Association of SARS susceptibility with single nucleic acid polymorphisms of OAS1 and MxA genes: a case-control study

Severe acute respiratory syndrome (SARS) is an emerging infectious disease caused by a novel positive-stranded RNA virus, termed SARS-CoV [1]. The global outbreak of SARS has killed 774 and infected more than 8098 people in 29 countries and regions, as reported by WHO [2]. In Beijing, more than 2523 cases and 181 deaths occurred including health care workers and family members who had closely contacted with SARS patients [3]. It was noticed that despite similar levels of exposure to the SARS-CoV some people were infected while others were still free from infection, suggesting a role of host genetic factors in susceptibility to SARS-CoV infection [4]. Human-Leukocyte-Antigen Class I and Class II Genotypes were previously reported to be associated with the development of SARS [5, 6]. The polymorphism of angiotensin converting enzyme 1 gene was supposed to be involved in the progression of the disease [7].

Resent study showed that type I interferons (IFN-c/d) could inhibit SARS-CoV replication both in vivo and in vitro [8‐10]. Type I interferons induce numerous proteins with antiviral activities, including 2'-5' oligoadenylate synthetase 1 (OAS1) and myxovirus resistance 1 (MxA). OAS1 protein can bind double-stranded RNA and polymerize ATP into PPP (A2'P5'A) N oligomers, which activate the latent RNAse L that, when activated, cleave single-stranded RNAs [11]. MxA protein belongs to the dynamin family and shows activity against several viruses [12‐14]. However, the precise mechanism of antiviral action has not been elucidated. Host genetic factors that influence antiviral effects of interferons were well demonstrated in viral hepatitis, especially of hepatitis C virus infection [15]. It was reported that the polymorphisms at position 347 in the 3'-untranslated region (3'-UTR) of the OAS1 gene and at position 88 in the MxA gene promoter region seemed associated with hepatitis C virus infection [16]. The polymorphisms of these two genes that might affect the susceptibility to SARS were reported in Vietnamese population [17]. The G allele frequencies of OAS1 at the position 347 in the 3'-UTR were estimated to be 41.7% among European-Americans, 10.9% among African-Americans, 18.2% among Japanese and 36.4% among Chinese. The T allele frequencies of MxA at position 88 in the promoter region were estimated to be 19.6% among Caucasian and 33% among Vietnamese population. The objective of the study was to investigate if single nucleic acid polymorphisms (SNPs) of OAS1 and MxA genes were associated with the development of SARS in Chinese Han Population.

Many factors may influence the outcome and clinical manifestations of SARS infection, such as exposure levels, differences in SARS-CoV strain virulence, host health status, smoking, other virus co-infection, and host genetic patterns. In order to study the association between gene polymorphisms and SARS-CoV infection, we took several measures to control for possible confounders. Firstly, cases and controls were selected from the same hospital to minimize the impact of virulence differences in SARS-CoV strains. Secondly, each control had to have close contact history with SARS patients to reduce differences in exposure levels between cases and controls. Thirdly, multivariate analyses were conducted to adjust the factors which were statistically significant in univariate analyses. In fact, mask-, gown- and goggles-wearing were found to be protective against SARS-CoV infection in this study as previously reported [18‐20].

We found that both the AG and GG genotypes of OAS1 gene at the 3'-UTR region were more common among controls compared to cases. The G allele frequency among controls (39.8%) was comparable to that found in the 45 Han Chinese in Beijing. Our findings indicate that the G allele in the OAS1 gene conferred protection against SARS-CoV infection. Knapp reported that this A/G allele was associated with the outcome of HCV infection, OAS1 playing a greater role in mediating self-limiting versus persistent HCV infection, rather than when viral persistence is established [16]. A nonsense mutation described in the gene encoding the mouse 2'-5' OAS1 isoform was associated with West Nile virus susceptibility in laboratory mice [21, 22]. There are two isoforms in human OAS1 gene (E16 and E18), which share identical N-terminal sequence but diverge at exon 7 [23]. Isoform E16 contains 7 exons and a hydrophobic C-terminus. Isoform E18 contains an additional exon and is very hydrophilic as compared to transcript variant E16. Production of the two mRNAs isoforms of OAS1 depends on processing that involves the recognition site located within the 3'-terminal exon [24]. The polymorphisms in the 3'UTR may be related with the expression of the two isoforms. The mechanisms of the protective effect of G allele need to be elucidated.

Regarding the G/T polymorphism at position 88 in promoter region of MxA gene, the GT genotype was found to be more frequent in patients than in controls. Other researchers found no difference in SARS-CoV replication in Vero cells that were stably expressing MxA [25, 26]. The patients with GT genotype may express more MxA protein when induced by IFNs. Consequently, the MxA protein can promote the establishment of persistent SARS-CoV infection. MxA protein may have dual functions in reaction to virus infection. Torisu reported that T allele with a high MxA-producing capability was more frequently seen in subacute sclerosing panencephalitis (SSPE) patients [27]. A luciferase reporter assay revealed that the MxA promoter sequence of T haplotype had higher promoter activity than that of G haplotype [28]. Arcas et al. reported that GT and TT genotype expressed higher amount of MxA mRNA than GG genotype in IFN-treated peripheral blood mononuclear cells in vitro [29]. In our study, 88 SNP in MxA was related to disease susceptibility. The mechanism of the MxA gene with SARS susceptibility needs to be further investigated. Since the polymorphisms in the two genes were associated with the SARS-CoV infection, we tried to find out if combined action between mutations could affect the susceptibility to SARS-CoV infection. But the sample size of this study is not sufficient to do the interaction analysis. More SARS cases and controls are required to do so.

In conclusion, we showed that the SNPs in OAS1 and MxA genes were associated with SARS-CoV infection in Chinese Han population. These findings may lead to a better understanding of IFN-induced antiviral response to SARS infection.