, MVA+ArA (18h); 133C, MVA-HI (18h); 134N, mock infection. Following incubation at room temperature for 1 h, the reaction was quenched with hydroxylamine to a final concentration of
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[110, 112, 885, 300]]<|/det|>
+0.3% (v/v). TMT- labeled samples were combined at a 1:1:1:1:1:1:1:1:1:1:1;1:1:1:1:1:1:1 ratio. The sample was vacuum- centrifuged to near dryness and subjected to C18 SPE (Sep- Pak, Waters). An unfractionated singleshot was analysed initially to ensure similar peptide loading across each TMT channel, thus avoiding the need for excessive electronic normalization. As all normalisation factors were \(>0.5\) and \(<2\) , data for each singleshot experiment was analysed with data for the corresponding fractions to increase the overall number of peptides quantified. Normalisation is discussed in 'Data Analysis', and high pH reversed- phase (HpRP) fractionation is discussed below.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 318, 343, 336]]<|/det|>
+## Offline HpRP fractionation  
+
+<|ref|>text<|/ref|><|det|>[[110, 341, 885, 770]]<|/det|>
+TMT- labelled tryptic peptides were subjected to HpRP fractionation using an Ultimate 3000 RSLC UHPLC system (Thermo Fisher Scientific) equipped with a 2.1 mm internal diameter (ID) \(\times 25\) cm long, \(1.7 \mu \mathrm{m}\) particle Kinetix Evo C18 column (Phenomenex). Mobile phase consisted of A: \(3\%\) acetonitrile (MeCN), B: MeCN and C: \(200 \mathrm{mM}\) ammonium formate pH 10. Isocratic conditions were \(90\%\) A / \(10\%\) C, and C was maintained at \(10\%\) throughout the gradient elution. Separations were conducted at \(45^{\circ}\mathrm{C}\) . Samples were loaded at \(200 \mu \mathrm{l} / \mathrm{min}\) for 5 min. The flow rate was then increased to \(400 \mu \mathrm{l} / \mathrm{min}\) over 5 min, after which the gradient elution proceed as follows: \(0 - 19\%\) B over 10 min, \(19 - 34\%\) B over 14.25 min, \(34 - 50\%\) B over 8.75 min, followed by a 10 min wash at \(90\%\) B. UV absorbance was monitored at \(280 \mathrm{nm}\) and 15 s fractions were collected into 96- well microplates using the integrated fraction collector. Fractions were recombined orthogonally in a checkerboard fashion, combining alternate wells from each column of the plate into a single fraction, and commencing combination of adjacent fractions in alternating rows. Wells were excluded prior to the start or after the cessation of elution of peptide- rich fractions, as identified from the UV trace. This yielded two sets of 12 combined fractions, A and B, which were dried in a vacuum centrifuge and resuspended in \(10 \mu \mathrm{l}\) MS solvent ( \(4\%\) MeCN / \(5\%\) formic acid) prior to LC- MS3. 12 set 'A' fractions were used for MS3 analysis of all experiments, and half of each sample was subjected to MS3 analysis, with the other half subjected to real- time search (RTS) MS3 analysis<sup>48</sup> (described below).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 785, 505, 804]]<|/det|>
+## LC-MS3 for TMT and TMT/SILAC experiments  
+
+<|ref|>text<|/ref|><|det|>[[115, 809, 884, 876]]<|/det|>
+For 'singleshot' analysis of unfractionated peptides from each experiment, mass spectrometry data was acquired using an Orbitrap Lumos. Subsequently, fractions were acquired using an Orbitrap Eclipse.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 885, 492]]<|/det|>
+For Orbitrap Lumos analyses: An Ultimate 3000 RSLC nano UHPLC equipped with a \(300\mu \mathrm{m}\) ID \(\times 5 \mathrm{mm}\) Acclaim PepMap \(\mu\) - Precolumn (Thermo Fisher Scientific) and a \(75\mu \mathrm{m}\) ID \(\times 50 \mathrm{cm} 2.1 \mu \mathrm{m}\) particle Acclaim PepMap RSLC analytical column was used. Loading solvent was \(0.1\%\) FA, analytical solvent A: \(0.1\%\) FA and B: \(80\%\) MeCN \(+0.1\%\) FA. All separations were carried out at \(40^{\circ}\mathrm{C}\) . Samples were loaded at \(5 \mu \mathrm{L} / \mathrm{min}\) for \(5 \mathrm{min}\) in loading solvent before beginning the analytical gradient. The following gradient was used: \(3 - 7\%\) B over \(2 \mathrm{min}\) , \(7 - 37\%\) B over \(173 \mathrm{min}\) , followed by a \(4 \mathrm{min}\) wash at \(95\%\) B and equilibration at \(3\%\) B for \(15 \mathrm{min}\) . Each analysis used a MultiNotch MS3- based TMT method \(^{49}\) . The following settings were used: MS1: 380- 1500 Th, 120,000 resolution, \(2 \times 10^{5}\) automatic gain control (AGC) target, \(50 \mathrm{ms}\) maximum injection time. MS2: Quadrupole isolation at an isolation width of \(m / z 0.7\) , CID fragmentation (normalised collision energy (NCE) 34) with ion trap scanning in turbo mode, with \(1.5 \times 10^{4}\) AGC target and \(120 \mathrm{ms}\) maximum injection time. MS3: In Synchronous Precursor Selection mode the top 10 MS2 ions were selected for HCD fragmentation (NCE 45) and scanned in the Orbitrap at 60,000 resolution with an AGC target of \(1 \times 10^{5}\) and a maximum accumulation time of \(150 \mathrm{ms}\) . Ions were not accumulated for all parallelisable time. The entire MS/MS/MS cycle had a target time of \(3 \mathrm{s}\) . Data analysis is discussed below.  
+
+<|ref|>text<|/ref|><|det|>[[111, 508, 885, 865]]<|/det|>
+For Orbitrap Eclipse analyses: The FAIMS Pro interface (ThermoFisher Scientific, San Jose, CA) was coupled to a Proxeon EASY- nLC 1200 liquid chromatograph (LC) (ThermoFisher Scientific, San Jose, CA). Peptides were separated on a \(100 \mu \mathrm{m}\) inner diameter micropapillary column packed with \(\sim 35 \mathrm{cm}\) of Accucore150 resin (2.6 \(\mu \mathrm{m}\) , \(150 \mathrm{\AA}\) , ThermoFisher Scientific, San Jose, CA). For each analysis, \(1 - 2 \mu \mathrm{g}\) of peptide was loaded onto the column and fractionated over a \(90 \mathrm{min}\) gradient of \(7\) to \(27\%\) acetonitrile in \(0.125\%\) formic acid at a flow rate of \(\sim 600 \mathrm{nL} / \mathrm{min}\) . Mass spectrometric data for each sample were collected using two distinct data acquisition modes (SPS- MS3 and RTS- MS3), all with FAIMS. SPS- MS3 data were collected with a FAIMS compensation voltage (CV) set of - 40 V, - 60 V, and - 80 V, while RTS- MS3 data were collected with a FAIMS CV set of - 30 V, - 50 V, and - 70 V, with each segment as a 1 s TopSpeed method. For both acquisition methods, the scan sequence began with an MS1 spectrum (Orbitrap analysis; resolution, 60,000; mass range, 350- 1350 Th; automatic gain control (AGC) target \(100\%\) ; maximum injection time, auto). Precursors (charge states 2- 5; precursor fit \(50\%\) at \(0.7 \mathrm{Th}\) ; minimum intensity of \(15 \mathrm{K}\) ) were then selected for MS2/MS3 analysis \(^{50}\) . MS2 analysis consisted of collision- induced dissociation (CID) with quadrupole ion trap analysis, using the following
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 884, 254]]<|/det|>
+parameters: scan speed, turbo; AGC target, \(100\%\) ; NCE, 35; q- value, 0.25; maximum injection time, 35 ms; and isolation window, 0.5 Th. MS3 precursors were fragmented by HCD and analyzed using the Orbitrap with the following parameters: resolution, 50,000; NCE, 55; AGC, \(250\%\) ; maximum injection time, 86 ms; maximum synchronous precursor selection (SPS) ions, 10; and isolation window, 1.2 Th. RTS- MS3 data were collected with the "close- out" parameter set to two. Dynamic exclusion was set at 120 sec with \(+ / - 10\) ppm error tolerance.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 273, 232, 291]]<|/det|>
+## Data analysis  
+
+<|ref|>text<|/ref|><|det|>[[112, 309, 884, 424]]<|/det|>
+In the following description, the first report in the literature for each relevant algorithm is listed. Mass spectra were processed using Sequest- based in house software suite, a software pipeline for quantitative proteomics, developed by Professor Steven Gygi's laboratory at Harvard Medical School. MS spectra were converted to mzXML using an extractor built upon Thermo Fisher's RAW File Reader library (version 4.0.26).  
+
+<|ref|>text<|/ref|><|det|>[[111, 440, 884, 725]]<|/det|>
+After converting RAW files to mzxml format, MS data were analyzed to correct errors in monoisotopic peak assignment using a published algorithm<sup>51</sup> and then searched against a database of protein sequences. A combined database was constructed from (a) the human Uniprot database (accessed \(20^{\text{th}}\) June 2022, UP000005640), (b) the MVA proteome (accessed \(20^{\text{th}}\) June 2022, UP000172909), (c) common contaminants such as porcine trypsin and endoproteinase LysC. The combined database was concatenated with a reverse database composed of all protein sequences in reversed order. Sequest searches were performed using a 20 ppm precursor ion tolerance<sup>52,53</sup>. Product ion tolerance was set to 1 Da. Oxidation of methionine residues (15.99492 Da) was set as a variable modification. TMT tags on lysine residues and peptide N termini (304.207145 Da) and carbamidomethylation of cysteine residues (57.02146 Da) were included as static modifications. Peptides were assumed to be fully tryptic with up to two missed cleavages.  
+
+<|ref|>text<|/ref|><|det|>[[111, 741, 884, 884]]<|/det|>
+To control the fraction of erroneous protein identifications, a target- decoy strategy was employed<sup>54</sup>. Peptide spectral matches (PSMs) were filtered to an initial peptide- level false discovery rate (FDR) of \(1\%\) with subsequent filtering to attain a final protein- level FDR of \(1\%\) . PSM filtering was performed using a linear discriminant analysis, as described previously<sup>54</sup>. This distinguishes correct from incorrect peptide IDs in a manner analogous to the widely used Percolator algorithm (https://noble.gs.washington.edu/proj/percolator/), though employing a
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 884, 230]]<|/det|>
+distinct machine learning algorithm. The following parameters were considered: XCorr, \(\Delta \mathsf{Cn}\) , missed cleavages, peptide length, charge state, and precursor mass accuracy. Peptides shorter than seven amino acids in length or with XCorr less than 1.0 were excluded prior to LDA filtering. Protein assembly was guided by principles of parsimony to produce the smallest set of proteins necessary to account for all observed peptides (algorithm described previously54).  
+
+<|ref|>text<|/ref|><|det|>[[111, 245, 885, 625]]<|/det|>
+Proteins were quantified by summing TMT reporter ion counts across all matching peptide- spectral matches, as described previously49. Briefly, a 0.003 Th window around the theoretical m/z of each reporter ion (126, 127n, 127c, 128n, 128c, 129n, 129c, 130n, 130c, 131n, 131c, 132n, 132c, 133n, 133c, 134n) was scanned for ions, and the maximum intensity nearest to the theoretical m/z was used. The primary determinant of quantitation quality is the number of TMT reporter ions detected in each MS3 spectrum, which is directly proportional to the signal- to- noise (S:N) ratio observed for each ion. Conservatively, every individual peptide used for quantitation was required to contribute sufficient TMT reporter ions (minimum of \(\sim 1250\) per spectrum) so that each on its own could be expected to provide a representative picture of relative protein abundance49. An isolation specificity filter with a cutoff of \(50\%\) was employed to minimise peptide co- isolation49. Peptide- spectral matches with poor quality MS3 spectra (more than 15 TMT channels missing and/or a combined S:N ratio of less than 250 across all TMT reporter ions) or no MS3 spectra at all were excluded from quantitation. Peptides meeting the stated criteria for reliable quantitation were then summed by parent protein, in effect weighting the contributions of individual peptides to the total protein signal based on their individual TMT reporter ion yields. Protein quantitation values were exported for further analysis in Excel.  
+
+<|ref|>text<|/ref|><|det|>[[111, 641, 885, 851]]<|/det|>
+For protein quantitation, reverse and contaminant proteins were removed, then each reporter ion channel was summed across all quantified proteins and normalised assuming equal protein loading across all channels. For further analysis and display, fractional TMT signals were used (i.e. reporting the fraction of maximal signal observed for each protein in each TMT channel, rather than the absolute normalized signal intensity). This effectively corrected for differences in the numbers of peptides observed per protein. In three instances (for MVA proteins MVA045 and MVA131 and human protein DZIP1L), the S:N in the mock sample was zero. For the purposes of calculating fold change of each infected sample compared to mock, these mock values were set to \(2\%\) of the maximum S:N for any sample in the same WCL experiment.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 884, 181]]<|/det|>
+For analysis of the viral proteome, the MVA proteome (UP000172909) was matched with the VACV- WR proteome (UP000000344) using the PathoSystems Resource Integration Center (PATRIC)'s proteome comparison tool with default settings<sup>55</sup>.  
+
+<|ref|>text<|/ref|><|det|>[[112, 198, 884, 265]]<|/det|>
+Hierarchical centroid clustering based on uncentered Pearson correlation, and k- means clustering were performed using Cluster 3.0 (Stanford University) and visualised using Java Treeview (http://jtreeview.sourceforge.net) unless otherwise noted.  
+
+<|ref|>sub_title<|/ref|><|det|>[[114, 283, 279, 301]]<|/det|>
+## Statistical analysis  
+
+<|ref|>text<|/ref|><|det|>[[112, 307, 884, 420]]<|/det|>
+Two biological replicates were employed throughout. P- values were estimated using the methods of Significance A or B as indicated in the figure legends, which consider two sides of a normal distribution with non- equal standard deviations, with multiple hypothesis correction using the method of Benjamini- Hochberg in Perseus version 1.5.1.6<sup>56,57</sup>. Blinding or sample- size estimation was not appropriate for this study. There were no inclusion criteria and no data was excluded.  
+
+<|ref|>sub_title<|/ref|><|det|>[[114, 440, 266, 457]]<|/det|>
+## Pathway analysis  
+
+<|ref|>text<|/ref|><|det|>[[112, 462, 884, 625]]<|/det|>
+The Database for Annotation, Visualisation and Integrated Discovery (DAVID) was used to determine pathway enrichment<sup>20</sup>. A given cluster was always searched against a background of all proteins quantified within the relevant experiment. Downregulated proteins were defined as those factors that decreased \(>2\) - fold in abundance compared to the mock sample at \(\geq 1\) time point during infection. Proteins that were not downregulated were defined as factors that decreased \(< 1.25\) - fold in abundance compared to the mock sample at all times of infection. Upregulated proteins were defined in a similar manner.  
+
+<|ref|>sub_title<|/ref|><|det|>[[114, 644, 285, 661]]<|/det|>
+## Interaction analysis  
+
+<|ref|>text<|/ref|><|det|>[[112, 666, 884, 783]]<|/det|>
+Interaction analysis was performed on groups of proteins defined by changes in abundance during MVA, MVA+ArAC, MVA- HI and VACV infection indicated in Figures 2c and 3b. Network analyses were performed using interactions from BioPlex 3.0<sup>23</sup>. For this analysis, both 293T and HCT116 networks were merged. All analyses were performed in Mathematica 13.1 (Wolfram Research).  
+
+<|ref|>text<|/ref|><|det|>[[112, 800, 884, 866]]<|/det|>
+To assess the tendency for sets of differentially expressed proteins to interact with each other, each was mapped onto the combined BioPlex network after converting all protein identifiers to Entrez GeneIDs. Graph assortativity was then calculated and the process repeated using 1000
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[113, 112, 883, 156]]<|/det|>
+randomized protein sets of equal size. Scores from randomized protein sets were then fit to a Gaussian distribution and Z- scores and p- values were estimated.  
+
+<|ref|>text<|/ref|><|det|>[[112, 173, 884, 386]]<|/det|>
+To produce interaction networks shown in Figures 2d and Extended data Fig.4d, proteins were grouped according to their differential expression under each experimental condition. Each group was then mapped onto the BioPlex network and network propagation was used to identify additional nodes that closely associate with proteins in each group. Network propagation was calculated via random walk with restart as described previously<sup>58</sup> with the restart probability set to 0.5. After 40 iterations, proteins were sorted by descending weights. In Figure 2e the top 100 highest ranking proteins were selected for display. In Extended data Fig.4d the top 150 highest ranking proteins were selected for display, though for clarity those proteins that did not interact with any other proteins on this list were excluded.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 403, 257, 420]]<|/det|>
+## Data Availability  
+
+<|ref|>text<|/ref|><|det|>[[115, 426, 884, 517]]<|/det|>
+The mass spectrometry proteomics data will be deposited to the ProteomeXchange Consortium (http://www.proteomexchange.org/) via the PRIDE<sup>59</sup> partner repository. All materials described in this manuscript, and any further details of protocols employed can be obtained on request from the corresponding author by email to mpw1001@cam.ac.uk.  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 535, 263, 552]]<|/det|>
+## Code Availability  
+
+<|ref|>text<|/ref|><|det|>[[113, 558, 884, 722]]<|/det|>
+The data analysis pipeline as a whole may be licensed from Harvard Medical School, and constituent components have been published and referenced above. These include a RAW file to mzXML conversion tool, built upon Thermo Fisher's RAW File Reader library (version 4.0.26); a monoisotopic mass assignment tool<sup>51</sup>; the Sequest algorithm<sup>53</sup>; strategies for target- decoy analysis, peptide spectral match filtering and protein assembly<sup>54</sup>, and methods for protein quantification and isolation specificity filtering<sup>49</sup> are recapitulated in Proteome Discoverer Software (Thermo Fisher, UK).  
+
+<|ref|>sub_title<|/ref|><|det|>[[115, 739, 277, 757]]<|/det|>
+## Acknowledgments  
+
+<|ref|>text<|/ref|><|det|>[[113, 774, 884, 865]]<|/det|>
+This research was funded in part by Medical Research Council Project Grant MR/W025647/1 to M.P.W., R01 GM132129 to J.A.P., R021 GM67945 to S.P.G., National Institute of Health grant U24HG006673 to E.L.H and S.P.G. This study was additionally supported by the Cambridge Biomedical Research Centre, UK. For the purpose of open access, the author has applied a CC
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[111, 112, 884, 156]]<|/det|>
+BY public copyright licence to any Author Accepted Manuscript version arising from this submission.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 199, 300, 217]]<|/det|>
+## Author Contributions  
+
+<|ref|>text<|/ref|><|det|>[[112, 234, 884, 303]]<|/det|>
+Conceptualization: M.P.W. Investigation: J.D.A., M.O., J.K., J.A.P., R.A. Data analysis: J.D.A., M.O., J.K., E.H., M.P.W. Funding acquisition: S.P.G, E.L.H, G.L.S, M.P.W. Supervision: S.P.G, E.L.H, G.L.S, M.P.W. Writing: J.D.A., M.O., J.K., J.A.P., R.A., S.P.G, E.L.H, G.L.S, M.P.W.  
+
+<|ref|>sub_title<|/ref|><|det|>[[113, 320, 375, 339]]<|/det|>
+## Competing Interest Statement  
+
+<|ref|>text<|/ref|><|det|>[[113, 356, 468, 375]]<|/det|>
+The authors declare no competing interests.
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 150, 886, 831]]<|/det|>
+510 1 Isidro, J. et al. Phylogenomic characterization and signs of microevolution in the 2022 multi- 511 country outbreak of monkeypox virus. Nat Med, doi:10.1038/s41591- 022- 01907- y (2022). 512 2 Weekes, M. P. et al. Quantitative temporal viromics: an approach to investigate host- pathogen 513 interaction. Cell 157, 1460- 1472 (2014). 514 3 Volz, A. & Sutter, G. Modified Vaccinia Virus Ankara: History, Value in Basic Research, and Current 515 Perspectives for Vaccine Development. Adv Virus Res 97, 187- 243, 516 doi:10.1016/bs.avir.2016.07.001 (2017). 517 4 Soday, L. et al. Quantitative temporal proteomic analysis of vaccinia virus infection reveals 518 regulation of histone deacetylases by an interferon antagonist. Cell Rep 27, 1920- 1933 e1927, 519 doi:10.1016/j.celrep.2019.04.042 (2019). 520 5 Yinka- Ogunleye, A. et al. Outbreak of human monkeypox in Nigeria in 2017- 18: a clinical and 521 epidemiological report. Lancet Infect Dis 19, 872- 879, doi:10.1016/S1473- 3099(19)30294- 4 522 (2019). 523 6 Babkin, I. V., Babkina, I. N. & Tikunova, N. V. An Update of Orthopoxvirus Molecular Evolution. 524 Viruses 14, doi:10.3390/v14020388 (2022). 525 7 World Health Organization (WHO). Multi- country monkeypox outbreak: situation update (2022). 526 8 Bunge, E. M. et al. The changing epidemiology of human monkeypox- A potential threat? A 527 systematic review. PLoS Negl Trop Dis 16, e0010141, doi:10.1371/journal.pntd.0010141 (2022). 528 9 Fenner, F., Henderson, D. A., Arita, I., Jezek, Z. & Ladnyi, I. D. Smallpox and its eradication. World 529 Health Organisation, Geneva (1988). 530 10 Sanchez- Sampedro, L. et al. The evolution of poxvirus vaccines. Viruses 7, 1726- 1803, 531 doi:10.3390/v7041726 (2015). 532 11 Stickl, H. & Hochstein- Mintzel, V. [Intracutaneous smallpox vaccination with a weak pathogenic 533 vaccinia virus ("MVA virus")]. Munch Med Wochenschr 113, 1149- 1153 (1971). 534 12 Jacobs, B. L. et al. Vaccinia virus vaccines: past, present and future. Antiviral Res 84, 1- 13, 535 doi:10.1016/j.antiviral.2009.06.006 (2009). 536 13 Greenberg, R. N. et al. Safety, immunogenicity, and surrogate markers of clinical efficacy for 537 modified vaccinia Ankara as a smallpox vaccine in HIV- infected subjects. The Journal of infectious 538 diseases 207, 749- 758, doi:10.1093/infdis/jis753 (2013). 539 14 Overton, E. T. et al. Immunogenicity and safety of three consecutive production lots of the non 540 replicating smallpox vaccine MVA: A randomised, double blind, placebo controlled phase III trial. 541 PLoS One 13, e0195897, doi:10.1371/journal.pone.0195897 (2018). 542 15 Volkmann, A. et al. The Brighton Collaboration standardized template for collection of key 543 information for risk/benefit assessment of a Modified Vaccinia Ankara (MVA) vaccine platform. 544 Vaccine 39, 3067- 3080, doi:10.1016/j.vaccine.2020.08.050 (2021). 545 16 Soh, T. K. et al. Herpes simplex virus- 1 pUL56 degrades GOPC to alter the plasma membrane 546 proteome. Cell Reports 33, 108235 (2020). 547 17 Nightingale, K. et al. High- definition analysis of host protein stability during Human 548 Cytomegalovirus infection reveals antiviral factors and viral evasion mechanisms. Cell Host 549 Microbe 12, 447- 460, doi:10.1016/j.chom.2018.07.011 (2018).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 110, 886, 864]]<|/det|>
+550 18 Li, J. et al. TMTpro reagents: a set of isobaric labeling mass tags enables simultaneous proteome- 551 wide measurements across 16 samples. Nat Methods 17, 399- 404, doi:10.1038/s41592- 020- 552 0781- 4 (2020). 553 19 Navarrete- Perea, J., Yu, Q., Gygi, S. P. & Paulo, J. A. Streamlined Tandem Mass Tag (SL- TMT) 554 Protocol: An Efficient Strategy for Quantitative (Phospho)proteome Profiling Using Tandem Mass 555 Tag- Synchronous Precursor Selection- MS3. J Proteome Res 17, 2226- 2236, 556 doi:10.1021/acs.jproteome.8b00217 (2018). 557 20 Huang da, W., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene 558 lists using DAVID bioinformatics resources. Nature protocols 4, 44- 57, 559 doi:10.1038/nprot.2008.211 (2009). 560 21 Sortino, O. et al. The Effects of Recombinant Human Lactoferrin on Immune Activation and the 561 Intestinal Microbiome Among Persons Living with Human Immunodeficiency Virus and Receiving 562 Antiretroviral Therapy. The Journal of infectious diseases 219, 1963- 1968, 563 doi:10.1093/infdis/jiz042 (2019). 564 22 Sivan, G. et al. Human genome- wide RNAi screen reveals a role for nuclear pore proteins in 565 poxvirus morphogenesis. Proc Natl Acad Sci U S A 110, 3519- 3524, doi:10.1073/pnas.1300708110 566 (2013). 567 23 Huttlin, E. L. et al. Dual proteome- scale networks reveal cell- specific remodeling of the human 568 interactome. Cell 184, 3022- 3040 e3028, doi:10.1016/j.cell.2021.04.011 (2021). 569 24 Kanth, S. M., Gairhe, S. & Torabi- Parizi, P. The Role of Semaphorins and Their Receptors in Innate 570 Immune Responses and Clinical Diseases of Acute Inflammation. Frontiers in immunology 12, 571 672441, doi:10.3389/fimmu.2021.672441 (2021). 572 25 Vallabhapurapu, S. & Karin, M. Regulation and function of NF- kappaB transcription factors in the 573 immune system. Annu Rev Immunol 27, 693- 733, doi:10.1146/annurev.immunol.021908.132641 574 (2009). 575 26 Arthur, J. S. & Ley, S. C. Mitogen- activated protein kinases in innate immunity. Nat Rev Immunol 576 13, 679- 692, doi:10.1038/nri3495 (2013). 577 27 Tang, J. et al. Human T- cell responses to vaccinia virus envelope proteins. J Virol 80, 10010- 10020, 578 doi:10.1128/JVI.00601- 06 (2006). 579 28 Kilcher, S. et al. siRNA screen of early poxvirus genes identifies the AAA+ ATPase D5 as the virus 580 genome- uncoating factor. Cell Host Microbe 15, 103- 112, doi:10.1016/j.chom.2013.12.008 581 (2014). 582 29 Lu, Y. et al. Histone deacetylase 4 promotes type I interferon signaling, restricts DNA viruses, and 583 is degraded via vaccinia virus protein C6. Proc Natl Acad Sci U S A 116, 11997- 12006, 584 doi:10.1073/pnas.1816399116 (2019). 585 30 Hajj Hussein, I. et al. Vaccines Through Centuries: Major Cornerstones of Global Health. Front 586 Public Health 3, 269, doi:10.3389/fpubh.2015.00269 (2015). 587 31 Sutter, G. & Moss, B. Nonreplicating vaccinia vector efficiently expresses recombinant genes. Proc 588 Natl Acad Sci U S A 89, 10847- 10851, doi:10.1073/pnas.89.22.10847 (1992). 589 32 Pulendran, B. & Ahmed, R. Translating innate immunity into immunological memory: implications 590 for vaccine development. Cell 124, 849- 863, doi:10.1016/j.cell.2006.02.019 (2006). 591 33 Richer, M. J., Nolz, J. C. & Harty, J. T. Pathogen- specific inflammatory milieux tune the antigen 592 sensitivity of CD8(+) T cells by enhancing T cell receptor signaling. Immunity 38, 140- 152, 593 doi:10.1016/j.immuni.2012.09.017 (2013).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[55, 110, 888, 864]]<|/det|>
+594 34 Liu, R., Olano, L. R., Mirzakhanyan, Y., Gershon, P. D. & Moss, B. Vaccinia Virus Ankyrin-Repeat/F- 595 Box Protein Targets Interferon-Induced IFITs for Proteasomal Degradation. Cell Rep 29, 816- 828 596 e816, doi:10.1016/j.celrep.2019.09.039 (2019). 597 35 Antoine, G., Scheiflinger, F., Dorner, F. & Falkner, F. G. The complete genomic sequence of the 598 modified vaccinia Ankara strain: comparison with other orthopoxviruses. Virology 244, 365- 396, 599 doi:10.1006/viro.1998.9123 (1998). 600 36 Strambio- De- Castillia, C., Niepel, M. & Rout, M. P. The nuclear pore complex: bridging nuclear 601 transport and gene regulation. Nat Rev Mol Cell Biol 11, 490- 501, doi:10.1038/nrm2928 (2010). 602 37 De Jesus- Gonzalez, L. A. et al. The Nuclear Pore Complex Is a Key Target of Viral Proteases to 603 Promote Viral Replication. Viruses 13, doi:10.3390/v13040706 (2021). 604 38 Turnbull, M. L. et al. Role of the B Allele of Influenza A Virus Segment 8 in Setting Mammalian Host 605 Range and Pathogenicity. J Virol 90, 9263- 9284, doi:10.1128/JVI.01205- 16 (2016). 606 39 Ersing, I. et al. A Temporal Proteomic Map of Epstein- Barr Virus Lytic Replication in B Cells. Cell 607 Rep 19, 1479- 1493, doi:10.1016/j.celrep.2017.04.062 (2017). 608 40 Bojkova, D. et al. Proteomics of SARS- CoV- 2- infected host cells reveals therapy targets. Nature 609 583, 469- 472, doi:10.1038/s41586- 020- 2332- 7 (2020). 610 41 Naamati, A. et al. Functional proteomic atlas of HIV infection in primary human CD4+ T cells. eLife 611 8, doi:10.7554/eLife.41431 (2019). 612 42 Ficarelli, M., Neil, S. J. D. & Swanson, C. M. Targeted Restriction of Viral Gene Expression and 613 Replication by the ZAP Antiviral System. Annu Rev Virol 8, 265- 283, doi:10.1146/annurev- virology- 0191919- 104213 (2021). 614 43 Peng, C. et al. Zinc- finger antiviral protein (ZAP) is a restriction factor for replication of modified 615 vaccinia virus Ankara (MVA) in human cells. PLoS Pathog 16, e1008845, 616 doi:10.1371/journal.ppat.1008845 (2020). 617 44 McSharry, B. P., Jones, C. J., Skinner, J. W., Kipling, D. & Wilkinson, G. W. Human telomerase 618 reverse transcriptase- immortalized MRC- 5 and HCA2 human fibroblasts are fully permissive for 619 human cytomegalovirus. J Gen Virol 82, 855- 863, doi:10.1099/0022- 1317- 82- 4- 855 (2001). 620 45 Blanchard, T. J., Alcami, A., Andrea, P. & Smith, G. L. Modified vaccinia virus Ankara undergoes 621 limited replication in human cells and lacks several immunomodulatory proteins: implications for 622 use as a human vaccine. J Gen Virol 79 ( Pt 5), 1159- 1167, doi:10.1099/0022- 1317- 79- 5- 1159 623 (1998). 624 46 Law, M. & Smith, G. L. Antibody neutralization of the extracellular enveloped form of vaccinia 625 virus. Virology 280, 132- 142, doi:10.1006/viro.2000.0750 (2001). 626 47 Cao, H. et al. Innate immune response of human plasmacytoid dendritic cells to poxvirus infection 627 is subverted by vaccinia E3 via its Z- DNA/RNA binding domain. PLoS One 7, e36823, 628 doi:10.1371/journal.pone.0036823 (2012). 629 48 Schweppe, D. K. et al. Full- Featured, Real- Time Database Searching Platform Enables Fast and 630 Accurate Multiplexed Quantitative Proteomics. J Proteome Res 19, 2026- 2034, 631 doi:10.1021/acs.jproteome.9b00860 (2020). 632 49 McAlister, G. C. et al. MultiNotch MS3 enables accurate, sensitive, and multiplexed detection of 633 differential expression across cancer cell line proteomes. Anal Chem 86, 7150- 7158, 634 doi:10.1021/ac502040v (2014). 635 50 Ting, L., Rad, R., Gygi, S. P. & Haas, W. MS3 eliminates ratio distortion in isobaric multiplexed 636 quantitative proteomics. Nat Methods 8, 937- 940, doi:10.1038/nmeth.1714 (2011).
+
+<--- Page Split --->
+<|ref|>text<|/ref|><|det|>[[57, 113, 886, 457]]<|/det|>
+638 51 Rad, R. et al. Improved Monoisotopic Mass Estimation for Deeper Proteome Coverage. J Proteome Res 20, 591- 598, doi:10.1021/acs.jproteome.0c00563 (2021).639 52 Haas, W. et al. Optimization and use of peptide mass measurement accuracy in shotgun proteomics. Mol Cell Proteomics 5, 1326- 1337, doi:10.1074/mcp.M500339- MCP200 (2006).640 53 Yates, J. R., 3rd. Pivotal role of computers and software in mass spectrometry - SEQUEST and 20 years of tandem MS database searching. J Am Soc Mass Spectrom 26, 1804- 1813, doi:10.1007/s13361- 015- 1220- 0 (2015).641 54 Huttlin, E. L. et al. A tissue- specific atlas of mouse protein phosphorylation and expression. Cell 143, 1174- 1189, doi:10.1016/j.cell.2010.12.001 (2010).642 55 Davis, J. J. et al. The PATRIC Bioinformatics Resource Center: expanding data and analysis capabilities. Nucleic acids research 48, D606- D612, doi:10.1093/nar/gkz943 (2020).643 56 Cox, J. & Mann, M. MaxQuant enables high peptide identification rates, individualized ppb- range mass accuracies and proteome- wide protein quantification. Nature Biotechnology 26, 1367- 1372, doi:10.1038/nbt.1511 (2008).644 57 Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B- Methodol. 57, 289- 300 (1995).645 58 Cowen, L., Ideker, T., Raphael, B. J. & Sharan, R. Network propagation: a universal amplifier of genetic associations. Nat Rev Genet 18, 551- 562, doi:10.1038/nrg.2017.38 (2017).646 59 Vizcaino, J. A. et al. 2016 update of the PRIDE database and its related tools. Nucleic acids research 647 44, D447- 456, doi:10.1093/nar/gkv1145 (2016).
+
+<--- Page Split --->
+<|ref|>sub_title<|/ref|><|det|>[[44, 42, 311, 70]]<|/det|>
+## Supplementary Files  
+
+<|ref|>text<|/ref|><|det|>[[44, 93, 765, 113]]<|/det|>
+This is a list of supplementary files associated with this preprint. Click to download.  
+
+<|ref|>text<|/ref|><|det|>[[60, 131, 312, 230]]<|/det|>
+Supplementarytable1. xlsx Supplementarytable2. xlsx Supplementarytable3. xlsx Supplementarytable4. xlsx
+
+<--- Page Split --->