Influence of HLA-G 3' Untranslated Region Haplotypes and SNP +3422 Gene Variants as Host Genetic Factors on the Outcomes of SARS-CoV-2 Infection During Acute and Post-Acute Phases in a German Cohort

Abstract

HLA-G, an important immune-checkpoint (IC) molecule that exerts inhibitory signalling on immune effector cells, has been suggested to represent a key player in regulating the immune response to Severe Acute Respiratory Syndrome Coronavirus Type 2 (SARS-CoV-2). Since specific single-nucleotide polymorphisms (SNP) in the HLA-G 3'untranslated region (UTR), which arrange as haplotypes, are crucial for the regulation of HLA-G expression, we analysed the contribution of these genetic variants as host factors in SARS-CoV-2 infection during acute and post-acute phases. HLA-G gene polymorphisms in the 3'UTR were investigated by sequencing in an unvaccinated Coronavirus Disease 2019 (COVID-19) cohort during acute SARS-CoV-2 infection (N = 505) and in the post-acute phase (N = 253). The HLA-G 3'UTR haplotype known as UTR-3 (p = 0.002) and the variant rs17875408 (also known as +3422) T variant (p = 0.004) are independent prognostic risk factors for fatal COVID-19. The +3422T variant (p = 0.006) predicted also the early loss of neutralising SARS-CoV-2 antibodies. In addition, the HLA-G 3'UTR haplotype UTR-7 (p = 0.023) emerged as an independent prognostic factor for increased susceptibility to Long-COVID symptoms after SARS-CoV-2 infection. Our study highlights that due to the variability of the 3'UTR genetic background, HLA-G has the potential to contribute to the progression of SARS-CoV-2 infection, extending to the development of Long-COVID symptoms, despite the likely alterations in the microenvironment and associated HLA-G-specific regulatory elements over the course of the disease. By spotlighting HLA-G, the importance of the genetic background of IC and their pivotal role in modulating immune responses during and after COVID-19 are emphasised.

Keywords: COVID‐19; HLA‐G; HLA‐G 3′UTR; SARS‐CoV‐2; long‐COVID.

FIGURE 1

Linkage disequilibrium (LD) of 10 analysed polymorphic variations in or closed to the HLA‐G 3′UTR. The image was generated in the Haploview 4.2. D′ pairwise correlation between polymorphic sites. Areas in red indicated strong LD (LOD ≥ 2, D′ = 1), areas in light red indicate moderate LD (LOD ≥ 2, D′ < 1), areas in blue indicate LD with a lack of statistical evidence (LOD ≤ 2, D′ = 1), while areas in white indicate no LD (LOD ≤ 2, D′ < 1). D′ values different from 1.00 are represented inside the squares as percentages. LOD, log of the odds; D′, pairwise correlation between single‐nucleotide polymporphisms. (For interpretation of the references to colour in the figure legend, the reader is referred to the web version of this article).

FIGURE 2

Association between the SNP +3422C/T and the HLA‐G 3′UTR haplotypes: The +3422C variant was associated with the 7 UTR haplotypes found in our patient cohort (UTR‐1, UTR‐2, UTR‐4, UTR‐7, UTR‐10, UTR‐17 and UTR‐18), whereas both the +3422C variant and the T variant were associated with UTR5 and 6. Notably, UTR 3 and the rare UTR‐13 were exclusively linked to the +3422T variant.

FIGURE 3

Association between HLA‐G 3′UTR‐1, ‐7 or ‐3 haplotypes or the SNP +3422T allele and CD3+ or CD8+ T‐cell counts and the CD4/CD8 ratio. During acute COVID‐19, specific HLA‐G 3′UTR haplotypes were not directly linked to T‐cell counts. After COVID‐19, reduced CD3+ and CD8+ T‐cell counts were seen in patients carrying UTR‐1 or UTR‐7, but only UTR‐7 carriers exhibited a significant increase in CD4/CD8 ratio. Interestingly, among patients with acute COVID‐19, UTR‐3 or +3422T carriers exhibited higher CD4/CD8 ratios, but this finding was reversed after COVID‐19, a finding indicating changing immune profiles across the disease phases. Solid line indicates median value of cell count. Patient groups are distinguished by colour codes: Red for patients during COVID‐19 and blue for patients after COVID‐19 (here, post‐COVID‐19); p: Significance difference between carriers and non‐carriers of a certain HLA‐G 3′UTR haplotype or SNP allele by by Mann–Whitney tests.

FIGURE 4

Association of UTR‐3 Haplotype and SNP +3422T carriers with 30‐day mortality rates among COVID‐19 patients. Kaplan–Meier plot analysis combined with log‐rank test revealed a reduced 30‐day OS probability for acute COVID‐19 patients carrying the (A) UTR‐3 haplotype of HLA‐G or (B) SNP +3422T variant. Tables under Kaplan–Meier plots show corresponding numbers at risk. Patient groups are distinguished by colour codes as follows: (A) blue for UTR‐3 negative and red for UTR‐3 positive patients; (B) blue for SNP +3422CC carriers and red for SNP +3422TT/CT carriers.

FIGURE 5

Forest plot of risk factors for 30‐day mortality of COVID‐19 patients. The forest plots show the results of multivariate analyses of the following covariates for 30‐day mortality: Diabetes, sex, age ≥ 60 years of age, obesity, hypertension, cardiovascular disease, immunosuppression, and (A) HLA‐G 3′UTR carrier status or (B) SNP +3422T carrier status. 95% CI, 95% confidence interval; HR, hazard ratio.

FIGURE 6

Association SNP +3422T carrier status with the early loss of SARS‐CoV‐2 neutralising antibodies. (A) Kaplan–Meier plot analysis combined with log‐rank test revealed a significant reduced detection of anti‐SARS‐CoV‐2 for patients carrying the SNP +3422T variant post infection. Patient groups are distinguished by colour codes: Blue for SNP +3422CC carriers and red for SNP +3422 TT/CT carriers. Tables under Kaplan–Meier plots show corresponding numbers at risk. (B) Forest plot of risk factors for loss of neutralising anti‐SARS‐CoV‐2 antibodies in COVID‐19 patients. The forest plots show the results of multivariate analyses of the following covariates for neutralising antibody loss: Sex, age ≥ 60 years of age, obesity, hypertension, cardiovascular disease, immunosuppression and SNP +3422T carrier status. 95% CI, 95% confidence interval; HR, hazard ratio.

FIGURE 7

(A) Association of HLA‐G UTR‐7 carrier status with the presence of Long‐COVID‐symptoms. Patient groups are distinguished by colour codes: Blue for UTR‐7 negative and red for UTR‐7 positive patients (B) Forest plot of risk factors for Long‐COVID‐ symptoms. The forest plots show the results of multivariate analyses of the following covariates for Long‐COVID symptoms: Diabetes, sex, age ≥ 60 years of age, obesity, hypertension, cardiovascular disease, immunosuppression and HLA‐G UTR‐7 carrier status. 95% CI, 95% confidence interval; HR, hazard ratio.