The protective association of HLA-B*52:01, HLA-C*12:02, and DQB1*06:01 alleles with severe acute hepatitis of unknown origin in Japanese children

Outbreak of indeterminate acute liver failure in children with adenoviraemia – Not a new diseaseJournal of HepatologyVol. 79Issue 1PreviewIn the year 2022, an outbreak of indeterminate acute hepatitis and indeterminate paediatric acute liver failure (ID-PALF) in association with adenoviraemia in immunocompetent children was reported in the UK. We postulate that this association is not a new disease in immunocompetent children. Full-Text PDF We read with great interest the recent article on the subject of an outbreak of indeterminate acute hepatitis and indeterminate pediatric acute liver failure in the United Kingdom from Jagadisan et al [1]Jagadisan B. Verma A. Deheragoda M. Deep A. Grammatikopoulos T. Sudhanva M. Bansal S. Hadzic N. Vimalesvaran S. Heaton N. Dhawan A. Outbreak of indeterminate acute liver failure in children with adenoviraemia - Not a new disease.J Hepatol. 2023 Feb 21; (00101-0): S0168-8278Google Scholar. Their data do not support direct viral cytotoxicity and concluded that this disease is probably mediated by immunological injury directed toward adenovirus (ADV) infection and/or adeno-associated virus-2(AAV-2). Three recent case-control metagenomic studies of unexplained acute hepatitis in children also suggested the association with and presence of AAV2 under coinfection with other viruses such as ADV 2Ho A. Orton R. Tayler R. Asamaphan P. Herder V. Davis C. et al.Adeno-associated virus 2 infection in children with non-A-E hepatitis.Nature. 2023 Mar 30; https://doi.org/10.1038/s41586-023-05948-2Crossref Scopus (3) Google Scholar, 3Servellita V. Gonzalez A.S. Lamson D.M. Foresythe A. Huh H.J. Bazinet A.L. et al.Adeno-associated virus type 2 in US children with acute severe hepatitis.Nature. 2023 Mar 30; https://doi.org/10.1038/s41586-023-05949-1Crossref Scopus (3) Google Scholar, 4Morfopoulou S. Buddle S. Montaguth O.E.T. Atkinson L. Guerra-Assunção J.A. Marjaneh M.M. et al.Genomic investigations of unexplained acute hepatitis in children.Nature. 2023 Mar 30; https://doi.org/10.1038/s41586-023-06003-wCrossref Scopus (4) Google Scholar, but AAV2 has not been thought to be directly linked to disease, and the possibility of immune response effects was suggested in these reports. Additionally, Ho et al. suggested that the class II human leukocyte antigen (HLA)-DRB1*04:01 allele was identified in 93% in comparison to a background frequency of 15.6% in Scottish blood donors, suggesting increased susceptibility in pediatric non A-E hepatitis cases [2]Ho A. Orton R. Tayler R. Asamaphan P. Herder V. Davis C. et al.Adeno-associated virus 2 infection in children with non-A-E hepatitis.Nature. 2023 Mar 30; https://doi.org/10.1038/s41586-023-05948-2Crossref Scopus (3) Google Scholar. To investigate the possibility that HLA is associated with hepatitis, we reviewed genomic 4-digit HLA typing results in children of <10 years of age (n=25) who underwent liver transplantation for acute hepatitis of unknown origin (AHUO) at our institution between 2000 and 2021. This study was approved by the Institutional Review Board of Kyoto University Hospital, Japan (R-1473-6). Patient characteristics and HLA typing data are shown in Supplemental Table 1. All study patients had a prothrombin time (PT) of ≤40% or an international normal ratio of PT (PT-INR) of ≥1.5 with AST or ALT values of > 500 U/L within 8 weeks of the onset of initial symptoms and hepatitis was diagnosed based on the pathological examination of the explanted native liver. The preoperative examination showed no evidence of active A-to E-type hepatitis virus, herpes virus, cytomegalovirus, or Epstein‒Barr virus infection, and the possibility of autoimmune-mediated, drug-induced, and toxic liver injury was ruled out; thus, the cause was considered unknown. The frequencies of allele occurrence for the five loci of HLA-A*, HLA-B*, HLA-C*, HLA-DQB1*, and HLA-DRB1* in the AHUO cases were compared to those of previously reported Japanese healthy controls (n=873) [5]Ujiie H. Muramatsu K. Mushiroda T. Ozeki T. Miyoshi H. Iwata H. et al.HLA-DQB1*03:01 as a Biomarker for Genetic Susceptibility to Bullous Pemphigoid Induced by DPP-4 Inhibitors.J Invest Dermatol. 2018 May; 138: 1201-1204Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar. In the HLA-B locus, there was no occurrence of HLA-B*52:01 in the AUHO group, and in the HLA-C locus, there was also no expression of HLA-C*12:02. Moreover, at the HLA-DQB1 locus, the incidence of HLA-DQB1*06:01 in AUHO patients was found to have a significantly lower incidence than in healthy controls (Table 1a and Supplemental Table 2). On the other hand, HLA-DRB1*04:01 was present in only 1 out of 50 alleles in our study cohort.Table 1aSummary of HLA alleles frequency of AHUO case and control.HLA BCase (n=50)Control (n=1746)AlelleNo of CasesFrequency(%)No of CasesFrequency(%)PPcOdds ratio95%CIB*52:0100.024213.91.14 x10-33.99 x10-20,8610.845-0.878HLA CCaseControlAlelleNo of CasesFrequency(%)No of CasesFrequency(%)PPcOdds ratio95%CIC*12:0200.024514.01.16 x10-32.08 x10-20,8600.844-0.876HLA DQB1CaseControlAlelleNo of CasesFrequency(%)No of CasesFrequency(%)PPcOdds ratio95%CIDQB1*06:0136.040323.12.98x10-34.76 x10-20,2130.066-0.687HLA DRB1CaseControlAlelleNo of CasesFrequency(%)No of CasesFrequency(%)PPcOdds ratio95%CIDRB1*15:0212.024113.80.010.32-- Open table in a new tab In this background, we performed multiple alignments of HLA-B, HLA-C, and HLA-DQB1 proteins to determine whether or not there were amino acid residues that were significantly associated with AHUO. Amino acid sequences were obtained for each HLA-B, HLA-C, and HLA-DQB1 allele for four-digit resolution from the IMGT database (http://www.ebi.ac.uk/ipd/imgt/hla/). The frequency of amino acid variants in acute hepatitis cases was compared with that of healthy controls for 52 positions for HLA-B*, 50 positions for HLA-C*, and 44 positions for HLA-DQB1*. Five amino acid residues comprising HLA-B* sequences were found to be significantly associated with AHUO (Table 1b and Supplemental Table 3), and one amino acid residue comprising HLA-C* was found to be associated with AHUO (Table 1b and Supplemental Table 4). No significant differences in HLA-DQB1* residues were found in our study (Supplemental Table 5).Table 1bSummary of amino acid residues significantly associated with AHUO.HLA-B proteinCase (n=50)Control (n=1746)PositionAmino acidFrequency(%)Frequency(%)PPcOdds ratio95%CI131Arg52.027.31.27 x10-46.60 x10-32,8821.639-5.069163Glu58.035.28.98 x10-44.67 x10-22,5461.440-4.503282Val72.046.23.16 x10-41.64 x10-22,9921.602-5.587305Ala72.046.23.16 x10-41.64x10-22,9921.602-5.587325Cys68.041.82.14 x10-41.13 x10-22,9651.624-5.411HLA C proteinCase (n=50)Control (n=1746)PositionAmino acidFrequency(%)Frequency(%)PPcOdds ratio95%CI156Trp0.015.17.29 x10-43.65 x10-20,8490.833-0.866The association between AHUO and HLA alleles was investigated using 2×2 contingency tables and tested for statistical significance using the χ2 test or Fisher’s exact test for alleles with a frequency less than 1% in either the cases or the controls. Crude P values (P) were calculated, and significance levels were set according to Bonferroni’s correction. The corrected P value (Pc) was calculated by multiplying the P value by the number of HLA alleles compared and by the number of amino acid positions compared. A statistically significant difference was determined when the Pc was less than 0.05. For factors with statistically significant differences, odds ratios (ORs) and 95% confidence intervals (CIs) were also calculated. All statistical analyses were conducted using the SPSS software program (version 27.0, IBM, Armonk, NY). Open table in a new tab The association between AHUO and HLA alleles was investigated using 2×2 contingency tables and tested for statistical significance using the χ2 test or Fisher’s exact test for alleles with a frequency less than 1% in either the cases or the controls. Crude P values (P) were calculated, and significance levels were set according to Bonferroni’s correction. The corrected P value (Pc) was calculated by multiplying the P value by the number of HLA alleles compared and by the number of amino acid positions compared. A statistically significant difference was determined when the Pc was less than 0.05. For factors with statistically significant differences, odds ratios (ORs) and 95% confidence intervals (CIs) were also calculated. All statistical analyses were conducted using the SPSS software program (version 27.0, IBM, Armonk, NY). Finally, we confirmed the HLA allele and haplotype frequencies of healthy Japanese individuals reported by the HLA Institute (https://hla.or.jp/med/frequency_search/en/haplo/). HLA-B*52:01 was found in 11.153% of healthy Japanese individuals, ranking first in frequency. HLA-C*12:02 was found in 11.293% of normal Japanese individuals, ranking 5th in frequency. HLA-DQB1*06:01 was also the most common allele, found in 19.084% of healthy Japanese individuals. The HLA-B*52:01 and HLA-C*12:02 alleles are known to be in strong linkage disequilibrium in Japanese individuals [6]Nakajima F. Nakamura J. Yokota T. Analysis of HLA haplotypes in Japanese, using high resolution allele typing.MHC. 2001; 8: 1-32Crossref Google Scholar, and the HLA-B*52:01-C*12:02 haplotype is the most common haplotype. However, our study population did not include any cases with this haplotype. Taken together, these results indicate that HLA-B*52:01, HLA-C*12:02, and HLA-DQB1*06:01, which are major alleles in Japanese individuals, may have protective effects against severe AHUO morbidity that leads to liver transplantation. The small sample size was a limitation of the present study. Although we found several amino acid residues in the HLA proteins associated with this effect, it was difficult to identify the residues that were specifically involved. Moreover, we were unable to investigate the infection status of ADV and AAV2 in our cases. Therefore, the AHUO in this study differs from that of AHUO in a recent outbreak. This may be the reason why our results differ from those of Ho et al. However, we believe that further epidemiological studies including HLA typing of AHUO cases on a global scale will contribute to the elucidation of the pathogenesis of this disease. The authors declare no conflicts of interest in association with the present study. The authors have no financial relationships relevant to this article to disclose.