Having genetic risk variants in the ABO gene might significantly increase the chances of developing COVID-19

0
695

Having genetic risk variants in the ABO gene might significantly increase the chances of developing COVID-19, and other genes may also increase COVID-19 risk, according to research presented at the ATS 2021 International Conference.

Much about COVID-19 remains a medical mystery, including whether certain genes place individuals at greater risk of contracting the SARS-CoV-2 virus, which causes COVID-19. Ana Hernandez Cordero, Ph.D., postdoctoral fellow with the Centre for Heart Lung Innovation, University of British Columbia, and colleagues used integrative genomics combined with proteomics to identify these genes.

Genomic research identifies specific genes that may play a role in biological processes such as the development of disease, while proteomics does the same for proteins. Researchers can get a fuller picture of disease processes by integrating tools to investigate both.

“DNA is a big, complex molecule and so, genetic associations alone cannot pinpoint the exact gene responsible for COVID-19,” said Dr. Hernandez.

“However, by combining COVID-19 genetic information with gene expression and proteomic datasets, we can figure out which genes are driving the relationship with COVID-19.”

The researchers combined genetic information with an examination of lung gene expression to identify genetic variants that were controlling gene expression in the lung that were responsible for COVID-19. The researchers identified specific genes’ markers that share their effects on gene expression and protein levels with COVID-19 susceptibility.

For the analysis, they used bioinformatics to integrate: (1) a genomic dataset obtained from patients who were infected with SARS-CoV-2 as well as non-infected individuals (controls); (2) lung and blood tissue gene expression datasets from clinical populations (non-COVID-19); and (3) a proteome dataset obtained from blood donors (non-COVID-19).

By doing this, they found that several genes responsible for the immune system’s response to COVID-19 are also involved in COVID-19 susceptibility. What they discovered was supported by the findings of previous research.

Looking for candidate genes in blood proteins, they were able to go one step further in connecting the effects of genes to susceptibility to COVID-19. Blood proteomics can also help identify markers in the blood that can be easily measured to indicate disease status, and potentially, to monitor the disease.

“By harnessing the power of genomic information, we identified genes that are related to COVID-19,” said Dr. Hernandez. “In particular, we found that the ABO gene is a significant risk factor for COVID-19. Of particular note was the relationship between the blood group ABO and COVID-19 risk. We showed that the relationship is not just an association but causal.”

In addition to the ABO gene, Dr. Hernandez and colleagues found that people carrying certain genetic variants for SLC6A20, ERMP1, FCER1G and CA11 have a significantly higher risk of contracting COVID-19.

“These individuals should use extreme caution during the pandemic. These genes may also prove to be good markers for disease as well as potential drug targets.”

Several of the genes identified in the researchers’ analysis have already been linked with respiratory diseases. For example, ERMP1 has been linked to asthma. CA11 may also elevate COVID-19 risk for people with diabetes.

Genetic associations for COVID-19 and gene and protein expression were combined using integrative genomics (IG). IG aims to identify mechanisms (for example: gene expression levels) that connect the effects of the genetic code to a complex disease.

These methods, although complex, are also fast and their outcomes can help researchers to prioritize candidate genes for in vitro (in the lab) and in vivo (in living organisms) testing.

Dr. Hernandez added, “Our research has progressed since the time that we first conducted this analysis. We have now identified even more interesting candidates for COVID-19 such as IL10RB, IFNAR2 and OAS1.

These genes have been linked to severe COVID-19. Their role in the immune response to viral infections and mounting evidence suggest that these candidates and their role in COVID-19 should be further investigated.”


The clinical presentation of SARS-CoV-2 infection in humans can range from severe respiratory failure to disease that is very mild or without symptoms1. Although hyperactivation of various cellular components of the immune system have been observed in patients with severe COVID-19 illness2,3, the host genetic factors that determine susceptibility to severe COVID-19 illness are not well understood. Genome-wide association studies (GWAS) addressing this question have identified a number of genetic variants associated with COVID-19 susceptibility and severity4–7.

However, their target genes and the immune cell types where their effects are most prominent are not known. The DICE database of immune cell gene expression, epigenomics and expression quantitative trait loci (eQTLs) (http://dice-database.org) was established to precisely answer these questions as well as to help narrow down functional variants in dense haploblocks linked to disease susceptibility8,9. Here, we utilize the DICE database and 3D cis-interactome maps to provide a list of target genes and cell types most affected by genetic variants linked to severity of COVID-19 illness.

We systematically assessed the effects of 679 COVID-19-risk variants (defined by the COVID-19 Host Genetics Initiative; release 4 from 20 October 20204; GWAS association P value < 5×10−8) on gene expression in 13 different immune cell types and 2 activation conditions (Supplementary Tables 1 and 2).

The expression of 11 protein-coding genes and 1 non-coding RNA (referred here as eGenes) was associated with the genetic variants linked to severe COVID-19 illness requiring hospitalization (Fig. 1a and Extended data Fig. 1). Notably, the majority of the eGenes associated with severe COVID-19 illness showed prominent effects in specific immune cell types (Fig. 1b).

Applying a more liberal GWAS association P value threshold of 1×10−5, we identified 41 additional eGenes that were associated with genetic variants non-significantly linked to severe COVID-19 illness (Supplementary Table 3). Some of these variants are likely to reach statistical significance (GWAS association P value < 5×10−8) as more donors with severe COVID-19 illness are included in the subsequent analysis phases.

An external file that holds a picture, illustration, etc.
Object name is nihpp-2020.12.01.407429-f0001.jpg
Figure 1.
COVID-19-risk variants with eQTL activity in human immune cell types.
(a) Genes and cell types influenced by GWAS SNPs linked to severe COVID-19 illness requiring hospitalization. For each cell type (columns), the adj. association P value for the peak GWAS cis-eQTL associated with the indicated eGenes (rows) is shown. (b) Fractions of GWAS eGenes linked to severe COVID-19 illness identified in varying numbers of cell types.

Genetic variants in the 3p21.31 locus have been linked to severity of COVID-19 illness by multiple GWAS studies4–7. These severe COVID-19-risk variants are inherited as a dense >300 kb haploblock that was shown to have entered the human population >50,000 years ago from Neanderthals10.

Populations with higher frequency of this Neanderthal-origin COVID-19-risk haplotype have higher risk of severe COVID-19 illness10. The severe COVID-19-risk variants in the 3p21.31 locus contains 17 known protein-coding genes (Fig. 2a), including SLC6A20, LZTFL1, CCR9, FYCO1, CXCR6, XCR1, CCR1, CCR3, CCR2 and CCR5.

Among these genes CCR2 (encoding for C-C type chemokine receptor, also known as monocyte chemoattractant protein 1 receptor) expression showed the strongest association with 3p21.31 severe COVID-19-risk variants identified by GWAS studies4 (Fig. 2a).

Importantly, the risk variants were associated with expression of CCR2 in certain CD4+ memory T cell subsets (TH17, TH1/17) and classical monocytes (Fig. 1a, ​,2a2a and Supplementary Tables 1 and 2). Although the CCR2 promoter did not directly overlap the risk variants, we found that the peak eQTL (rs6808074), located >200kb upstream, directly overlapped an intergenic cis-regulatory region that specifically interacted (H3K27ac HiChIP) with CCR2 promoter in classical monocytes (Fig. 2b).

These findings suggested that the severe COVID-19-risk variant (rs6808074) likely perturbs the function of a distal enhancer that is important for regulating CCR2 expression in monocytes. Thus, genetic evidence points to an important role of CCR2 pathway in the pathogenesis of COVID-19. Patients with severe COVID-19 illness were shown to have increased CCR2 expression in circulating monocytes as well as very high levels of CCR2 ligand (CCL2) in bronchoalveolar lavage fluid11, supporting the hypothesis that excessive recruitment of CCR2-expressing monocytes may drive pathogenic lung inflammation in COVID-19.

An external file that holds a picture, illustration, etc.
Object name is nihpp-2020.12.01.407429-f0002.jpg
Figure 2.
Promoter interacting distal cis-eQTLs regulate CCR2 promoter activity specifically in classical monocytes.
(a) Genes and cell types most susceptible to the effects of severe COVID-19-risk variants (all with GWAS association P value < 5×10−8) in the 3p21.31 locus. The adj. association P value for the peak GWAS cis-eQTL associated with the indicated eGenes in each cell type and activation condition is shown (left). Right, mean expression levels (TPM) of CCR2 gene in classical monocytes (* adj. association P value: 5.94×10−3), from subjects (n=91) categorized based on the genotype at the indicated GWAS cis-eQTL (each symbol represents an individual subject; adj. association P value calculated by Benjamini-Hochberg method). (b) WashU Epigenome browser tracks for the 3p21.31 locus, severe COVID-19-risk associated GWAS variants (based on GWAS study, see Extended Data Figure 1a; red color bars are lead GWAS SNPs, black color bars are SNPs in linkage disequilibrium), adj. association P value for GWAS cis-eQTLs associated with expression of CCR2 expression in classical monocytes (dark red) and naïve B cells (green), recombination rate tracks27,28, H3K27ac ChIP-seq tracks, and H3K27ac HiChIP interactions in classical monocytes and naïve B cells.

Defects in the type 1 interferon pathway have been reported in patients with severe COVID-19 illness12–15. We found many severe COVID-19-risk variants in chromosome 21, overlapping the IFNAR2 gene that encodes for interferon receptor 2, were associated with the expression of IFNAR2 in multiple immune cell types (Fig. 3a). H3K27ac HiChIP-based chromatin interaction maps in this locus showed that the severe COVID-19-risk variants overlapping the IFNAR2 gene promoter and an intronic enhancer interacted with the promoter of a neighboring gene, IL10RB and also influenced its expression levels (Fig. 3b). The effects of these risk variants were most prominent in NK cells (rs2284551, adj. association P value = 8.99×10−7) (Fig. 3a). IL10RB, encodes for IL-10 receptor beta, and given the immunomodulatory role of IL-1016, it is likely that the lower expression on the IL10RB in NK cells may perturb their responsiveness to IL-10. Thus, our findings point to a potentially important role for IL-10 signaling and NK cells in influencing susceptibility to severe COVID-19 illness.

An external file that holds a picture, illustration, etc.
Object name is nihpp-2020.12.01.407429-f0003.jpg
Figure 3.
COVID-19-risk variants show cell-type-restriction of their effects on gene expression.
(a) Mean expression levels (TPM) of selected severe COVID-19-risk associated GWAS eGenes (all with GWAS association P value < 5×10−8) in the indicated cell types from subjects (n=91) categorized based on the genotype at the indicated peak GWAS cis-eQTL; each symbol represents an individual subject, * adj. association P value < 0.05. (b) WashU Epigenome browser tracks for the IFNAR2 and IL10RB loci, severe COVID-19-risk associated GWAS variants (based on GWAS study, see Extended Data Figure 1a; red color bars are lead GWAS SNPs, black color bars are SNPs in linkage disequilibrium), adj. association P value for GWAS cis-eQTLs associated with expression of IL10RB in NK cells, recombination rate tracks27,28, H3K27ac ChIP-seq tracks, and H3K27ac HiChIP interactions in NK cells.

The expression of two interferon-inducible genes (OAS1 and OAS3) was also influenced by severe COVID-19-risk variants in chromosome 12. OAS1 and OAS3, encode for oligoadenylate synthase family of proteins that degrades viral RNA and activate antiviral responses17.

OAS1 showed a peak COVID-19-risk eQTL (rs2057778, adj. association P value = 1.77×10−7) specifically in patrolling non-classical monocytes, whereas OAS3 showed prominent eQTLs in T cells (Fig. 4a), further highlighting cell-type-restricted effects of severe COVID-19-risk variants. Interestingly, we found that a severe COVID-19-risk variant (rs2010604, adj. association P value = 4.50×10−2) in the OAS1/OAS3 locus also influenced the expression of a neighboring gene (DTX1) specifically in naïve B cells (Fig. 4a).

Active chromatin interaction maps in naïve B cells showed that a cis-regulatory region near the eQTL (rs2010604) interacts with the promoter of DTX1, located >80kb away, and likely modulates its transcriptional activity (Fig. 4b). This notion is supported by recent reports showing that promoters can interact with neighboring gene promoters and regulate their expression9,18.

DTX1, encodes for a ubiquitin ligase Deltex1 that regulates NOTCH activity in B cells19. Deltex1 has also been shown to promote anergy, a functionally hypo-responsive state, in T cells20; if Deltex1 has a similar functions in B cells, then genetic modulation of DTX1 levels may have a profound impact on the function of B cells in COVID-19 illness.

An external file that holds a picture, illustration, etc.
Object name is nihpp-2020.12.01.407429-f0004.jpg
Figure 4.
Target genes of severe COVID-19-risk variants in chromosome 12.
(a) Mean expression levels (TPM) of selected severe COVID-19-risk associated GWAS eGenes (all with GWAS association P value < 5×10−8) in the indicated cell types from subjects (n=91) categorized based on the genotype at the indicated peak GWAS cis-eQTL; each symbol represents an individual subject, * adj. association P value < 0.05. (b) WashU Epigenome browser tracks for the extended DTX1 locus, severe COVID-19-risk associated GWAS variants (based on GWAS study, see Extended Data Figure 1a; red color bars are lead GWAS SNPs, black color bars are SNPs in linkage disequilibrium), adj. association P value for GWAS cis-eQTLs associated with expression of DTX1 in naïve B cells, recombination rate tracks27,28, H3K27ac ChIP-seq tracks, and H3K27ac HiChIP interactions in naïve B cells.

Several COVID-19 risk variants located in the promoter region of TCF19 were associated with its expression in multiple lymphocyte subsets but not in classical or non-classical monocytes (Fig. 1a). TCF19 encodes for a transcription factor TCF19 that has been shown to regulate activation of T cells21 and also involved in cell proliferation22,23. A noteworthy example of a highly cell-specific severe COVID-19-risk eGene in regulatory T cells (TREG) was PDE4A (Extended Data Fig. 2). This gene encodes for phosphodiesterase 4A, which has been shown to reduce the levels of cAMP, and thus influence T cell activity to module immune responses24.

In summary, several severe COVID-19-risk variants show cell-type-restriction of their effects on gene expression, and thus have the potential to impact the function of diverse immune cell types and gene pathways. Our analysis of eQTLs and cis-interaction maps in multiple immune cell types enabled a precise definition of the cell types and genes that drive genetic susceptibility to severe COVID-19 illness, potentially contributing to the different clinical outcomes. Our study also highlights how information about common genetic polymorphisms can be used to define molecular pathways and cell types that play a role in disease pathogenesis.

reference link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7724655/


More information: A. I. Hernandez Cordero et al. Integrative Genomic Analysis Highlights Potential Genetic Risk Factors for Covid-19. conference.thoracic.org/progra … search.php?sid=P9325

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Questo sito utilizza Akismet per ridurre lo spam. Scopri come vengono elaborati i dati derivati dai commenti.