Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the Epstein-Barr Virus Proteome

Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the... Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the Epstein-Barr Virus Proteome Zhiwei Liu, Anna E Coghill, Ruth M Pfeiffer, Carla Proietti, Wan-Lun Hsu, Yin-Chu Chien, Lea Lekieffre, Lutz Krause, Kelly J Yu, Pei-Jen Lou, Cheng-Ping Wang, Jason Mulvenna, Jaap M Middeldorp, Jeff Bethony, Chien-Jen Chen, Denise L Doolan, Allan Hildesheim Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 The Journal of Infectious Diseases MAJOR ARTICLE Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the Epstein-Barr Virus Proteome 1,a 1,a 1 3,4 5,6 5,8 3 3 1 7 Zhiwei Liu, Anna E. Coghill, Ruth M. Pfeiffer, Carla Proietti, Wan-Lun Hsu, Yin-Chu Chien, Lea Lekieffre, Lutz Krause, Kelly J. Yu, Pei-Jen Lou, 7 3 9 2 5,6 3,4,b 1,b Cheng-Ping Wang, Jason Mulvenna, Jaap M. Middeldorp, Jeff Bethony, Chien-Jen Chen, Denise L. Doolan, and Allan Hildesheim 1 2 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, Maryland; Department of Microbiology, Immunology, and Tropical Medicine, George 3 4 Washington University Medical Center, Washington, D. C.; QIMR Berghofer Medical Research Institute, Brisbane, and Centre for Biosecurity and Tropical Infectious Diseases, 5 6 15 Australian Institute of Tropical Health and Medicine, James Cook University, Cairns, Australia; Genomics Research Center, Academia Sinica, Graduate Institute of Epidemiology and Prevention Medicine, College of Public Health, National Taiwan University, and Department of Otolaryngology, National Taiwan University Hospital and College of Medicine, 8 9 Taipei, and National Institute of Cancer Research, National Health Research Institute, Miaoli, Taiwan; and Department of Pathology, VU University Medical Center, Amsterdam, June Netherlands Background. Little is known about variation in antibody responses targeting the full spectrum of Epstein-Barr virus (EBV) proteins and how such patterns inform disease risk. Methods. We used a microarray to measure immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses against 199 EBV protein sequences from 5 EBV strains recovered from 289 healthy adults from Taiwan. We described positivity patterns, estimated the correlation between antibodies, and investigated the associations between environmental and genetic risk factors and variations in antibody responses. Results. Healthy adults were more likely to mount IgG antibody responses to EBV proteins (median positivity frequency, 46.5% –46 for IgG and 17.3% for IgA; P  =  1.6 ×  10 , by the Wilcoxon rank sum test). Responses against glycoproteins were particularly prevalent. The correlations between antibody responses of the same class were higher than correlations across classes. The mucosal exposure to proteins involved in EBV reactivation (as determined by the IgA response) was associated with smoking (P = .002, by the sequence kernel association test–combined), and approximately one quarter of adults displayed antibody responses associated with EBV-related cancer risk. Conclusions. es Th e data comprehensively define the variability in human IgG and IgA antibody responses to the EBV proteome. Patterns observed can serve as the foundation for elucidating which individuals are at highest risk of EBV-associated clinical condi- tions and for identifying targets for effective immunodiagnostic tests. Keywords. Antibody microarray, antibody response, Epstein-Barr virus, Taiwan, IgA, IgG Epstein-Barr virus (EBV) is a ubiquitous herpesvirus that infects infected individuals; such diseases include hematopoietic >90% of individuals worldwide [1]. Primary infection with the malignancies (eg, Burkitt and Hodgkin lymphomas) and epi- virus typically occurs in early childhood [2, 3], aer w ft hich EBV thelial malignancies (eg, gastric cancer and nasopharyngeal establishes lifelong latency in human B cells. During the course carcinoma [NPC]) [4]. EBV is etiologically involved in approx- of lifelong infection, EBV periodically reactivates in B cells or imately 200 000 cancers worldwide each year [5]. Despite its epithelial cells, expressing a broader set of lytic proteins required ubiquity and association with various chronic conditions, there for viral replication [4]. This ongoing replication in epithelial are currently no proven ways to prevent EBV infection from cells of the oral cavity contributes to shedding of the virus into occurring or to predict with sufficient accuracy which infected saliva, facilitating person-to-person transmission of EBV. individuals will develop serious chronic diseases associated Although established EBV infection is generally asymp- with this virus [5, 6]. tomatic, infection manifests as clinical disease in a subset of One way to monitor patterns of EBV exposure in the general population is to measure levels of circulating anti-EBV antibod- Received 11 January 2018; editorial decision 26 February 2018; accepted 1 March 2018; ies that reflect responses to viral proteins. For example, immu - published online March 2, 2018. noglobulin G (IgG) antibodies appear within several weeks aer ft Z. L. and A. E. C. contributed equally to this work. primary infection and remain in circulation for many years [7]. D. L. D. and A. H. contributed equally to this work. Correspondence: Z. Liu, PhD, Infections and Immunoepidemiology Branch, Division of Cancer Prior studies aimed at defining rates of EBV infection by age Epidemiology and Genetics, National Cancer Institute, 9609 Medical Center Dr, Rockville, MD across populations have used positivity for IgG antibodies tar- 20850 (zhiwei.liu@nih.gov). The Journal of Infectious Diseases 2018;217:1923–31 geting EBV proteins such as the viral capsid antigen (VCA) and Published by Oxford University Press for the Infectious Diseases Society of America 2018. EBV nuclear antigen 1 (EBNA1) [8] to classify individuals as This work is written by (a) US Government employee(s) and is in the public domain in the US. EBV positive (ie, as ever exposed to EBV). IgG antibodies can DOI: 10.1093/infdis/jiy122 Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1923 Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 also be used to better understand the nature of an individual’s Aer p ft rinting, anti-polyHis probing confirmed the presence of EBV infection. For example, a subset of long-term EBV carri- 99.5% of sequences; anti-HA probes confirmed the presence of ers displays elevated levels of IgG antibody targeting the early 98.5% of sequences. High coverage was achieved across the 5 antigen protein (EA) [9]. This EA complex is expressed as part EBV strains, with 97% of the predicted sequences from each of the EBV lytic cycle; therefore, elevated levels of EA indicate strain represented on the microarray at ≥99% homology. ongoing viral lytic activity (reactivation) and a potential lack of A positive antibody response in a healthy adult was defined control over the virus [10]. as a response greater than that individual’s person-specific back - In contrast to IgG, immunoglobulin A  (IgA) responses ground (mean of 4 no-DNA-control spots plus 1.5 standard devi- to viral exposure are believed to reflect recent reactivation ations) [17]. As previously reported [23], we included blinded at mucosal sites (eg, the oral epithelium) [11]. This type of quality control replicates during testing and observed good immune response is of interest because of the oral route of EBV reproducibility for antibodies measured using this custom pro- transmission. During primary infection, IgA responses against tein microarray. The average percentage agreement for classifying EBV in the oral cavity can be strong but are short lived [12]. an individual’s antibody response as positive versus negative was e d Th urability and spectrum of EBV-directed IgA responses in high for both IgA (84%; range, 52%–100%) and IgG (79%; range, healthy adults are largely uncharacterized. However, it has been 48%–100%), respectively. e Th antibody level (determined on the shown that individuals at greatest risk of developing NPC, an basis of the standardized signal intensity output from the array) EBV-associated cancer, have elevated anti-EBV IgA responses was further grouped into quartiles (4 categories) based on the to VCA and EBNA1 years before a cancer diagnosis [8, 13, 14]. population distribution of the response for each given antibody. Although informative, prior studies of humoral responses to Based on coefficients of variation reported from the previous EBV infection [15, 16] have largely focused on measuring levels study [23], we excluded from analysis the 2 IgG array spots out of of IgG and IgA antibodies to a handful (<10) of the approxi- 199 evaluated that had coefficients of variation of >20% (BZLF1/ mately 90 EBV protein targets [8]. With the advent of protein IgG [array protein sequence CAA24861.1-103155-102655] and array systems capable of screening for antibody responses BXRF1/IgG [array protein sequence CAA24798.1-144860- against the full complement of EBV proteins [17–19], it is now 145606]). Finally, we included 3 synthetic EBV peptides that are feasible to evaluate humoral responses to EBV infection in a putative cancer biomarkers (VCAp18, EBNA1, and EAdp47) on more comprehensive manner. Thus, we evaluated IgG and IgA the microarray, bringing the total number of anti-EBV antibod- anti-EBV antibody responses against 199 predicted sequences ies for analysis to 402 (200 IgG and 202 IgA antibodies). and 3 synthetic antigens from 86 EBV proteins involved in Ethics Statement various stages of the EBV life cycle (eg, latent and lytic cycles) All human subjects included were adults, and all studies were among 289 healthy individuals from Taiwan, a region with high approved by appropriate human subjects committees in Taiwan rates of EBV-associated cancer. and the United States. This study was reviewed/approved by the National Cancer Institute Special Studies Institutional METHODS Review Board and the National Taiwan University Institutional A total of 289 healthy, asymptomatic adults who served as Review Board)\. Written informed consent was obtained for all controls for 2 previously published Taiwanese studies were participants. selected for this report [20, 21]. Details are described in the Supplementary Materials. A  previously described EBV pro- Statistical Analysis tein microarray targeting IgG and IgA antibodies against 199 Visualization and statistical analyses were conducted using R sta- predicted EBV protein sequences (86 EBV proteins) from 5 tistical sowa ft re. Details on the statistical analyses are described EBV strains (AG876, Akata, B95-8, Mutu, and Raji) was used in the Supplementary Materials. Briefly, to pool data across the [17, 18, 22, 23]. Briefly, these 199 sequences represent nonre - 2 study populations, we used population-specific cut points to dundant open reading frames from 5 EBV strains representing generate similar underlying antibody distributions and then both African (Mutu, Raji, and AG876) and Asian (Akata) vari- compared individual response rates (ie, percentage positivity ants, as well as known splice variants identified in the litera - values) across immunoglobulin class and EBV life cycle. We esti- ture. Eighty-five percent of 199 predicted sequences represented mated the correlation between antibodies across immunoglobu- complete transcripts of EBV genes. The remaining 30 predicted lin classes (ie, IgG against protein X versus IgA against protein sequences represented linear segments from 8 EBV genes X) and within each immunoglobulin class (ie, IgG against pro- >1000  bp long. Each of the protein sequences cloned into the tein X versus IgG against protein Y), using Spearman correla- pXT7 expression vector prior to printing onto the microarray tion coefficients. All individuals were clustered according to the included N-terminal 10x histidine (His) and C-terminal hem- positive responses across the full EBV proteome, using unsuper- agglutinin (HA) tags to confirm expression on the microarray. vised hierarchical clustering with binary distance and complete 1924 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 linkage [24]. To examine the association between environmen- e in Th dividual with the narrowest pattern responded to 38 EBV tal and genetic risk factors (ie, smoking, alcohol drinking, and targets on the array (28 IgG antibodies and 10 IgA antibodies), HLA alleles) and biological groupings of antibody response (eg, representing 21 EBV proteins (Figure  1A). The individual with responses directed against a select EBV life cycle), we used the the broadest response had a positive response against 329 EBV sequence kernel association test–combined (SKAT-C) from the targets on the array (163 IgG antibodies and 166 IgA antibodies), single-nucleotide polymorphism set [25], using binary variables representing 84 of 86 EBV proteins evaluated. Positive IgG (ie, (with a positive antibody response assigned a value of 1 and a long-term) responses against EBV proteins were markedly more negative response assigned a value of 0). frequent than positive IgA (ie, more-recent) responses (Table 1). On average, individuals mounted IgG antibody responses RESULTS against 93 (46.5%) of 200 EBV targets on the array and IgA anti- Anti-EBV IgG and IgA Antibody Responses Mounted by Healthy body responses against 35 (17.3%) of 202 EBV targets. Individuals Additional evidence of broad anti-EBV IgG reactivity was All individuals had measurable (ie, positive) antibody responses provided by the observation that almost all individuals (≥95%) against EBV (Figures  1A–C; Supplementary Figures  1–10). in our study mounted a positive IgG response against 14 IgG IgA Subjects BC Subjects Positive for IgG Antibody Subjects Positive for IgA Antibody BZLF2:CAA24860.1–102116–101445 BZLF2:YP_001129466.1–90630–89959 100% 96% BRRF2:YP_001129470.1–94844–96457 BZLF2:CAA24860.1–102116–101445 100% 94% BDLF3:AFY97964.1–118644–117940 BRRF2:AFY97943.1–93884–95497 100% 93% BRRF2:YP_001129470.1–94844–96457 100% BRRF2:AFY97943.1–93884–95497 93% BSLF1:YP_001129457.1–74727–72103 BFRF3 (VCA_p18):YP_001129448.1–49335–49865 83% 100% BFRF3 (VCA_p18):YP_001129448.1–49335–49865 BILF1:YP_001129506.1–154125–153187 100% 80% BFRF3 (VCA_p18):AFY97924.1–49199–49729 BZLF2:YP_001129466.1–90630–89959 100% 80% BDLF4:YP_001129488.1–117560–116883 BDLF3:AFY97964.1–118644–117940 100% 79% BFRF3 (VCA_p18):AFY97924.1–49199–49729 BLRF2 (VCA_p23):YP_001129461.1–76771–77259 99% 77% BLRF2 (VCA_p23):YP_001129461.1–76771–77259 BFRF3 (VCA_p18):CAA24838.1–61507–62037 99% 70% BFRF3 (VCA_p18):CAA24838.1–61507–62037 BDLF4:YP_001129488.1–117560–116883 99% 70% BBRF3:YP_001129479.1–107679–108896 BFRF1:YP_001129446.1–46719–47729 98% 65% VCA_p18:synthetic antigen EBNA3B:YP_001129464.1–83509–86532–1 98% 65% BMRF2:YP_001129455.1–68964–70037 98% EBNA3B:YP_001129464.1–83509–86532–1 65% EBNA3A:AFY97915.1–80252–82747 97% BMRF2:YP_001129455.1–68964–70037 64% EBNA3A:YP_401669.1–80382–82877 97% EBNA1:synthetic antigen 64% BPFL1:CAA24839.1–71527–62078–2 62% VCA_p18:synthetic antigen 96% BDLF3:YP_001129490.1–119605–118901 BBRF3:YP_001129479.1–107679–108896 96% 61% BGLF5 (ALK.EXONUCLEASE):YP_001129480.1–109516–109289 BDLF3:YP_001129490.1–119605–118901 96% 60% BGLF5 (ALK.EXONUCLEASE):AFY97956.1–108555–108328 LMP2B:AFY97910.1–1026–1196 94% 55% BcLF1 (VCA_p160):YP_001129493.1–126005–121860–2 EBNA–LP:AFY97917.1–35572–35676 10% 2% EBNA–LP:YP_001129440.1–35558–35662 BKRF4:YP–001129474.1–99676–100329 9% 2% BALF5 (DNA polymerase):YP_001129507.1–157772–154725–1 CAPSID:YP_001129451.1–63084–64178 8% 2% BGLF3:YP_001129483.1–112496–112035 8% BDLF3:YP_001129489.1–117772–117539 2% EBNA1:AFY97842.1–95349–97142 8% BALF5 (DNA polymerase):AFY97894.1–154809–155094 1% 8% BBRF2:YP_001129477.1–104490–105326 1% BcLF1 (VCA_p160):YP_001129493.1–126005–121860–2 7% EBNA–LP:AFY97917.1–35572–35676 1% BALF2 (EA(D)_p138):YP_001129510.1–165796–162410–1 7% EBNA3C:AFY97856.1–85691–86050 1% BNLF2B:AFY97989.1–166696–166400 BALF5 (DNA polymerase):CAA24805.1–156746–153699–1 BGLF5 (DNAse):YP_001129481.1–110883–109471 6% 1% FGAM:YP_001129438.1–1736–5692–1 BZLF1 (Zta):YP_001129467.1–91697–91197 6% 1% BNLF2B:CAA24811.1–167303–166998 LF2:YP_001129504.1–151808–150519 6% 1% BORF2:YP_001129452.1–64253–66733 FGAM:YP_001129438.1–1736–5692–1 6% 0% BcLF1 (VCA_p160):AFY97883.1–124729–120584–2 BGLF4 (protein kinase):CAA24828.1–123692–122328 4% 0% EBNA3C:AFY97856.1–86125–88794 BALF5 (DNA polymerase):CAA24805.1–156746–153699–1 4% 0% LMP1:YP_001129515.1–170457–170190 4% BRFR1A:AFY97836.1–46138–46545 0% 4% EAdp47:synthetic antigen 0% BFRF1A:YP_001129445.1–46352–46759 3% BFLF2:YP_001129443.1–44763–43807 0% BBRF2:YP_001129477.1–104490–105326 2% BGRF1/BDRF1:YP_001129485.1–117754–118890 0% BGRF1/BDRF1:YP_001129485.1–117754–118890 2% BFRF1A:AFY97921.1–46216–46623 0% BFRF1A:AFY97921.1–46216–46623 BGLF4 (protein kinase):CAA24828.1–123692–122328 0% BZLF1 (Zta):YP_001129467.1–90855–90724 1% Subjects Subjects Most/Least Prevalent IgG Responses Most/Least Prevalent IgA Responses Figure 1. Anti–Epstein-Barr virus (EBV) immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses among healthy individuals in Taiwan. A, Number of pos- itive antibody responses in 289 healthy individuals. Blue indicates IgG antibody positivity, and red indicates IgA positivity. The individual with the narrowest response had antibodies against 38 EBV targets on the array (28 IgG antibodies and 10 IgA antibodies), representing 21 EBV proteins. In contrast, the individual with the broadest response had antibodies against 329 EBV targets (163 IgG antibodies and 166 IgA antibodies), representing 84 of 86 tested EBV proteins. B and C, The 20 IgG (B) and IgA (C) antibodies with the highest antibody positivity rates and the 20 with the lowest antibody positivity rates. Percentages indicate the antibody positivity rate for the given target antigen in this population. Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1925 Number of Positive Antibody Responses (Sum of IgG and IgA Antibody Responses) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Table  1. Number of Positive Antibody Responses to Epstein-Barr Virus of broad, long-term exposure to proteins from all stages in the (EBV) Proteins Among Healthy Individuals in Taiwan, by Antibody Type EBV life cycle among healthy adults. and Stage of EBV Life Cycle Targeted IgA antibody responses against EBV glycoproteins (median a IgA positivity frequency, 42.1%; Table  1) was also more fre- Antibody, Target Protein Antibodies, No. Median (% ) IQR Range quently observed than responses against proteins from other IgG stages of the EBV life cycle. The median positivity frequen - Overall 200 93 (46.5) 67–132 22–188 Latent 58 30 (51.7) 21–40 21–58 cies observed for IgA antibodies across proteins from latent Immediate early lytic 11 4 (36.4) 2–7 0–11 and lytic phases of the EBV life cycle were similar (range, Early lytic 44 20 (45.5) 13–28 5–42 15.6%–17.2%; Table  1), with 1 notable exception: IgA anti- Late lytic 47 20 (42.6) 14–30 7–45 body responses against EBV proteins involved in the switch Glycoprotein 19 13 (68.4) 11–16 4–19 from latency to lytic phases were particularly infrequent, with Other/unknown 21 7 (33.3) 4–12 0–20 a median positivity frequency of 0% (0 of 12; range, 0%–75%). IgA Overall 202 35 (17.3) 18–85 4–166 This is particularly interesting because IgA responses to some Latent 58 10 (17.2) 4–20 0–48 of these proteins have been shown to be predictive of the risk of Immediate early lytic 12 0 (0.0) 0–1 0–9 the EBV-associated malignancy NPC [8]. Early lytic 45 7 (15.6) 3–15 0–35 Late lytic 47 8 (17.0) 6–14 0–39 Clustering of Antibody Responses by Immunoglobulin Class Glycoprotein 19 8 (42.1) 3–15 0–19 Consistent with the observation that IgG responses to EBV were Other/unknown 21 1 (4.8) 0–6 0–17 more frequently observed than IgA responses, we noted that Abbreviations: IgA, immunoglobulin A; IgG, immunoglobulin G; IQR, interquartile range. IgG responses to a particular antigenic target had, on average, a Calculated on the basis of the median number of positive antibodies among the total number of antibodies evaluated in each group. higher correlation with IgG responses to other antigenic targets, rather than with IgA responses against the same target (median Spearman correlation, 0.495 [range, −0.093–0.978]; 85.9% different EBV proteins, including 2 latent proteins (EBNA3B of comparisons had a Bonferroni-corrected P value of  <  .05; and EBNA1), 4 early lytic proteins (BILF1, BDLF4, BLRF2, Figures  2A and B and Supplementary Table  3). Practically, and BGLF5), 4 late lytic proteins (BRRF2, BFRF3, VCA-p18, this meant that an individual with elevated IgG titers against and BPFL1), and 4 glycoproteins (BZLF2, BDLF3, BBRF3, and a specific EBV protein (eg, protein X) was more likely to have BMRF2; Figure  1B and Supplementary Table 1). This IgG pat - elevated IgG titers against another EBV protein (eg, protein Y), tern stands in contrast to that of IgA, for which only 1 target rather than elevated IgA responses to protein X. The same phe - on the array (representing BZLF2) elicited a uniformly positive nomenon was evident for IgA responses, with IgA responses response (Figure 1C; Supplementary Table 1), with a plurality of against a particular target being more strongly correlated with study participants (40.5% [117]) positive for <25 array targets. IgA responses to other antigenic targets (median Spearman Only 8 EBV targets were weakly immunogenic at eliciting IgG correlation, 0.366 [range, −0.293–0.904]; 64.5% of comparisons responses (ie, <5% of individuals tested positive), including 1 had a Bonferroni-corrected P value of < .05), rather than with targeting a latent protein (EBNA3C), 1 early lytic protein (EAD- IgG responses to the same antigen. p47: BMRF1), and 4 late lytic proteins (BGRF1/BDRF1, BFLF2, While correlations tended to be higher within rather than BRFR1A, and BcLF1; Figure  1B and Supplementary Table  2). across antibody class, interimmunoglobulin correlations (ie, This contrasts with 33 EBV proteins that were weakly immuno - IgG vs IgA correlations) were higher when considering the same genic for IgA (Figure 1C; Supplementary Table 2). protein target (median Spearman correlation, 0.226; 23.0% of correlations had a Bonferroni-corrected P value of  <  .05), Antibody Responses Against EBV Proteins Expressed at Different Stages rather than different EBV proteins (median Spearman correla - of the EBV Life Cycle tion, 0.127 [range, −0.334–0.563]; 1.9% of the correlations had Although IgG responses were more common than IgA responses, a Bonferroni-corrected P value of < .05). Similar patterns were the biological function of the target protein was also an important observed across all EBV life cycle stages (Supplementary Table 3 determinant of the host antibody response (Table 1). IgG against and Supplementary Figures 11A–E). EBV glycoproteins (median IgG positivity frequency, 68.4%) was more frequently observed than IgG targeting proteins from Clustering of Healthy, Asymptomatic Adults According to Anti-EBV other stages of the EBV life cycle. Robust IgG responses to EBV Antibody Responses latent proteins and proteins involved in the switch from latent In addition to understanding the breadth of the antibody to lytic infection were also observed (median positivity frequen- response marking EBV exposure in healthy adults, hierarchical cies, 51.7% and 45.5%, respectively). Responses to EBV proteins clustering identified 4 distinct patterns among study partici - involved in immediate early lytic proteins (median positivity fre- pants that could be distinguished by their average number of quency, 36.4%) were less widespread but still frequent, evidence 1926 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 AB Antibody type IgA IgG EBV life cycle Latent Early lytic Glycoprotein Immediate early lytic Late lytic Other/unknown –0.5 0.0 0.5 1.0 Spearman correlation Comparison IgA IgG Spearman correlation IgA antibody vs IgA antibody IgA antibody vs IgA antibody (di erent target) –1 –0.5 0 0.5 1 IgG antibody vs IgA antibody (same target) IgG antibody vs IgG antibody Correlation between the average IgA and IgG antibody response Interantibody correlations according to immunoglobulin type Figure 2. Spearman correlation between the average immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses in this population. A, Unsupervised hierar- chical clustering of IgG and IgA antibody responses, based on the correlation between the average antibody output for any 2 given Epstein-Barr virus (EBV) targets (Spearman correlation coefficient). Red depicts a strong positive correlation, and blue indicates a negative correlation. B, Distribution of Spearman correlation coefficients according to the type of comparison being made (eg, between 2 IgA antibodies or between IgG and IgA antibodies targeting a selected antigen on the array). Each row/column represents the average antibody response to a given EBV target in the study population. Each histogram plots the set of average antibody responses illustrated in panel A. positive IgG and IgA antibody responses (Figures  3A and B). IgG and IgA responses observed among individuals in group We observed 4 groups: group 1 comprised participants with 4 were broadly comparable to those observed among NPC low IgG and IgA responses, group 2 comprised those with a cases (Figure  3B). When we estimated a cancer risk score for low IgG response and an elevated IgA response, group 3 com- asymptomatic individuals in our study by using our previously prised those with an intermediate IgG response and a low IgA reported 14-antibody NPC risk stratification signature [ 23], response, and group 4 comprised those with high IgG and IgA 79% of individuals in group 4 had scores that ranked in the responses. Group  1 participants (9.0% [26]) mounted 46 IgG highest quartile of cancer risk, compared with only 7% of indi- -14 2 and 15 IgA antibody responses, on average. Group  4 partici- viduals in groups 1–3 (P  =  2.2 ×  10 , by the test; Table  2). pants (26.0% [75]) mounted 152 IgG and 69 IgA responses, on es Th e findings suggest that healthy individuals can be classified average. Interestingly, although IgA responses were less frequent as having distinct patterns of EBV responses and that a propor- than IgG responses overall, group 2 and 4 participants (39% tion of healthy individuals are characterized by high IgG and [114]) displayed a high level of IgA reactivity, indicating that IgA responses typically observed among individuals at highest ongoing EBV exposure at mucosal surfaces is not an infrequent risk of NPC development. event in healthy, Taiwanese adults previously exposed to EBV. We further examined whether group 4 participants were more Smoking Is Associated With Differential EBV-Directed IgA Antibody likely to respond to targets from a specific stage of the EBV life Responses cycle and observed that these subjects had elevated responses e Th extent to which the effect of known NPC risk factors is to all stages (Supplementary Figures  12A and B). Notably, this mediated through their ability to modulate immune responses high response pattern included higher IgA antibody reactivity to EBV is not fully understood. We observed that individuals against immediate early lytic proteins (median number of pos- from group 4 also had the highest proportion of current smok- itivity  among individuals in this group, 1; interquartile range, ers, compared with other groups (72% vs 45%; P  =  0.02, by 0–2.5) compared to the other groups (median number of posi- the test). In addition, when considering the proteome-wide -7 tivity, 0; interquartile range, 0–0; P = 1.9 × 10 , by the rank test). antibody response, the total level of the IgA response but not We further compared the antibody pattern in group 4 to that the IgG response was significantly different between smokers previously described among patients with NPC (n = 175) [23]. and nonsmokers in our population (P = .002, by the SKAT-C; Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1927 IgA IgG Density Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 AB 0.8 0.4 0.0 #1 #2 #4 #3 Group 1 (26) Group 2 (39) Group 3 (149) Group 4 (75) NPC Group (175) 150 100 Group 1 (26) Group 2 (39) Group 3 (149) Group 4 (75) NPC Group (175) Clustering of Individuals on the Basis of the Average Number of Positive Antibody Responses Total Number of Positive Anti-EBV Antibody Responses in 4 Groups of Disease-Free, EBV-Infected Adults and Patients With NPC Figure 3. Clustering of individuals defined according to their immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody response. A, Dendrogram showing clusters of individuals defined according to the number of positive antibody responses (with a positive antibody response assigned a value of 1 and a negative response assigned a value of 0) across the full spectrum of evaluated IgG and IgA antibody responses, using unsupervised hierarchical clustering with binary distance and complete linkage. B, Box plot showing number of positive IgG/IgA responses according to 4 groups/clusters based on antibody response, compared with individuals with nasopharyngeal carcinoma (NPC) who had a blood sample collected at the time of cancer diagnosis. Group 1 comprised participants with low IgG and IgA responses, group 2 comprised those with a low IgG response and an elevated IgA response, group 3 comprised those with an intermediate IgG response and a low IgA response, and group 4 comprised those with high IgG and IgA responses. See Methods and Results for additional details. Figure  4A). Notably, when evaluating anti-EBV antibody proteins differed significantly by smoking behavior (Bonferroni- responses targeting proteins across different stages of the EBV corrected P value of < .05, by the SKAT-C). life cycle, the EBV-directed IgA response to latent and early lytic Further, in logistic regression models that evaluated the associa- tion between smoking and each marker individually, we observed Table  2. Nasopharyngeal Carcinoma (NPC) Risk Scores Among suggestive associations (P < .05) for multiple anti-EBV IgA anti- Controls From a Cancer Screening Population Cohort in Taiwan, by bodies from all stages of the EBV life cycle (37.9% [22 of 58] were Immunoglobulin A (IgA) and Immunoglobulin G (IgG) Responses associated with latent proteins, 8.3% [1 of 12] were associated with immediate early proteins, 20.0% [9 of  45] were associated with Control Group, Participants, No. (%) early lytic proteins, 6.4% [3 of 47] were associated with late lytic Group 1 Group 2 Group 3 Group 4 proteins, and 21.0% [4 of 19] were associated with glycoproteins; Risk Score (n = 16) (n = 21) (n = 49) (n = 28) Supplementary Tables 4–8). The strongest individual association Quartile 1 11 (68.7) 2 (9.5) 14 (28.6) 2 (7.1) Quartile 2 2 (12.5) 5 (23.8) 19 (38.8) 2 (7.1) was observed between smoking and IgA targeting LMP1(odds Quartile 3 2 (12.5) 11 (52.4) 14 (28.6) 2 (7.1) ratio, 4.64; P = .001; Figure 4B), an EBV oncoprotein expressed in Quartile 4 1 (6.2) 3 (14.3) 2 (4.1) 22 (78.6) NPC tumors that is normally considered a weakly immunogenic Estimated using our previously reported 14-antibody NPC risk stratification signature [ 23]. protein [26, 27]. Taken together, these findings suggest that smok - Quartile 1 denotes the lowest NPC risk, and quartile 4 denotes the highest NPC risk. b ing may alter the anti-EBV mucosal antibody response. Group 1 comprised participants with low IgG and IgA responses, group 2 comprised those with a low IgG response and an elevated IgA response, group 3 comprised those with an We also evaluated the association between anti-EBV antibody intermediate IgG response and a low IgA response, and group 4 comprised those with high IgG and IgA responses. See Methods and Results for additional details. responses, alcohol drinking, and HLA alleles that modulate 1928 • JID 2018:217 (15 June) • Liu et al Number of Positive Number of Positive Antibody Responses (IgA) Antibody Responses (IgG) Height Average Number of Positive Average Number of Positive Antibody Responses (IgA) Antibody Responses (IgG) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 A B Smoking 2.0 –5 P = 9.89 × 10 All IgA Latent Immediate Early Lytic 1.5 Early Lytic Late Lytic Glycoprotein 1.0 All IgG Latent Immediate Early Lytic 0.5 Early Lytic IgG Late Lytic IgA Glycoprotein 0.0 Never (42) Former (14) Current (60) 01 2 34 Smoking Status (Among Male Participants) –log10 (P ) Figure  4. Associations between anti–Epstein-Barr virus (EBV) antibody responses and smoking status among males. A, Association between anti-EBV immunoglobulin A (IgA) antibodies against proteins expressed during different EBV life cycles and smoking status (current vs former/never). P values were obtained from sequence kernel association test–C tests. The dashed line represents the Bonferroni-corrected statistically significant P value threshold of .005. B, Box plot showing the association between LMP1/IgA reactivity and smoking status. The P value was obtained from linear regression, adjusted for age. NPC risk (ie, HLA-A*0207 and HLA-A*1101; Supplementary in this study displayed antibody response patterns similar to Figures  13A–C). HLA-A*0207 was nominally associated with NPC cases, including positive responses reflecting exposure IgA responses against immediate early lytic proteins (P  =  .02). to immediate early antigenic targets, such as Zta and Rta. However, no statistically significant associations persisted aer ft Importantly, our data indicate that smoking is associated with correction for multiple comparisons. altered IgA antibody patterns that could reflect poor control over EBV lytic activity, suggesting a possible mechanism to DISCUSSION explain the role of smoking in the etiology of NPC. Our findings represent the first comprehensive evaluation of Previous studies have evaluated responses against a handful natural variation in the host antibody response to the full spec- of anti-EBV antibodies by using techniques including immuno- trum of EBV proteins. This unique characterization was made blot and microarray chips targeting 10–15 EBV proteins [28, 29]. possible through the use of protein microarray technology in Zheng et  al used a microarray to test 82 EBV open-reading a population of healthy, asymptomatic adults from Taiwan. frames in a study that included 10 nondiseased adults and Because the antibody repertoire can shed light on exposure showed that these adults mounted IgG responses against an to viral proteins, this virus-wide description of host response average of 17 proteins (at 1:100 serum dilution) and IgA anti- to the EBV proteome represents an important step forward bodies against an average of 13 proteins (at 1:1000 serum dilu- in understanding population-level variation that might have tion) [19]. This is in line with our findings of more frequent implications for multiple EBV-related diseases. IgG reactivity, rather than IgA reactivity. In the present study, We found that all adults evaluated in our study had evidence we expanded testing to 199 EBV open-reading frames among of exposure to EBV proteins, that these EBV-positive adults 289 asymptomatic adults. Because our array was not designed were more likely to mount IgG rather than IgA antibody to detect antibodies to conformational epitopes that require gly- responses to EBV, and that responses against glycoproteins cosylation, future studies will be required to better understand were particularly prevalent. If the IgG response targeting 1 EBV patterns of response to those epitopes and how they are related protein (eg, protein X) was elevated in an individual, they were to those reported herein. more likely to mount an IgG response targeting other antigens, Our finding that antibody responses to some glycoproteins rather than an IgA response against protein X, likely reflecting were the most prevalent is biologically plausible because these differential timing of exposure (longer-term exposure for IgG proteins are expressed on the viral surface and therefore readily and more-recent exposure for IgA) and separate triggering accessible as targets for the immune system [30, 31]. Given that mechanisms in different compartments (eg, oral mucosa vs IgA responses are thought to reflect recent pathogen exposure, systemic circulation). Approximately one quarter of the subjects our finding that IgA responses to glycoproteins were also more Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1929 Antibody Level (LMP1/IgA) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Notes frequently observed than responses against proteins from other stages of the EBV life cycle suggests that the presence of EBV in Financial support. This work was supported by the the oral cavity is not a rare event [32]. It was notable that indi- National Cancer Institute Intramural Research Program, the viduals in groups 2 and 4 (39% [114]) mounted a substantial, National Health and Medical Research Council of Australia, proteome-wide IgA response. This supports data from studies and the National Science Council of Taiwan. reporting EBV viral load shedding in the saliva of the majority Potential coni fl cts of interest. J. M. M. received payments as of healthy adults measured [32, 33], as well as EBV reactivation owner and chief executive officer of Cyto-Barr. All other authors rates that are much more frequent than those for other herpes- report no potential conflicts. All authors have submitted the viruses, such as cytomegalovirus [34]. ICMJE Form for Disclosure of Potential Conflicts of Interest. An important public health implication of our data is that Conflicts that the editors consider relevant to the content of the smoking, a modifiable behavior, was associated with elevated manuscript have been disclosed. IgA responses against EBV. An association between smoking and EBV reactivation has been posited by one study, which References evaluated VCA/IgA in individuals from southern China [35]. In 1. Kieff E, Rickinson AB. Epstein-Barr virus and its replication. the present study, we expand beyond VCA/IgA alone to demon- In Fields virology. 5th ed. ed. Philadelphia, PA: Lippincott strate large-scale differences in the EBV-directed IgA repertoire Williams & Wilkins, 2007. in smokers. The most significant smoking association was 2. Kangro HO, Osman HK, Lau YL, Heath RB, Yeung CY, Ng observed for IgA responses against LMP1, an important NPC MH. Seroprevalence of antibodies to human herpesviruses oncogene that has cell-transforming ability in rodent b fi roblasts in England and Hong Kong. J Med Virol 1994; 43:91–6. [36]. In addition to these individual antibody associations, we 3. Chen CY, Huang KY, Shen JH, Tsao KC, Huang YC. A large- observed that individuals in group 4 also had the highest pro- scale seroprevalence of Epstein-Barr virus in Taiwan. PLoS portion of current smokers. Taken together, these findings sug - One 2015; 10:e0115836. gest that smoking may negatively impacts an individual’s ability 4. Young LS, Rickinson AB. Epstein-Barr virus: 40  years on. to control EBV infection, leading to an increased frequency of Nat Rev Cancer 2004; 4:757–68. viral reactivation/exposure in the host. However, we did not 5. Cohen JI, Fauci AS, Varmus H, Nabel GJ. Epstein-Barr assess the degree to which the established association between virus: an important vaccine target for cancer prevention. smoking and NPC is mediated by interindividual variation in Sci Transl Med 2011; 3:107fs7. the immune response to this ubiquitous virus. The underlying 6. Kanekiyo M, Bu W, Joyce MG, et al. Rational design of an biological mechanisms merit further investigation. Epstein-Barr virus vaccine targeting the receptor-binding In conclusion, we described the pattern of the IgG and IgA site. Cell 2015; 162:1090–100. responses to EBV among asymptomatic, healthy adults, using a 7. Hjalgrim H, Friborg J, Melbye M. The epidemiology of protein microarray chip targeting 199 EBV protein sequences. EBV and its association with malignant disease. In: Arvin Given that the generation of antibodies is a complex process A, Campadelli-Fiume G, Mocarski E, et  al., eds. Human requiring interplay between viral antigens and numerous com- herpesviruses: biology, therapy, and immunoprophylaxis. partments of the immune system [37], our study represents Cambridge: Cambridge University Press, 2007. an initial but important step in understanding the immune 8. Coghill AE, Hildesheim A. Epstein-Barr virus antibodies response to EBV infection. In the future, our approach could be and the risk of associated malignancies: review of the litera- applied to other populations to characterize the similarities and ture. Am J Epidemiol 2014; 180:687–95. differences observed in the humoral response to EBV in indi - 9. Melbye M, Ebbesen P, Levine PH, Bennike T. Early primary viduals predisposed to different EBV-associated conditions (eg, infection and high Epstein-Barr virus antibody titers in sub-Saharan Africans at higher risk of Burkitt lymphoma). Such Greenland Eskimos at high risk for nasopharyngeal carci- studies could elucidate subsets of EBV-infected individuals at noma. Int J Cancer 1984; 34:619–23. highest risk of EBV-associated clinical conditions and help iden- 10. Henle W, Henle G. Epstein-Barr virus-specific serology in tify targets for the development of effective immunodiagnostics. immunologically compromised individuals. Cancer Res Supplementary Data 1981; 41:4222–5. Supplementary materials are available at e Th Journal of Infectious 11. Brandtzaeg P. Potential of nasopharynx-associated lym- Diseases online. Consisting of data provided by the authors to phoid tissue for vaccine responses in the airways. Am J benefit the reader, the posted materials are not copyedited and Respir Crit Care Med 2011; 183:1595–604. are the sole responsibility of the authors, so questions or com- 12. Bhaduri-McIntosh S, Landry ML, Nikiforow S, Rotenberg ments should be addressed to the corresponding author. M, El-Guindy A, Miller G. Serum IgA antibodies to 1930 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Epstein-Barr virus (EBV) early lytic antigens are present in of rare and common variants. Am J Hum Genet 2013; primary EBV infection. J Infect Dis 2007; 195:483–92. 92:841–53. 13. Chien YC, Chen JY, Liu MY, et  al. Serologic markers of 26. Dawson CW, Port RJ, Young LS. The role of the EBV- Epstein-Barr virus infection and nasopharyngeal carci- encoded latent membrane proteins LMP1 and LMP2 in the noma in Taiwanese men. N Engl J Med 2001; 345:1877–82. pathogenesis of nasopharyngeal carcinoma (NPC). Semin 14. Cao SM, Liu Z, Jia WH, et al. Fluctuations of Epstein-Barr Cancer Biol 2012; 22:144–53. virus serological antibodies and risk for nasopharyngeal 27. Meij P, Vervoort MB, Aarbiou J, et  al. Restricted low-level carcinoma: a prospective screening study with a 20-year human antibody responses against Epstein-Barr virus (EBV)- follow-up. PLoS One 2011; 6:e19100. encoded latent membrane protein 1 in a subgroup of patients 15. Henle W, Henle G, Zajac BA, Pearson G, Waubke R, Scriba with EBV-associated diseases. J Infect Dis 1999; 179:1108–15. M. Differential reactivity of human serums with early 28. Fachiroh J, Schouten T, Hariwiyanto B, et  al. Molecular antigens induced by Epstein-Barr virus. Science 1970; diversity of Epstein-Barr virus IgG and IgA antibody 169:188–90. responses in nasopharyngeal carcinoma: a comparison of 16. Ginsburg CM, Henle W, Henle G, Horwitz CA. Infectious Indonesian, Chinese, and European subjects. J Infect Dis mononucleosis in children. Evaluation of Epstein-Barr 2004; 190:53–62. virus-specific serological data. JAMA 1977; 237:781–5. 29. Loebel M, Eckey M, Sotzny F, et al. Serological profiling of 17. Doolan DL. Plasmodium immunomics. Int J Parasitol 2011; the EBV immune response in Chronic Fatigue Syndrome 41:3–20. using a peptide microarray. PLoS One 2017; 12:e0179124. 18. Davies DH, Duffy P, Bodmer JL, Felgner PL, Doolan DL. 30. Hutt-Fletcher LM. EBV glycoproteins: where are we now? Large screen approaches to identify novel malaria vaccine Future Virol 2015; 10:1155–62. candidates. Vaccine 2015; 33:7496–505. 31. Li Q, Turk SM, Hutt-Fletcher LM. The Epstein-Barr virus 19. Zheng D, Wan J, Cho YG, et  al. Comparison of humoral (EBV) BZLF2 gene product associates with the gH and gL immune responses to Epstein-Barr virus and Kaposi’s homologs of EBV and carries an epitope critical to infection sarcoma-associated herpesvirus using a viral proteome of B cells but not of epithelial cells. J Virol 1995; 69:3987–94. microarray. J Infect Dis 2011; 204:1683–91. 32. Hadinoto V, Shapiro M, Sun CC, Thorley-Lawson DA. The 20. Chen CJ, Yang HI, Su J, et al.; REVEAL-HBV Study Group. dynamics of EBV shedding implicate a central role for epi- Risk of hepatocellular carcinoma across a biological gra- thelial cells in amplifying viral output. PLoS Pathog 2009; dient of serum hepatitis B virus DNA level. JAMA 2006; 5:e1000496. 295:65–73. 33. Yao QY, Rickinson AB, Epstein MA. A re-examination of 21. Hildesheim A, Apple RJ, Chen CJ, et al. Association of HLA the Epstein-Barr virus carrier state in healthy seropositive class  I  and II alleles and extended haplotypes with naso- individuals. Int J Cancer 1985; 35:35–42. pharyngeal carcinoma in Taiwan. J Natl Cancer Inst 2002; 34. Ling PD, Lednicky JA, Keitel WA, et  al. The dynamics of 94:1780–9. herpesvirus and polyomavirus reactivation and shedding in 22. Vigil A, Davies DH, Felgner PL. Defining the humoral healthy adults: a 14-month longitudinal study. J Infect Dis immune response to infectious agents using high-density 2003; 187:1571–80. protein microarrays. Future Microbiol 2010; 5:241–51. 35. Xu FH, Xiong D, Xu YF, et  al. An epidemiological and 23. Coghill AE, Pfeiffer RM, Proietti C, et al. Identification of a molecular study of the relationship between smoking, risk novel, EBV-based antibody risk stratification signature for of nasopharyngeal carcinoma, and Epstein-Barr virus acti- early detection of nasopharyngeal carcinoma in Taiwan. vation. J Natl Cancer Inst 2012; 104:1396–410. Clin Cancer Res 2018; 24:1305–14. 36. Wang D, Liebowitz D, Kieff E. An EBV membrane protein 24. Galili T. dendextend: an R package for visualizing, adjust- expressed in immortalized lymphocytes transforms estab- ing and comparing trees of hierarchical clustering. lished rodent cells. Cell 1985; 43:831–40. Bioinformatics 2015; 31:3718–20. 37. Nutt SL, Hodgkin PD, Tarlinton DM, Corcoran LM. The 25. Ionita-Laza I, Lee S, Makarov V, Buxbaum JD, Lin X. generation of antibody-secreting plasma cells. Nat Rev Sequence kernel association tests for the combined effect Immunol 2015; 15:160–71. Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1931 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Infectious Diseases Oxford University Press

Loading next page...
1
 
/lp/ou_press/patterns-of-interindividual-variability-in-the-antibody-repertoire-TudQld1lyP

References (42)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Infectious Diseases Society of America
ISSN
0022-1899
eISSN
1537-6613
DOI
10.1093/infdis/jiy122
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the Epstein-Barr Virus Proteome Zhiwei Liu, Anna E Coghill, Ruth M Pfeiffer, Carla Proietti, Wan-Lun Hsu, Yin-Chu Chien, Lea Lekieffre, Lutz Krause, Kelly J Yu, Pei-Jen Lou, Cheng-Ping Wang, Jason Mulvenna, Jaap M Middeldorp, Jeff Bethony, Chien-Jen Chen, Denise L Doolan, Allan Hildesheim Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 The Journal of Infectious Diseases MAJOR ARTICLE Patterns of Interindividual Variability in the Antibody Repertoire Targeting Proteins Across the Epstein-Barr Virus Proteome 1,a 1,a 1 3,4 5,6 5,8 3 3 1 7 Zhiwei Liu, Anna E. Coghill, Ruth M. Pfeiffer, Carla Proietti, Wan-Lun Hsu, Yin-Chu Chien, Lea Lekieffre, Lutz Krause, Kelly J. Yu, Pei-Jen Lou, 7 3 9 2 5,6 3,4,b 1,b Cheng-Ping Wang, Jason Mulvenna, Jaap M. Middeldorp, Jeff Bethony, Chien-Jen Chen, Denise L. Doolan, and Allan Hildesheim 1 2 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, Maryland; Department of Microbiology, Immunology, and Tropical Medicine, George 3 4 Washington University Medical Center, Washington, D. C.; QIMR Berghofer Medical Research Institute, Brisbane, and Centre for Biosecurity and Tropical Infectious Diseases, 5 6 15 Australian Institute of Tropical Health and Medicine, James Cook University, Cairns, Australia; Genomics Research Center, Academia Sinica, Graduate Institute of Epidemiology and Prevention Medicine, College of Public Health, National Taiwan University, and Department of Otolaryngology, National Taiwan University Hospital and College of Medicine, 8 9 Taipei, and National Institute of Cancer Research, National Health Research Institute, Miaoli, Taiwan; and Department of Pathology, VU University Medical Center, Amsterdam, June Netherlands Background. Little is known about variation in antibody responses targeting the full spectrum of Epstein-Barr virus (EBV) proteins and how such patterns inform disease risk. Methods. We used a microarray to measure immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses against 199 EBV protein sequences from 5 EBV strains recovered from 289 healthy adults from Taiwan. We described positivity patterns, estimated the correlation between antibodies, and investigated the associations between environmental and genetic risk factors and variations in antibody responses. Results. Healthy adults were more likely to mount IgG antibody responses to EBV proteins (median positivity frequency, 46.5% –46 for IgG and 17.3% for IgA; P  =  1.6 ×  10 , by the Wilcoxon rank sum test). Responses against glycoproteins were particularly prevalent. The correlations between antibody responses of the same class were higher than correlations across classes. The mucosal exposure to proteins involved in EBV reactivation (as determined by the IgA response) was associated with smoking (P = .002, by the sequence kernel association test–combined), and approximately one quarter of adults displayed antibody responses associated with EBV-related cancer risk. Conclusions. es Th e data comprehensively define the variability in human IgG and IgA antibody responses to the EBV proteome. Patterns observed can serve as the foundation for elucidating which individuals are at highest risk of EBV-associated clinical condi- tions and for identifying targets for effective immunodiagnostic tests. Keywords. Antibody microarray, antibody response, Epstein-Barr virus, Taiwan, IgA, IgG Epstein-Barr virus (EBV) is a ubiquitous herpesvirus that infects infected individuals; such diseases include hematopoietic >90% of individuals worldwide [1]. Primary infection with the malignancies (eg, Burkitt and Hodgkin lymphomas) and epi- virus typically occurs in early childhood [2, 3], aer w ft hich EBV thelial malignancies (eg, gastric cancer and nasopharyngeal establishes lifelong latency in human B cells. During the course carcinoma [NPC]) [4]. EBV is etiologically involved in approx- of lifelong infection, EBV periodically reactivates in B cells or imately 200 000 cancers worldwide each year [5]. Despite its epithelial cells, expressing a broader set of lytic proteins required ubiquity and association with various chronic conditions, there for viral replication [4]. This ongoing replication in epithelial are currently no proven ways to prevent EBV infection from cells of the oral cavity contributes to shedding of the virus into occurring or to predict with sufficient accuracy which infected saliva, facilitating person-to-person transmission of EBV. individuals will develop serious chronic diseases associated Although established EBV infection is generally asymp- with this virus [5, 6]. tomatic, infection manifests as clinical disease in a subset of One way to monitor patterns of EBV exposure in the general population is to measure levels of circulating anti-EBV antibod- Received 11 January 2018; editorial decision 26 February 2018; accepted 1 March 2018; ies that reflect responses to viral proteins. For example, immu - published online March 2, 2018. noglobulin G (IgG) antibodies appear within several weeks aer ft Z. L. and A. E. C. contributed equally to this work. primary infection and remain in circulation for many years [7]. D. L. D. and A. H. contributed equally to this work. Correspondence: Z. Liu, PhD, Infections and Immunoepidemiology Branch, Division of Cancer Prior studies aimed at defining rates of EBV infection by age Epidemiology and Genetics, National Cancer Institute, 9609 Medical Center Dr, Rockville, MD across populations have used positivity for IgG antibodies tar- 20850 (zhiwei.liu@nih.gov). The Journal of Infectious Diseases 2018;217:1923–31 geting EBV proteins such as the viral capsid antigen (VCA) and Published by Oxford University Press for the Infectious Diseases Society of America 2018. EBV nuclear antigen 1 (EBNA1) [8] to classify individuals as This work is written by (a) US Government employee(s) and is in the public domain in the US. EBV positive (ie, as ever exposed to EBV). IgG antibodies can DOI: 10.1093/infdis/jiy122 Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1923 Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 also be used to better understand the nature of an individual’s Aer p ft rinting, anti-polyHis probing confirmed the presence of EBV infection. For example, a subset of long-term EBV carri- 99.5% of sequences; anti-HA probes confirmed the presence of ers displays elevated levels of IgG antibody targeting the early 98.5% of sequences. High coverage was achieved across the 5 antigen protein (EA) [9]. This EA complex is expressed as part EBV strains, with 97% of the predicted sequences from each of the EBV lytic cycle; therefore, elevated levels of EA indicate strain represented on the microarray at ≥99% homology. ongoing viral lytic activity (reactivation) and a potential lack of A positive antibody response in a healthy adult was defined control over the virus [10]. as a response greater than that individual’s person-specific back - In contrast to IgG, immunoglobulin A  (IgA) responses ground (mean of 4 no-DNA-control spots plus 1.5 standard devi- to viral exposure are believed to reflect recent reactivation ations) [17]. As previously reported [23], we included blinded at mucosal sites (eg, the oral epithelium) [11]. This type of quality control replicates during testing and observed good immune response is of interest because of the oral route of EBV reproducibility for antibodies measured using this custom pro- transmission. During primary infection, IgA responses against tein microarray. The average percentage agreement for classifying EBV in the oral cavity can be strong but are short lived [12]. an individual’s antibody response as positive versus negative was e d Th urability and spectrum of EBV-directed IgA responses in high for both IgA (84%; range, 52%–100%) and IgG (79%; range, healthy adults are largely uncharacterized. However, it has been 48%–100%), respectively. e Th antibody level (determined on the shown that individuals at greatest risk of developing NPC, an basis of the standardized signal intensity output from the array) EBV-associated cancer, have elevated anti-EBV IgA responses was further grouped into quartiles (4 categories) based on the to VCA and EBNA1 years before a cancer diagnosis [8, 13, 14]. population distribution of the response for each given antibody. Although informative, prior studies of humoral responses to Based on coefficients of variation reported from the previous EBV infection [15, 16] have largely focused on measuring levels study [23], we excluded from analysis the 2 IgG array spots out of of IgG and IgA antibodies to a handful (<10) of the approxi- 199 evaluated that had coefficients of variation of >20% (BZLF1/ mately 90 EBV protein targets [8]. With the advent of protein IgG [array protein sequence CAA24861.1-103155-102655] and array systems capable of screening for antibody responses BXRF1/IgG [array protein sequence CAA24798.1-144860- against the full complement of EBV proteins [17–19], it is now 145606]). Finally, we included 3 synthetic EBV peptides that are feasible to evaluate humoral responses to EBV infection in a putative cancer biomarkers (VCAp18, EBNA1, and EAdp47) on more comprehensive manner. Thus, we evaluated IgG and IgA the microarray, bringing the total number of anti-EBV antibod- anti-EBV antibody responses against 199 predicted sequences ies for analysis to 402 (200 IgG and 202 IgA antibodies). and 3 synthetic antigens from 86 EBV proteins involved in Ethics Statement various stages of the EBV life cycle (eg, latent and lytic cycles) All human subjects included were adults, and all studies were among 289 healthy individuals from Taiwan, a region with high approved by appropriate human subjects committees in Taiwan rates of EBV-associated cancer. and the United States. This study was reviewed/approved by the National Cancer Institute Special Studies Institutional METHODS Review Board and the National Taiwan University Institutional A total of 289 healthy, asymptomatic adults who served as Review Board)\. Written informed consent was obtained for all controls for 2 previously published Taiwanese studies were participants. selected for this report [20, 21]. Details are described in the Supplementary Materials. A  previously described EBV pro- Statistical Analysis tein microarray targeting IgG and IgA antibodies against 199 Visualization and statistical analyses were conducted using R sta- predicted EBV protein sequences (86 EBV proteins) from 5 tistical sowa ft re. Details on the statistical analyses are described EBV strains (AG876, Akata, B95-8, Mutu, and Raji) was used in the Supplementary Materials. Briefly, to pool data across the [17, 18, 22, 23]. Briefly, these 199 sequences represent nonre - 2 study populations, we used population-specific cut points to dundant open reading frames from 5 EBV strains representing generate similar underlying antibody distributions and then both African (Mutu, Raji, and AG876) and Asian (Akata) vari- compared individual response rates (ie, percentage positivity ants, as well as known splice variants identified in the litera - values) across immunoglobulin class and EBV life cycle. We esti- ture. Eighty-five percent of 199 predicted sequences represented mated the correlation between antibodies across immunoglobu- complete transcripts of EBV genes. The remaining 30 predicted lin classes (ie, IgG against protein X versus IgA against protein sequences represented linear segments from 8 EBV genes X) and within each immunoglobulin class (ie, IgG against pro- >1000  bp long. Each of the protein sequences cloned into the tein X versus IgG against protein Y), using Spearman correla- pXT7 expression vector prior to printing onto the microarray tion coefficients. All individuals were clustered according to the included N-terminal 10x histidine (His) and C-terminal hem- positive responses across the full EBV proteome, using unsuper- agglutinin (HA) tags to confirm expression on the microarray. vised hierarchical clustering with binary distance and complete 1924 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 linkage [24]. To examine the association between environmen- e in Th dividual with the narrowest pattern responded to 38 EBV tal and genetic risk factors (ie, smoking, alcohol drinking, and targets on the array (28 IgG antibodies and 10 IgA antibodies), HLA alleles) and biological groupings of antibody response (eg, representing 21 EBV proteins (Figure  1A). The individual with responses directed against a select EBV life cycle), we used the the broadest response had a positive response against 329 EBV sequence kernel association test–combined (SKAT-C) from the targets on the array (163 IgG antibodies and 166 IgA antibodies), single-nucleotide polymorphism set [25], using binary variables representing 84 of 86 EBV proteins evaluated. Positive IgG (ie, (with a positive antibody response assigned a value of 1 and a long-term) responses against EBV proteins were markedly more negative response assigned a value of 0). frequent than positive IgA (ie, more-recent) responses (Table 1). On average, individuals mounted IgG antibody responses RESULTS against 93 (46.5%) of 200 EBV targets on the array and IgA anti- Anti-EBV IgG and IgA Antibody Responses Mounted by Healthy body responses against 35 (17.3%) of 202 EBV targets. Individuals Additional evidence of broad anti-EBV IgG reactivity was All individuals had measurable (ie, positive) antibody responses provided by the observation that almost all individuals (≥95%) against EBV (Figures  1A–C; Supplementary Figures  1–10). in our study mounted a positive IgG response against 14 IgG IgA Subjects BC Subjects Positive for IgG Antibody Subjects Positive for IgA Antibody BZLF2:CAA24860.1–102116–101445 BZLF2:YP_001129466.1–90630–89959 100% 96% BRRF2:YP_001129470.1–94844–96457 BZLF2:CAA24860.1–102116–101445 100% 94% BDLF3:AFY97964.1–118644–117940 BRRF2:AFY97943.1–93884–95497 100% 93% BRRF2:YP_001129470.1–94844–96457 100% BRRF2:AFY97943.1–93884–95497 93% BSLF1:YP_001129457.1–74727–72103 BFRF3 (VCA_p18):YP_001129448.1–49335–49865 83% 100% BFRF3 (VCA_p18):YP_001129448.1–49335–49865 BILF1:YP_001129506.1–154125–153187 100% 80% BFRF3 (VCA_p18):AFY97924.1–49199–49729 BZLF2:YP_001129466.1–90630–89959 100% 80% BDLF4:YP_001129488.1–117560–116883 BDLF3:AFY97964.1–118644–117940 100% 79% BFRF3 (VCA_p18):AFY97924.1–49199–49729 BLRF2 (VCA_p23):YP_001129461.1–76771–77259 99% 77% BLRF2 (VCA_p23):YP_001129461.1–76771–77259 BFRF3 (VCA_p18):CAA24838.1–61507–62037 99% 70% BFRF3 (VCA_p18):CAA24838.1–61507–62037 BDLF4:YP_001129488.1–117560–116883 99% 70% BBRF3:YP_001129479.1–107679–108896 BFRF1:YP_001129446.1–46719–47729 98% 65% VCA_p18:synthetic antigen EBNA3B:YP_001129464.1–83509–86532–1 98% 65% BMRF2:YP_001129455.1–68964–70037 98% EBNA3B:YP_001129464.1–83509–86532–1 65% EBNA3A:AFY97915.1–80252–82747 97% BMRF2:YP_001129455.1–68964–70037 64% EBNA3A:YP_401669.1–80382–82877 97% EBNA1:synthetic antigen 64% BPFL1:CAA24839.1–71527–62078–2 62% VCA_p18:synthetic antigen 96% BDLF3:YP_001129490.1–119605–118901 BBRF3:YP_001129479.1–107679–108896 96% 61% BGLF5 (ALK.EXONUCLEASE):YP_001129480.1–109516–109289 BDLF3:YP_001129490.1–119605–118901 96% 60% BGLF5 (ALK.EXONUCLEASE):AFY97956.1–108555–108328 LMP2B:AFY97910.1–1026–1196 94% 55% BcLF1 (VCA_p160):YP_001129493.1–126005–121860–2 EBNA–LP:AFY97917.1–35572–35676 10% 2% EBNA–LP:YP_001129440.1–35558–35662 BKRF4:YP–001129474.1–99676–100329 9% 2% BALF5 (DNA polymerase):YP_001129507.1–157772–154725–1 CAPSID:YP_001129451.1–63084–64178 8% 2% BGLF3:YP_001129483.1–112496–112035 8% BDLF3:YP_001129489.1–117772–117539 2% EBNA1:AFY97842.1–95349–97142 8% BALF5 (DNA polymerase):AFY97894.1–154809–155094 1% 8% BBRF2:YP_001129477.1–104490–105326 1% BcLF1 (VCA_p160):YP_001129493.1–126005–121860–2 7% EBNA–LP:AFY97917.1–35572–35676 1% BALF2 (EA(D)_p138):YP_001129510.1–165796–162410–1 7% EBNA3C:AFY97856.1–85691–86050 1% BNLF2B:AFY97989.1–166696–166400 BALF5 (DNA polymerase):CAA24805.1–156746–153699–1 BGLF5 (DNAse):YP_001129481.1–110883–109471 6% 1% FGAM:YP_001129438.1–1736–5692–1 BZLF1 (Zta):YP_001129467.1–91697–91197 6% 1% BNLF2B:CAA24811.1–167303–166998 LF2:YP_001129504.1–151808–150519 6% 1% BORF2:YP_001129452.1–64253–66733 FGAM:YP_001129438.1–1736–5692–1 6% 0% BcLF1 (VCA_p160):AFY97883.1–124729–120584–2 BGLF4 (protein kinase):CAA24828.1–123692–122328 4% 0% EBNA3C:AFY97856.1–86125–88794 BALF5 (DNA polymerase):CAA24805.1–156746–153699–1 4% 0% LMP1:YP_001129515.1–170457–170190 4% BRFR1A:AFY97836.1–46138–46545 0% 4% EAdp47:synthetic antigen 0% BFRF1A:YP_001129445.1–46352–46759 3% BFLF2:YP_001129443.1–44763–43807 0% BBRF2:YP_001129477.1–104490–105326 2% BGRF1/BDRF1:YP_001129485.1–117754–118890 0% BGRF1/BDRF1:YP_001129485.1–117754–118890 2% BFRF1A:AFY97921.1–46216–46623 0% BFRF1A:AFY97921.1–46216–46623 BGLF4 (protein kinase):CAA24828.1–123692–122328 0% BZLF1 (Zta):YP_001129467.1–90855–90724 1% Subjects Subjects Most/Least Prevalent IgG Responses Most/Least Prevalent IgA Responses Figure 1. Anti–Epstein-Barr virus (EBV) immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses among healthy individuals in Taiwan. A, Number of pos- itive antibody responses in 289 healthy individuals. Blue indicates IgG antibody positivity, and red indicates IgA positivity. The individual with the narrowest response had antibodies against 38 EBV targets on the array (28 IgG antibodies and 10 IgA antibodies), representing 21 EBV proteins. In contrast, the individual with the broadest response had antibodies against 329 EBV targets (163 IgG antibodies and 166 IgA antibodies), representing 84 of 86 tested EBV proteins. B and C, The 20 IgG (B) and IgA (C) antibodies with the highest antibody positivity rates and the 20 with the lowest antibody positivity rates. Percentages indicate the antibody positivity rate for the given target antigen in this population. Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1925 Number of Positive Antibody Responses (Sum of IgG and IgA Antibody Responses) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Table  1. Number of Positive Antibody Responses to Epstein-Barr Virus of broad, long-term exposure to proteins from all stages in the (EBV) Proteins Among Healthy Individuals in Taiwan, by Antibody Type EBV life cycle among healthy adults. and Stage of EBV Life Cycle Targeted IgA antibody responses against EBV glycoproteins (median a IgA positivity frequency, 42.1%; Table  1) was also more fre- Antibody, Target Protein Antibodies, No. Median (% ) IQR Range quently observed than responses against proteins from other IgG stages of the EBV life cycle. The median positivity frequen - Overall 200 93 (46.5) 67–132 22–188 Latent 58 30 (51.7) 21–40 21–58 cies observed for IgA antibodies across proteins from latent Immediate early lytic 11 4 (36.4) 2–7 0–11 and lytic phases of the EBV life cycle were similar (range, Early lytic 44 20 (45.5) 13–28 5–42 15.6%–17.2%; Table  1), with 1 notable exception: IgA anti- Late lytic 47 20 (42.6) 14–30 7–45 body responses against EBV proteins involved in the switch Glycoprotein 19 13 (68.4) 11–16 4–19 from latency to lytic phases were particularly infrequent, with Other/unknown 21 7 (33.3) 4–12 0–20 a median positivity frequency of 0% (0 of 12; range, 0%–75%). IgA Overall 202 35 (17.3) 18–85 4–166 This is particularly interesting because IgA responses to some Latent 58 10 (17.2) 4–20 0–48 of these proteins have been shown to be predictive of the risk of Immediate early lytic 12 0 (0.0) 0–1 0–9 the EBV-associated malignancy NPC [8]. Early lytic 45 7 (15.6) 3–15 0–35 Late lytic 47 8 (17.0) 6–14 0–39 Clustering of Antibody Responses by Immunoglobulin Class Glycoprotein 19 8 (42.1) 3–15 0–19 Consistent with the observation that IgG responses to EBV were Other/unknown 21 1 (4.8) 0–6 0–17 more frequently observed than IgA responses, we noted that Abbreviations: IgA, immunoglobulin A; IgG, immunoglobulin G; IQR, interquartile range. IgG responses to a particular antigenic target had, on average, a Calculated on the basis of the median number of positive antibodies among the total number of antibodies evaluated in each group. higher correlation with IgG responses to other antigenic targets, rather than with IgA responses against the same target (median Spearman correlation, 0.495 [range, −0.093–0.978]; 85.9% different EBV proteins, including 2 latent proteins (EBNA3B of comparisons had a Bonferroni-corrected P value of  <  .05; and EBNA1), 4 early lytic proteins (BILF1, BDLF4, BLRF2, Figures  2A and B and Supplementary Table  3). Practically, and BGLF5), 4 late lytic proteins (BRRF2, BFRF3, VCA-p18, this meant that an individual with elevated IgG titers against and BPFL1), and 4 glycoproteins (BZLF2, BDLF3, BBRF3, and a specific EBV protein (eg, protein X) was more likely to have BMRF2; Figure  1B and Supplementary Table 1). This IgG pat - elevated IgG titers against another EBV protein (eg, protein Y), tern stands in contrast to that of IgA, for which only 1 target rather than elevated IgA responses to protein X. The same phe - on the array (representing BZLF2) elicited a uniformly positive nomenon was evident for IgA responses, with IgA responses response (Figure 1C; Supplementary Table 1), with a plurality of against a particular target being more strongly correlated with study participants (40.5% [117]) positive for <25 array targets. IgA responses to other antigenic targets (median Spearman Only 8 EBV targets were weakly immunogenic at eliciting IgG correlation, 0.366 [range, −0.293–0.904]; 64.5% of comparisons responses (ie, <5% of individuals tested positive), including 1 had a Bonferroni-corrected P value of < .05), rather than with targeting a latent protein (EBNA3C), 1 early lytic protein (EAD- IgG responses to the same antigen. p47: BMRF1), and 4 late lytic proteins (BGRF1/BDRF1, BFLF2, While correlations tended to be higher within rather than BRFR1A, and BcLF1; Figure  1B and Supplementary Table  2). across antibody class, interimmunoglobulin correlations (ie, This contrasts with 33 EBV proteins that were weakly immuno - IgG vs IgA correlations) were higher when considering the same genic for IgA (Figure 1C; Supplementary Table 2). protein target (median Spearman correlation, 0.226; 23.0% of correlations had a Bonferroni-corrected P value of  <  .05), Antibody Responses Against EBV Proteins Expressed at Different Stages rather than different EBV proteins (median Spearman correla - of the EBV Life Cycle tion, 0.127 [range, −0.334–0.563]; 1.9% of the correlations had Although IgG responses were more common than IgA responses, a Bonferroni-corrected P value of < .05). Similar patterns were the biological function of the target protein was also an important observed across all EBV life cycle stages (Supplementary Table 3 determinant of the host antibody response (Table 1). IgG against and Supplementary Figures 11A–E). EBV glycoproteins (median IgG positivity frequency, 68.4%) was more frequently observed than IgG targeting proteins from Clustering of Healthy, Asymptomatic Adults According to Anti-EBV other stages of the EBV life cycle. Robust IgG responses to EBV Antibody Responses latent proteins and proteins involved in the switch from latent In addition to understanding the breadth of the antibody to lytic infection were also observed (median positivity frequen- response marking EBV exposure in healthy adults, hierarchical cies, 51.7% and 45.5%, respectively). Responses to EBV proteins clustering identified 4 distinct patterns among study partici - involved in immediate early lytic proteins (median positivity fre- pants that could be distinguished by their average number of quency, 36.4%) were less widespread but still frequent, evidence 1926 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 AB Antibody type IgA IgG EBV life cycle Latent Early lytic Glycoprotein Immediate early lytic Late lytic Other/unknown –0.5 0.0 0.5 1.0 Spearman correlation Comparison IgA IgG Spearman correlation IgA antibody vs IgA antibody IgA antibody vs IgA antibody (di erent target) –1 –0.5 0 0.5 1 IgG antibody vs IgA antibody (same target) IgG antibody vs IgG antibody Correlation between the average IgA and IgG antibody response Interantibody correlations according to immunoglobulin type Figure 2. Spearman correlation between the average immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody responses in this population. A, Unsupervised hierar- chical clustering of IgG and IgA antibody responses, based on the correlation between the average antibody output for any 2 given Epstein-Barr virus (EBV) targets (Spearman correlation coefficient). Red depicts a strong positive correlation, and blue indicates a negative correlation. B, Distribution of Spearman correlation coefficients according to the type of comparison being made (eg, between 2 IgA antibodies or between IgG and IgA antibodies targeting a selected antigen on the array). Each row/column represents the average antibody response to a given EBV target in the study population. Each histogram plots the set of average antibody responses illustrated in panel A. positive IgG and IgA antibody responses (Figures  3A and B). IgG and IgA responses observed among individuals in group We observed 4 groups: group 1 comprised participants with 4 were broadly comparable to those observed among NPC low IgG and IgA responses, group 2 comprised those with a cases (Figure  3B). When we estimated a cancer risk score for low IgG response and an elevated IgA response, group 3 com- asymptomatic individuals in our study by using our previously prised those with an intermediate IgG response and a low IgA reported 14-antibody NPC risk stratification signature [ 23], response, and group 4 comprised those with high IgG and IgA 79% of individuals in group 4 had scores that ranked in the responses. Group  1 participants (9.0% [26]) mounted 46 IgG highest quartile of cancer risk, compared with only 7% of indi- -14 2 and 15 IgA antibody responses, on average. Group  4 partici- viduals in groups 1–3 (P  =  2.2 ×  10 , by the test; Table  2). pants (26.0% [75]) mounted 152 IgG and 69 IgA responses, on es Th e findings suggest that healthy individuals can be classified average. Interestingly, although IgA responses were less frequent as having distinct patterns of EBV responses and that a propor- than IgG responses overall, group 2 and 4 participants (39% tion of healthy individuals are characterized by high IgG and [114]) displayed a high level of IgA reactivity, indicating that IgA responses typically observed among individuals at highest ongoing EBV exposure at mucosal surfaces is not an infrequent risk of NPC development. event in healthy, Taiwanese adults previously exposed to EBV. We further examined whether group 4 participants were more Smoking Is Associated With Differential EBV-Directed IgA Antibody likely to respond to targets from a specific stage of the EBV life Responses cycle and observed that these subjects had elevated responses e Th extent to which the effect of known NPC risk factors is to all stages (Supplementary Figures  12A and B). Notably, this mediated through their ability to modulate immune responses high response pattern included higher IgA antibody reactivity to EBV is not fully understood. We observed that individuals against immediate early lytic proteins (median number of pos- from group 4 also had the highest proportion of current smok- itivity  among individuals in this group, 1; interquartile range, ers, compared with other groups (72% vs 45%; P  =  0.02, by 0–2.5) compared to the other groups (median number of posi- the test). In addition, when considering the proteome-wide -7 tivity, 0; interquartile range, 0–0; P = 1.9 × 10 , by the rank test). antibody response, the total level of the IgA response but not We further compared the antibody pattern in group 4 to that the IgG response was significantly different between smokers previously described among patients with NPC (n = 175) [23]. and nonsmokers in our population (P = .002, by the SKAT-C; Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1927 IgA IgG Density Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 AB 0.8 0.4 0.0 #1 #2 #4 #3 Group 1 (26) Group 2 (39) Group 3 (149) Group 4 (75) NPC Group (175) 150 100 Group 1 (26) Group 2 (39) Group 3 (149) Group 4 (75) NPC Group (175) Clustering of Individuals on the Basis of the Average Number of Positive Antibody Responses Total Number of Positive Anti-EBV Antibody Responses in 4 Groups of Disease-Free, EBV-Infected Adults and Patients With NPC Figure 3. Clustering of individuals defined according to their immunoglobulin G (IgG) and immunoglobulin A (IgA) antibody response. A, Dendrogram showing clusters of individuals defined according to the number of positive antibody responses (with a positive antibody response assigned a value of 1 and a negative response assigned a value of 0) across the full spectrum of evaluated IgG and IgA antibody responses, using unsupervised hierarchical clustering with binary distance and complete linkage. B, Box plot showing number of positive IgG/IgA responses according to 4 groups/clusters based on antibody response, compared with individuals with nasopharyngeal carcinoma (NPC) who had a blood sample collected at the time of cancer diagnosis. Group 1 comprised participants with low IgG and IgA responses, group 2 comprised those with a low IgG response and an elevated IgA response, group 3 comprised those with an intermediate IgG response and a low IgA response, and group 4 comprised those with high IgG and IgA responses. See Methods and Results for additional details. Figure  4A). Notably, when evaluating anti-EBV antibody proteins differed significantly by smoking behavior (Bonferroni- responses targeting proteins across different stages of the EBV corrected P value of < .05, by the SKAT-C). life cycle, the EBV-directed IgA response to latent and early lytic Further, in logistic regression models that evaluated the associa- tion between smoking and each marker individually, we observed Table  2. Nasopharyngeal Carcinoma (NPC) Risk Scores Among suggestive associations (P < .05) for multiple anti-EBV IgA anti- Controls From a Cancer Screening Population Cohort in Taiwan, by bodies from all stages of the EBV life cycle (37.9% [22 of 58] were Immunoglobulin A (IgA) and Immunoglobulin G (IgG) Responses associated with latent proteins, 8.3% [1 of 12] were associated with immediate early proteins, 20.0% [9 of  45] were associated with Control Group, Participants, No. (%) early lytic proteins, 6.4% [3 of 47] were associated with late lytic Group 1 Group 2 Group 3 Group 4 proteins, and 21.0% [4 of 19] were associated with glycoproteins; Risk Score (n = 16) (n = 21) (n = 49) (n = 28) Supplementary Tables 4–8). The strongest individual association Quartile 1 11 (68.7) 2 (9.5) 14 (28.6) 2 (7.1) Quartile 2 2 (12.5) 5 (23.8) 19 (38.8) 2 (7.1) was observed between smoking and IgA targeting LMP1(odds Quartile 3 2 (12.5) 11 (52.4) 14 (28.6) 2 (7.1) ratio, 4.64; P = .001; Figure 4B), an EBV oncoprotein expressed in Quartile 4 1 (6.2) 3 (14.3) 2 (4.1) 22 (78.6) NPC tumors that is normally considered a weakly immunogenic Estimated using our previously reported 14-antibody NPC risk stratification signature [ 23]. protein [26, 27]. Taken together, these findings suggest that smok - Quartile 1 denotes the lowest NPC risk, and quartile 4 denotes the highest NPC risk. b ing may alter the anti-EBV mucosal antibody response. Group 1 comprised participants with low IgG and IgA responses, group 2 comprised those with a low IgG response and an elevated IgA response, group 3 comprised those with an We also evaluated the association between anti-EBV antibody intermediate IgG response and a low IgA response, and group 4 comprised those with high IgG and IgA responses. See Methods and Results for additional details. responses, alcohol drinking, and HLA alleles that modulate 1928 • JID 2018:217 (15 June) • Liu et al Number of Positive Number of Positive Antibody Responses (IgA) Antibody Responses (IgG) Height Average Number of Positive Average Number of Positive Antibody Responses (IgA) Antibody Responses (IgG) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 A B Smoking 2.0 –5 P = 9.89 × 10 All IgA Latent Immediate Early Lytic 1.5 Early Lytic Late Lytic Glycoprotein 1.0 All IgG Latent Immediate Early Lytic 0.5 Early Lytic IgG Late Lytic IgA Glycoprotein 0.0 Never (42) Former (14) Current (60) 01 2 34 Smoking Status (Among Male Participants) –log10 (P ) Figure  4. Associations between anti–Epstein-Barr virus (EBV) antibody responses and smoking status among males. A, Association between anti-EBV immunoglobulin A (IgA) antibodies against proteins expressed during different EBV life cycles and smoking status (current vs former/never). P values were obtained from sequence kernel association test–C tests. The dashed line represents the Bonferroni-corrected statistically significant P value threshold of .005. B, Box plot showing the association between LMP1/IgA reactivity and smoking status. The P value was obtained from linear regression, adjusted for age. NPC risk (ie, HLA-A*0207 and HLA-A*1101; Supplementary in this study displayed antibody response patterns similar to Figures  13A–C). HLA-A*0207 was nominally associated with NPC cases, including positive responses reflecting exposure IgA responses against immediate early lytic proteins (P  =  .02). to immediate early antigenic targets, such as Zta and Rta. However, no statistically significant associations persisted aer ft Importantly, our data indicate that smoking is associated with correction for multiple comparisons. altered IgA antibody patterns that could reflect poor control over EBV lytic activity, suggesting a possible mechanism to DISCUSSION explain the role of smoking in the etiology of NPC. Our findings represent the first comprehensive evaluation of Previous studies have evaluated responses against a handful natural variation in the host antibody response to the full spec- of anti-EBV antibodies by using techniques including immuno- trum of EBV proteins. This unique characterization was made blot and microarray chips targeting 10–15 EBV proteins [28, 29]. possible through the use of protein microarray technology in Zheng et  al used a microarray to test 82 EBV open-reading a population of healthy, asymptomatic adults from Taiwan. frames in a study that included 10 nondiseased adults and Because the antibody repertoire can shed light on exposure showed that these adults mounted IgG responses against an to viral proteins, this virus-wide description of host response average of 17 proteins (at 1:100 serum dilution) and IgA anti- to the EBV proteome represents an important step forward bodies against an average of 13 proteins (at 1:1000 serum dilu- in understanding population-level variation that might have tion) [19]. This is in line with our findings of more frequent implications for multiple EBV-related diseases. IgG reactivity, rather than IgA reactivity. In the present study, We found that all adults evaluated in our study had evidence we expanded testing to 199 EBV open-reading frames among of exposure to EBV proteins, that these EBV-positive adults 289 asymptomatic adults. Because our array was not designed were more likely to mount IgG rather than IgA antibody to detect antibodies to conformational epitopes that require gly- responses to EBV, and that responses against glycoproteins cosylation, future studies will be required to better understand were particularly prevalent. If the IgG response targeting 1 EBV patterns of response to those epitopes and how they are related protein (eg, protein X) was elevated in an individual, they were to those reported herein. more likely to mount an IgG response targeting other antigens, Our finding that antibody responses to some glycoproteins rather than an IgA response against protein X, likely reflecting were the most prevalent is biologically plausible because these differential timing of exposure (longer-term exposure for IgG proteins are expressed on the viral surface and therefore readily and more-recent exposure for IgA) and separate triggering accessible as targets for the immune system [30, 31]. Given that mechanisms in different compartments (eg, oral mucosa vs IgA responses are thought to reflect recent pathogen exposure, systemic circulation). Approximately one quarter of the subjects our finding that IgA responses to glycoproteins were also more Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1929 Antibody Level (LMP1/IgA) Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Notes frequently observed than responses against proteins from other stages of the EBV life cycle suggests that the presence of EBV in Financial support. This work was supported by the the oral cavity is not a rare event [32]. It was notable that indi- National Cancer Institute Intramural Research Program, the viduals in groups 2 and 4 (39% [114]) mounted a substantial, National Health and Medical Research Council of Australia, proteome-wide IgA response. This supports data from studies and the National Science Council of Taiwan. reporting EBV viral load shedding in the saliva of the majority Potential coni fl cts of interest. J. M. M. received payments as of healthy adults measured [32, 33], as well as EBV reactivation owner and chief executive officer of Cyto-Barr. All other authors rates that are much more frequent than those for other herpes- report no potential conflicts. All authors have submitted the viruses, such as cytomegalovirus [34]. ICMJE Form for Disclosure of Potential Conflicts of Interest. An important public health implication of our data is that Conflicts that the editors consider relevant to the content of the smoking, a modifiable behavior, was associated with elevated manuscript have been disclosed. IgA responses against EBV. An association between smoking and EBV reactivation has been posited by one study, which References evaluated VCA/IgA in individuals from southern China [35]. In 1. Kieff E, Rickinson AB. Epstein-Barr virus and its replication. the present study, we expand beyond VCA/IgA alone to demon- In Fields virology. 5th ed. ed. Philadelphia, PA: Lippincott strate large-scale differences in the EBV-directed IgA repertoire Williams & Wilkins, 2007. in smokers. The most significant smoking association was 2. Kangro HO, Osman HK, Lau YL, Heath RB, Yeung CY, Ng observed for IgA responses against LMP1, an important NPC MH. Seroprevalence of antibodies to human herpesviruses oncogene that has cell-transforming ability in rodent b fi roblasts in England and Hong Kong. J Med Virol 1994; 43:91–6. [36]. In addition to these individual antibody associations, we 3. Chen CY, Huang KY, Shen JH, Tsao KC, Huang YC. A large- observed that individuals in group 4 also had the highest pro- scale seroprevalence of Epstein-Barr virus in Taiwan. PLoS portion of current smokers. Taken together, these findings sug - One 2015; 10:e0115836. gest that smoking may negatively impacts an individual’s ability 4. Young LS, Rickinson AB. Epstein-Barr virus: 40  years on. to control EBV infection, leading to an increased frequency of Nat Rev Cancer 2004; 4:757–68. viral reactivation/exposure in the host. However, we did not 5. Cohen JI, Fauci AS, Varmus H, Nabel GJ. Epstein-Barr assess the degree to which the established association between virus: an important vaccine target for cancer prevention. smoking and NPC is mediated by interindividual variation in Sci Transl Med 2011; 3:107fs7. the immune response to this ubiquitous virus. The underlying 6. Kanekiyo M, Bu W, Joyce MG, et al. Rational design of an biological mechanisms merit further investigation. Epstein-Barr virus vaccine targeting the receptor-binding In conclusion, we described the pattern of the IgG and IgA site. Cell 2015; 162:1090–100. responses to EBV among asymptomatic, healthy adults, using a 7. Hjalgrim H, Friborg J, Melbye M. The epidemiology of protein microarray chip targeting 199 EBV protein sequences. EBV and its association with malignant disease. In: Arvin Given that the generation of antibodies is a complex process A, Campadelli-Fiume G, Mocarski E, et  al., eds. Human requiring interplay between viral antigens and numerous com- herpesviruses: biology, therapy, and immunoprophylaxis. partments of the immune system [37], our study represents Cambridge: Cambridge University Press, 2007. an initial but important step in understanding the immune 8. Coghill AE, Hildesheim A. Epstein-Barr virus antibodies response to EBV infection. In the future, our approach could be and the risk of associated malignancies: review of the litera- applied to other populations to characterize the similarities and ture. Am J Epidemiol 2014; 180:687–95. differences observed in the humoral response to EBV in indi - 9. Melbye M, Ebbesen P, Levine PH, Bennike T. Early primary viduals predisposed to different EBV-associated conditions (eg, infection and high Epstein-Barr virus antibody titers in sub-Saharan Africans at higher risk of Burkitt lymphoma). Such Greenland Eskimos at high risk for nasopharyngeal carci- studies could elucidate subsets of EBV-infected individuals at noma. Int J Cancer 1984; 34:619–23. highest risk of EBV-associated clinical conditions and help iden- 10. Henle W, Henle G. Epstein-Barr virus-specific serology in tify targets for the development of effective immunodiagnostics. immunologically compromised individuals. Cancer Res Supplementary Data 1981; 41:4222–5. Supplementary materials are available at e Th Journal of Infectious 11. Brandtzaeg P. Potential of nasopharynx-associated lym- Diseases online. Consisting of data provided by the authors to phoid tissue for vaccine responses in the airways. Am J benefit the reader, the posted materials are not copyedited and Respir Crit Care Med 2011; 183:1595–604. are the sole responsibility of the authors, so questions or com- 12. Bhaduri-McIntosh S, Landry ML, Nikiforow S, Rotenberg ments should be addressed to the corresponding author. M, El-Guindy A, Miller G. Serum IgA antibodies to 1930 • JID 2018:217 (15 June) • Liu et al Downloaded from https://academic.oup.com/jid/article/217/12/1923/4917688 by DeepDyve user on 18 July 2022 Epstein-Barr virus (EBV) early lytic antigens are present in of rare and common variants. Am J Hum Genet 2013; primary EBV infection. J Infect Dis 2007; 195:483–92. 92:841–53. 13. Chien YC, Chen JY, Liu MY, et  al. Serologic markers of 26. Dawson CW, Port RJ, Young LS. The role of the EBV- Epstein-Barr virus infection and nasopharyngeal carci- encoded latent membrane proteins LMP1 and LMP2 in the noma in Taiwanese men. N Engl J Med 2001; 345:1877–82. pathogenesis of nasopharyngeal carcinoma (NPC). Semin 14. Cao SM, Liu Z, Jia WH, et al. Fluctuations of Epstein-Barr Cancer Biol 2012; 22:144–53. virus serological antibodies and risk for nasopharyngeal 27. Meij P, Vervoort MB, Aarbiou J, et  al. Restricted low-level carcinoma: a prospective screening study with a 20-year human antibody responses against Epstein-Barr virus (EBV)- follow-up. PLoS One 2011; 6:e19100. encoded latent membrane protein 1 in a subgroup of patients 15. Henle W, Henle G, Zajac BA, Pearson G, Waubke R, Scriba with EBV-associated diseases. J Infect Dis 1999; 179:1108–15. M. Differential reactivity of human serums with early 28. Fachiroh J, Schouten T, Hariwiyanto B, et  al. Molecular antigens induced by Epstein-Barr virus. Science 1970; diversity of Epstein-Barr virus IgG and IgA antibody 169:188–90. responses in nasopharyngeal carcinoma: a comparison of 16. Ginsburg CM, Henle W, Henle G, Horwitz CA. Infectious Indonesian, Chinese, and European subjects. J Infect Dis mononucleosis in children. Evaluation of Epstein-Barr 2004; 190:53–62. virus-specific serological data. JAMA 1977; 237:781–5. 29. Loebel M, Eckey M, Sotzny F, et al. Serological profiling of 17. Doolan DL. Plasmodium immunomics. Int J Parasitol 2011; the EBV immune response in Chronic Fatigue Syndrome 41:3–20. using a peptide microarray. PLoS One 2017; 12:e0179124. 18. Davies DH, Duffy P, Bodmer JL, Felgner PL, Doolan DL. 30. Hutt-Fletcher LM. EBV glycoproteins: where are we now? Large screen approaches to identify novel malaria vaccine Future Virol 2015; 10:1155–62. candidates. Vaccine 2015; 33:7496–505. 31. Li Q, Turk SM, Hutt-Fletcher LM. The Epstein-Barr virus 19. Zheng D, Wan J, Cho YG, et  al. Comparison of humoral (EBV) BZLF2 gene product associates with the gH and gL immune responses to Epstein-Barr virus and Kaposi’s homologs of EBV and carries an epitope critical to infection sarcoma-associated herpesvirus using a viral proteome of B cells but not of epithelial cells. J Virol 1995; 69:3987–94. microarray. J Infect Dis 2011; 204:1683–91. 32. Hadinoto V, Shapiro M, Sun CC, Thorley-Lawson DA. The 20. Chen CJ, Yang HI, Su J, et al.; REVEAL-HBV Study Group. dynamics of EBV shedding implicate a central role for epi- Risk of hepatocellular carcinoma across a biological gra- thelial cells in amplifying viral output. PLoS Pathog 2009; dient of serum hepatitis B virus DNA level. JAMA 2006; 5:e1000496. 295:65–73. 33. Yao QY, Rickinson AB, Epstein MA. A re-examination of 21. Hildesheim A, Apple RJ, Chen CJ, et al. Association of HLA the Epstein-Barr virus carrier state in healthy seropositive class  I  and II alleles and extended haplotypes with naso- individuals. Int J Cancer 1985; 35:35–42. pharyngeal carcinoma in Taiwan. J Natl Cancer Inst 2002; 34. Ling PD, Lednicky JA, Keitel WA, et  al. The dynamics of 94:1780–9. herpesvirus and polyomavirus reactivation and shedding in 22. Vigil A, Davies DH, Felgner PL. Defining the humoral healthy adults: a 14-month longitudinal study. J Infect Dis immune response to infectious agents using high-density 2003; 187:1571–80. protein microarrays. Future Microbiol 2010; 5:241–51. 35. Xu FH, Xiong D, Xu YF, et  al. An epidemiological and 23. Coghill AE, Pfeiffer RM, Proietti C, et al. Identification of a molecular study of the relationship between smoking, risk novel, EBV-based antibody risk stratification signature for of nasopharyngeal carcinoma, and Epstein-Barr virus acti- early detection of nasopharyngeal carcinoma in Taiwan. vation. J Natl Cancer Inst 2012; 104:1396–410. Clin Cancer Res 2018; 24:1305–14. 36. Wang D, Liebowitz D, Kieff E. An EBV membrane protein 24. Galili T. dendextend: an R package for visualizing, adjust- expressed in immortalized lymphocytes transforms estab- ing and comparing trees of hierarchical clustering. lished rodent cells. Cell 1985; 43:831–40. Bioinformatics 2015; 31:3718–20. 37. Nutt SL, Hodgkin PD, Tarlinton DM, Corcoran LM. The 25. Ionita-Laza I, Lee S, Makarov V, Buxbaum JD, Lin X. generation of antibody-secreting plasma cells. Nat Rev Sequence kernel association tests for the combined effect Immunol 2015; 15:160–71. Variability in Abs Targeting EBV Proteins • JID 2018:217 (15 June) • 1931

Journal

The Journal of Infectious DiseasesOxford University Press

Published: May 25, 2018

Keywords: immunoglobulin g; antibodies; taiwan

There are no references for this article.