Diversity And Evolution Of Computationally Predicted T Cell Epitopes Against Human Respiratory Syncytial Virus

Abstract 

Human respiratory syncytial virus (RSV) is a major cause of lower respiratory infection. Despite more than 60 years of research, there is no licensed vaccine. While B cell response is a major focus for vaccine design, the T cell epitope profile of RSV is also important for vaccine development. Here, we computationally predicted putative T cell epitopes in the Fusion protein (F) and Glycoprotein (G) of RSV wild circulating strains by predicting Major Histocompatibility Complex (MHC) class I and class II binding affinity. We limited our inferences to conserved epitopes in both F and G proteins that have been experimentally validated. We applied multidimensional scaling (MDS) to construct T cell epitope landscapes to investigate the diversity and evolution of T cell profiles across different RSV strains. We find the RSV strains are clustered into three RSV-A groups and two RSV-B groups on this T epitope landscape. These clusters represent divergent RSV strains with potentially different immunogenic profiles. In addition, our results show a greater proportion of F protein T cell epitope content conservation among recent epidemic strains, whereas the G protein T cell epitope content was decreased. Importantly, our results suggest that RSV-A and RSV-B have different patterns of epitope drift and replacement and that RSV-B vaccines may need more frequent updates. Our study provides a novel framework to study RSV T cell epitope evolution. Understanding the patterns of T cell epitope conservation and change may be valuable for vaccine design and assessment.

Benefits of cistanche tubulosa-strengthen immune syste

Author Summary 

Lower respiratory infections caused by human respiratory syncytial virus (RSV) are a global health challenge. B cell epitope immune response has been the major focus of RSV vaccine and therapeutic development. However, T cell epitope-induced immunity plays an important role in the resolution of RSV infection. While RSV genetic diversity has been widely reported, few studies focus on RSV T cell epitope diversity, which can influence vaccine effectiveness. Here, we use computationally predicted T cell epitope profiles of circulating strains to characterize the diversity and evolution of the T cell epitope of RSV A and B. We systematically evaluate the T cell epitope profile of RSV F and G proteins. We provide a T cell epitope landscape visualization that shows the co-circulation of three RSV-A groups and two RSV-B groups, suggesting potentially distinct T cell immunity. Furthermore, our study shows different levels of F and G protein T cell epitope content conservation, which may be important to correlate with the duration of vaccine protection. This study provides a novel framework to study RSV T cell epitope evolution, infer RSV T cell immunity at population levels, and monitor RSV vaccine effectiveness. 

Introduction

Human respiratory syncytial virus (RSV) is a negative-strand RNA virus that is classified in the Orthopneumovirus genus of the family Pneumoviridae. It is a major cause of lower respiratory disease in young infants, immunocompromised individuals, and elderly people, resulting in annual epidemics worldwide [1]. The single-stranded RNA genome of RSV is approximately 15.2 kb and encodes 11 viral proteins [2]. The Fusion (F) and Glycoprotein (G) proteins are the two major surface proteins [3]. F protein is generally thought to be conserved and therefore it is the focus of most current RSV vaccine designs. Although the G protein is highly variable, its contribution to disease pathogenesis and its role in the biology of infection suggest it can also be an effective RSV vaccine antigen [4]. Despite the significant burden of RSV infection worldwide, there is no licensed vaccine. The only approved intervention is passive immuno-prophylaxis with palivizumab, which is achieved by administering the monoclonal antibody (mAb) to a highly restricted group of infants under the age of 24 months and treatment must be repeated monthly during the RSV season due to the relatively short half-life of the antibody [5,6]. Due to the high cost of monoclonal antibody treatments, this intervention is limited to high-risk infants and is generally unavailable in developing countries. An RSV vaccine is an urgent global healthcare priority, and it is likely that different strategies are needed for the various high-risk groups.

cistanche benefits-strengthen immune system

A number of research teams have worked on the development of the RSV vaccine since its isolation and characterization in 1956 [7,8]. However, vaccination with the formalin-inactivated, alum-precipitated RSV (FI-RSV) vaccine in RSV-naïve infants and young children, led to the development of vaccine-enhanced disease (VED) that hampered vaccine development for decades to follow [9]. Many studies have been conducted to explain this undesirable outcome. It is likely that formalin fixation led to a vaccine that mostly presented the post-fusion conformation of RSV F protein, leading to an excess of non-neutralizing antibodies and immune complex formation [10–12]. Other studies indicated that an impaired T cell response with Th2 skewing [13,14], as well as complement deposition in the lungs, contributed to enhanced neutrophil recruitment [12]. Recent developments, including the resolution of the F protein [15] and the development of RSV rodent models [16] have contributed to a number of vaccine candidates with novel designs and formulations currently in clinical trials [3,17,18]. While most current RSV vaccination strategies focus on a B-cell-induced neutralization immune response, T-cell immunity also plays a major role in the resolution of virus infection and is essential for RSV vaccine development [17,18]. Once RSV infection of the lower airways is established, CD8 T cells play an important part in viral clearance and CD4 helper T cells can orchestrate cellular immune responses and stimulate B cells to produce antibodies. However, Th2-biased responses have been associated with animal models of RSV VED, and measurement of Th1 and Th2 responses are considered important to predict the safety of vaccine candidates [12]. Therefore, induction of a balanced cell-mediated immune response through vaccination would promote RSV clearance, but caution must be taken to avoid the potential for immunopathology. Taken together, a closer examination of T cell immunity and the virus sequences that induce T cell responses are needed for RSV vaccine development.

The human respiratory syncytial virus has a complex circulation pattern in the human population. Within two antigenic groups, RSV-A and RSV-B, different genotypes can co-circulate within the same community, while novel RSV genotypes with high genomic diversity may arise and potentially replace the previously dominant genotypes [19]. In recent years, several unique genetic modifications in RSV have been identified, including a 72-nucleotide (nt) duplication (ON genotype) in RSV-A G gene and another with a 60-nt duplication (BA genotype) in RSV-B at a similar region [20]. The observed RSV genetic diversity has raised a question about whether it is necessary for an RSV vaccine to include several different strains to be effective. Most current RSV vaccine developments are based on an RSV A2 laboratory strain, which is a chimeric strain that belongs to subtype A [21]. While these treatments hold promise, there is the possibility of viral strains developing escape mutations. For example, palivizumab-resistant strains have been isolated from both RSV rodent models and humans [17,22]. Several lines of evidence also suggest antigenic variation may play a role in the ability of RSV to escape immune response and established infections [23]. While highly conserved T cell epitopes in the RSV vaccine may not provide complete protection against infection when cross-protective antibody responses are lacking, highly conserved T cell epitopes in the vaccine may still reduce the severity of the illness and limit the spread of the virus. However, amino-acid variation at the T cell epitope level and the potential emergence of novel T cell epitopes of recent RSV circulating strains have been reported [24], and further studies are needed to illustrate the effects of amino acid variations on T cell recognition. Hence, characterizing T cell epitope profiles across different strains is very important to understand RSV evolution and can be important for RSV vaccine development.

cistanche plant-increasing immune system

Click here to view  Cistanche Enhance Immunity products

【Ask for more】 Email:cindy.xue@wecistanche.com /  Whats App:  0086 18599088692 /  Wechat:  18599088692

In this study, we utilize immunoinformatic approaches that are implemented in the iVAX toolkit [25] to predict T cell epitopes in RSV across different strains with a focus on the two major surface proteins F and G. With the analysis of a comprehensive dataset, we evaluate the lineage-specific T cell epitope profile of RSV. We also create sequence-based T cell epitope landscapes based on epitope content comparison across different strains and further correlate RSV T cell immunity change with virus evolution. The proportion of cross-conserved T cell epitope content between vaccine candidate strains that developed earlier and RSV circulating strains with different isolation years and locations was also calculated. These analyses may aid in understanding RSV T cell immunity across different strains and contribute to current vaccine design efforts. 

Results 

Distribution of T cell epitopes in RSV surface proteins

We evaluated the T cell immunogenic potential across RSV surface proteins by scanning 9 residue regions to predict the binding probability to MHC class I and class II molecules (Fig 1). The epitope density of RSV surface proteins was evaluated using a normalized epitope density score, which is computed by summing up the predicted peptide-MHC binding score across the protein and normalizing it with the protein length. The score for randomly generated proteins is set to zero and vaccine antigens generally score above +20 on this scale [25]. F protein has an epitope density score greater than +20 for both the class I and class II immunogenicity scale analysis, indicating significant immunogenic potential [25]. This contrasts with lower G protein class I and class II epitope density protein scores for both RSV subtypes. The class I epitope density score of G protein was greater than +10 in both subtypes but the class II density was lower than random expectation in the analysis of RSV-B (Fig 1A). This result suggests that RSV surface proteins are likely to have the potential to stimulate T cells that are required for protective immunity. We then investigated the distribution of T cell immunogenicity across the proteins and found that there are regions with relatively high T cell immunogenic potential (Fig 1B). The distribution of T cell immunogenicity of F protein was mapped onto its protein structure and overlap between protein sequence regions with high T cell immunity potential and the antibody neutralizing targets was observed at antigenic site F and site II.

Fig 1. T cell immunogenic potential for RSV surface proteins based on MHC binding prediction

Lineage-specific T cell epitope profiles 

We then extended T cell epitope predictions from RSV representative strains to multiple wild circulating strains. The distribution and diversity of T cell epitopes across different strains are illustrated in heatmaps with the corresponding time-scaled phylogenies (S1 and S2 Figs). Both F and G proteins contain epitopes that were conserved across all RSV strains in almost 100% of sampled isolates, suggesting that they could serve as high-quality T cell epitope candidates for vaccine design. In contrast, some epitopes were mutated in selected strains, and those epitopes that only occurred in certain clades within the phylogeny could be interpreted as clade-specific "fingerprints".

The G gene duplication events in RSV, which are unique gene signatures, can either shift the position of epitopes (locations are different but the amino acids of epitopes are identical to the G protein isolates without duplication) or cause the emergence of novel epitopes. Two novel class I epitopes, (no. 31 and no. 40 in S2A Fig), were found in RSV-A strains that contain G gene duplication. In addition, an emergent class II epitope (no. 25 in S2A Fig) was identified in RSV-A sequences that contain G gene duplication, which was a shift from an epitope (no. 24) that has been observed in other strains. From RSV-B strains that contain the G gene duplication event, we also observed multiple lineage-specific class I T cell epitopes, which are caused by a 2-amino acid (aa) deletion (aa157 and aa158) in these strains instead of directly due to the 60-nt duplication event. RSV-B G proteins that have the duplication event contain multiple novel epitopes (no. 22, 23, 26, 28, 30, 37) but do not contain several epitopes (no. 24, 25, 27, 29, 31, 38) that are identified in other strains (S2B Fig). To further determine whether the computationally predicted T cell epitopes with high MHC binding potential are immunogenic, we utilized the JanusMatrix [27] algorithm to identify the T cell epitopes that are likely to be cross-conserved with human peptides and thereby tolerated by the immune system. Based on this analysis, 6.45% of putative class I epitopes and 1.12% of putative class II epitopes of RSV major surface proteins are cross-conserved with human proteome-derived epitopes at T cell receptor (TCR)-facing residues. As these peptides have similar HLA binding preferences that are contained in human proteins (S3 Fig), they were therefore assumed not to be immunogenic. After excluding the high-JanusMatrix score epitopes identified above, we were able to identify T cell epitopes that were conserved in more than 60% of currently circulating RSV strains. We searched the IEDB epitope database to determine if these epitopes were related to experimentally validated RSV T cell epitopes or HLA ligands (Table A in S1 Text). The conserved RSV T cell epitope sequences that may be important for future vaccine development are shown in Tables 1 and 2 (Tables B and C in S1 Text).

Table 1. Experimentally validated conserved MHC class I epitopes peptides in RSV major surface proteins

Predicted RSV T cell epitope landscapes 

To investigate the evolution of RSV on T cell immunity profiles, we use a multidimensional scaling (MDS) approach to visualize the T cell immunity profile of multiple RSV strains on a landscape. We performed a T cell epitope content pairwise comparison between RSV strains using in silico predicted peptide-HLA allele binding affinity. The pairwise T-cell epitope distances were then calculated using the algorithm reported in this study (Eq 1). We then applied a multidimensional scaling (MDS) approach using these estimated pair-wise T epitope distances to map RSV strains to a landscape to characterize their T-cell immunity profile. We found both class I and class II T cell immunity profiles of F and G proteins of different RSV strains were clustered into groups on this T cell epitope landscape (S4 Fig). Combining the class I and class II T-cell epitope binding profiles, RSV-A major surface protein isolates can be divided into three clusters and RSV-B major surface protein isolates can be divided into two clusters (Figs 2 and S5). We observe that the G gene sequence isolates that contain 72-nt (RSV-A) or 60-nt (RSV-B) duplications clustered together with other sequences instead of forming isolated groups. To further investigate the T cell epitope diversity, we correlated this clustering pattern with the phylogenetic histories (Fig 2B). The phylogenetic tree topologies of the RSV-A F gene and the G gene are similar. The F gene cluster 1 is paraphyletic, while clusters 2 and 3 are monophyletic. Cluster 1 is the closest to the ancestral sequence and mapping this group onto the phylogeny shows that this cluster has a basal relationship with clusters 2 and 3 indicating that the phylogenetic divergence occurred prior to epitope drift. The RSV-B F and G gene genealogies are very different. In particular, the RSV-B F gene topologies are indicative of strong immune selection, similar to observed human influenza A virus or within host HIV phylogenies [28]. In contrast, the RSV-B G gene phylogeny shows the co-circulation of multiple lineages, though this could reflect the sequencing bias of G genes (Fig 2B). We then calculated the T-cell epitope immune distance of each strain from a reconstructed ancestral sequence (Fig 2C). These distances were then plotted against the year of isolation and colored according to the cluster identified in Fig 2A. RSV-A shows that multiple predicted immune phenotypes co-circulate and persist for long periods (>2 decades). Analysis of RSV-B shows a turnover of the predicted immune phenotypes with short periods of co-circulation (<5 years) for F and G protein T cell epitopes. The limited periods of co-circulation are again consistent with phenotype patterns observed for viruses under strong immune selection (e.g. H3N2 influenza A virus) [29,30]. In contrast, genetic distances from the reconstructed ancestral sequence plotted against year of isolation show patterns typical of gradual genetic drift, except in the G gene where a 72-nt and 60-nt insertion is present (Fig 2D). Taken together, these results suggest that genetic and predicted T-cell epitope immune diversity are different and may be an important factor to consider when evaluating RSV vaccine efficacy.

Table 2. Experimentally validated conserved MHC class II epitopes peptides in RSV major surface proteins

Fig 2. Predicted T cell epitope landscapes and genetic evolution of RSV surface proteins

There are multiple methods available to predict T cell epitopes [31], which may result in different reconstructed landscapes if there is a systematic bias in the prediction method. We used the NetMHCpan method [32] to predict T cell epitopes and perform the same landscape reconstruction using MHC class I binding predictions for RSV-A F protein. Our analysis showed a consistent clustered pattern of RSV T epitope profile on the landscape regardless of the T cell epitope prediction method (S7 Fig).

Fig 3. Evaluation of previously used RSV vaccine candidate strains with T cell epitope content of circulating strains

Assessment of vaccine candidate strains with T cell epitope content 

T cell epitopes that are similar between vaccine strains and wild strains (cross-conserved T epitopes, which are defined as epitopes that share identical T cell receptor-facing residues and are restricted by the same alleles [33]) may be responsible for the T cell immune protection of the vaccine. To quantitatively evaluate whether it might be necessary to include multiple RSV strains to prepare an effective vaccine, two live attenuated RSV strains that are previously considered as vaccine candidates, CP248, a recombinant virus that belongs to subtype A, and CP52, which is a recombinant RSV-B strain, were included in our analysis and we evaluated their T cell epitope conservation with different RSV wild-type strains. We calculated the average proportion of cross-conserved T cell epitope content between the selected vaccine strains and wild-circulating strains from different isolation years and WHO regional groups (Figs 3 and S8). Different proportions of cross-conserved T cell epitope content against isolates from two different subtypes, A and B, were observed in both the F and G protein analyses. In the comparison of the vaccine strains and wild strains belonging to the same subtype, the proportion of cross-conserved T cell epitope in RSV F protein is relatively stable in different groups, all are higher than 78% for RSV-A and higher than 85% for RSV-B. In contrast, changes in the proportion of cross-conserved T cell epitopes were detected among groups within the same RSV subtype, especially in different temporal groups in the G protein analysis (Fig 3). Vaccine strain CP248 appears to have a relatively higher proportion of cross-conserved T cell epitopes within G protein when compared to the RSV-A strains that were isolated before 1991 (> 70%) and a relatively lower degree of conservation against recently isolated strains. A similar decrease in T cell epitope conservation with time was identified for vaccine strain CP52 among circulating RSV-B strains. 

Discussion

Although both CD4 and CD8 T cells contribute to protection against RSV-induced disease following primary infection [16,34], T cell epitopes have received limited attention in the RSV research effort. We demonstrate RSV surface proteins appear to have significant potential to drive T cell immunity using a computational approach, based on their T cell epitope density scores as determined by MHC molecular binding prediction. The relatively high putative T cell epitope density might make F protein a good target for the RSV vaccine. In addition to the analysis of T cell epitope density and distribution in RSV major surface proteins, we also demonstrated lineage-specific variations in T cell epitope content. Even though RSV F protein is believed to be well conserved, epitope mutations are observed across different lineages within the F protein, suggesting that studying the lineage-specific T cell epitopes in RSV can provide insight into the impact of immune selection on viral diversity and persistence. In contrast to the conserved F protein, RSV G protein is reported to be highly variable, we still observed potential conserved T cell epitopes across different strains, which suggests a great interest in the G protein conserved domain as potential vaccine targets on T cell immune protection. While experimental validation is needed, this analysis highlights the importance of understanding population-level epitope conservation as it may provide important insight into the development of T cell epitope-driven vaccines against RSV infection.

A major focus of our work is the development of a sequence-based method to map the evolution of T cell immunity across different stains. Following a previously pivotal work that used the MDS method to map the evolutionary adaptation of influenza A virus-induced by CD8 T cells using the presence and absence of MHC class I epitopes [35], we constructed RSV T cell immunity landscapes using immune distances that were generated by T cell epitope cross-conservation analyses, which allows for easy visualization and intuitive understanding of the potential for T cell immunity relationships among different strains. When comparing across strains, we found that the T cell epitope content of RSV surface proteins from different strains can be clustered, as has been observed for the antigenic relationship reported in other pathogens [36,37]. Our results also demonstrate the correspondence between RSV T cell immunity clusters and their corresponding phylogeny, with sequences in the same clade generally belonging to the same T cell immune cluster. Importantly, we also observe different patterns of T-cell epitope evolution of RSV wild strains compared with their genetic evolution. We find the RSV strains with G gene duplications can still cluster with previous RSV isolates on the T cell epitope MDS space. Our results highlight the importance of characterizing T cell epitope changes in RSV.

cistanche tubulosa-improve immune system

We identified highly conserved RSV T cell epitopes in this study, some of which have already been experimentally validated and published in the IEDB database. However, we also identified several other conserved T-cell epitopes that have not been previously described. These may be valuable for vaccine design, although experimental validation will be needed. Furthermore, the homology of selected RSV epitopes to human peptides suggests that some predicted RSV T cell epitopes might be tolerated by the human immune system or could induce a harmful cross-reactive immune response against human proteins when administered with an adjuvant [27]. Certain aspects of immunity to RSV were not addressed by this study. For example, neutralizing antibody responses are currently considered to be the most important correlate of immunity. While neutralizing antibodies would not directly be elicited by a T cell epitope-driven vaccine, helper (CD4) T cell epitopes are required to generate high affinity, high specificity antibodies. We also note that we have limited our focus on the two major RSV surface proteins in our current analysis, but other RSV proteins like N, M, or M2-2 proteins might also contribute to vaccine efficacy [38].

cistanche tubulosa-improve immune system

An effective vaccine against variable viruses should contain T-cell epitopes that are highly conserved among circulating strains [39]. Vaccine efficacy can be diminished if T cell epitopes in a vaccine strain do not match when new strains of pathogens emerge. In this study, we used an immunoinformatic-based approach to estimate cross-conserved T cell epitope contents between two live attenuated vaccine candidate strains and RSV circulating wild strains. We found that there was a low proportion of cross-conserved T cell epitope content with vaccine strains that belonged to different antigenic groups, which indicates the risk of using a single subtype strain in RSV vaccines. In addition, we observed a lower proportion of cross-conserved G-protein T cell epitope content between vaccine strains and recent circulating strains in the same antigenic group, which suggests that including T cell epitopes from different strains in the same antigenic group might also be important for RSV vaccine development. Although we did not observe a significant change in cross-conserved T cell epitope content in F protein, we cannot rule out the possibility that variation of F protein in the future could render a single-strain-based vaccine less effective. Our current analysis is based on reduced datasets due to the heavy computational capacity required to perform epitope content comparison. We constructed these representative datasets by randomly subsampling the complete datasets according to geographical regions and isolated years. Our findings may reflect the T cell epitope diversity of publicly available RSV strains, however, additional RSV surveillance efforts may be required to get a full picture of the T cell epitope variability of RSV.

Our current study is limited by lacking experimental validation of T cell epitope prediction. We focus on computationally predicted MHC binding to identify T-cell epitopes. Although the strength of MHC binding is the key parameter that determines a peptide's immunogenicity, but not sufficient for a module to be immunogenic [40]. Other aspects associated with pathogen-induced T cell immune response, such as appropriate antigen-processing [41] and T cell receptor recognition [40] are not considered in this study, which might cause bias in the computational-based T cell epitope landscapes. However, the observed clustered pattern of RSV surface proteins on the T cell epitope landscape in this study reflects the diversity of T cell epitopes within different strains. This finding provides valuable insights into virus evolution in the aspects of T-cell immunity and can contribute to strain selection for vaccine design. Overall, this study provides a focused analysis of T cell epitopes in RSV major surface proteins using computational tools. We performed a comprehensive T cell epitope prediction for RSV showing the immunological relationship of T cell epitopes in RSV surface proteins. This study demonstrates that T cell epitope evolution may differ from genetic variation and provides a framework for developing an integrated epitope-based RSV vaccine and evaluation methods that could be used to optimize vaccination strategies.

References

1. Crowe JE Jr., Williams JV. Paramyxoviruses: Respiratory Syncytial Virus and Human Metapneumovirus: Viral Infections of Humans. 2014 Feb 27:601–27. https://doi.org/10.1007/978-1-4899-7448-8_26 

2. Yun MR, Kim AR, Lee HS, Kim DW, Lee WJ, Kim K, et al. Complete genome sequences of human respiratory syncytial virus genotype A and B isolates from South Korea. Genome Announc. 2015; 3(2). Epub 2015/04/25. https://doi.org/10.1128/genomeA.00332-15 PMID: 25908140; PubMed Central PMCID: PMC4408341.

3. McLellan JS, Ray WC, Peeples ME. Structure and function of respiratory syncytial virus surface glycoproteins. Curr Top Microbiol Immunol. 2013; 372:83–104. Epub 2013/12/24. https://doi.org/10.1007/ 978-3-642-38919-1_4 PMID: 24362685; PubMed Central PMCID: PMC4211642. 

4. Lee J, Klenow L, Coyle EM, Golding H, Khurana S. Protective antigenic sites in respiratory syncytial virus G attachment protein outside the central conserved and cysteine noose domains. PLOS Pathogens. 2018; 14(8):e1007262. https://doi.org/10.1371/journal.ppat.1007262 PMID: 30142227 

5. Updated guidance for palivizumab prophylaxis among infants and young children at increased risk of hospitalization for respiratory syncytial virus infection. Pediatrics. 2014; 134(2):e620–38. Epub 2014/ 07/30. https://doi.org/10.1542/peds.2014-1666 PMID: 25070304. 

6. Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants. The IMpact-RSV Study Group. Pediatrics. 1998; 102(3 Pt 1):531–7. Epub 1998/09/17. PMID: 9738173. 

7. Schwarz TF, Johnson C, Grigat C, Apter D, Csonka P, Lindblad N, et al. Three Dose Levels of a Maternal Respiratory Syncytial Virus Vaccine Candidate Are Well Tolerated and Immunogenic in a Randomized Trial in Nonpregnant Women. J Infect Dis. 2022; 225(12):2067–76. Epub 2021/06/20. https://doi.org/10.1093/infdis/jiab317 PMID: 34146100; PubMed Central PMCID: PMC9200160. 

8. Biagi C, Dondi A, Scarpini S, Rocca A, Vandini S, Poletti G, et al. Current State and Challenges in Developing Respiratory Syncytial Virus Vaccines. Vaccines (Basel). 2020; 8(4). Epub 2020/11/15. https://doi.org/10.3390/vaccines8040672 PMID: 33187337; PubMed Central PMCID: PMC7711987. 

9. Kim HW, Canchola JG, Brandt CD, Pyles G, Chanock RM, Jensen K, et al. Respiratory syncytial virus disease in infants despite prior administration of antigenic inactivated vaccine. Am J Epidemiol. 1969; 89(4):422–34. Epub 1969/04/01. https://doi.org/10.1093/oxfordjournals.aje.a120955 PMID: 4305198. 

10. Murphy BR, Sotnikov AV, Lawrence LA, Banks SM, Prince GA. Enhanced pulmonary histopathology is observed in cotton rats immunized with formalin-inactivated respiratory syncytial virus (RSV) or purified F glycoprotein and challenged with RSV 3–6 months after immunization. Vaccine. 1990; 8(5):497–502. Epub 1990/10/01. https://doi.org/10.1016/0264-410x(90)90253-i PMID: 2251875. 

11. Killikelly AM, Kanekiyo M, Graham BS. Pre-fusion F is absent on the surface of formalin-inactivated respiratory syncytial virus. Scientific Reports. 2016; 6(1):34108. https://doi.org/10.1038/srep34108 PMID: 27682426 

12. Polack FP, Teng MN, Collins PL, Prince GA, Exner M, Regele H, et al. A role for immune complexes in enhanced respiratory syncytial virus disease. J Exp Med. 2002; 196(6):859–65. Epub 2002/09/18. https://doi.org/10.1084/jem.20020781 PMID: 12235218; PubMed Central PMCID: PMC2194058. 

13. Connors M, Giese NA, Kulkarni AB, Firestone CY, Morse HC 3rd, Murphy BR. Enhanced pulmonary histopathology induced by respiratory syncytial virus (RSV) challenge of formalin-inactivated RSV immunized BALB/c mice is abrogated by depletion of interleukin-4 (IL-4) and IL-10. J Virol. 1994; 68 (8):5321–5. Epub 1994/08/01. https://doi.org/10.1128/JVI.68.8.5321-5325.1994 PMID: 8035532; PubMed Central PMCID: PMC236482. 

14. Waris ME, Tsou C, Erdman DD, Zaki SR, Anderson LJ. Respiratory syncytial virus infection in BALB/c mice previously immunized with formalin-inactivated virus induces enhanced pulmonary inflammatory response with a predominant Th2-like cytokine pattern. J Virol. 1996; 70(5):2852–60. Epub 1996/05/01. https://doi.org/10.1128/JVI.70.5.2852-2860.1996 PMID: 8627759; PubMed Central PMCID: PMC190142. 

15. Gilman MSA, Furmanova-Hollenstein P, Pascual G, BvtW A, Langedijk JPM, McLellan JS. Transient opening of trimeric prefusion RSV F proteins. Nat Commun. 2019; 10(1):2105. Epub 2019/05/10. https://doi.org/10.1038/s41467-019-09807-5 PMID: 31068578; PubMed Central PMCID: PMC6506550 RSV, has received research funding from MedImmune and Janssen, has been a paid consultant for MedImmune, and is on the scientific advisory board for Calder Biosciences. M.S.A.G. is a named inventor on a patent application for RSV F-directed single-domain antibodies. P.F.-H., G.P., A.vW., and J.P. M.L. are employees of Janssen, a pharmaceutical company of Johnson & Johnson. 

16. Taylor G. Animal models of respiratory syncytial virus infection. Vaccine. 2017; 35(3):469–80. Epub 2016/12/03. https://doi.org/10.1016/j.vaccine.2016.11.054 PMID: 27908639; PubMed Central PMCID: PMC5244256. 

17. Blunck BN, Rezende W, Piedra PA. Profile of respiratory syncytial virus prefusogenic fusion protein nanoparticle vaccine. Expert Rev Vaccines. 2021; 20(4):351–64. Epub 2021/03/19. https://doi.org/10. 1080/14760584.2021.1903877 PMID: 33733995. 

18. Mazur NI, Higgins D, Nunes MC, Melero JA, Langedijk AC, Horsley N, et al. The respiratory syncytial virus vaccine landscape: lessons from the graveyard and promising candidates. Lancet Infect Dis. 2018; 18(10):e295–e311. Epub 2018/06/20. https://doi.org/10.1016/S1473-3099(18)30292-5 PMID: 29914800. 

19. Liang X, Liu DH, Chen D, Guo L, Yang H, Shi YS, et al. Gradual replacement of all previously circulating respiratory syncytial virus A strain with the novel ON1 genotype in Lanzhou from 2010 to 2017. Medicine (Baltimore). 2019; 98(19):e15542. Epub 2019/05/15. https://doi.org/10.1097/MD.0000000000015542 PMID: 31083208; PubMed Central PMCID: PMC6531141. 

20. Ahmed A, Haider SH, Parveen S, Arshad M, Alsenaidy HA, Baaboud AO, et al. Co-Circulation of 72bp Duplication Group A and 60bp Duplication Group B Respiratory Syncytial Virus (RSV) Strains in Riyadh, Saudi Arabia in 2014. PLOS ONE. 2016; 11(11):e0166145. https://doi.org/10.1371/journal.pone. 0166145 PMID: 27835664 

21. Tian D, Battles MB, Moin SM, Chen M, Modjarrad K, Kumar A, et al. Structural basis of respiratory syncytial virus subtype-dependent neutralization by an antibody targeting the fusion glycoprotein. Nature Communications. 2017; 8(1):1877. https://doi.org/10.1038/s41467-017-01858-w PMID: 29187732 

22. Hashimoto K, Hosoya M. Neutralizing epitopes of RSV and palivizumab resistance in Japan. Fukushima J Med Sci. 2017; 63(3):127–34. Epub 2017/09/05. https://doi.org/10.5387/fms.2017-09 PMID: 28867684; PubMed Central PMCID: PMC5792496.

23. Sullender WM. Respiratory syncytial virus genetic and antigenic diversity. Clin Microbiol Rev. 2000; 13 (1):1–15, table of contents. Epub 2000/01/11. https://doi.org/10.1128/CMR.13.1.1 PMID: 10627488; PubMed Central PMCID: PMC88930. 

24. Chen X, Xu B, Guo J, Li C, An S, Zhou Y, et al. Genetic variations in the fusion protein of respiratory syncytial virus isolated from children hospitalized with community-acquired pneumonia in China. Scientific Reports. 2018; 8(1):4491. https://doi.org/10.1038/s41598-018-22826-4 PMID: 29540836 

25. De Groot AS, Moise L, Terry F, Gutierrez AH, Hindocha P, Richard G, et al. Better Epitope Discovery, Precision Immune Engineering, and Accelerated Vaccine Design Using Immunoinformatics Tools. Front Immunol. 2020; 11:442. Epub 2020/04/23. https://doi.org/10.3389/fimmu.2020.00442 PMID: 32318055; PubMed Central PMCID: PMC7154102. 

26. Gilman MS, Castellanos CA, Chen M, Ngwuta JO, Goodwin E, Moin SM, et al. Rapid profiling of RSV antibody repertoires from the memory B cells of naturally infected adult donors. Sci Immunol. 2016; 1 (6). Epub 2017/01/24. https://doi.org/10.1126/sciimmunol.aaj1879 PMID: 28111638; PubMed Central PMCID: PMC5244814. 

27. He L, De Groot AS, Gutierrez AH, Martin WD, Moise L, Bailey-Kellogg C. Integrated assessment of predicted MHC binding and cross-conservation with self reveals patterns of viral camouflage. BMC Bioinformatics. 2014; 15 Suppl 4(Suppl 4):S1. Epub 2014/08/12. https://doi.org/10.1186/1471-2105-15-S4- S1 PMID: 25104221; PubMed Central PMCID: PMC4094998. 

28. Grenfell BT, Pybus OG, Gog JR, Wood JL, Daly JM, Mumford JA, et al. Unifying the epidemiological and evolutionary dynamics of pathogens. Science. 2004; 303(5656):327–32. Epub 2004/01/17. https:// doi.org/10.1126/science.1090727 PMID: 14726583. 

29. Smith DJ, Lapedes AS, de Jong JC, Bestebroer TM, Rimmelzwaan GF, Osterhaus AD, et al. Mapping the antigenic and genetic evolution of influenza virus. Science. 2004; 305(5682):371–6. Epub 2004/06/ 26. https://doi.org/10.1126/science.1097211 PMID: 15218094. 

30. Bedford T, Suchard MA, Lemey P, Dudas G, Gregory V, Hay AJ, et al. Integrating influenza antigenic dynamics with molecular evolution. eLife. 2014; 3:e01914. https://doi.org/10.7554/eLife.01914 PMID: 24497547 

31. Korber B, LaBute M, Yusim K. Immunoinformatics Comes of Age. PLOS Computational Biology. 2006; 2(6):e71. https://doi.org/10.1371/journal.pcbi.0020071 PMID: 16846250 

32. Reynisson B, Alvarez B, Paul S, Peters B, Nielsen M. NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Research. 2020; 48(W1):W449–W54. https://doi.org/10.1093/nar/ gkaa379 PMID: 32406916

33. Gutie´rrez AH, Rapp-Gabrielson VJ, Terry FE, Loving CL, Moise L, Martin WD, et al. T-cell epitope content comparison (EpiCC) of swine H1 influenza A virus hemagglutinin. Influenza Other Respir Viruses. 2017; 11(6):531–42. Epub 2017/10/21. https://doi.org/10.1111/irv.12513 PMID: 29054116; PubMed Central PMCID: PMC5705686. 

34. Russell CD, Unger SA, Walton M, Schwarze J. The Human Immune Response to Respiratory Syncytial Virus Infection. Clin Microbiol Rev. 2017; 30(2):481–502. Epub 2017/02/10. https://doi.org/10.1128/ CMR.00090-16 PMID: 28179378; PubMed Central PMCID: PMC5355638. 

35. Woolthuis RG, van Dorp CH, Keşmir C, de Boer RJ, van Boven M. Long-term adaptation of the influenza A virus by escaping cytotoxic T-cell recognition. Scientific Reports. 2016; 6(1):33334. https://doi. org/10.1038/srep33334 PMID: 27629812 

36. Katzelnick LC, Fonville JM, Gromowski GD, Arriaga JB, Green A, James SL, et al. Dengue viruses cluster antigenically but not as discrete serotypes. Science. 2015; 349(6254):1338–43. https://doi.org/10. 1126/science.aac5017 PMID: 26383952 

37. Tuju J, Mackinnon MJ, Abdi AI, Karanja H, Musyoki JN, Warimwe GM, et al. Antigenic cartography of immune responses to Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1). PLOS Pathogens. 2019; 15(7):e1007870. https://doi.org/10.1371/journal.ppat.1007870 PMID: 31260501 

38. Liu J, Ruckwardt TJ, Chen M, Johnson TR, Graham BS. Characterization of Respiratory Syncytial Virus M- and M2-Specific CD4 T Cells in a Murine Model. Journal of Virology. 2009; 83(10):4934–41. https:// doi.org/10.1128/JVI.02140-08 PMID: 19264776

39. Viboud C, Gnostic K, Nelson MI, Price GE, Perofsky A, Sun K, et al. Beyond clinical trials: Evolutionary and epidemiological considerations for development of a universal influenza vaccine. PLOS Pathogens. 2020; 16(9):e1008583. https://doi.org/10.1371/journal.ppat.1008583 PMID: 32970783 

40. Schaap-Johansen AL, Vujović M, Borch A, Hadrup SR, Marcatili P. T Cell Epitope Prediction and Its Application to Immunotherapy. Front Immunol. 2021; 12:712488. Epub 2021/10/05. https://doi.org/10. 3389/fimmu.2021.712488 PMID: 34603286; PubMed Central PMCID: PMC8479193.

41. Mettu RR, Charles T, Landry SJ. CD4+ T-cell epitope prediction using antigen processing constraints. J Immunol Methods. 2016; 432:72–81. Epub 2016/02/20. https://doi.org/10.1016/j.jim.2016.02.013 PMID: 26891811; PubMed Central PMCID: PMC5321161.

42. Chen J, Qiu X, Avadhanula V, Shepard SS, Kim D-K, Hixson J, et al. Novel and extendable genotyping system for human respiratory syncytial virus based on whole-genome sequence analysis. Influenza Other Respir Viruses. 2022; 16(3):492–500. https://doi.org/10.1111/irv.12936 PMID: 34894077 

43. Shepard SS, Davis CT, Bahl J, Rivailler P, York IA, Donis RO. LABEL: Fast and Accurate Lineage Assignment with Assessment of H5N1 and H9N2 Influenza A Hemagglutinins. PLOS ONE. 2014; 9(1): e86921. https://doi.org/10.1371/journal.pone.0086921 PMID: 24466291 

44. Organization UNbtWH. World Health Statistics 2011 2011. 

45. Rozewicki J, Li S, Amada KM, Standley DM, Katoh K. MAFFT-DASH: integrated protein sequence and structural alignment. Nucleic Acids Research. 2019; 47(W1):W5–W10. https://doi.org/10.1093/nar/ gkz342 PMID: 31062021 

46. Rice P, Longden I, Bleasby A. EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 2000; 16(6):276–7. Epub 2000/05/29. https://doi.org/10.1016/s0168-9525(00)02024-2 PMID: 10827456. 

47. Whitehead SS, Hill MG, Firestone CY, St Claire M, Elkins WR, Murphy BR, et al. Replacement of the F and G proteins of respiratory syncytial virus (RSV) subgroup A with those of subgroup B generates chimeric live attenuated RSV subgroup B vaccine candidates. J Virol. 1999; 73(12):9773–80. Epub 1999/ 11/13. https://doi.org/10.1128/JVI.73.12.9773-9780.1999 PMID: 10559287; PubMed Central PMCID: PMC113024. 

48. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014; 30(9):1312–3. Epub 2014/01/24. https://doi.org/10.1093/bioinformatics/btu033 PMID: 24451623; PubMed Central PMCID: PMC3998144. 

49. Sagulenko P, Puller V, Neher RA. TreeTime: Maximum-likelihood phylodynamic analysis. Virus Evolution. 2018; 4(1). https://doi.org/10.1093/ve/vex042 PMID: 29340210 

50. Yu G, Smith DK, Zhu H, Guan Y, Lam TT-Y. agree on an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods in Ecology and Evolution. 2017; 8(1):28–36. https://doi.org/10.1111/2041-210X.12628 

51. Southwood S, Sidney J, Kondo A, del Guercio MF, Appella E, Hoffman S, et al. Several common HLADR types share largely overlapping peptide binding repertoires. J Immunol. 1998; 160(7):3363–73. Epub 1998/04/08. PMID: 9531296. 

52. Sette A, Sidney J. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics. 1999; 50(3–4):201–12. Epub 1999/12/22. https://doi.org/10.1007/ s002510050594 PMID: 10602880. 

53. UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res. 2021; 49(D1): D480–d9. Epub 2020/11/26. https://doi.org/10.1093/nar/gkaa1100 PMID: 33237286; PubMed Central PMCID: PMC7778908. 

54. Moise L, Gutierrez A, Kibria F, Martin R, Tassone R, Liu R, et al. iVAX: An integrated toolkit for the selection and optimization of antigens and the design of epitope-driven vaccines. Human Vaccines & Immunotherapeutics. 2015; 11(9):2312–21. https://doi.org/10.1080/21645515.2015.1061159 PMID: 26155959 

55. Zhu Q, McLellan JS, Kallewaard NL, Ulbrandt ND, Palaszynski S, Zhang J, et al. A highly potent extended half-life antibody as a potential RSV vaccine surrogate for all infants. Sci Transl Med. 2017; 9 (388). Epub 2017/05/05. https://doi.org/10.1126/scitranslmed.aaj1928 PMID: 28469033. 

56. McLellan JS, Yang Y, Graham BS, Kwong PD. The structure of respiratory syncytial virus fusion glycoprotein in the postfusion conformation reveals the preservation of neutralizing epitopes. J Virol. 2011; 85 (15):7788–96. Epub 2011/05/27. https://doi.org/10.1128/JVI.00555-11 PMID: 21613394; PubMed Central PMCID: PMC3147929. 

57. Karosiene E, Lundegaard C, Lund O, Nielsen M. NetMHCcons: a consensus method for the major histocompatibility complex class I predictions. Immunogenetics. 2012; 64(3):177–86. https://doi.org/10. 1007/s00251-011-0579-8 PMID: 22009319 

58. GOWER JC. Some distance properties of latent root and vector methods are used in multivariate analysis. Biometrika. 1966; 53(3–4):325–38. https://doi.org/10.1093/biomet/53.3–4.325 

59. Sa´nchez J. MARDIA K. V., J. T. KENT J. M. BIBBY: Multivariate Analysis. Academic Press, LondonNew York-Toronto-Sydney-San Francisco 1979. xv, 518 pp., $ 61.00. Biometrical Journal. 1982;24 (5):502-. https://doi.org/10.1002/bimj.4710240520.

60. Charrad M, Ghazzali N, Boiteau V, Niknafs A. NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set. Journal of Statistical Software. 2014; 61(6):1–36. https://doi.org/10. 18637/jss.v061.i06 

61. de Leeuw J, Mair P. Multidimensional Scaling Using Majorization: SMACOF in R. Journal of Statistical Software. 2009; 31(3):1–30. https://doi.org/10.18637/jss.v031.i03