HLA Class II Genes HLA-DRB1, HLA-DPB1, and HLA-DQB1 Are Associated With the Antibody Response to Inactivated Japanese Encephalitis Vaccine

Aim: The objective of this study was to evaluate the association of the human leukocyte antigen (HLA) class II genes HLA-DRB1, HLA-DPB1, and HLA-DQB1 with the humoral immune response elicited by inactivated Japanese encephalitis (JE) vaccine (IJEV). Methods: A total of 373 individuals aged 3–12 years in the Inner Mongolia Autonomous Region in China, who received two doses of IJEV at 0 and 7 days, were enrolled in the current study. Based on the individuals' specific JE virus (JEV)-neutralizing antibodies (NAbs), they were divided into a seropositive and a seronegative group. HLA-DRB1, HLA-DPB1, and HLA-DQB1 were genotyped using a sequencing-based typing method. Next, the association of the HLA class II genes and their haplotypes with antibody response was evaluated. Results: Based on NAbs, a total of 161 individuals were classified as seropositive and 212 as seronegative. DQB1*02:01 was significantly associated with JEV seropositivity (P < 0.001, OR = 0.364, 95% CI: 0.221–0.600), while DQB1*02:02 was significantly associated with JEV seronegativity (P = 5.03 × 10−6, OR = 7.341, 95% CI: 2.876–18.736). The haplotypes DRB1*07:01-DPB1*04:01-DQB1* 02:01, DRB1*15:01-DPB1*02:01-DQB1*06:02, DRB1*07:01-DQB1*02:01, and DPB1*02:01-DQB1*06:02 were very frequent in the seropositive group, while DRB1*07:01-DPB1*17:01-DQB1*02:02, DRB1*07:01-DQB1*02:02, and DPB1*17:01-DQB1*02:02 were very frequent in the seronegative group. The presence of DRB1*01:01, DRB1*04:05, DRB1*09:01, DRB1*12:02, DRB1*13:02, and DRB1*14:01 was associated with a higher geometric mean titer (GMT) of NAbs than that of DRB1*11:01 at the DRB1 locus (P < 0.05). At the DPB1 locus, the presence of DPB1*05:01 was associated with higher GMTs than that of DPB1*02:01 and DPB1*13:01 (P < 0.05), and the presence of DPB1*04:01 and DPB1*09:01 was associated with higher GMTs than that of DPB1*13:01 (P < 0.05). Conclusions: The present study suggests that HLA class II genes may influence the antibody response to IJEV.


INTRODUCTION
Japanese encephalitis (JE) is one of the most serious mosquitoborne infectious diseases, with approximately 67,900 individuals being infected by the JE virus (JEV) annually (1). Approximately 75% of these individuals are under 14 years of age, and 50% of the infections occur in China (2,3).
Vaccination is an efficient method of controlling JEV infection. Four different types of JE vaccine are available in affected countries, namely inactivated mouse brain-derived, live attenuated cell culture-derived, inactivated cell culture-derived, and genetically engineered live attenuated chimeric vaccine. The Vero cell-derived inactivated JE vaccine (IJEV) has been widely used in China, Japan, the US, Europe, Canada, Australia, Hong Kong, Switzerland, and India (4)(5)(6). A JEV-neutralizing antibody (NAb) titer of at least 10 has been established as a correlate for protection against JEV, while positive serum conversion rate and geometric mean titer (GMT) have been used as alternative markers of efficacy of JE vaccines (7,8). After immunization with the JE vaccine, the positive serum conversion rate ranges from 60 to 100% (1,9). Vaccine efficacy may be influenced by factors such as the type of vaccine and the vaccinated person's age, gender, and nutritional status (6,10). Several studies have reported that the efficacy of attenuated JE vaccine has reached 85-99.26% in Chinese, South Korean, and Nepalese children; however, it exhibited only 67.2% efficacy in Indians after primary immunization (11)(12)(13)(14). These results indicate that different genetic backgrounds of hosts could play an important role in the efficacy of JE vaccines.
As one of the key immune gene complexes, the human leukocyte antigen (HLA) genes play an important role in the adaptive immune response to viruses and vaccines. HLA molecules are divided into three classes: class I, II, and III. Among them, HLA class II molecules (HLA-DR, -DQ, and -DP) bind to extracellular viral antigen peptides and display them on the surface of antigen-presenting cells to CD4 + cells to stimulate their multiplication, which, in turn, stimulate antibodyproducing B cells to produce specific antibodies (15,16). HLA genes exhibit extraordinary polymorphisms, and different alleles can affect the peptide-binding properties of the HLA molecular pocket, which subsequently influences the immune response to a vaccine. In 2005 and 2006, Ovsyannikova et al. (17,18) observed that HLA-DPB1 * 03:01, HLA-DPB1 * 04:01, and HLA-DPB1 * 15:01 are associated with rubella vaccineinduced antibodies. On the other hand, the HLA-DRB1 * 15/16-DQB1 * 06-DPB1 * 13 haplotype has been associated with high levels of measles antibody response, but low levels of rubella antibody response.
In order to evaluate the association of HLA class II genes HLA-DRB1, HLA-DPB1, and HLA-DQB1 and JEV-NAbs with the humoral immune response to IJEV, this study examined Mongolian Chinese individuals who had been administered IJEV.

Subjects and Vaccination
A randomized, double-blinded, positive-control, non-inferiority IJEV trial was implemented in the Inner Mongolia Autonomous Region of China from August 2012 to September 2013. The IJEV (lot: 20101201) was manufactured in a GMP-accredited facility of the Institute of Medical Biology at the Chinese Academy of Medical Sciences (IMBCAMS) and verified by the National Institute for Food and Drug Control (China, approval no. 2010L02035). Briefly, JEV P3 strains were grown on Vero cell microcarriers in a 75 L bioreactor. The virus suspension was harvested, inactivated with ultra-concentrated formalin, and purified by Sepharose 6FF and DEAE Sepharose FF. The resulting vaccine contained 0.5 mL per dose with ≥ 0.6 IU/mL JEV antigens. The clinical study procedure was approved by the Ethics Committee of the Inner Mongolia Autonomous Region Center for Disease Control and Prevention. The IJEV control (lot: 201012B02-1) was manufactured by Liaoning Chengda Biotechnology (Shenyang, China), containing the same concentration of antigens as the vaccine made by IMBCAMS. A total of 1,200 individuals aged 8 months−12 years in the Inner Mongolia Autonomous Region were enrolled to receive two doses of IJEV at 0 and 7 days. They were vaccinated with either the IJEV made by IMBCAMS or the IJEV control at a 1:1 ratio. The inclusion criteria were that the individual was in good health, was not infected by JEV, had not been inoculated with other vaccines within 7 days, and had not been inoculated with attenuated JE vaccine within 1 month. The peripheral blood samples were collected before vaccine administration and 30 days after the second dose received for the detection of neutralization antibody. Considering the limited blood sample volume and the consistency of the test, only individuals of 3-12 years of age, who were negative for NAbs before vaccination, were selected for further HLA genotyping. Finally, after vaccination, 212 individuals negative for NAbs were included in the seronegative group, and 161 individuals positive for NAbs were randomly selected and included in the seropositive group.

Japanese Encephalitis Vaccine Neutralization Antibody Detection
JEV-specific NAbs were determined by the National Institute for Food and Drug Control using the 50% plaque-reduction neutralization test according to the requirement of the Pharmacopeia of the People's Republic of China (19). Briefly, BHK-21 cells were initially inoculated at 10 6 cells/well in 24-well tissue culture plates and propagated for 48 h at 37 • C in a CO 2 incubator. The serum samples were inactivated for 30 min in a 56 • C water bath, diluted 10-fold, and then serially diluted 2-fold from 1:10 to 1:1280 in Minimum Essential Medium (GIBCO, Grand Island, NY, USA) containing 2% fetal bovine serum and 1% penicillin/streptomycin. The diluted serum and the positive conference serum were mixed with an equal volume of diluted challenge virus (P3 strain, lot 20151102, 500 PFU/mL). The suspensions were kept in a 37 • C water bath for 30 min. Afterwards, 0.1 mL aliquots of the virus-serum mixtures were dispensed separately into each well of the 24-well microplates with the BHK-21 cells. The cells were overlaid with medium containing 1% methylcellulose. The cells in the wells were stained after inoculation at 37 • C for 5 days in a 5% CO 2 incubator, and the plaques were counted. The NAb titer was defined as the reciprocal value of the last serum dilution that showed 50% or greater plaque reduction compared with the plaque counts in the virus-only control wells. NAbs at 50% plaque reduction neutralization titer (PRNT50) <10 or PRNT 50 increased <4-fold after vaccination were considered as negative seroconversion, while PRNT 50 >10, or at least a 4-fold increase after vaccination, was considered to be positive seroconversion. The antibody titers were determined by calculating the GMT as follows: GMT = Log−1 [(LogX 1 + LogX 2 + . . . LogX n )/n].
HLA-DRB1, HLA-DPB1, and HLA-DQB1 Genotyping Genomic DNA was extracted from peripheral lymphocytes using the QIAamp Blood Kit (Qiagen, Hilden, Germany). HLA-DRB1, HLA-DPB1, and HLA-DQB1 were genotyped using a high-resolution sequencing-based typing method (Applied Biosystems, Foster City, CA, USA). Briefly, exons 2 and 3 of DRB1 and DQB1 as well as all exons of DPB1 were amplified, and the PCR products were sequenced using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Finally, the sequence was analyzed with the 3730xl DNA Analyzer (Applied Biosystems), and the HLA alleles were identified using the SBTengine (Applied Biosystems).

Statistical Analysis
The differences in age and sex between the seropositive and seronegative group were determined using Student's t-test or a χ 2 test. The HLA-DRB1, -DPB1 and -DQB1 allele frequencies were calculated using the PyPop or PyHLA software based on the genotyping results (20)(21)(22). The Hardy-Weinberg equilibrium was assessed using the Guo and Thompson method (23). The haplotypes were constructed based on the genotyping results using the expectation-maximization algorithm (20)(21)(22). The χ² test was used to determine differences in allele and haplotype frequencies between the seropositive and seronegative group. The odds ratios (ORs) and associated 95% confidence intervals (CIs) were also calculated for allele-specific risks. False discovery rate (FDR) correction was used for the multiple comparisons (20). For each gene, the amino acid sequences for all alleles were aligned together. If there was more than one amino acid at one position, a test was performed for each amino acid to examine whether it is distributed differently in the seropositive and seronegative group using PyHLA software (20). Fisher's exact test was used to analyze the association, and the odds ratio was calculated with Haldane's correction of Woolf 's method (20). The association between HLA-DRB1, -DPB1, -DQB1 alleles and antibody levels was analyzed through the analysis of variance using GraphPad Prism 7.0. P-values of < 0.05 were considered statistically significant. Table 1 lists the characteristics of the enrolled subjects. They were randomly selected from the two vaccination groups, which had no difference in seroconversion or GMT (P > 0.05). All subjects were negative for NAbs before vaccination. After vaccination, 161 individuals with PRNT 50 > 10 were included in the seropositive group, while 212 individuals with PRNT 50 < 10 were included in the seronegative group. Table 2 shows that there were no age or gender differences between the seropositive and seronegative group (P > 0.05). In addition, there was no difference in NAbs titers according to age and gender (P > 0.05) in the seropositive group (data not shown). Moreover, there was no difference in HLA allele distribution according to gender (P > 0.05).

DISCUSSION
Vaccines are one of the greatest advances in controlling infectious diseases in the past 300 years. The humoral immune response induced by a vaccine produces NAbs, so the seroconversion rate and GMT are widely used to evaluate vaccine efficacy. In the present study, we examined the association of HLA class II genes with the IJEV antibody response to reveal the role of the genetic variation in the HLA class II genes in the IJEV immune response.
HLA class II molecules present viral antigens in the form of peptides derived from the extracellular processing of vaccine peptides, which plays an important role in the humoral immune response to vaccines (24,25). In an inactivated vaccine, the extracellular vaccine antigens are degraded into smaller peptides and integrated with the HLA class II molecule to constitute the HLA class II peptide complex, which plays a major role in stimulating the differentiation of CD4 + T cells into Th1 and Th2 cells; in turn, the Th2 cells can interact with B cells to promote differentiation into antibody-secreting plasma cells, thus secreting a specific antibody against the vaccine antigen (18).
In addition to DQB1 * 02:01 and DQB1 * 02:02, other HLA class II genes are reportedly associated with the antibody response to vaccines. For example, Jafarzadeh et al. (27) reported that DRB1 * 01:01, DRB1 * 13:01, DRB1:15:01, and DQB1 * 04:01 were positively associated with HBV antibody response, while DRB1 * 03:01, DRB1 * 07:01, and DQB1 * 02:01 were negatively associated. However, other than DQB1 * 02:01, no HLA alleles have been associated with IJEV in the present study. One of the reasons for different HLA alleles being associated with antibody responses could be a distinct immune response to different vaccines or pathogens. Most previous association studies have been performed with attenuated vaccines (mumps, measles, rubella vaccine, etc.) or virus-like particle-based vaccines (HBV), which could induce both HLA class I-and class IImediated immune response to generate an immune response. However, in an inactivated vaccine like IJEV, the humoral immune response mediated by HLA class II molecules would be key in generating JEV-NAbs. Thus, the difference in the immune response mechanism between the inactivated vaccine and the attenuated vaccine may be caused by the association with different HLA alleles (18,38). Another reason could be a population-specific difference in HLA distribution, as with HLA genes and their motifs, even if the populations were administered   the same vaccine. For example, in 2015, Jafarzadeh et al. (27) reported that non-responsiveness to HBV is associated with HLA-A1, -B15, and -B40 in Indians, HLA-A1, -A2, and -B8 in Caucasians, HLA-B54 in Chinese, and HLA-A10 and -Cw4  Southwest Asia, and rare in North Africa and South America (39). To the best of our knowledge, this HLA allele diversity was generated in the long evolutionary interaction between hosts and pathogens, which makes it encode adequate products to generate immune responses against different pathogens. Thus, different HLA alleles were formed as an outcome of specific pathogen infections and are therefore associated with different infectious diseases (40)(41)(42). As such, the mechanism of the immune response to different vaccine antigens could vary based on different vaccines. In addition to the seroconversion rate, GMT is an important factor in evaluating vaccine efficacy. In the present study, we evaluated the relationship between GMTs and HLA alleles. Interestingly, we found that the HLA alleles associated with an antibody response were different from the HLA alleles associated with GMTs. These results indicate that HLA alleles have different roles in the host immune response.
In summary, we investigated the association between HLA class II genes and antibody response after IJEV administration, determining that HLA-DQB1 * 02:01 and HLA-DQB1 * 02:02 were associated with NAb seroconversion. Furthermore, certain HLA-DRB1 and -DPB1 alleles were associated with higher GMTs than others. The present study suggests that HLA class II genes may influence the antibody response to IJEV. However, as only 161 individuals were examined in the present study, future studies should comprehensively analyze larger samples.

ETHICS STATEMENT
This study was carried out in accordance with the recommendations of the ethical standards of the Responsible Committee on Human Experimentation of the Ethics Committee of the Guangxi Centre for Disease Control and Prevention, with written informed consent obtained from all subjects in accordance with the Declaration of Helsinki. The protocol was approved by the Inner Mongolia Autonomous Region Center for Disease Control and Prevention.