key: cord-0301262-cz42bee9
authors: Polyiam, Kanokporn; Ruengjitchatchawalya, Marasri; Mekvichitsaeng, Phenjun; Kaeoket, Kampon; Hoonsuwan, Tawatchai; Joiphaeng, Pichai; Roshorm, Yaowaluck Maprang
title: Immunodominant and neutralizing linear B cell epitopes spanning the spike and membrane proteins of Porcine Epidemic Diarrhea Virus
date: 2021-10-07
journal: bioRxiv
DOI: 10.1101/2021.10.05.463270
sha: 9d6e96166e2db1c79faeee9bbfba31dd92b075ba
doc_id: 301262
cord_uid: cz42bee9

Porcine Epidemic Diarrhea Virus (PEDV) is the causative agent of PED, an enteric disease that causes high mortality rates in piglets. PEDV is an alphacoronavirus that has high genetic diversity. Insights into neutralizing B cell epitopes of all genetically diverse PEDV strains are of importance, particularly for designing a vaccine that can provide broad protection against PEDV. In this work, we aimed to explore the landscape of linear B cell epitopes on the spike (S) and membrane (M) proteins of global PEDV strains. All amino acid sequences of the PEDV S and M proteins were retrieved from the NCBI database and grouped. Immunoinformatics-based methods were next developed and used to identify putative linear B cell epitopes from 14 and 5 consensus sequences generated from distinct groups of the S and M proteins, respectively. ELISA testing predicted peptides with PEDV-positive sera revealed 9 novel immunodominant epitopes on the S protein. Importantly, 7 of these novel immunodominant epitopes and other subdominant epitopes were demonstrated to be neutralizing epitopes by neutralization-inhibition assay. Additionally, our study shows the first time that M protein is also the target of neutralizing antibodies as 7 neutralizing epitopes in the M protein were identified. Conservancy analysis revealed that epitopes in the S1 subunit are more variable than those in the S2 subunit and M protein. In this study, we offer the immunoinformatics approach for linear B cell epitope identification and a more complete profile of linear B cell epitopes across the PEDV S and M proteins, which may contribute to the development of a greater PEDV vaccine as well as peptide-based immunoassays.

This is a provisional file, not the final typeset article from different consensus sequences are located on the same regions. Impressively, our 147

immunoinformatics methods could identify all published epitopes in the S protein as indicated in 148 Table 1 and Supplementary Figure 5 and 6. 149 150 B cell epitopes on the S2 subunit and M protein are more conserved than those on the S1 151 subunit 152 We further analyzed the strain coverage of each predicted epitope, as information towards epitope 153 conservancy could be beneficial to the design and development of a vaccine, particularly a universal 154

vaccine. For the M protein, 6 out of 7 predicted epitopes cover higher than 70% of 560 accession 155 numbers with 4 epitopes (M-3, M-4, M-5 and M-6) having strain coverage greater than 90% (Fig.  156 5A). Analyzed with 924 accession numbers of the PEDV S protein, 34 out of 50 predicted epitopes 157 of the S protein exhibited strain coverage greater than 70% (Fig. 5B) . Among these 34 epitopes, 24 158

showed strain coverage higher than 80%, of which 8 and 16 are located on the S1 and S2 subunits, 159 respectively, suggesting that the S1 subunit is more variable than the S2 subunit. 160 161

Predicted B cell epitopes are recognized by PEDV-positive sera 162

To further validate the predicted linear B cell epitopes, ELISA was performed. PEDV-positive sera 163

were prepared from F1 and F3 pigs raised in the farms and naturally infected with PEDV. 164

Neutralizing activity of the pig sera were investigated using neutralization assay and only the serum 165 samples with neutralizing antibody titers at least 32 (reciprocal serum dilution) were further used in 166 ELISA analysis. A total of 12 F1 pigs and 11 F3 pigs showed neutralizing activity that passes the 167 criteria ( Fig. 6) and were subject to ELISA. Control sera were prepared from 10 uninfected F3 pigs 168 raised in the animal research facility and their neutralizing antibody titer was confirmed to be lower 169 than 32 by neutralization assay (Supplementary Table 1) . 170 Three well known epitopes, SS2, SS6 and 2C10 (21, 24) , were included in ELISA as 171 reference epitopes. Reactivity of the peptides with their target antibodies was determined based on 172 statistical analysis comparing OD450 value of the F1 and F3 pig sera to that of the control group (p < 173 0.05). All 7 peptides from the M protein showed reactivity with sera from at least one pig strain (Fig. 174 7A). Out of 50, 48 peptides of the S protein showed reactivity with the sera from at least one strain 175 pig strain (Fig. 7C) . Based on the analyses to determine immunodominant and subdominant epitopes 176 (described in materials and methods), the epitopes S1B-1, S1B-2.2, S1B-3, S1B-9.2, S1B-14.2, S1B-177

19.1, S1B-19.3, S2B-22, S2B-25, S2B-29, and S2B-30.2 on the S protein and the epitope M-2 on the 178 M protein were categorized as immunodominant epitopes, while the epitopes S1B-2, S1B-5, S1B-179 9.1, S1B-15, S1B-16, S1B-19.2, S2B-21, S2B-24, S2B-26.1 and S2B-32 were categorized as 180 subdominant epitopes ( Fig. 7A -D, and Table 2) . Two subdominant epitopes, S1B-15 and S1B- 16, 181 are located on the COE domain. 182

However, some of the B cell epitopes we identified here have already been reported. The 183 S1B-19 epitope (aa. 719-772) is a part of the S1D domain (aa. 636-789) containing neutralizing 184 epitopes (32). The S1B19.1 epitope (aa. 719-730) overlaps with the known epitope, S1D SE16 (33), 185

while the S1B-19.3 epitope (aa. 748-772) consists of 2 known epitopes, SS2 (21) and SS6 (21). The 186 S1B-8 epitope (aa. 203-211) corresponds to the known epitope, Peptide M (22), while the epitopes 187 S2B-33 (aa. 1345-1379) contains the 2C10 epitope (34). The epitope S1B-14.1 (aa. 457-477) entirely 188 overlaps with the neutralizing epitope (aa. 432-481) (35). In the COE region, all three epitopes (S1B-189 15 to S1B-17) are parts of the conformational epitope T-636) (35) as underlined. Additionally, the S1B-16 epitope also has 3 amino acids (GYP) 192 overlapping with the C2-1 epitope (TSLLASACTIDLFGYP) (36). Besides these known epitopes, 193 other B cell epitopes identified in our study are considered novel. Thus, epitopes S1B-1, S1B-2.2, 194 S1B-3, S1B-9.2, S1B-14.2, S2B-22, S2B-25, S2B-29 and S2B-30.2 can be recognized as novel 195 immunodominant epitopes. For the M protein, one B cell epitope named M14 (8), has been reported; 196 however, this epitope does not overlap with the epitopes identified in our study. Therefore, all 7 B 197 cell epitopes in the M protein we identified are novel epitopes. 198 199 Novel neutralizing epitopes are verified by neutralization-inhibition assay 200 Neutralization ability of the antibodies against each epitope were further tested using neutralization-201 inhibition assay as shown in schematic representation in Fig. 8A . Four sera including F1 1-3, F1 1-4, 202 F1 1-5 and F3 1-10 were used in the assay. Peptides corresponding to the known epitopes SS2, SS6 203 and 2C10 were used as references, and irrelevant peptides N(MHC) and E(CTL) were included as 204 negative controls. All 3 reference peptides exhibited their ability to inhibit neutralizing activity of the 205 sera, while irrelevant peptides did not (Fig. 8B) . Notably, peptides from the M protein were tested 206 with only 2 sera, F1 1-4, F1 1-5. While the F1 1-4 serum lost its neutralizing activity when incubated 207

with peptides M-1, M-2, M-3, M-4 and M-6, neutralizing activity of serum F1 1-5 was depleted in 208 the presence of peptides M-1, M-2, M-,5 M-6 and M-7 ( Fig. 8B and Supplementary Fig. 8-9) . A 209 combined result from both sera suggested that all 7 epitopes of the M protein had potential as 210 neutralizing epitopes. 211

In the presence of the S peptides, four pig sera showed different patterns of neutralization 212 inhibition as summarized in Table 2 and Supplementary Fig. 7-10 . In Fig. 8B , neutralizing epitopes 213

were concluded based on the combined results of all 4 sera. The peptide that inhibited neutralizing 214 activity of at least one serum was considered a neutralizing epitope. Almost all peptides, except S1B-215 7, S1B-11, S1B-14.2, S1B-19.1, S1B-19.2 and S2B-22, could inhibit neutralizing activity of at least 216 one serum. Hence, the epitopes S1B-1, S1B-2.2, S1B-3, S1B-9.2, S1B-19.3, S2B-22, S2B-25, S2B-217 29 and S2B-30.2 posed as both neutralizing epitopes as well as immunodominant epitopes. 218

Importantly, antibodies recognizing the 3 epitopes in the COE domain, which are S1B-15, S1B-16 219 and S1B-17, were demonstrated with neutralization ability. 220 221

Depiction of the identified B cell epitopes in the 3-D structure of the PEDV S protein 222

Surface representation of all epitopes is depicted on the prefusion structure of the trimeric S protein 223 of the PEDV strain USA/Colorado/2013 (PDB: 6VV5) (37) using PyMOL 2.3.4. As shown in Fig.  224 9A, most of the epitopes are exposed on the surface, one main feature associated with B cell epitopes. 225

Although coil is generally recognized as one main characteristic of linear B cell epitopes, B cell 226 epitopes identified in our study were found to be in various structures including coil, alpha helix and 227 beta sheet (Fig. 9B) . Close-ups of the immunodominant epitopes and neutralizing epitopes in the 228 COE revealed that these epitopes are either partly or entirely composed of coil structure and some 229 epitopes also consist of other structures either alpha helix or beta sheet ( Fig. 9C-D next to the signal sequence, (ii) CTD, upstream region of the S1/S2 cleavage site, (iii) FP region, and 256 (iv) upstream region of the HR2, respectively, which have also been indicated to be the main sites of 257 immunodominant epitopes on the SARS-CoV-2 S protein studied by our group (manuscript 258 submitted). and Li et.al. (2021) (40) . Moreover, the epitopes S1B-15, S1B-16 and S1B-17, which are 259 located within the COE domain, also mirror the 3 most dominant epitopes within the S RBD of 260 SARS-CoV-2 identified in our previous work (manuscript submitted). Additionally, the C-terminal 261 endodomain of the S protein is another main site accommodating the immunodominant B cell 262 epitope of SARS-CoV-2 (40-42) as well as the neutralizing epitope 2C10 with the GPRLQPY motif 263

(epitope S2B-33 in this study) of PEDV (34). 264

Neutralizing activity of the antibodies targeting these novel epitopes was addressed using 265 neutralization-inhibition assay, which has been previously used to identify neutralizing epitopes from 266 SARS-CoV-2 (41) and coxsackievirus A16, a causative agent of human, foot, hand and mouth 267 disease (43). However, we observed that inhibitory effect of peptide on different sera were different 268

for some peptides. This may be because the levels of antibodies against each epitope vary in different 269 sera. Surprisingly, we did not see inhibitory effect of 3 immunodominant epitopes, S1B-14.2, S1B-270

19.1, and S2B-22, on any sera. This result can be explained by 2 possibilities: (i) these epitopes are 271

indeed non-neutralizing epitopes recognized by non-neutralizing antibodies, and (ii) the peptides 272 cannot compete with the S protein present on the virus particle in binding to their cognate antibodies; 273 thus, neutralizing activity of the serum is maintained. 274

Neutralizing epitopes we identified are located at the regions that are functionally important 275 during virus infection. Epitopes S1B-1 to S1B-9 are located at the N-terminal domain (NTD) of the 276 S1 subunit, the region that involves in binding to sialoglycoconjugates receptor on the host cell (44, 277 45). The neutralizing epitope S1B-10 are located at the S1 region that plays an important role in 278 PEDV attachment to host cell using sugar-binding activity (46). Moreover, the neutralizing epitope 279 S1B 14.1 overlaps with the C-terminal sequence of the conformational epitope (aa. 435-485) 280

previously identified in the PEDV PT strain (35), suggesting that the overlapping sequence (457-281 RILYCDDPVSQLKCSQVAFDL-477) may function as a core neutralizing epitope in that region. 282

Within COE domain, 3 neutralizing epitopes we identified, S1B-15, S1B-16 and S1B-17, may be the 283 main targets for neutralizing antibody recognition in the conformational and neutralizing epitope (aa. 284 572-636) previously identified (35). The S1D domain located in the S1 CTD, which also includes our 285 neutralizing epitope S1B-19, is well documented as an immunodominant epitope/domain and a target 286 of neutralizing antibodies (10, 21, 47). 287

While most of the known neutralizing B cell epitopes are located in the S1 subunit, 288

information of B cell epitopes in the S2 subunit are limited with only one neutralizing epitope 289

identified. The 2C10 epitope located at the C-terminus of the S2 domain has been identified as 290 neutralizing epitope (23, 24) . In this study, in addition to the 2C10 epitope, more neutralizing B cell 291 epitopes in the S2 subunit were identified. The neutralizing epitope S2B-20 is located next to the 292 S1/S2 cleavage site; thus, antibody binding to this region may interfere S1/S2 cleavage, resulting in a 293

loss of viral infectivity. Coronavirus S2 subunit consists of Fusion peptide (FP) and Heptad repeat 294 regions (HR) that play important roles in cell membrane fusion (48). In a betacoronavirus, SARS-295

CoV-2, the epitopes located in the FP and HR2 regions have been identified as immunodominant and 296 neutralizing epitope (49). In our study, the neutralizing epitopes S2B-21 and S2B-23 are located in 297 close proximity to the FP, a region responsible for protease cleavage and membrane fusion (45); thus, 298 antibody binding to this region could affect these processes during viral cell entry. Neutralizing 299 epitopes S2B-24, 25 and 26 are located within the HR1 region, whereas neutralizing epitopes S2B-300 27, 28, 29, 30 and 31 are located in the upstream region of the HR2 and S2B-32 is located within the 301 HR2. The HR1 and HR2 domains of coronaviruses play important roles in membrane fusion (45); 302 thus, blocking these regions with antibodies may result in inhibition of the membrane fusion process. 303

As our B cell epitopes are predicted based on consensus sequence of each group of the S and 304 M proteins, the epitope sequence may not 100% match with all PEDV strains/variants. However, 305 some epitopes are highly conserved as evidenced by high percentage of strain coverage among global 306 PEDV strains. Compared to the epitopes in the S protein, epitopes in the M protein were found to be 307 more conserved. Within the S protein, epitopes in the S2 subunit are more conserved than those in 308 the S1 subunit. These conserved epitopes may serve as candidates for development of a multi-epitope 309 universal vaccine. 310

Taken together, the method combining immunoinformatics with immunoassays enabled 311 identification of novel neutralizing linear B cell epitopes on the PEDV S and M proteins. Even 312 though these B cell epitopes are derived from consensus sequence, some of which are highly 313 conserved among the global PEDV strains, which represent a promising vaccine target for 314 development of a universal epitope-based vaccine as well as for antibody detection. Importantly, the 315 immunoinformatics method developed in this study can serve as a useful tool for prediction of linear 316 B cell epitopes from any protein of interest. 317 318

Protein sequence retrieval and phylogenetic analysis 320

Amino acid sequences of the PEDV S and M proteins were retrieved from National Center for 321

Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/protein/). Retrieved protein 322 sequences were analyzed individually based on a defined name and length using Sublime Text3 323

program. Based on sequence similarity, the amino acid sequences of the S and M protein were 324 grouped using alignment and phylogenetic tree tools in SeaView program (version 4). In SeaView 325 program, these amino acid sequences were aligned using MUSCLE and grouped using tree based on 326 distance analysis, NJ method, ignore gap and bootstrap 1000 replication. Consensus sequence of each 327 group was generated using MUSCLE method in Unipro UGENE program (version 1.24.2; September 328 1, 2016) (50, 51). These consensus sequences were then subjected to epitope prediction. 329 330 B cell epitope prediction 331

Immunoinformatics tools were exploited to predict B cell epitopes and properties of the amino acid 332 residues. Firstly, linear B cell epitope prediction was performed using BepiPred-2.0 333 (http://www.cbs.dtu.dk/services/BepiPred/), which predicts linear B cell epitopes from a protein 334 sequence, using a Random Forest algorithm trained on epitopes and non-epitope amino acids 335 determined from crystal structures, followed by performing sequential prediction smoothing (52). In 336 our study, the residues with the threshold score of epitope probability above 0.5 were considered as 337

parts of a B cell epitope. Putative B cell epitopes were selected based on the region with at least 6 338 consecutive residues predicted to have epitope probability above 0.5. Notably, BepiPred-2.0 also 339 predicts and provides accessibility and coil probability of each amino acid residue. In addition, other 340 This is a provisional file, not the final typeset article properties of the proteins were also characterized using multiple immunoinformatics tools provided 341 by IEDB (http://tools.iedb.org/bcell/), except IUPred. IUPred (https://iupred.elte.hu/) was used to 342 predict intrinsically unstructured proteins which infers to coil probability (53). The method of Emini 343

(54) was used to predict surface accessibility, while the method of Kolaskar & Tongaonkar, semi-344 empirical method (55), was used to predict antigenicity determinant. Hydrophilicity of the protein 345 was predicted using the method of Parker. The results of prediction from each method were aligned 346 and 3 methods for identifying and selecting candidate B cell epitopes were created based on the 4 347 known B cell epitopes derived from PEDV protein as described in the result part. 348 349

Epitope conservancy analysis 350

As epitopes were predicted based on consensus sequences of each group of the PEDV S (14 groups) 351

and M (5 groups) proteins. Epitope conservancy was analyzed by calculating percent strain coverage 352 of the predicted epitopes in comparison to all of the sequences in the group using the formula below. 353

Peptide synthesis and preparation 355

Peptides corresponding to immunoassay were chemically synthesized (Mimotopes). All synthetic 356

peptides were (i) 57 predicted epitopes, (ii) 3 known reference B cell epitopes, SS2 (YSNIGVCK) 357

(21), SS6 (LQDGQVKI) (21), and 2C10 (GPRLQPY) (24) were collected from 13 female F1 pigs and 14 either male or female pigs raised in 3 different farms. 376

Control sera were prepared from 10 uninfected F3 pigs (age 7 weeks old) raised in Mahidol 377

University's animal research facility. Bloods were left to clot at room temperature for approximately 378 2 h, followed by centrifugation at 1,500 x g for 10 min in a refrigerated centrifuge to separate the 379 serum. Serum was aliquoted and separated to a small portion in new tubes and stored at -20 °C until 380 used. 381 382

Cell and PEDV preparation 383 % strain coverage = Number of accession sequences with100% match to the predicted epitope × 100 Total accession number in the group PEDV used in this study is a lineage of PEDV CV777. Vero cells were cultured in complete DMEM 384 medium supplemented with 10% fetal bovine serum (FBS, HyClone) and 1% penicillin/streptomycin 385 (Pen/Strep, gibco), and then incubated at 37 °C with 5% CO 2 . The seed virus was sequentially 386

propagated in Vero cell with PEDV medium (DMEM supplemented with 0.02% yeast extract 387 (HIMEDIA), 10 μg/ml of trypsin (SIGMA), and 0.3% tryptose phosphate broth (tryptose, 20 g/L; 388 dextrose, 2 g/L; sodium chloride, 5 g/L; disodium hydrogen phosphate, 2.5 g/L) (35). To prepare 389 PEDV stock, Vero cells were cultured in 10 flasks of T175. When the cell confluency was over 90%, 390 old medium was removed and the cells were infected with PEDV resuspended in 20 ml of PEDV 391 medium. When cytopathic effect (CPE) reached 70%, medium was harvested and pooled. Sodium 392 chloride (NaCl) was added to the medium at a final concentration of 0.5 M and polyethylene glycol 393

(PEG) was then added to a final concentration of 8%. After incubated overnight, PEDV was 394 harvested by centrifugation at 8,000 rpm, 20 min. The viral pellet was next resuspended in 5 ml of 395 endotoxin-free PBS, aliquoted in a small volume to new tubes and stored at -80 °C. Next, virus titer 396 was determined using 50% tissue culture infective dose (TCID 50 ) method. Virus suspension was 397 diluted in ten-fold serial dilutions and pipetted into ten wells of confluent Vero cells. Neutralization-inhibition assay 438

Neutralization-inhibition assay was conducted as previously described by Shi et al. (2013) (43) with 439 some modifications. Sera from 3 F1 pigs (F1 1-3, F-1 1-4 and F1 1-5) and one F3 pig (F3 1-10) were 440

subject to the assay. Serum was 2-fold serially diluted with DMEM containing 1% antibiotic to the 441 dilution ranging from 1:32 to 1:256. PEDV was diluted in PEDV medium to 2,000 TCID 50 /ml. 442

Synthetic peptides were mixed with the diluted serum to a final concentration of 200 nmol/ml in a 443 volume of 50 μl. After an incubation at 37 °C for 1 h, PEDV (100 TCID 50 in 50 μl) was added to the 444 peptide-serum mixture. After an incubation at 37 °C for 1 h, the mixture (100 μl) was transferred to 445

Vero cells grown in 96-well plates. Following an incubation at 37 °C with 5% CO 2. for 3 h, the virus 446 suspension was removed from the well and new medium was added into the wells. In addition, the 447 conditions that the Vero cells were incubated with (i) medium alone, (ii) serum alone, (iii) PEDV 448 alone, and (iv) the mixture of PEDV and serum, were also included in the study as control conditions. 449

All conditions were performed in duplicate. Following a 3-day (F-1 1-4 and F1 1-5) or 5-day (F1 1-3 450

and F3 1-10) incubation, PEDV infection in Vero cell was investigated using immunofluorescence 451 staining assay. Neutralization inhibition mediated by a peptide was determined at the endpoint 452 neutralizing antibody titer. 453 454

Labelling of neutralizing B cell epitopes 455

Putative neutralizing B cell epitopes were aligned and compared with full-length S protein of PEDV 456 USA/Colorado/2013 strain for identification of epitope position. To localize each epitope within 457 trimeric PEDV S protein, prefusion structure of PEDV spike named 6VV5 (37) was used as a model. 458

All predicted epitopes were depicted PyMOL 2.3.4 program. 459 460

Statistical analysis 461

The statistical significance of the samples in different groups was analyzed using SPSS 22 for 462

Windows software (SPSS, USA). Data analyses were performed using non-parametric Mann-463

Whitney U test. The difference between the 2 groups was determined based on p < 0.05. 464

Immunodominant and subdominant epitopes were defined with following criteria. This is a provisional file, not the final typeset article This is a provisional file, not the final typeset article M-1 9/9 (n=9) 6/11 nt. nt. M-2 9/9 (n=9) 11/11 Immunodominant nt. nt. M-3 9/9 (n=9) 8/11 nt. ⎯ nt. M-4 2/9 (n=9) 7/11 nt. ⎯ nt. M-5 9/9 (n=9) 4/11 nt. ⎯ nt. M-6 9/9 (n=9) 9/11 subdominant nt. nt. M-7 5/9 (n=9) 3/11 nt. ⎯ nt. S1B-1 12/12 11/11 Immunodominant nt. nt. S1B-2 8/12 6/11 subdominant ⎯ ⎯ S1B-2.1 9/12 11/11 ⎯ ⎯ ⎯ S1B-2.2 12/12 11/11 Immunodominant S1B-3 12/12 11/11 Immunodominant ⎯ ⎯ S1B-4 12/12 1/11 ⎯ S1B-5 12/12 2/11 subdominant nt. nt. S1B-6 4/12 11/11 ⎯ ⎯ ⎯ S1B-7 7/12 11/11 ⎯ ⎯ ⎯ ⎯ S1B-8 5/12 11/11 ⎯ S1B-9 7/12 4/11 ⎯ ⎯ S1B-9.1 11/12 11/11 subdominant ⎯ ⎯ S1B-9.2 12/12 11/11 Immunodominant ⎯ S1B-10 5/12 11/11 ⎯ ⎯ S1B-11 10/12 11/11 ⎯ ⎯ ⎯ S1B-12 6/12 11/11 ⎯ S1B-13 4/12 11/11 ⎯ S1B-14 0/12 0/11 ⎯ ⎯ ⎯ S1B-14.1 1/12 2/11 nt. ⎯ nt. S1B-14.2 12/12 11/11 Immunodominant nt. ⎯ ⎯ nt. S1B-15 5/12 11/11 subdominant ⎯ ⎯ ⎯ S1B-16 12/12 9/11 subdominant ⎯ S1B-17 1/12 1/11 ⎯ S1B-18 4/12 2/11 ⎯ ⎯ ⎯ ⎯ S1B-19 1/12 3/11 ⎯ ⎯ ⎯ S1B-19. 

A and C) Original phylogenetic tree of 666 M and S proteins generated by SeaView program. (B and D) Groups of the M and S proteins 667 generated by FigTree tool

Figure 2. Immunoinformatics prediction and methods for B cell epitopes identification

Immunoinformatics prediction and mapping of the known epitopes to the peptides predicted by 706 different prediction tools. Predicted peptides obtained from each tool were mapped to 4 published 707 epitopes. (B) Methods for identification of B cell epitopes. Three methods (A, B and C) were 708 generated based on mapping results of the 4

Predicted B cell epitopes of the M protein. B cell epitopes were predicted from consensus 740 sequences of 5 groups of the M protein. Epitopes resulting from prediction of each group are shown 741 and epitopes selected for further experimental evaluation are indicated in the

Predicted B cell epitopes of the PEDV S protein. B cell epitope prediction was carried out 819 using consensus sequences derived from 14 groups of the S protein

Pig sera were 2-fold serially diluted and incubated with 892 PEDV (10 TCID 50 ) before adding to Vero cells. PEDV infection was investigated by 893 immunofluorescence staining using anti-PEDV monoclonal antibody at 3 days post-inoculation

Surface representation and locations of the B cell epitopes on the S protein

Left panel demonstrates 1107 immunodominant epitopes; right panel demonstartes subdominant epitopes. (B) Localization and 1108 structure of the epitopes on the monomeric S protein. (C) Close-ups of immunodominant epitopes. 1109 (D) Close-ups of neutralizing epitopes in the COE domain

* Positions of amino acid residues are based on the M and S amino acid sequences of PEDV

proteins, respectively. ELISA was performed with PEDV-positive F1 (12 samples) and F3 (11 971 samples) pig sera whose neutralizing activity against PEDV had been confirmed. Sera from 972 uninfected F3 pigs were used as a control. ELISA plates were coated with synthetic peptides 973 corresponding to predicted epitopes as indicated in the X axis. Serum was diluted 1:50 and used in 974ELISA. The optical density at the wavelength of 450 nm was measured. The responses of serum 975 samples with neutralizing antibody titer equal or higher than 64 are shown in yellow circle. Non-976parametric Mann-Whitney U test was used in statistical analysis. * (red asterisk) indicates p < 0.05; * 977 (black asterisk) indicates p < 0.001. 984  985  986  987  988  989  990  991  992  993  994  995  996  997  998  999  1000  1001  1002  1003  1004  1005  1006  1007  1008  1009  1010  1011  1012  1013  1014  1015  1016  This is a provisional file, not the final typeset article   1017  1018  1019  1020  1021  1022  1023  1024  1025  1026  1027  1028  1029  1030  1031  1032  1033  1034  1035  1036  1037  1038  1039  1040  1041  1042  1043  1044  1045  1046  1047  1048  1049  1050  1051  1052  1053  1054