rs231806 — CTLA4

Upstream regulatory variant in the CTLA4 MH30 region associated with latent autoimmune diabetes in adults (LADA); the G allele reduces soluble CTLA-4 expression and is linked to the disease-risk haplotype

See your personal result for CTLA4

Upload your DNA data to find out which genotype you carry and what it means for you.

Upload your DNA data

Works with 23andMe, AncestryDNA, and other DNA test exports. Results in under 60 seconds.

CTLA4 MH30 — The Upstream Sentinel for Adult-Onset Autoimmune Diabetes

CTLA-4 is one of the immune system's most critical brakes — a checkpoint receptor expressed on activated T cells that prevents them from attacking the body's own tissues. While much attention has focused on coding and 3'UTR variants in CTLA4, the far-upstream regulatory landscape also influences how much CTLA-4 the immune system produces. The MH30 variant (rs231806), located approximately 23 kilobases upstream of the CTLA4 coding sequence, sits within a long-range regulatory element11 regulatory element
Distal regulatory elements can loop physically to gene promoters and modulate transcription factor access across tens of kilobases that influences CTLA4 gene expression. This variant has been specifically implicated in latent autoimmune diabetes in adults (LADA) — a slowly progressing form of autoimmune diabetes that is frequently misdiagnosed as type 2 diabetes at onset.

The Mechanism

The MH30 region lies within a chromatin-accessible, distal regulatory domain upstream of CTLA4. Regulatory variants at this distance can influence transcription factor binding and long-range chromatin looping that brings upstream elements into contact with the gene's core promoter. The G allele at rs231806 is the population-major allele globally (~60% frequency) and is part of a disease-associated haplotype (GCC — combining the G allele at MH30 with specific alleles at the -1147 position (rs16840252) and -318 position (rs5742909)22 -1147 position (rs16840252) and -318 position (rs5742909)
These three promoter-region SNPs were studied together in the Estonian LADA cohort as part of CTLA4 haplotype analysis).

The functional consequence appears to be reduced production of the soluble isoform of CTLA-4 (sCTLA-4)33 soluble isoform of CTLA-4 (sCTLA-4)
Soluble CTLA-4 is secreted into the extracellular space and acts as a decoy receptor that blocks B7 costimulatory ligands, suppressing T cell activation independently of cell surface CTLA-4. In a study of newly diagnosed type 1 diabetes patients, sCTLA4 mRNA expression was significantly lower in individuals homozygous for the G allele (GG genotype, p=0.039). Less soluble CTLA-4 means a less effective extracellular dampening signal on T cell activation — potentially lowering the threshold for autoreactive T cells to escape regulation and attack pancreatic beta cells.

The Evidence

The primary association between rs231806 and LADA comes from a study of 61 LADA patients and 230 controls44 study of 61 LADA patients and 230 controls
Douroudis, Prans, and Uibo, Human Immunology, 2009 in an Estonian population. The MH30 GG genotype was significantly more frequent in LADA patients (p=0.0051), and the G allele was overrepresented among cases (p=0.0023). Haplotype analysis reinforced the finding: the GCC haplotype (spanning rs231806, rs16840252, and rs5742909) was significantly elevated in LADA cases (p=0.000073), while the CCC haplotype was protective (p=0.0019).

A follow-up population study comparing LADA, T1D, T2D, and controls55 follow-up population study comparing LADA, T1D, T2D, and controls
Kisand and Uibo, Gene, 2012; 65 LADA, 154 T1D, 260 T2D, 229 controls found that LADA had a distinct genetic risk profile from classical T1D: LADA was associated with protective HLA haplotypes and with CTLA4 haplotypes (including rs231806) rather than the classical T1D HLA risk alleles. A model using CTLA4 region and HLA-DQB1 predicted LADA with AUC 0.693, confirming the CTLA4 region as a genuine LADA susceptibility locus.

Expression data from a study of B7/CD28 family genes in 49 newly diagnosed T1D children66 study of B7/CD28 family genes in 49 newly diagnosed T1D children
Pruul et al., Human Immunology, 2013 showed that sCTLA4 mRNA was lowest in GG individuals at rs231806 (p=0.039), providing a mechanistic link between genotype and reduced immune regulation.

Importantly, the -318C/T (rs5742909) and -1147 (rs16840252) SNPs studied alongside rs231806 in the same Estonian LADA cohort showed no individual associations — only the MH30 (rs231806) G allele reached significance on its own, while the three-SNP GCC haplotype provided the strongest combined signal.

The evidence level is emerging — the LADA association comes from a single case-control study with a small sample (n=61 LADA). This is a real and biologically plausible signal with mechanistic support, but independent replication in larger cohorts is needed to confirm the effect size and establish the risk quantitatively.

Practical Implications

The GG genotype is the most common globally (~36% of people) but carries the highest MH30-associated LADA risk. LADA is frequently misdiagnosed as type 2 diabetes at onset because patients are often adults and not immediately insulin-dependent — but unlike T2D, LADA involves progressive beta cell destruction by autoreactive T cells, requiring insulin therapy within a few years. Awareness of CTLA4 risk genotypes may support earlier antibody testing (GAD65, IA-2) in adults presenting with apparent type 2 diabetes who don't respond well to oral agents.

For individuals who already have LADA or are being evaluated for possible LADA, CTLA4 genotype is one of several risk markers alongside islet autoantibodies, HLA type, and clinical presentation. No CTLA4-genotype-based intervention currently alters LADA management, but abatacept (a CTLA-4-Ig fusion protein)77 abatacept (a CTLA-4-Ig fusion protein)
Abatacept mimics CTLA-4 function and is being studied in preserving beta cell function in early autoimmune diabetes is under investigation as a beta cell-preserving agent in early T1D and LADA.

Interactions

rs231806 is part of a three-SNP CTLA4 haplotype studied together with rs16840252 (-1147) and rs5742909 (-318C/T). The GCC haplotype combining risk alleles across all three upstream variants showed the strongest LADA signal (p=0.000073), much stronger than any single variant alone. This suggests epistatic cooperation within the upstream CTLA4 regulatory architecture.

The other two established CTLA4 variants — rs3087243 (CT60, 3'UTR) and rs231775 (+49A/G, coding) — provide complementary and stronger associations with autoimmune thyroid disease, T1D, and rheumatoid arthritis. Together, rs231806 + rs3087243 + rs231775 form a fuller picture of CTLA4 autoimmune susceptibility spanning promoter, coding, and 3'UTR regions. LADA risk is particularly sensitive to the upstream regulatory haplotype (rs231806), while thyroid and rheumatoid disease are more strongly predicted by the downstream variants.

Genotype Interpretations

What each possible genotype means for this variant.

CC Protective Haplotype Normal

Two copies of the protective C allele — lowest LADA risk associated with the MH30 position

The CC genotype at rs231806 sits at the protective end of the MH30 haplotype spectrum. Mechanistic data suggest that C allele carriers maintain higher sCTLA4 mRNA expression compared to GG individuals, meaning more soluble CTLA-4 is available to dampen T cell activation in the extracellular space. In the Estonian LADA cohort (PMID 19735686), the CCC haplotype (C at MH30, C at -1147, C at -318) was significantly underrepresented in LADA cases, confirming that this multi-locus protective configuration is associated with reduced adult-onset autoimmune diabetes risk.

CC is relatively rare globally (~16%) because the G allele predominates in most populations — particularly in East Asians (~74% G allele frequency) and Europeans (~51% G allele frequency). Being CC represents a protective minority genotype for LADA at this specific locus.

CG Intermediate Risk Intermediate

One copy of the risk G allele — intermediate LADA susceptibility at the MH30 position

CG heterozygosity at rs231806 places one copy of each allele in your regulatory architecture. The G allele participates in the GCC haplotype that was elevated in LADA patients (p=0.000073), while the C allele participates in the protective CCC haplotype. For soluble CTLA-4 expression, heterozygotes are expected to have intermediate sCTLA4 mRNA levels between CC (highest) and GG (lowest) based on the dose-response pattern described by Pruul et al. (PMID 23911738). The LADA association for single G allele carriers follows a codominant or additive model per the Douroudis study (G allele p=0.0023), but the per-genotype OR for CG was not separately reported in the published abstract.

GG High Risk Haplotype High Risk

Two copies of the risk G allele — highest MH30-associated LADA susceptibility and reduced soluble CTLA-4 expression

GG homozygosity at rs231806 places you in the highest-risk category at this specific CTLA4 locus for LADA. The GCC haplotype (G at rs231806, C at rs16840252, C at rs5742909) was significantly overrepresented in LADA patients at p=0.000073 — a much stronger signal than any individual variant. The expression consequence — reduced sCTLA4 mRNA in GG individuals — provides biological plausibility: soluble CTLA-4 acts as a decoy receptor in extracellular fluid that competes with B7 costimulatory ligands, suppressing T cell activation without requiring direct cell contact. Lower sCTLA4 means this extracellular braking mechanism is less effective.

LADA is a slowly progressive form of autoimmune diabetes that presents in adults (typically age 30-50), is initially non-insulin-dependent, but involves ongoing beta cell destruction by autoreactive T cells. It is positive for islet autoantibodies (particularly GAD65) and progresses to insulin dependence over 2-6 years. It is frequently misdiagnosed as type 2 diabetes. Genotypic CTLA4 risk at MH30 is one factor in a multigenic risk landscape including HLA alleles and other immune-regulatory variants.

Critically, the evidence base for this specific variant is emerging — one primary study of 61 LADA patients. The signal is real and biologically plausible but needs replication in larger independent cohorts before clinical genotyping can be routinely recommended.