VP2 Dominated CD4+ T Cell Responses against 71 and Cross-Reactivity against Coxsackievirus A16 and in a Healthy Population This information is current as of October 1, 2021. Shuguang Tan, Xiaojuan Tan, Xiaoman Sun, Guangwen Lu, Chun-Chi Chen, Jinghua Yan, Jun Liu, Wenbo Xu and George F. Gao J Immunol 2013; 191:1637-1647; Prepublished online 17

July 2013; Downloaded from doi: 10.4049/jimmunol.1301439 http://www.jimmunol.org/content/191/4/1637

Supplementary http://www.jimmunol.org/content/suppl/2013/07/17/jimmunol.130143 http://www.jimmunol.org/ Material 9.DC1 References This article cites 59 articles, 13 of which you can access for free at: http://www.jimmunol.org/content/191/4/1637.full#ref-list-1

Why The JI? Submit online.

by guest on October 1, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

VP2 Dominated CD4+ T Cell Responses against Enterovirus 71 and Cross-Reactivity against Coxsackievirus A16 and Polioviruses in a Healthy Population

Shuguang Tan,*,† Xiaojuan Tan,‡ Xiaoman Sun,*,† Guangwen Lu,* Chun-Chi Chen,* Jinghua Yan,*,† Jun Liu,*,‡ Wenbo Xu,‡ and George F. Gao*,†,‡,x

Enterovirus 71 (EV71)–associated hand-foot-mouth disease has become a major threat to public health in the Asia–Pacific region. Although T cell immunity is closely correlated with clinical outcomes of EV71 infection, little is known about T cell immunity baseline against EV71 and T cell immunogenecity of EV71 Ags in the population, which has restricted our understanding of immunoprotection mechanisms. In this study, we investigated the cellular immune responses against the four structural Ags of EV71 and determined the immunohierarchy of these Ags in healthy adults. A low frequency of EV71-responsive T cells was detected circulating in peripheral blood, and broad T cell immune responses could be identified in most of the subjects after Downloaded from in vitro expansion. We demonstrated that the VP2 Ag with broad distribution of immunogenic peptides dominates T cell responses against EV71 compared with VP1, VP3, and VP4. Furthermore, the responses were illuminated to be mainly single IFN-g– secreting CD4+ T cell dependent, indicating the previous natural acute viral infection of the adult population. Conservancy analysis of the immunogenic peptides revealed that moderately variant peptides were in the majority in coxsackievirus A16 (CV-A16) whereas most of the peptides were highly variant in polioviruses. Less efficient cross-reactivity against CV-A16 might broadly exist among individuals, whereas influences derived from vaccination would be limited. Our findings suggest http://www.jimmunol.org/ that the significance of VP2 Ag should be addressed in the future EV71-responsive immunological investigations. And the findings concerning the less efficient cross-reactivity against CV-A16 and limited influences from poliovirus vaccination in EV71-contacted population would contribute to a better understanding of immunoprotection mechanisms against . The Journal of Immunology, 2013, 191: 1637–1647.

ince its first identification in California in 1969, enterovirus in Fuyang City, , has caused ∼5000 cases and .188 deaths 71 (EV71)–associated hand-foot-mouth disease (HFMD) in mainland China (9). During the following years, .1 million S epidemics have been reported throughout the world, with incidences and hundreds of deaths have been reported annually in by guest on October 1, 2021 large outbreaks mainly occurring in the Asia–Pacific region since China, and EV71 has become a major threat to public health (http:// 1990s (1–7). Although EV71 and coxsackievirus A16 (CV-A16) www.moh.gov.cn/publicfiles/business/htmlfiles/wsb/pyqxx/list. are the most common causative agents of HFMD, infection by EV71 htm). In the first half of 2012, the so-called “mystery disease” in usually carries a higher risk of developing central nervous compli- children that caused .50 deaths in Cambodia was also found to cations and fatalities (8). In 2008, a nationwide EV71 outbreak started be EV71-assosiated HFMD (10). A continued lack of therapeutic medication and an effective vaccine for EV71 will likely lead to persistent periodical epidemics and dissemination in the near future *Chinese Academy of Sciences Key Laboratory of Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese Academy of Sciences, (11). Thus, effective vaccines that would prevent the prevalence of 100101, China; †University of Chinese Academy of Sciences, Beijing 100049, China; EV71 infection in at-risk populations are in urgent need for the ‡National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing 102206, China; and xResearch Network of control of HFMD around the world. Nevertheless, the mechanisms Immunity and Health, Beijing Institutes of Life Science, Chinese Academy of Sci- of EV71 pathogenesis remain unclear, and child-orientated epi- ences, Beijing 100101, China demics and severe neurologic complications with unknown causes Received for publication May 31, 2013. Accepted for publication June 6, 2013. have made the study of the immunogenicity of EV71 Ags and the This work was supported by the China National Grant Science and Technology roles they play in the protection against infection of significant Special Project (Grants 2012ZX10004201-009 and 2013ZX10004608-002). The fun- ders had no role in study design, data collection and analysis, decision to publish, or importance. preparation of the manuscript. G.F.G. is a leading principal investigator of the Na- As a major member of the Enterovirus genus of the family Picor- tional Natural Science Foundation of China Innovative Research Group (Grant naviridae, EV71 is a nonenveloped ssRNA coated with an 81021003). icosahedral capsid containing 60 copies of four structural proteins, Address correspondence and reprint requests to Dr. George F. Gao or Dr. Wenbo Xu, Chinese Academy of Sciences Key Laboratory of Pathogenic Microbiology and Im- VP1, VP2, VP3, and the inner VP4 (12). The four structural pro- munology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, teins are the major Ags of EV71 and have thus been the focus of China (G.F.G.) or National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing 102206, China (W.X.). E-mail vaccine development and applied as the main effective constituents addresses: [email protected] (G.F.G.) or [email protected] (W.X.) in vaccine effectiveness evaluations (13, 14). In particular, VP1 is The online version of this article contains supplemental material. highly exposed, as well as considered the main target of neutral- Abbreviations used in this article: CV-A16, coxsackievirus A16; EV71, enterovirus izing Abs, and has thus been broadly used for subunit vaccine 71; HFMD, hand-foot-mouth disease; ICS, intracellular cytokine staining; SFC, spot- development (15–17). The lower prevalence of neutralizing Abs forming cell. accompanied by higher transmission rates in children and infants Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 indicates that neutralizing Abs are important for the prevention of www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301439 1638 T CELL IMMUNITY AGAINST EV71 IN ADULTS

EV71 infection (8, 18). However, 80% of EV71-infected patients Neutralizing Ab detection become positive for EV71-specific neutralizing Abs 1 d after Neutralizing Ab against EV71 and CV-A16 were detected, respectively, the onset of illness, and the magnitude of Ab responses are not with microneutralization test on human rhabdomyosarcoma cell on 96-well correlated with disease severity and outcome (19). In contrast, flat-bottom cell culture microtiter plate as described previously (31). Sera documented clues imply that cellular but not humoral immune were inactivated at 56˚C for 30 min at first and prepared by 4-fold serial responses are correlated with disease progression and clinical dilutions ranged from 1:8 to 1:256. Serum sample (50 ml) was added to each well with duplicated wells for each dilution. Then, 50 mlvirus outcome (20, 21). (EV71 or CV-A16) with 100 50% cell culture infective dose were dis- Current studies on host immune responses against EV71 in- tributed to each well. The EV71 isolate (subgenotype C4a; GenBank dicate that T cell immunity plays a critical role in protection accession number EU703812) and the CV-A16 isolate (subgenotype B1b; against EV71 infection and control of the disease (20, 22). Pre- GenBank accession number GQ429229) were used in the analysis. After thorough mixture, virus and serum were incubated at 36˚C for 2 h. After vious work demonstrates that decreased cellular immunity and that, the virus would be neutralized and lose infectivity to cells if specific lower IFN-g and other cytokine/chemokine responses are corre- Ab exists in serum. Then, 100 ml cell suspension containing 2 3 105/ml lated with more severe clinical outcome of EV71 infection, rhabdomyosarcoma cells was added to each well, and the plate was cul- whereas the neutralizing Ab titers display no difference between tured in CO2 incubator at 36˚C for 7 d. Cytopathogenic effect was ex- mild, severe, and even fatal cases (22). Altered cellular immunity amined on day 7, and the dilution of serum, which could protect 50% cell from appearance of cytopathogenic effect, was designated as the titer of associated with polymorphisms of CTLA-4 is suggested to be the neutralizing Ab to specific virus. Serum control was set for each correlated with disease severity (20). Likewise, studies in animal specimen to observe any nonspecific toxic effect to cell at 1:8 dilution, models reveal that cellular immune-related lymphocytes or cyto- with virus replaced with cell maintain medium. To make sure the virus kines reduce the lethality of EV71-infected mice by decreasing used to attack was infective, virus control was set for each plate, with serum replaced with maintain medium. The method was previously calibrated Downloaded from viral loads in tissues (23–26). These observations lead to the hy- using reference sera of EV71 and CV-A16, and no cross-reaction was found pothesis that T cells might be crucial in protection and immunity (31). Finally, neutralizing titration $ 1:8 was considered to be positive against EV71. (9, 31, 32). Despite its importance, our knowledge of human cellular im- Synthetic peptides and peptide matrix pool design munity toward EV71 is still limited. To date, only a few studies have assessed cellular immunity against EV71 in small cohorts of To evaluate T cell responses against EV71, each of the four structural pro- subjects based on in silico predicted peptides (27–29). Thus, the teins of EV71 from the 2010 Henan strain (GenBank accession number http://www.jimmunol.org/ ADX87405.1) was selected as a template sequence for overlapping peptide magnitude of human T cell responses to EV71 and the immuno- design. A total of 110 15- to 19-mer C-terminal–adapted peptides over- hierarchy of EV71 Ags remain unclear. Furthermore, the pheno- lapping by 10 aa and spanning all of the structural proteins (VP1, VP2, type and functionalities of EV71-responsive T cells in humans VP3, and VP4) were designed and synthesized (purity . 90%; China- remain to be elucidated. A better understanding of all of these Peptides) as described previously (Supplemental Table I) (33). The peptides from VP1–VP3 were further assigned to two pools according to the se- aspects would be highly valuable to vaccine development and quence position (designated as the N- and C-terminal pools) for each Ag to understanding EV71 pathogenesis. minimum the matrix pools needed to screen the positive peptides in the pool To investigate human cellular immunity to EV71 and the im- (Table I). munohierarchy of the four structural proteins of EV71, we synthe- ELISPOT assay by guest on October 1, 2021 sized overlapping peptides spanning all the four structural proteins (VP1–VP4) and performed a study of viral Ag-specific T cell The Ag-specific T cell responses were detected with an IFN-g–secreting 3 5 responses in adults. In addition, we also determined the phenotype ELISPOT assay (Quantobio) (34, 35). A total of 2.5 10 freshly iso- lated PBMCs from donors in 100 ml RPMI 1640 medium supplemented and functional features of dominant Ag-specific T cells and their with 10% FBS were seeded in each well of 96-well ELISPOT plate cross-reactivity with different genotypes of EV71, CV-A16, and membranes precoated with 10 mg/ml anti–IFN-g mAb. To detect polioviruses at the individual peptide level. These data will aid in in vitro–cultured T cell lines, 5 3 104 cells were added to each well. To the understanding of host protection mechanisms against EV71 and stimulate the effector cells, individual peptide (with final concentration of 10 mg/ml) or peptide pools (with final concentration of 5 mg/ml) for effective vaccine development. each peptide diluted in 100 ml RPMI 1640 medium supplemented with 10% FBS were added to each well and incubated at 37˚C with 5% CO2 Materials and Methods for 18 h. PHA (36) was added as a positive control for nonspecific Study subjects stimulation. Cells incubated without stimulator were employed as a negative control, which produced less than five spots in 90% of the Thirty healthy subjects (age, 20–51 y) from Beijing, China, were recruited experiments. All of the wells were duplicated to minimize possible in this study. All of the subjects displayed no symptoms of HFMD and discrepancies. Then, the cells were removed, and the plates were pro- declared no contact with EV71-infected individuals during the sampling cessed according to the manufacturer’s instructions. Finally, the colored period. Written informed consent was obtained from all of the subjects, spots, which represent epitope-specific T cells, were counted and analyzed and the study was approved by the Ethics Review Committee of the In- using an automatic ELISPOT reader (CTL). Responses were considered stitute of Microbiology, Chinese Academy of Sciences. The study was to be positive when the spots-forming cells (SFCs) in the target well were conducted in accordance with the principles of the Declaration of Helsinki, more than five spots and greater than two times the average negative the standards of Good Clinical Practice (as defined by the International control value. The adjusted average SFCs after subtracting the mean neg- Conference on Harmonization). ative control values are presented as SFCs per 106 or 105 PBMCs in the determination of freshly isolated PBMCs or in vitro–expanded T cell lines, Lymphocyte purification and in vitro generation of polyclonal respectively. T cell lines CD4+ and CD8+ T cell depletion PBMCs were isolated from the whole blood of donors by density gradient centrifugation using Ficoll–Hypaque (TBD Science) and washed twice in Initially confirmed positive Ag pools or peptide-specific responses were RPMI 1640 medium containing 10% FBS (Life Technologies) (30). subsequently characterized by CD4 or CD8 depletion of expanded T cell Freshly isolated PBMCs in RPMI 1640 medium supplemented with 10% lines using anti–CD4- or anti–CD8-coated magnetic beads (MACS) (37). FBS were incubated with peptide pool (with final concentration of 5 mg/ml Briefly, polyclonal T cell lines generated by expansion in the presence of 6 for each peptide) at 37˚Cin 5% CO2 at a density of 2 3 10 cells/ml in corresponding peptide pools or peptide were incubated with anti–CD4- or a 24-well culture plate. rIL-2 (20 IU/ml) was added to the culture medium anti–CD8-coated magnetic beads for 15 min at 4˚C. Then, the mixture was on day 3. Half of the medium was replaced with fresh medium supple- washed and resuspended in PBS containing 0.5% FBS and 2 mM EDTA- mented with 20 IU/ml rIL-2 on day 7. Peptide-specific T cells were tested Na and passed through a strong magnetic field for depletion of CD4 or via ELISPOT assays on day 10. CD8 populations. The CD4- or CD8-depleted T cell lines were collected The Journal of Immunology 1639 for ELISPOT assays to analyze the CD4 and CD8 distribution of the peptide pool– or peptide-specific responses.

Intracellular cytokine staining and flow cytometry After in vitro culture, T cell lines were rested for 2 h and then stimulated with a specific peptide pool (5 mg/ml for each peptide) or peptide (10 mg/ml) for 2 h and incubated for an additional 4 h with Golgistop/monesin (BD Bio- sciences) at 37˚C in 5% CO2 (30). Cells cultured with medium alone or PHA were used as negative and positive controls, respectively. To determine CD4 and CD8 phenotypes, the cells were harvested and stained with anti- CD4 allophycocyanin and/or anti-CD8 FITC surface markers, fixed, and permeabilized in permeabilizing buffer (BD Biosciences), and stained with anti–IFN-g PE-Cy7 (BD Biosciences). To analyze the functional properties of the peptide-specific peptides, T cell lines were stained with anti-CD4 FITC as a surface marker and then intracellularly stained with anti–IFN-g PE-Cy7 and anti–IL-2 allophycocyanin. All fluorescent lymphocytes were gated on a FACSCalibur flow cytometer (BD Biosciences) and analyzed with FlowJo software.

Statistical analysis Differences in mean values were evaluated for statistical significance (p , 0.05 or , 0.01) by the Student two-tailed t test. Data were assembled and Downloaded from statistically calculated using Excel (Microsoft) or Origin 8.0 (OriginLab).

Results Overall humoral and cellular immunohierarchy in the peripheral blood of healthy donors against EV71 and CV-A16

Blood samples from 30 healthy adults were collected to evaluate http://www.jimmunol.org/ their humoral and cellular immune responses against EV71 (male: female = 19:11, age 28 6 6.1 y). Humoral immune responses were evaluated by determining neutralizing Ab titers against EV71 and CV-A16. Generally, EV71-dominated neutralizing Ab responses FIGURE 1. Ex vivo determination of humoral and cellular immunity were demonstrated among the study subjects with titers of $ 1:8 against EV71 in healthy individuals. (A) Neutralizing Ab titers against and showed an overall positive rate of 80% (24 of 30) (Fig. 1A). EV71 and CV-A16 in the study subjects. Neutralizing Ab titers against In contrast, less than half of the subjects (47%; 14 of 30) were EV71 are indicated on the y-axis, whereas the responses to CV-A16 are positive for CV-A16–specific neutralizing Abs. A significantly expressed on the x-axis. Each spot represents neutralizing Ab titers against higher responsive magnitudes were also demonstrated against EV71 EV71 and CV-A16 in an individual, and different size of the spot repre- by guest on October 1, 2021 (average of 1:36) than CV-A16 (average of 1:9) among the sub- sents number of subjects with same neutralizing Ab patterns, as indicated jects (p , 0.01). Specifically, EV71-only responses were detected on the right. Two circles were presented to show subjects with EV71 only B in 41.7% (10 of 24) of the subjects who possess undetectable CV- neutralizing Abs (left) or EV71 superior neutralizing Ab titers (right). ( ) A16 humoral responses, whereas superior (5 of 24) or equal (9 of Ex vivo assessment of T cell responses against the structural Ag peptide pools. The magnitude of responses to the seven peptide pools is repre- 24) responses against EV71, compared with CV-A16, were detected sented on the y-axis. Each diamond-shaped spot represents the mean among the other subjects. In contrast, no CV-A16–only or CV-A16– responses of two duplicate wells to the corresponding peptide pool in an superior responses were detected among the study subjects. These individual. Negative control and are presented as “M” group on the x-axis. findings are in accordance with the EV71-dominated epidemics in The top and bottom of each rectangular box denote the SE of each group, China (31, 38, 39). EV71-dominated neutralizing Ab responses with the mean responses of the group shown inside the box. Lines among the study subjects suggested a previous contact of EV71 stretching from the box represent 10–90% of the entire responses. virus in most of the study subjects and ensured that the investi- gation of EV71 responsive cellular immunity among these subjects the others had no detectable neutralizing Abs. These results revealed was reasonable and feasible. that mild T cell responses against EV71 could be detected in To evaluate T cell responses against EV71 in a healthy popu- a healthy population with or without the presence of EV71- lation, we synthesized overlapping peptides spanning the entire neutralizing Abs. The magnitude of responses against VP2 were sequence of the four structural proteins of EV71. A more conserved significantly higher than that against the other Ags (p , 0.01), EV71/Henan/DC/2010 strain was selected as the template strain for which indicated that VP2 dominated the T cell responses against overlapping peptide design. As a result, 110 C-terminal–adapted EV71 (Fig. 1B). 15- to 19-mer peptides overlapping by 10 aa were designed and further assigned to seven pools (Table I). Freshly isolated PBMCs The dominant role of VP2-specific T cell responses from 30 donors were separately challenged with the seven peptide To investigate the possible low-frequency T cells against EV71, the pools, and T cell responses were determined by ex vivo ELISPOT remainder of the PBMCs collected from all of the donors were assays and analyzed as SFCs per 106 PBMCs. According to the individually cultured in the presence of the seven peptide pools positive criteria (see Materials and Methods), mild T cell responses in vitro for 9 d, and T cell responses against EV71 were then against EV71 were detected in seven donors (with a mean response evaluated from the expanded T cells. Unexpectedly, on the basis of of 25.4 6 4.7 SFCs/106 PBMCs). Interestingly, all seven positive the criteria, we used to score a “positive” response (see Materials responses were against the VP2 Ag: four to the N-terminal pool and Methods), 28 of the subjects (93%) displayed significant positive and the other three to the C-terminal pool. Among these indi- responses to at least one of the four structural Ags (Fig. 2A, 2B). viduals, four had neutralizing Ab titers $ 1:16 against EV71, but Similar to the ex vivo ELISPOT results, dominant responses were 1640 T CELL IMMUNITY AGAINST EV71 IN ADULTS

Table I. Summary of designed overlapping peptide pools spanning the responses as a whole with neutralizing Ab titers against EV71 or four structural Ags of EV71 CV-A16 (p . 0.05). Next, we analyzed the immunohierarchy of the four structural Length Amino Acids No. of Ags in the subjects. The level of responses from in vitro–expanded Aga (amino acids) Peptide Pool Spectrumb Peptides PBMCs was both related to the proliferation rate of Ag-specific VP1 297 VP1-N-pool 1–162 20 T cells in different individuals and determined by the memory VP1-C-pool 145–297 19 precursors and phenotypes of the virus-specific T cells in different VP2 254 VP2-N-pool 1–126 16 VP2-C-pool 111–241 16 donors. From the analysis, we observed that the VP2 Ag elicited 5 VP3 242 VP3-N-pool 1–132 16 a mean of 171 SFCs/10 PBMCs among the 28 subjects positive VP3-C-pool 115–254 15 against EV71, which was much higher compared with 73, 67, and VP4 69 VP4 1–69 8 44 SFCs/105 PBMCs for the VP1, VP3, and VP4 Ags, respectively aOverlapping peptides were designed using EV71/HENAN/DC/2010 (GenBank (p , 0.01) (Fig. 2C). Moreover, the magnitudes of the VP2- accession number ADX87405.1) strain as template. specific responses among different individuals were well corre- bAmino acids spectrum of each peptide pool was presented from N- to C-terminal of each Ag. lated with the overall T cell responses to the four structural Ags as a whole (r = 0.84; p , 0.01) (Fig. 2D). Taken together, these findings demonstrated that the VP2 Ag displays a higher recog- VP2 specific, and positive responses against VP2 were detected in nition frequency and immunohierarchy than the other structural 26 subjects (i.e., 87% of the study population (Fig. 2B)). More Ags in cellular responses against EV71, demonstrating that the specifically, 57% (17 of 30) and 67% (20 of 30) of the responses VP2 Ag dominates T cell responses against EV71 in an adult were specific to the VP2-N- or -C-terminal peptide pools, respec- population. Downloaded from tively. In contrast, the recognition frequency for VP1 (70%) and VP3 (77%) was much lower, with the lowest recognition rate for Uncommon diverse distribution of T cell immunogenic regions VP4 (47%) (Fig. 2B). Furthermore, we analyzed the correlation in VP2 protein between the magnitudes of EV71 or CV-A16–neutralizing Ab We next sought to determine the shorter immunodominant regions titers with T cell responsive magnitudes against the four Ags. No toward which the immunogenicity of the four structural Ags was significant correlation was observed in any of the Ag or overall specified. In this section, only subjects who displayed a positive http://www.jimmunol.org/

FIGURE 2. Immunohierarchy of four EV71 struc- by guest on October 1, 2021 tural Ags and the dominance of the VP2 Ag. (A) All participants are represented on the x-axis, and the total magnitudes of the IFN-g ELISPOT responses against the four Ags are represented on the y-axis. The results represent the average SFCs per 105 PBMCs from two duplicate wells, after subtracting the mean negative control values in the individual. Each colored segment represents the source Ag of the corresponding respon- ses. (B) Differential recognition of four Ags among the individuals. According to the positive criterion for ELISPOT analysis of in vitro–expanded PBMCs, the overall recognition frequency of the four Ags in the study cohort is presented on the y-axis. The shadow on each colored bar represents the frequency of dual rec- ognition to the N and C terminus of the correspond- ing Ag. The upper parts represent the recognition of the N-terminal peptide pool, and the lower parts represent the recognition of the C-terminal peptide pool of a given Ag. Overall recognition frequency was designated as “EV71.” (C) Magnitudes of the average responses to the four Ags among the 28 individuals who showed positive responses. The SE was represented on each Ag bar. **p , 0.01. (D) The magnitude of VP2-specific re- sponses was highly correlated with the overall responses to the four Ags as a whole (r =0.84,p , 0.01). The Journal of Immunology 1641 response to a particular peptide pool were involved to determine donor 5. From the above ELISPOT assays, we identified that the more specific responses. Our results demonstrated that the responses responses to VP2 in donor 5 were widely distributed on VP2, to VP1, VP3, and VP4 were mainly focused on the central region whereas those in donor 1 were dominated by the VP2–26 (192–209) of the Ag, although the C terminus of VP1 was also recognized in peptide (Fig. 4A, 4B). In both of the donors, we found that the some of the subjects (Fig. 3A, 3C, 3D). Interestingly, the responses T cells responsive to VP2 were mainly CD4+ T cells, although against VP4 among all of the responsive subjects were highly fo- mild CD8+ T cell responses were detected within region 153–196 cused on two overlapped peptides in a 23-aa central region (VP4 in both of the subjects. In particular, T cell responses against the (18-41)). Unlike the other three Ags, responses to VP2 were widely VP2–26 (192–209) peptide in both of the donors were completely distributed from the N- to the C terminus of the protein in the study CD4+ T cell dependent. subjects (Fig. 3B). The phenotype and functional properties of VP2-specific T cells Next, two-dimensional matrix peptide pools were designed to in donor 1 were further analyzed with intracellular cytokine determine the particular individual peptides responsible for the staining (ICS) assays by flow cytometry. ICS assays further con- positive responses in the pools, and promiscuous peptides were firmed that T cell responses to VP2 were mainly CD4+ T cell further determined by repeated ELISPOT assays using individual dependent. VP2-specific CD4+ T cells accounted for 95% of the peptides with the same T cell line. As a whole, 31 peptides con- IFN-g–secreting cells, whereas only 5% were CD8+ dependent taining T cell epitopes were identified in the four structural Ags, (Fig. 4C). In contrast, the VP2–26 (192–209) peptide-specific including two HLA-DR–restricted epitopes previously identified T cells were completely CD4+ T cell dependent. In addition, by other groups (Table II) (27). The most abundant peptides (13 two major CD4+ T cell functional cytokines, IFN-g and IL-2, were peptides) were defined within the VP2 Ag, indicating that VP2 analyzed to characterize the functional properties of CD4+ T cells had the highest epitope density (5.4 responsive peptides per 100 at the individual peptide level. The results demonstrated that the Downloaded from aa) compared with VP1, VP3, and VP4 (3, 2.8, and 2.9 responsive VP2–26 (192–209) peptide-specific T cells were mainly single peptides per 100 aa, respectively). Distribution of the immuno- IFN-g–producing T cells (5.05% of the total CD4+ T cells), with genic peptides on the four structural Ags was presented together a low level of polyfunctional T cells simultaneously secreting with the identified B cell epitopes from previous literatures IFN-g and IL-2 (0.14% of the total CD4+ T cells) (Fig. 4D). (Supplemental Fig. 1). Given all of the above data, the uncommon

Less efficient cross-reactivity against moderately variant http://www.jimmunol.org/ diversely distributed T cell responses spanning the entire Ag in- peptides from CV-A16 dicated a dominant role for VP2 in T cell responses against EV71 via points: 1) a higher quantity of T cell epitopes relative to the EV71 is most genetically relatedtoCV-A16,whichisanother other Ags; and 2) more epitopes with different HLA restrictions, major causative pathogen of HFMD; the amino acid identity of which may cover a broader population. all of the structural Ags is ∼78% between the two (40). + Moreover, EV71 comprises three major genotypes, and each ge- Single IFN-g–secreting CD4 T cell responses against VP2 notype contains various subgenogroups circulating in different Subsequently, we characterized the phenotypes and functional regions around the world (41). Thus, we performed conservancy properties of VP2-specific T cells. ELISPOT assays using CD4+-or analysis of the identified peptides between EV71 and CV-A16 CD8+-depleted PBMCs with anti-CD4 or anti-CD8 Ab-conjugated viruses and among different EV71 genotypes. Of the 31 con- by guest on October 1, 2021 magnetic beads were performed to determine the phenotypes of firmed peptides containing EV71 T cell epitope regions, 29 dis- the T cell responses of the in vitro expanded VP2-specific poly- played varied amino acid substitutions compared with CV-A16 clonal T cell lines from two representative donors, donor 1 and (with an average variation of 2.8 aa per peptide) (Table II). We

FIGURE 3. Diverse distribution of VP2-specific T cell immunogenic regions. The T cell responses were specified to a region of 40–50 aa for VP1 (A), VP2 (B), VP3 (C), and VP4 (D). Four or five peptides were mixed together to constitute a peptide pool, which represented the responses of a specific region of the Ag. The magnitude of responses is represented as SFCs per 105 PBMCs on the y-axis, whereas the peptide spanning region and subjects are represented on the x-andz-axes, respectively. Unlike the highly focused distribution of responses to the central region in VP1, VP3, and VP4, the responses to VP2 were more diversely distributed along the Ag (B). 1642 T CELL IMMUNITY AGAINST EV71 IN ADULTS

Table II. Summary of the identified immunogenic peptides

Recognition Source Ag Peptides Amino Acid Location EV71 CV-A16 Equivalent Regiona Frequencyb VP1 9 VP1–11(75–90) TAETTLDSFFSRAGLVc –Q––AIGN–––––––– 1/7 VP1–16(112–129) DITGYAQMRRKVELFTYM –LM––––––––C–––––– 4/7 VP1–17(120–137) RRKVELFTYMRFDAEFTF –––C–––––––––––––– 2/7 VP1–19(137–154) FVACTPTGGVVPQLLQYM ––VAK–N––L–––––––– 1/7 VP1–20(145–162)d EVVPQLLQYMFVPPGAPK –L––––––––Y––––––– 1/7 VP1–23(169–187) AWQTATNPSVFVKLSDPPA –––––––––I––––T–––– 1/5 VP1–33(244–261)d KYPLVVRIYMRMKHVRAW PHSITL–V–––I–––––– 2/5 VP1–34(252–267) YMRMKHVRAWIPRPMR –––I––––––––––L– 1/5 VP1–37(271–285) YLFKANPNYAGNSIK ––––T–––K–––D–– 2/5 VP2 13 VP2–3(14–31) AQLTIGNSTITTQEAANI –––––––––––––––––– 3/7 VP2–5(29–47) ANIIVGYGEWPSYCSDDDA ––––IA–––––E—K–A––– 1/7 VP2–9(62–79) RFYTLDTKLWEKSSKGWY ––F–––––S–A–D––––– 2/7 VP2–10(70–86) LWEKSSKGWYWKFPDVL S–A–D–––––––––––– 2/7 VP2–11(77–92) GWYWKFPDVLTETGVF ––––––––––––V––– 1/7 VP2–12(83–100) PDVLTETGVFGQNAQFHY ––––––V––––––––––– 2/7 VP2–15(107–224) CIHVQCNASKFHQGALLV –V–––––––––––––––– 1/7 VP2–16(115–132) SKFHQGALLVAVLPEYVI –––––––––––I–––––L 2/7 VP2–19(139–155) TGTEDTHPPYKQTQPGA D–N–NS––––VT––––Q 1/6 VP2–21(152–169) PGADGFELQHPYVLDAGI ––––––––TN–––––––– 1/6 Downloaded from VP2–24(176–193)d TVCPHQWINLRTNNCATI –––––––––––––––––– 3/6 VP2–25(184–201) NLRTNNCATIIVPYINAL ––––––––––––––M–TV 1/6 VP2–26(192–209) TIIVPYINALPFDSALNH ––––––M–TV–––––––– 3/6 VP3 7 VP3–6(41–58) RNLLELCQVETILEVNNV –––––I–R–––––––––L 1/2 VP3–7(49–66) VETILEVNNVPTNATSLM –––––––––LQS–E–TP– 1/2 VP3–14(99–115) TMLGQLCGYYTQWSGSL –I–––––R––––––––– 1/2 TFMFTGSFMATGKMLIAY ––––A–––I–––––––––

vp3–17(118–135) 1/5 http://www.jimmunol.org/ VP3–18(126–143) MATGKMLIAYTPPGGPLP I––––––––––––––GV– 5/5 VP3–20(142–169) LPKDRATAMLGTHVIWDF V–A—L––––––––––––– 1/5 vp3–23(173–190) SSVTLVIPWISNTHYRAH ––––––V––––––––––– 1/5 VP4 2 VP4(18–34) SATEGSTINYTTINYYK ––S–––––––––––––– 4/7 VP4(29–41) INYTTINYYKDSYAATA –––––––––––A–––S– 7/7 aThe dashes represent residues that are identical to those in EV71, whereas residues in CV-A16 that differ from those in EV71 are shown. bRecognition frequency of the EV71 peptides was detected in some of the study subjects. Numbers of total studied subjects and peptide-positive subjects were presented as denominator and numerator, respectively. cThese three peptides have been determined to be promiscuous HLA-DR–restricted peptides (27, 29). dResidues in EV71 that are different from that in CV-A16 are presented in underlined boldface. The accession number of the CV-A16 strain compared

in the table is National Center for Biotechnology Information RefSeq NP_042242.1. by guest on October 1, 2021 then analyzed the variation levels of the immunogenic peptides whereas no cross-reactivity was detected in donor 5 who showed and the proportions of responses against peptides with different weak responses against VP2–26-EV71C. EC50 of the responses levels of variation among the detected subjects. The results against VP2–26-EV71C (0.038 mg/ml) in donor 1 was lower demonstrated that peptides with limited variations (0–3 aa) take than VP2–26-EV71A/B (0.0608 mg/ml), whereas highest EC50 over the major portion of the identified immunogenic peptides, (0.127mg/ml) was detected against VP2–26-CV16, which indi- which constitute ∼80% of all (Fig. 5A). As expected, the peptides cated the less efficient cross-priming of T cell responses by the were more conserved among EV71 genotypes than compared with variant peptide from CV-A16. Responses in donor 25 showed that in CV-A16, with only one or few amino acid variations in similar lower EC50 (0.065 mg/ml) against VP2–26-EV71C than 39% (12 of 31) of the peptides (Supplemental Table II). VP2–26-CV16 (0.214 mg/ml). Although neutralizing Ab titers Subsequently, cross-reactivity against variant peptides derived specific to EV71 and CV-A16 were both detected in these donors, from different EV71 genotypes and CV-A16 was studied in three our findings implied that T cells existed in the peripheral blood individuals against a representative peptide with moderate varia- were likely elicited from natural infection by EV71 genotype C tions, the VP2–26 (192–209). The VP2–26 (192–209) peptide of strains, which is in accordance with the EV71 C4–dominated epi- the VP2 Ag was highly conserved in genotype C (VP2–26-EV71C), demics in mainland China (38, 41). All these results indicated that including the most prevalent C2 and C4 strains, whereas moderate cross-reactivities against the moderately variant peptides from CV- variant of two or three amino acids was found in EV71 genotypes A16 broadly existed among different subjects, although they may be A/B (VP2–26-EV71A/B) and in CV-A16 strains (VP2–26-CV16), less efficient in priming T cell responses than the variant peptides respectively (Fig. 5B). IFN-g ELISPOT assays were performed derived from EV71. along a gradient peptide concentration to mimic in vivo Ag con- centrations and minimize possible false-positive cross-reactivity in Limited influences to EV71-responsive cellular immunity derived from vaccination of polioviruses the presence of high peptide concentrations. Besides, EC50 also was calculated to determine the sensitivity of cross-reactivity against Poliovirus was another member of Enterovirus genus and polio- variant peptides. Cross-reactivity against the variant peptides could virus vaccine strains shared ∼42% identity in capsid Ags with that be detected though the responses differed among different sub- of EV71. Programmed vaccination of poliovirus vaccines was jects (Fig. 5B). Moderate to strong cross-reactivity against variant carried out national wide among children in China, and most of VP2–26-EV71A/B and VP2–26-CV16 was detected in donor 1 and the subjects enrolled in this study claimed to have received polio- donor 25 who showed robust responses against VP2–26-EV71C, virus vaccination previously. Thus, we next sought to investigate the The Journal of Immunology 1643

FIGURE 4. Phenotype and functional characteriza- tion of VP2-specific T cells. The CD4/CD8 T cell de- pendency of VP2-specific T cell responses was determined by CD4/CD8-depleted ELISPOT analysis (A, B)andICS assays (C). The responses of CD4- or CD8-depleted or untreated T cells from donor 1 are designated as D1_CD4, D1_CD8, and D1_PBMC (respectively) and are represented with different shaped bars (A); the responses in donor 5 are likewise represented (B). “M” is the negative control cultured with medium alone, and Downloaded from “PHA” is the positive control cultured with PHA. In ICS analysis with IFN-g, the phenotypic dependency of VP2-specific responses was further confirmed to be CD4-dominated (upper), although low frequency CD8 T cells could be detected (59), whereas responses to the immunodominant peptide VP2–26 (192–209) were completely CD4+ T cell dependent (C). Functional http://www.jimmunol.org/ properties of VP2–26 (192–209)-specific T cells were determined through staining of IFN-g (y-axis) and IL-2 (x-axis). The T cell populations with different cytokine combination properties were gated on CD4+ T cells and represented in different quadrants on the plot (D). For VP2–26 (192–209) peptide–specific T cells, dominant single IFN-g–secreting T cells are shown in the upper left quadrant, whereas low-frequency dual cytokine–

secreting T cells are shown in the lower right quadrant. by guest on October 1, 2021 Numbers in the quadrants of each plot represent the frequency of positive T cells in each quadrant. Cells cultured with medium alone or PHA were used as negative or positive controls, respectively. These results are representative of three independent experiments.

influence of T cell responses derived from poliovirus vaccination subjects were cultured in the presence of VP2–26-EV71C or VP2– to the EV71-responsive T cells identified in the current study. 26-Polio, separately, and then the expanded polyclonal T cells were We first analyzed variation levels of the corresponding peptides in detected against VP2–26-EV71C, VP2–26-CV16, and VP2–26- polioviruses compared with the immunogenic peptides identified Polio. The results showed that T cells derived from VP2–26-EV71C from EV71 among the detected subjects. All of the immunogenic stimulation could be responsive to variant peptides VP2–26-EV71C peptides identified in EV71 contained more than three amino acid and VP2–26-CV16 but failed to response against VP2–26-Polio variations in all the three poliovirus vaccine strains, and more than (Fig. 6B). Besides, no responses against any of the variant peptides 80% of the peptides were highly varied with 5–16 aa variations and could be detected from VP2–26-Polio–stimulated T cells (Fig. 6C). an average of 8.9 aa variations within each peptide (Fig. 6A). Cross- Taken together, T cell responses against EV71 were dominated reactivity of EV71-responsive T cells against polioviruses were by immunogenic peptides, which were highly varied from that of further determined in three subjects with the representative peptide, polioviruses, and thus, the influences derived from previous po- VP2–26 (192–209), which had 10 aa variations in polioviruses liovirus vaccination to EV71-responsive cellular immunity would (VP2–26-Polio, TLVLPYVNSLSIDSMVKH). PBMCs from the be limited. 1644 T CELL IMMUNITY AGAINST EV71 IN ADULTS

of T cells responsive to EV71 was detected in a small portion of the subjects (23%; 7 of 30) via direct ex vivo detection, broad EV71-responsive cellular responses were detected in most of the subjects (93%) after in vitro expansion, which is in accordance with the high prevalence of neutralizing Abs detected in the subjects. Our findings suggest that a low frequency of EV71- responsive memory T cells circulating in the peripheral blood of most people can react in the presence of Ag and the presence of these T cells might assist the control of recurrent EV71 infection, considering that the virus persistently circulated in China since 2008 (38). Previous studies of cellular immunity to enterovirus mostly pay greater attention to the VP1 Ag (48–50). Nonetheless, a study of the immunohierarchy of conserved peptides from four structural Ags from enteroviruses, including Poliovirus 1, coxsackie B3, B5, B6, and A9, and Echovirus 30, on a small cohort of subjects revealed that cross-reactive T cell epitopes are mainly located in VP2andVP3(andtoalesserextentinVP1)(51).ForEV71, investigation of T cell immunity has also mainly focused on VP1, and several HLA-class II–restricted T cell epitopes have been Downloaded from determined within this Ag (27, 29). Moreover, a collection of studies demonstrate that VP1 is the dominant target of neutral- izing Abs in EV71, and thus, VP1 has been broadly explored for subunit vaccine development, priming strong humoral immunity in a mouse model (15, 52, 53). However, compared with an N-terminal–biased neutralizing determinants distribution in http://www.jimmunol.org/ humans, distribution of neutralizing Ab epitopes in mouse model reveals no N- or C-terminal bias, and numbers of epitopes are identified in the C-terminal of VP1 in mouse model (52, 54, 55). FIGURE 5. Conservancy analysis and determination of cross-reactivity Besides, the pathology of EV71 infection in mice and humans is between EV71 and CV-A16. (A) The conservation of the 31 identified immu- nogenic peptides in our study was analyzed between EV71 and CV-A16. also quite different (56). Thus, the immunity and protection ef- Each colored sector of the pie shows proportion to the number of peptides ficiency of subunit vaccines in the mouse model cannot be with the same variation level from all the immunogenic peptides. All the analogized to that in humans. Thus, Ag candidates for an effec- peptides were classified to three groups according to the number of vari- tive vaccine should be rich in both B and T cell epitopes. Al- by guest on October 1, 2021 able amino acids in each peptide, low group (0–1 varied amino acids), though VP1 is the major target of neutralizing Abs, our present moderate group (2–4 varied amino acids), and high group ($5 varied amino study demonstrated that T cell responses were dominated by acids). A majority of the peptides belong to the moderate and low group the VP2 Ag. Wei et al. (57) also identified a highly conserved B peptides. ( ) Cross-reactivity of the typical peptide VP2–26 (192–209), dominant T cell epitope in VP2 Ag from an in silico–predicted which represents the peptides with moderate number (2, 3) of mutated peptide pool which is included in the identified peptides in the amino acids from EV71 C genotype (VP2–26-EV71C), genotype A/B current study (VP2–24[176–193] peptide) (Table II) (29). In the (VP2–26-EV71A/B), or CV-A16 (VP2–26-CV16) was determined along a serial peptide concentration (x-axis) among three subjects (D1, D5, and current study, we demonstrated that the T cell responses against D25). Varied amino acids in VP2–26-EV71A/B or VP2–26-CV16 are the dominant T cell Ag of EV71 (i.e., VP2) were also mainly + highlighted in bold. Cells cultured with medium alone were enrolled as CD4 T cell dependent, although determining the roles that these negative control (NC). Each symbol represented the average spots from CD4+ T cells played in immunity against EV71 requires further duplicate wells, and the results are representative of two independent investigation. The incomparable lower CD8+ T cells detected in experiments. the current study might also be related with the less efficient expansion of CD8+ T cells against the overlapping peptides than CD4+ T cells (33). Furthermore, functional profiling with IL-2 Discussion and IFN-g revealed that the responding EV71-responsive T cell T cells are crucial both in the generation and maintenance of long- responses were dominated by single IFN-g–secreting cells, lived humoral immunity as well as to reduce or eliminate viral which may be correlated with the acute viral infection and high replication postinfection (42, 43). Previous studies demonstrate that viral load with quick disease progression characteristics of pri- subneutralizing Abs may be correlated with Ab-dependent en- mary EV71 infection and recurrent epidemics (57). Nonetheless, hancement in EV71 infection (44–46). Thus, T cells that are es- a complete understanding of the protection efficiency of VP2- sential in the selection and maturation of high-affinity Abs deserve specific T cells during natural infection and vaccination requires more attention for immunological investigation, and a compre- additional research in the future. hensive knowledge of EV71 T cell immunity would benefit our A major concern in the current study might be that whether the understanding of host protection mechanisms against EV71 (47). T cell responses detected were elicited from nature EV71 exposure, In the current study, we completed (for what, to our knowledge, or cross-primed by other enterovirus infection or polioviruses we believe to be the first time) a comprehensive investigation of vaccination, especially considering that a proportion of study T cell responses to EV71. Using a panel of 110 overlapping pep- subjects showed neutralizing titers to both EV71 and CV-A16, and tides spanning all the four structural Ags of EV71 allowed us to coinfection with EV71 and CV-A16 has been identified in previous study the profile of the existing T cell epitopes in 30 healthy adults studies (39, 58). Serial analyses were performed to elucidate this in a non–HLA-restricted manner. Although only a low frequency question based on the following findings. First, the neutralizing The Journal of Immunology 1645 Downloaded from http://www.jimmunol.org/

FIGURE 6. Analysis of conservancy and cross-reactivity between EV71 and poliovirus vaccine strains. (A) Conservancy of the 31 identified im- munogenic peptides was analyzed between EV71 and three widely used vaccine-related polioviruses, Poliovirus 1 strain Sabin (GenBank accession number CAA24465.1), Poliovirus 2 strain Sabin (GenBank accession number CAA25246.1), and Poliovirus 3 strain Sabin (UniProtKB/Swiss-Prot: Q84792; http://www.uniprot.org/uniprot/Q84792). As in Fig. 5A, proportion to the number of peptides with the same variation level from all the im- munogenic peptides was displayed as colored sector diagram. Peptides were classified into three groups with different levels of variations accordingtothe number of variable amino acids the peptide, moderate (2–4 aa variation), high (5–10 aa variations), and high+ (.10 aa variations). A majority of the peptides have high level of variation. (B and C) Cross-reactivity against the representative variant peptides of VP2–26 (192–209) from EV71 C genotype (VP2–26-EV71C), CV-A16 (VP2–26-CV16), and polioviruses (VP2–26-Polio) were detected in three subjects (D1, D5, and D25). PBMCs from the B C subjects were cultured under the stimulation of VP2–26-EV71C ( ) or its variant peptide VP2–26-Polio from polioviruses ( ), respectively. After the by guest on October 1, 2021 culturing, the expanded polyclonal T cells were detected using ELISPOT against variant VP2–26-EV71C, VP2–26-CV16, or VP2–26-Polio, respectively. Cells cultured with medium alone or PHA were used as negative or positive controls, respectively, and the results are representative of two independent experiments.

Abs detected in the study subjects showed EV71 superior responses dren despite the programmed nationwide poliovirus vaccination to CV-A16, which ensured that the investigation of EV71-specific also indicated that poliovirus vaccination contributed little to anti- T cell responses among these subjects was reasonable and reliable. EV71 immunity (31). However, further exploration is still needed Second, conservancy analysis of the immunogenic peptides iden- to determine the contribution of cross-reactive T cell responses tified in the current study demonstrated that peptides with mod- elicited from natural infection of EV71 or other enteroviruses to erate variations of two to four amino acid differences compared the control of disease. with CV-A16 constituted the largest proportion of the immuno- In conclusion, this is, to our knowledge, the first comprehensive genic peptides against EV71 and dominated the responses among study of EV71-responsive cellular immunity comparing the profile the study population. In addition, determination of the cross-re- of all of the structural Ags. We demonstrated that EV71-responsive activity from a representative peptide in CV-A16 suggested that T cells broadly exist in healthy adults, and our findings concerning cross-reactivity might have broadly existed among the population, the predominant immunogenicity of the VP2 Ag provide signifi- although the variant peptides from CV-A16 were less efficient in cant implications for the understanding of EV71 cellular immunity. priming T cell responses than that from EV71. The partially cross- Furthermore, the broadly existed less efficient cross-reactivity recognized variant peptides from CV-A16 might provide partial against CV-A16 and limited influences from poliovirus vacci- protection against CV-A16 and contribute to a less prevalence of nation to EV71-responsive cellular immunity would contribute CV-A16 among population (31). Third, although cross-reactivity to the understanding of immune protection mechanisms against of variant peptides with three amino acids from polioviruses could different enteroviruses. be detected from previous study (29), we found that peptides with moderate variations of three or four amino acids occupies ,20% Acknowledgments of all the immunogenic peptides from EV71. Furthermore, highly We thank Dr. Jianhua Zhang and Tong Zhao from the Institute of Micro- varied peptides with an average of 8.9 aa differences take over biology, Chinese Academy of Sciences, for technical assistance in flow .80% of the immunogenic peptides and dominate the responses cytometry analysis. against EV71. Thus, the influences of previous poliovirus vacci- nation to EV71-responsive cellular immunity would be limited. Disclosures Besides, millions of EV71 infections and complications in chil- The authors have no financial conflicts of interest. 1646 T CELL IMMUNITY AGAINST EV71 IN ADULTS

References 28. Dong, C., L. Liu, H. Zhao, J. Wang, Y. Liao, X. Zhang, R. Na, Y. Liang, L. Wang, and Q. Li. 2011. Immunoprotection elicited by an enterovirus type 71 experi- 1. Hagiwara, A., I. Tagaya, and T. Yoneyama. 1978. Epidemic of hand, foot and mental inactivated vaccine in mice and rhesus monkeys. Vaccine 29: 6269–6275. mouth disease associated with enterovirus 71 infection. Intervirology 9: 60–63. 29. Wei, R., C. Yang, M. Zeng, F. Terry, K. Zhu, C. Yang, R. Altmeyer, W. Martin, 2. Ho, M., E. R. Chen, K. H. Hsu, S. J. Twu, K. T. Chen, S. F. Tsai, J. R. Wang, and A. S. De Groot, and Q. Leng. 2012. A dominant EV71-specific CD4+ T cell S. R. Shih; Enterovirus Epidemic Working Group. 1999. An epidemic of epitope is highly conserved among human enteroviruses. PLoS One 7: e51957. enterovirus 71 infection in Taiwan. N. Engl. J. Med. 341: 929–935. 30. Liu, J., S. Zhang, S. Tan, Y. Yi, B. Wu, B. Cao, F. Zhu, C. Wang, H. Wang, J. Qi, ´ ´ ´ 3. Nagy, G., S. Takatsy, E. Kukan, I. Mihaly, and I. Do¨mo¨k. 1982. Virological and G. F. Gao. 2012. Cross-allele cytotoxic T lymphocyte responses against diagnosis of enterovirus type 71 infections: experiences gained during an epi- 2009 pandemic H1N1 influenza A virus among HLA-A24 and HLA-A3 demic of acute CNS diseases in Hungary in 1978. Arch. Virol. 71: 217–227. supertype-positive individuals. J. Virol. 86: 13281–13294. 4. Gilbert, G. L., K. E. Dickson, M. J. Waters, M. L. Kennett, S. A. Land, and 31. Zhu, Z., S. L. Zhu, X. B. Guo, J. T. Wang, D. Y. Wang, D. M. Yan, X. J. Tan, M. Sneddon. 1988. Outbreak of enterovirus 71 infection in Victoria, Australia, L. Y. Tang, H. Zhu, Z. H. Yang, et al. 2010. Retrospective seroepidemiology with a high incidence of neurologic involvement. Pediatr. Infect. Dis. J. 7: 484– indicated that human enterovirus 71 and coxsackievirus A16 circulated wildly in 488. central and southern China before large-scale outbreaks from 2008. Virol. J. 7: 5. Samuda, G. M., W. K. Chang, C. Y. Yeung, and P. S. Tang. 1987. Monoplegia 300. caused by Enterovirus 71: an outbreak in Hong Kong. Pediatr. Infect. Dis. J. 6: 32. Luo, S. T., P. S. Chiang, A. S. Chao, G. Y. Liou, R. Lin, T. Y. Lin, and M. S. Lee. 206–208. 2009. Enterovirus 71 maternal antibodies in infants, Taiwan. Emerg. Infect. Dis. 6. Fujimoto, T., M. Chikahira, S. Yoshida, H. Ebira, A. Hasegawa, A. Totsuka, and 15: 581–584. O. Nishio. 2002. Outbreak of central nervous system disease associated with 33. Draenert, R., M. Altfeld, C. Brander, N. Basgoz, C. Corcoran, A. G. Wurcel, hand, foot, and mouth disease in Japan during the summer of 2000: detection and D. R. Stone, S. A. Kalams, A. Trocha, M. M. Addo, et al. 2003. Comparison of molecular epidemiology of enterovirus 71. Microbiol. Immunol. 46: 621–627. overlapping peptide sets for detection of antiviral CD8 and CD4 T cell 7. Singh, S., V. T. Chow, M. C. Phoon, K. P. Chan, and C. L. Poh. 2002. Direct responses. J. Immunol. Methods 275: 19–29. detection of enterovirus 71 (EV71) in clinical specimens from a hand, foot, and 34. Liu, J., Y. Sun, J. Qi, F. Chu, H. Wu, F. Gao, T. Li, J. Yan, and G. F. Gao. 2010. mouth disease outbreak in Singapore by reverse transcription-PCR with uni- The membrane protein of severe acute respiratory syndrome coronavirus acts as versal enterovirus and EV71-specific primers. J. Clin. Microbiol. 40: 2823–2827. a dominant immunogen revealed by a clustering region of novel functionally and 8. Chang, L. Y., C. C. King, K. H. Hsu, H. C. Ning, K. C. Tsao, C. C. Li, structurally defined cytotoxic T-lymphocyte epitopes. J. Infect. Dis. 202: 1171– Y. C. Huang, S. R. Shih, S. T. Chiou, P. Y. Chen, et al. 2002. Risk factors of 1180. Downloaded from enterovirus 71 infection and associated hand, foot, and mouth disease/ 35. Liu, J., P. Wu, F. Gao, J. Qi, A. Kawana-Tachikawa, J. Xie, C. J. Vavricka, herpangina in children during an epidemic in Taiwan. Pediatrics 109: e88. A. Iwamoto, T. Li, and G. F. Gao. 2010. Novel immunodominant peptide pre- 9. Zhang, Y., Z. Zhu, W. Yang, J. Ren, X. Tan, Y. Wang, N. Mao, S. Xu, S. Zhu, sentation strategy: a featured HLA-A*2402‑restricted cytotoxic T-lymphocyte A. Cui, et al. 2010. An emerging recombinant human enterovirus 71 responsible epitope stabilized by intrachain hydrogen bonds from severe acute respiratory for the 2008 outbreak of hand foot and mouth disease in Fuyang city of China. syndrome coronavirus nucleocapsid protein. J. Virol. 84: 11849–11857. Virol. J. 7: 94. 36. Topham, D. J., R. A. Tripp, and P. C. Doherty. 1997. CD8+ T cells clear influenza 10. Seiff, A. 2012. Cambodia unravels cause of mystery illness. Lancet 380: 206. virus by perforin or Fas-dependent processes. J. Immunol. 159: 5197–5200. 11. Qiu, J. 2008. Enterovirus 71 infection: a new threat to global public health? 37. Lee, L. Y. H., L. A. Ha, C. Simmons, M. D. de Jong, N. V. V. Chau, http://www.jimmunol.org/ Lancet Neurol. 7: 868–869. R. Schumacher, Y. C. Peng, A. J. McMichael, J. J. Farrar, G. L. Smith, et al. 12. Yi, L., J. Lu, H. F. Kung, and M. L. He. 2011. The virology and developments 2008. Memory T cells established by seasonal human influenza A infection toward control of human enterovirus 71. Crit. Rev. Microbiol. 37: 313–327. cross-react with avian influenza A (H5N1) in healthy individuals. J. Clin. Invest. 13. Wang, X., W. Peng, J. Ren, Z. Hu, J. Xu, Z. Lou, X. Li, W. Yin, X. Shen, 118: 3478–3490. C. Porta, et al. 2012. A sensor-adaptor mechanism for enterovirus uncoating 38. Tan, X., X. Huang, S. Zhu, H. Chen, Q. Yu, H. Wang, X. Huo, J. Zhou, Y. Wu, from structures of EV71. Nat. Struct. Mol. Biol. 19: 424–429. D. Yan, et al. 2011. The persistent circulation of enterovirus 71 in People’s 14. Chou, A. H., C. C. Liu, C. P. Chang, M. S. Guo, S. Y. Hsieh, W. H. Yang, Republic of China: causing emerging nationwide epidemics since 2008. PLoS H. J. Chao, C. L. Wu, J. L. Huang, M. S. Lee, et al. 2012. Pilot scale production ONE 6: e25662. of highly efficacious and stable enterovirus 71 vaccine candidates. PLoS One 7: 39. Yan, X. F., S. Gao, J. F. Xia, R. Ye, H. Yu, and J. E. Long. 2012. Epidemic e34834. characteristics of hand, foot, and mouth disease in Shanghai from 2009 to 2010: 15. Wu, C. N., Y. C. Lin, C. Fann, N. S. Liao, S. R. Shih, and M. S. Ho. 2001. Pro- enterovirus 71 subgenotype C4 as the primary causative agent and a high inci- tection against lethal enterovirus 71 infection in newborn mice by passive immu- dence of mixed infections with coxsackievirus A16. Scand. J. Infect. Dis. 44: by guest on October 1, 2021 nization with subunit VP1 vaccines and inactivated virus. Vaccine 20: 895–904. 297–305. 16. Xu, J., S. Wang, W. Gan, W. Zhang, L. Ju, Z. Huang, and S. Lu. 2012. Expression 40. Brown, B. A., and M. A. Pallansch. 1995. Complete nucleotide sequence of and immunogenicity of novel subunit enterovirus 71 VP1 antigens. Biochem. enterovirus 71 is distinct from poliovirus. Virus Res. 39: 195–205. Biophys. Res. Commun. 420: 755–761. 41. Solomon, T., P. Lewthwaite, D. Perera, M. J. Cardosa, P. McMinn, and 17. Ch’ng, W. C., E. J. Stanbridge, K. C. Ong, K. T. Wong, K. Yusoff, and N. Shafee. M. H. Ooi. 2010. Virology, epidemiology, pathogenesis, and control of entero- 2011. Partial protection against enterovirus 71 (EV71) infection in a mouse virus 71. Lancet Infect. Dis. 10: 778–790. model immunized with recombinant Newcastle disease virus capsids displaying 42. Vinuesa, C. G., S. G. Tangye, B. Moser, and C. R. Mackay. 2005. Follicular B the EV71 VP1 fragment. J. Med. Virol. 83: 1783–1791. helper T cells in antibody responses and autoimmunity. Nat. Rev. Immunol. 5: 18. Chang, L. Y., K. C. Tsao, S. H. Hsia, S. R. Shih, C. G. Huang, W. K. Chan, 853–865. K. H. Hsu, T. Y. Fang, Y. C. Huang, and T. Y. Lin. 2004. Transmission and 43. Pantaleo, G., and A. Harari. 2006. Functional signatures in antiviral T-cell im- clinical features of enterovirus 71 infections in household contacts in Taiwan. munity for monitoring virus-associated diseases. Nat. Rev. Immunol. 6: 417–423. JAMA 291: 222–227. 44. Wang, S. M., I. C. Chen, L. Y. Su, K. J. Huang, H. Y. Lei, and C. C. Liu. 2010. 19. Yang, C., C. Deng, J. Wan, L. Zhu, and Q. Leng. 2011. Neutralizing antibody Enterovirus 71 infection of monocytes with antibody-dependent enhancement. response in the patients with hand, foot and mouth disease to enterovirus 71 and Clin. Vaccine Immunol. 17: 1517–1523. its clinical implications. Virol. J. 8: 306. 45. Wang, S. M., T. S. Ho, H. C. Lin, H. Y. Lei, J. R. Wang, and C. C. Liu. 2012. 20. Yang,K.D.,M.Y.Yang,C.C.Li,S.F.Lin,M.C.Chong,C.L.Wang,R.F.Chen, Reemerging of enterovirus 71 in Taiwan: the age impact on disease severity. Eur. and T. Y. Lin. 2001. Altered cellular but not humoral reactions in children with J. Clin. Microbiol. Infect. Dis. 31: 1219–1224. complicated enterovirus 71 infections in Taiwan. J. Infect. Dis. 183: 850–856. 46. Han, J. F., R. Y. Cao, Y. Q. Deng, X. Tian, T. Jiang, E. D. Qin, and C. F. Qin. 21. Chang, L. Y., I. S. Chang, W. J. Chen, Y. C. Huang, G. W. Chen, S. R. Shih, 2011. Antibody dependent enhancement infection of enterovirus 71 in vitro and J. L. Juang, H. M. Shih, C. A. Hsiung, T. Y. Lin, and L. M. Huang. 2008. HLA- in vivo. Virol. J. 8: 106. A33 is associated with susceptibility to enterovirus 71 infection. Pediatrics 122: 47. Schwickert, T. A., G. D. Victora, D. R. Fooksman, A. O. Kamphorst, M. R. Mugnier, 1271–1276. A. D. Gitlin, M. L. Dustin, and M. C. Nussenzweig. 2011. A dynamic T cell- 22. Chang, L. Y., C. A. Hsiung, C. Y. Lu, T. Y. Lin, F. Y. Huang, Y. H. Lai, limited checkpoint regulates affinity-dependent B cell entry into the germinal Y. P. Chiang, B. L. Chiang, C. Y. Lee, and L. M. Huang. 2006. Status of cellular center. J. Exp. Med. 208: 1243–1252. rather than humoral immunity is correlated with clinical outcome of enterovirus 48. Kutubuddin, M., J. Simons, and M. Chow. 1992. Identification of T-helper 71. Pediatr. Res. 60: 466–471. epitopes in the VP1 capsid protein of poliovirus. J. Virol. 66: 3042–3047. 23. Liu, M. L., Y. P. Lee, Y. F. Wang, H. Y. Lei, C. C. Liu, S. M. Wang, I. J. Su, 49. Mahon, B. P., K. Katrak, and K. H. Mills. 1992. Antigenic sequences of po- J. R. Wang, T. M. Yeh, S. H. Chen, and C. K. Yu. 2005. Type I interferons protect liovirus recognized by T cells: serotype-specific epitopes on VP1 and VP3 and mice against enterovirus 71 infection. J. Gen. Virol. 86: 3263–3269. cross-reactive epitopes on VP4 defined by using CD4+ T-cell clones. J. Virol. 66: 24. Lin, Y. W., K. C. Chang, C. M. Kao, S. P. Chang, Y. Y. Tung, and S. H. Chen. 7012–7020. 2009. Lymphocyte and antibody responses reduce enterovirus 71 lethality in 50. Simons, J., M. Kutubuddin, and M. Chow. 1993. Characterization of poliovirus- mice by decreasing tissue viral loads. J. Virol. 83: 6477–6483. specific T lymphocytes in the peripheral blood of Sabin-vaccinated humans. J. 25. Lin, Y. W., S. W. Wang, Y. Y. Tung, and S. H. Chen. 2009. Enterovirus 71 in- Virol. 67: 1262–1268. fection of human dendritic cells. Exp. Biol. Med. 234: 1166–1173. 51. Cello, J., O. Strannega˚rd, and B. Svennerholm. 1996. A study of the cellular 26. Shen, F. H., C. C. Tsai, L. C. Wang, K. C. Chang, Y. Y. Tung, I. J. Su, and immune response to enteroviruses in humans: identification of cross-reactive S. H. Chen. 2013. Enterovirus 71 infection increases expression of interferon- ‑ T cell epitopes on the structural proteins of enteroviruses. J. Gen. Virol. 77: g inducible protein 10 which protects mice by reducing viral burden in multiple 2097–2108. tissues. J. Gen. Virol. 94: 1019–1027. 52. Tan, C. S., and M. J. Cardosa. 2007. High-titred neutralizing antibodies to human 27. Foo, D. G., P. A. Macary, S. Alonso, and C. L. Poh. 2008. Identification of enterovirus 71 preferentially bind to the N-terminal portion of the capsid protein human CD4 T-cell epitopes on the VP1 capsid protein of enterovirus 71. Viral VP1. Arch. Virol. 152: 1069–1073. Immunol. 21: 215–224. The Journal of Immunology 1647

53.Sivasamugham,L.A.,M.J.Cardosa,W.S.Tan,andK.Yusoff.2006. neurotropism, causing neurological manifestations in a novel mouse model of Recombinant Newcastle Disease virus capsids displaying enterovirus 71 EV71 infection. J. Virol. 86: 2121–2131. VP1 fragment induce a strong immune response in rabbits. J. Med. Virol. 78: 57. Jiang, M., D. Wei, W. L. Ou, K. X. Li, D. Z. Luo, Y. Q. Li, E. Chen, and 1096–1104. G. M. Nong. 2012. Autopsy findings in children with hand, foot, and mouth 54. Foo, D. G., S. Alonso, M. C. Phoon, N. P. Ramachandran, V. T. Chow, and disease. N. Engl. J. Med. 367: 91–92. C. L. Poh. 2007. Identification of neutralizing linear epitopes from the 58. Lin, Y., K. Wen, Y. Pan, Y. Wang, X. Che, and B. Wang. 2011. Cross-reactivity of VP1 capsid protein of Enterovirus 71 using synthetic peptides. Virus Res. anti-EV71 IgM and neutralizing antibody in series sera of patients infected with 125: 61–68. Enterovirus 71 and Coxsackievirus A 16. J. Immunoassay Immunochem. 32: 233–243. 55. Man-Li, T., M. Szyporta, L. X. Fang, and J. Kwang. 2012. Identification and 59. Joosten, S. A., K. E. van Meijgaarden, P. C. van Weeren, F. Kazi, A. Geluk, characterization of a monoclonal antibody recognizing the linear epitope N. D. Savage, J. W. Drijfhout, D. R. Flower, W. A. Hanekom, M. R. Klein, and RVADVI on VP1 protein of enterovirus 71. J. Med. Virol. 84: 1620–1627. T. H. Ottenhoff. 2010. Mycobacterium tuberculosis peptides presented by HLA- 56. Khong, W. X., B. Yan, H. Yeo, E. L. Tan, J. J. Lee, J. K. Ng, V. T. Chow, and E molecules are targets for human CD8 T-cells with cytotoxic as well as regu- S. Alonso. 2012. A non-mouse-adapted enterovirus 71 (EV71) strain exhibits latory activity. PLoS Pathog. 6: e1000782. Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021