Rs394581

rs394581, a SNP cited to be associated with the risk of Rheumatoid Arthritis (RA), is situated on 6q25.3 in the 5' untranslated region of the TAGAP gene, which encodes the T-cell Activation Rho GTPase activating protein. This gene is highly expressed in immune cells such as B-lymphocytes, T-lymphocytes, dendritic cells, natural killer cells, and monocytes, yet its specific role in immune function is still unclear. Meta-analysis of GWAS in RA with validation in independent samples identified the TAGAP as an RA susceptibility locus.

Rheumatoid Arthritis and Genetics
RA is an autoimmune disease that is characterized by chronic and systemic inflammation. At least half the variance in RA risk is estimated to be genetic based on family studies. A twin study with nationwide British and Finnish data from both monozygotic and dizygotic twins concluded that there may be a 60% heritability for RA.

Study Summaries
A 2009 study took data from a previous meta-analysis of two GWAS data sets which contain 3393 cases and 12462 controls (European). They analyzed 370 SNPs from 179 independent loci with p < 0.001 and identified 22 loci with significant functional connectivity. 22 representative SNPs from these loci were replicated in 7957 cases and 11958 matched controls. They found statistically significant association of the rs394581 SNP with RA risk, reporting T as the risk allele and major allele, and C as the minor allele, with a minor allele odds ratio of 0.89 (p = 5.6E-4) in the discovery set, 0.92 (p = 1.5E-04) in the replication set, and 0.91 (p = 3.8E-07) combined. .

A follow-up study published in 2010 with 3962 patients and 3531 controls (European) used conditional logistic regression to investigate whether the effect from rs394581 was independent from the rs182429 SNP, which is in moderate linkage disequilibrium (R^2 = 0.32; D’ = 0.73) with the rs394581. P-value for the rs394581 went up to 0.9 after conditioning on rs182429, which lost statistical significance and implied that the effects found previously for rs394581 may have come from the rs182429 SNP. This study concluded that the association observed with rs394581 has a high probability to be from LD with rs182429.

Another study from 2010 conducted a GWAS meta-analysis with 5539 RA cases and 20169 controls, with replication in an independent set of 6768 cases and 8806 controls (European). It included rs394581 and found minor allele frequency (MAF) of 0.3, odds ratio of 0.91 (95% CI: 0.87 – 0.96), and GWAS p-value of 0.0006. Even though the GWAS p-value could not reach the genome-wide statistically significant value of 5E-08, it was said to be validated by replication in independent samples.

In the same year, another study that looked at the African American population, with 556 cases and 791 controls, found that rs394581 had odds ratio in the opposite direction from the European RA patients. The MAF reported here was 0.47 for patients and 0.43 for controls, with an odds ratio of 1.15 (95% CI: 0.98-1.34). They postulated that possible explanations include different LD structures in the African American vs. European populations or genetic heterogeneity.

In 2011, Chen et al., in a GWAS of 5500 RA cases and 22621 controls (European), found a new TAGAP SNP rs212389 that had 5 orders of magnitude more significant association than rs394581, and was in LD with rs394581 (R^2 = 0.59; D’ = 0.92). This new SNP remained highly significant after conditioning on the rs394581, while conditioning on the new SNP eliminated statistical significance from the rs394581 (P = 0.07). This is an additional study suggesting that the observed association of rs394581 is due to LD.

An interesting study in 2011 tried to use electronic health records with biospecimen data for multiple ethnic groups to obtain similar results as previous GWAS outcomes. With the rs394581 SNP, they found similar results as previous studies (minor allele odds ratio 0.86, 95% CI: 0.75 – 0.99) for the European population with a previous GWAS (OR = 0.91, 95% CI: 0.87 – 0.96). Using EHR for African, Asian, and Hispanic populations, they found odds ratios of 1.35, 0.14 and 1.52 respectively, suggesting that African and Hispanic populations may have opposite association from the European population, and the Asian population has strong association in the same direction as the European population.

In 2011, the rs394581 risk allele was included, among 39 validated genetic risk alleles, in an odds ratio-weighted genetic risk score system for developing RA. The weighting was based on published odds ratio values. The study looked at 542 cases and 551 controls (Caucasian), and found strong predictive power of their genetic risk score for seropositive, erosive RA (RF+ or CCP+), with lower and varying odds ratios for other RA phenotypes, suggesting heterogeneous genetic causality for different RA phenotypes.

In a recent 2013 study, the rs394581 was found to have a zero-inflated negative binomial combined p value of 0.024 during regression analysis for correlation with the Larsen score, which is a measure of bone damage, in RA patients. This suggests that the rs394581 is associated with disease severity in RA.

Conclusion
In summary, even with a few controversial results indicating that the association observed with the rs394581 may be due to LD with other SNPs, overall the majority of studies indicate that this SNP is associated with either the risk of developing RA or disease severity. Studies looking at different populations suggest that the major allele (T) has a higher likelihood of being the risk allele for European and Asian populations, and the minor allele (C) may be the risk allele for African and Hispanic populations.