In the present issue of Arthritis Research & Therapy, Dieguez-Gonzalez and colleagues report the results of a large case–control genetic study of the 6q23 region that contains the TNFAIP3 gene [1]. Their study reveals a complex association involving interactions between the TNFAIP3 locus and an intergenic region >150 kb upstream.

The upstream region was initially implicated in rheumatoid arthritis (RA) susceptibility in the Wellcome Trust Case Control Consortium study [2], and was subsequently replicated in large European and American populations [3, 4]. The biological explanation was not clear as the associated SNPs (rs6920220 and rs10499194) are located in a 60 kb linkage disequilibrium block not encoding any known genes or transcripts other than a pseudogene of PTPN11. The region is flanked by the OLIG3 and TNFAIP3 loci. The former encodes a protein involved in neuronal development and does not seem to be a plausible susceptibility gene for RA; however, the TNFAIP3 gene encodes a potent inhibitor of NF-κB signalling, resulting in downregulation of production of proinflammatory cytokines including TNF, IL-1 and IL-6, and mice with targeted deletion of TNFAIP3 develop cachexia widespread organ inflammation, including a destructive arthritis [5]. Additional loci encoding immune relevant proteins (IL22RA and IFNGR1) are also located within 1 Mb.

Recent genetic studies of 6q23 in systemic lupus erythematosus (SLE) have revealed a more complex pattern of associations than was initially reported for RA. A genome scan reported three independent associations arising from the upstream region (rs6920220) and two signals near the TNFAIP3 gene including a missense polymorphism in exon 3 (rs2230926), resulting in a phenylalanine to cysteine switch at amino acid 127 of TNFAIP3 [6]. These findings have been replicated in a case–control study that typed 129 SNPs spanning TNFAIP3; furthermore, examination of the functional effects of the lupus-associated Cys127 variant revealed reduced effectiveness at inhibiting TNF-induced NF-κB activity, suggesting a biological mechanism of association [7].

The study by Dieguez-Gonzalez and colleagues involved the use of haplotype tagging SNPs centred on two regions identified in the SLE studies. Surprisingly, only modest associations with RA were detected at both the TNFAIP3 and intergenic regions. This is a common finding in replication studies reflecting winner curse. On further analysis, however, strong evidence of interaction (epistasis) between the two regions with RA susceptibility was detected.

We can therefore conclude that the 6q23 region contains at least two RA susceptibility effects and perhaps three effects for SLE. The biological mechanism for the associations of the intergenic region is unclear since it is only known to contain the pseudogene for PTPN11. Intriguingly, however, the parent PTPN11 gene – located at 12q24.13 – lies within a region associated with susceptibility to RA, type 1 diabetes and inflammatory bowel disease [2]. Approximately 20% of pseudogenes are transcribed, and some generate siRNAs that target homologous genes [8]. Sequence variability in the sequence of the pseudogene could potentially affect expression of PTPN11, leading to autoimmune disease. Determining whether the pseudogene is transcribed and whether sequence variants of the transcript are present on disease-associated haplotypes will therefore be required as initial steps in determining its potential role in modulating PTPN11 expression. An alternative explanation is that a polymorphism within an enhancer sequence in the intergenic region may alter TNFAIP3 expression. The most plausible disease susceptibility variant identified so far is rs2230926, with experimental evidence that the SLE-associated variant is less biologically active [6].

The identification of the primary disease-related variants at 6q23 is likely to initially involve high-throughput DNA sequencing in a large number of patients and controls to more fully characterise the genetic structure of the region. This will be followed by well-powered genetic studies in both RA and SLE that will hopefully lead to the identification of the primary disease-related variants, some of which may arise from low-frequency alleles. Complementary functional studies should lead to a full understanding of the biological basis of the genetic associations of this region with autoimmune rheumatic diseases.