The G-to-A splice polymorphism rs10774671 in the OAS1 gene induces alternative splicing that creates the p46 and p42 isoforms. Most individuals of European origin carry the alternative A allele. The ancestral G allele is the major allele in African populations, and became fixed in the Neanderthal genome. The G allele increases production of the p46 OAS1 isoform, which is more active than p42 in certain viral infections. This polymorphism is in linkage disequilibrium with another polymorphism, rs4767027, which is also associated with higher OAS1 levels (summary by Zhou et al., 2021).
Bonnevie-Nielsen et al. (2005) found an unusual splice acceptor site polymorphism of the OAS1 gene associated with high constitutive activity of the OAS enzyme. Enzyme activity varied in a dose-dependent manner across the GG, GA, and AA genotypes, with the allele G having the higher frequency in persons with high enzyme activity. Because constitutive enzyme activity of 2-prime,5-prime AS had been shown to correlate with strength of response to viral infections (Bonnevie-Nielsen et al., 1989), Bonnevie-Nielsen et al. (2005) suggested that OAS1 is a strong candidate for determining susceptibility or resistance to viral infections.
Susceptibility to SARS-CoV-2
In large-scale cohorts of over 20,000 individuals of European descent, Zhou et al. (2021) found an association between increased circulating levels of OAS1 protein and protection from COVID-19 in a noninfectious state. The increased OAS1 levels protected against 3 measured outcomes: infection, hospitalization, and overall disease severity. The protection was likely due to increased levels of the p46 isoform, which is associated with the ancestral G allele of rs10774671, and has been shown to be more active in certain viral infections.
In detailed in vitro cellular studies, Wickenhagen et al. (2021) determined that a prenylated form of OAS1 isoform p46 is necessary to initiate a block to SARS-CoV-2. Prenylation of the C terminus of OAS1 targeted it to perinuclear structures rich in viral dsRNA, whereas nonprenylated OAS1 was diffusely localized and unable to initiate a detectable block to SARS-CoV-2 replication. The antiviral activity of OAS1 was effective only in the presence of RNase L (RNASEL; 180435). OAS1 was able to sense conserved dsRNA structures in the SARS-CoV-2 5-prime untranslated region; this antiviral activity was highly specific for SARS-CoV-2 and not other viruses tested. The authors noted that the p46 isoform can be expressed only by the G allele of rs10774671; thus, only individuals with at least one G allele are able to express the OAS1 isoform necessary to initiate a block to SARS-CoV-2. Transcriptome analysis of 499 hospitalized COVID-19 patients from the UK showed that absence of the prenylated form of OAS1 was associated with more severe disease. Individuals lacking the p46 transcript (212 patients, 42.5%) were more likely to have severe disease (95 patients, 44.8%) compared to those expressing the p46 transcript (287 patients, 57.5%), of whom 34.1% (98 patients) had severe disease (OR 1.57, p = 0.016).
Possible Association With Type I Diabetes Mellitus
In a case-sib control study, Field et al. (2005) found that the minor allele (G) of the intron 6 OAS1 SNP (rs10774671) was associated with type 1 diabetes (T1D; see 222100). Tessier et al. (2006) confirmed this T1D association in 784 nuclear families (2 parents and at least 1 affected offspring) by the transmission disequilibrium test (TDT; G:A = 386:329, p = 0.033). However, because of linkage disequilibrium within OAS1 and with the other 2 OAS genes, functional attribution of the association to this SNP could not be assumed. To help answer this question, they also genotyped 2 nonsynonymous SNPs in OAS1 exons 3 and 7. All 3 SNPs showed significant transmission distortion. Tessier et al. (2006) described this variant as an A-to-G substitution and gave the frequency of the minor allele (G) as 0.373. They confirmed the T1D association with the splicing variant but concluded that it cannot be attributed (solely) to the splicing variant; they considered the ser162-to-gly substitution (164350.0002) in OAS1 exon 3 as more likely a functional variant.
Hamosh (2018) found this variant (c.1039-1G-A) in 182,659 of 274,116 alleles and in 62,631 homozygotes in the combined populations of the gnomAD database, for an allele frequency of 0.6664 (July 3, 2018).
