Association between HLA gene polymorphisms and mortality of COVID-19: An in silico analysis

Abstract

Introduction: The emergence of SARS-CoV-2 has caused global public health and economic crisis. Human leukocyte antigen (HLA) is a critical component of the viral antigen presentation pathway and plays essential roles in conferring differential viral susceptibility and severity of diseases. However, the association between HLA gene polymorphisms and risk for COVID-19 has not been fully elucidated. We hypothesized that HLA genotypes might impact on the differences in morbidity and mortality of COVID-19 across countries.

Methods: We conducted in silico analyses and examined an association of HLA gene polymorphisms with prevalence and mortality of COVID-19 by using publicly available databases.

Results: We found that a possible association between HLA-A*02:01 and an increased risk for COVID-19. HLA-A*02:01 had a relatively lower capacity to present SARS-CoV-2 antigens compared with other frequent HLA class I molecules, HLA-A*11:01 or HLA-A*24:02.

Conclusion: This study suggests that individuals with HLA-A*11:01 or HLA-A*24:02 genotypes may generate efficiently T-cell-mediated antiviral responses to SARS-CoV-2 compared with HLA-A*02:01. The differences in HLA genotypes may potentially alter the course of the disease and its transmission.

Keywords: COVID-19; T cell; human leukocyte antigen; pandemic; severe acute respiratory syndrome coronavirus 2.

Figure 1

Association of human leukocyte antigen (HLA) class I gene polymorphisms with prevalence and mortality of COVID‐19. (A) Differences in total confirmed cases and deaths caused by COVID‐19 between countries in which HLA‐A*02:01 is the most frequent in the population (HLA‐A*02:01 group) and countries in which other HLA genotypes (HLA‐A*11:01 and HLA‐A*24:02) are the most frequent in the population (non‐HLA‐A*02:01 group) as of April 24, 2020. (B) Differences in total confirmed cases and deaths caused by COVID‐19 between HLA‐A*02:01 and non‐HLA‐A*02:01 groups as of August 15, 2020. Permuted Brunner–Munzel test was performed

Figure 2

A relationship between deaths per 10⁶ population, confirmed COVID‐19 cases per 10⁶ population, and the most common HLA‐A genotypes in each country. Upper panels indicate that the relationship between deaths confirmed, COVID‐19 cases, and countries in which HLA‐A*02:01 is the most frequent in the population as of April 24 (left panel) and August 15, 2020 (right panel). Pink dots indicate the distribution of Australia, Belgium, Brazil, France, Germany, Mexico, Spain, Turkey, England, U.S.A., Israel, and Italy. Middle panels indicate that the relationship between deaths, confirmed COVID‐19 cases, and countries in which HLA‐A*11:01 is the most frequent in the population. Blue dots indicate the distribution of China, Singapore, and Thailand. Lower panels indicate that the relationship between deaths, confirmed COVID‐19 cases, and countries in which HLA‐A*24:02 is the most frequent in the population. Green dots indicate the distribution of India, Iran, Japan, and South Korea. Spearman's correlation coefficients and the p value were denoted as “r” and “P,” respectively. COVID‐19, coronavirus disease 2019; HLA, human leukocyte antigen

Figure 3

Number of predicted SARS‐CoV‐2‐derived T‐cell antigens. A comparison of the numbers of predicted SARS‐CoV‐2‐derived T‐cell antigens between HLA‐A*02:01, HLA‐A*11:01, or HLA‐A*24:02 is shown. Epitope prediction was carried out using the reference SARS‐CoV‐2 isolate, Wuhan‐Hu‐1. Protein sequences from the entire SARS‐CoV‐2 proteome were obtained from the SARS‐CoV‐2 reference sequence (GenBank: MN908947.3). HLA binding affinity of all possible 8–11‐mer across the entire SARS‐CoV‐2 proteome was assessed. The SARS‐CoV‐2 protein sequences were run against HLA alleles using the NetMHCpan EL 4.0 algorithm available at the IEDB (http://tools.iedb.org/mhci/) and a size range of 8–11‐mers. Top 0.5% (red), 0.5% < percentile rank ≤ 1% (blue), and 1% < percentile rank ≤ 2% (green panel) epitopes ranked based on prediction score (high to low predicted binding affinity to HLA‐A*02:01, HLA‐A*11:01, or HLA‐A*24:02) were selected and the total number of predicted 8–11‐mer SARS‐CoV‐2 T‐cell epitopes per HLA class I genotype were counted. HLA, human leukocyte antigen; IEDB, Immune Epitope Database and Analysis Resource; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus 2