rs763361 — CD226 Gly307Ser

CD226 Gly307Ser — The T-Cell Co-Stimulation Checkpoint Variant

CD226 (also called DNAM-1, DNAX Accessory Molecule-1)¹¹ CD226 (also called DNAM-1, DNAX Accessory Molecule-1)
A co-stimulatory receptor expressed on T cells, natural killer cells, and myeloid cells. CD226 binds the ligands PVR (CD155) and Nectin-2 (CD112) on antigen-presenting cells and cancer cells, delivering a secondary activation signal that lowers the threshold for T-cell and NK-cell cytotoxicity occupies a central position in both antitumour immunity and autoimmune regulation. rs763361 is a missense variant in exon 7 of CD226 on chromosome 18q22 that substitutes glycine for serine at protein position 307. Unlike loss-of-function variants, the Ser307 form of the protein alters co-stimulatory signalling in a way that has been consistently — across more than 50 independent studies — associated with increased susceptibility to a broad spectrum of autoimmune diseases.

The Mechanism

The Gly307Ser substitution lies in the cytoplasmic tail of CD226, near a tyrosine-based signalling motif²² tyrosine-based signalling motif
A short amino acid sequence (typically YxxL or similar) within a receptor's intracellular domain that, when phosphorylated, recruits signalling kinases and adaptor proteins to propagate downstream immune activation. The Gly→Ser change at position 307 is thought to alter the phosphorylation state of this domain, affecting how efficiently CD226 recruits downstream kinases such as Grb2 and PI3K after ligand engagement. The net biological result is that T cells and NK cells carrying the Ser307 form have a subtly altered activation threshold: they respond more readily to self-antigens presented in the context of inflammatory signals, a pattern consistent with the variant's observed association with multiple autoimmune diseases that have T-cell hyperactivation as a core pathogenic feature.

CD226 also functions in a competitive balance with its inhibitory counterpart TIGIT³³ TIGIT
A co-inhibitory receptor on T cells that competes with CD226 for the same ligands (PVR, Nectin-2) but delivers an inhibitory rather than activating signal, normally dampening T-cell responses. When CD226 Ser307 shifts this co-stimulatory/co-inhibitory balance toward excess activation, the self-tolerance mechanism that TIGIT normally enforces is partially undermined. This mechanistic framework positions CD226 Gly307Ser alongside PTPN22 R620W (rs2476601) and CTLA4 variants (rs3087243, rs231775) as part of a broader network of T-cell checkpoint genes where single-nucleotide variants collectively tune the threshold at which T cells become self-reactive.

The Evidence

The association was first established by Hafler et al. 2009⁴⁴ Hafler et al. 2009
Hafler JP et al. CD226 Gly307Ser association with multiple autoimmune diseases. Genes Immun, 2009 in a European cohort, finding association with type 1 diabetes (P=3.46×10⁻⁹), multiple sclerosis (P=4.2×10⁻⁴), and rheumatoid arthritis (P=0.017). This multi-disease pattern immediately distinguished CD226 from disease-specific loci and placed it among the small number of pan-autoimmune susceptibility genes.

The most comprehensive meta-analysis to date, Bai et al. 2020⁵⁵ Bai et al. 2020
Bai et al. Role of CD226 rs763361 polymorphism in susceptibility to multiple autoimmune diseases. Immunol Invest, 2020; 29 reports, 51 studies, 18,157 cases, 29,904 controls, confirmed associations with rheumatoid arthritis, SLE, type 1 diabetes, and multiple sclerosis, while finding no stable association for systemic sclerosis. An earlier meta-analysis by Song et al. 2012⁶⁶ meta-analysis by Song et al. 2012
Song G et al. Association between CD226 rs763361 polymorphism and susceptibility to autoimmune diseases. Lupus, 2012; 17 studies, 8,900 cases, 10,295 controls quantified the per-allele risk: overall OR 1.162 (95% CI 1.097–1.230, P<10⁻⁸), with T1D showing a stronger signal (OR 1.353) than SLE (OR 1.150) or systemic sclerosis (OR 1.126).

A further meta-analysis by Qiu et al. 2013⁷⁷ Qiu et al. 2013
Qiu et al. CD226 Gly307Ser association with multiple autoimmune diseases: a meta-analysis. Hum Immunol, 2013 reported an overall OR of 1.19 (95% CI 1.12–1.27), with markedly higher risk in South American populations (OR 1.72) compared to Asian (OR 1.46) or European (OR 1.29) cohorts. Both the additive and codominant models were significant, confirming that each T allele contributes incrementally to risk — TT homozygotes face approximately OR² ≈ 1.42× the risk of CT heterozygotes relative to CC homozygotes.

Sex-specific analysis in a Brazilian T1D cohort⁸⁸ Brazilian T1D cohort
Mattana et al. 2014, PMID 24891767 found the TT genotype association was driven primarily by females (P=0.0012), and TT carriers showed both higher GAD65 autoantibody frequency (31.9% vs 24.5%, OR 1.57) and lower residual C-peptide levels — markers of more aggressive autoimmune beta-cell destruction.

Practical Implications

The per-allele OR of ~1.16–1.19 places rs763361 among the moderate-effect common autoimmune risk variants. For CT heterozygotes (approximately 50% of the population), the risk elevation is modest and the absolute lifetime risk of any specific disease remains low. For TT homozygotes (~28%), the cumulative risk is more meaningful, especially when combined with other autoimmune susceptibility variants in the T-cell checkpoint network.

The diseases most consistently associated — T1D, MS, SLE, and RA — all have well-defined autoantibody and inflammatory biomarkers that can detect subclinical disease years before clinical presentation. Targeted monitoring is the primary actionable response.

Interactions

CD226 rs763361 is one node in a T-cell checkpoint network that includes PTPN22 rs2476601 (LYP phosphatase, reduces TCR signalling), CTLA4 rs3087243 and rs231775 (inhibitory co-receptor on T cells), and IL2RA rs2104286 (IL-2 receptor alpha, T-regulatory cell survival). These four genes converge on the decision of whether a T cell becomes activated or tolerized when encountering antigen. Carrying risk alleles at multiple checkpoints compounds the tolerance deficit multiplicatively. The compound effect of CD226 T + PTPN22 A (rs2476601) on T1D risk in particular deserves a compound action: both variants shift the T-cell activation threshold in the same direction (lower tolerance, higher autoimmune activation), and their combined risk is plausibly additive or super-additive.