rs34536443 — TYK2 TYK2 p.Pro1104Ala

TYK2 P1104A — The Natural Brake on Autoimmune Signaling

Your immune system runs on a network of molecular switches that amplify responses to infection and then shut them off before they damage healthy tissue. TYK2¹¹ TYK2
Tyrosine kinase 2, a member of the Janus kinase (JAK) family that transduces cytokine signals from cell-surface receptors into gene expression changes is one of those amplifiers — it sits at the junction of the IL-12, IL-23, and type I interferon pathways, three signaling cascades that when overactivated drive most major autoimmune diseases. The rs34536443 variant (p.Pro1104Ala) is a naturally occurring partial loss-of-function that blunts this amplification. The result is a measurably reduced risk of rheumatoid arthritis, lupus, multiple sclerosis, psoriasis, type 1 diabetes, systemic sclerosis, and hypothyroidism — and the same variant is now actively mimicked by a blockbuster immunology drug.

The Mechanism

TYK2 mediates signaling downstream of the IL-12 receptor (driving Th1 responses), IL-23 receptor (driving Th17 responses), and type I interferon receptors (IFN-α/β, critical for antiviral defense and lupus pathogenesis). The Pro1104Ala substitution occurs in the pseudokinase domain²² pseudokinase domain
The pseudokinase (JH2) domain of TYK2 regulates the catalytic kinase (JH1) domain in an allosteric fashion; P1104A reduces this regulatory capacity without abolishing it — a region that normally stabilizes the active kinase domain through intramolecular contacts. Replacing proline (rigid, constrained) with alanine (flexible, small) disrupts an interdomain contact that is required for optimal signal amplification.

The key word is partial: TYK2-P1104A still functions. Carriers are not immunodeficient. Instead, Gorman et al.³³ Gorman et al. demonstrated that the variant specifically limits signaling when multiple autoimmune-relevant pathways are co-activated simultaneously — the precise situation in active autoimmune disease. Under normal immune conditions (single-pathway stimulation), the variant has little effect. Under autoimmune-like conditions (multiple pathways firing at once), it substantially reduces the generation of pathogenic T cell subsets: Th1, Th17, T follicular helper (Tfh), and double-positive IL-17+/IFNγ+ cells — the effectors responsible for tissue destruction in RA, lupus, and MS.

A 2025 study in PNAS identified a second layer of protection specific to multiple sclerosis⁴⁴ multiple sclerosis
MS is a demyelinating autoimmune disease of the central nervous system driven by autoreactive T cells and neuroinflammation: the P1104A effect appears to operate within the CNS itself, where TYK2-expressing microglia and reactive astrocytes propagate neuroinflammation. Brain-penetrant TYK2 inhibition dramatically outperformed peripherally-restricted inhibition in experimental models, explaining why partial central TYK2 impairment confers stronger MS protection than peripheral immune effects alone would predict.

The same mechanism that makes P1104A protective also explains a known trade-off: Yarmolinsky et al. (2022)⁵⁵ Yarmolinsky et al. (2022) found that each P1104A allele associates with modestly increased lung cancer risk (OR 1.15, 95% CI 1.09-1.23) and non-Hodgkin lymphoma risk (OR 1.18). This is consistent with TYK2's role in anti-tumor immunity through type I interferon and IL-12-driven NK and CD8+ T cell activation. The immune surveillance trade-off is modest for heterozygotes but is relevant for cancer screening discussions.

The Evidence

A meta-analysis of 34 studies⁶⁶ meta-analysis of 34 studies
Pellenz et al. 2021, encompassing multiple autoimmune diseases including MS, SLE, RA, Crohn's disease, psoriasis, and T1D confirmed that the rs34536443 C allele is significantly associated with protection across autoimmune diseases, consistent with the earlier meta-analysis by Tao et al. (2011)⁷⁷ Tao et al. (2011) (21,497 cases / 22,647 controls) finding OR 0.76 per C allele (95% CI 0.69-0.84, P<0.00001).

For systemic sclerosis (scleroderma), López-Isac et al.⁸⁸ López-Isac et al. (7,103 patients / 12,220 controls) confirmed P1104A protection (OR 0.80, P=2.28×10⁻³), reinforcing TYK2's IL-12 pathway role in SSc pathophysiology. A Mendelian randomization study⁹⁹ Mendelian randomization study using rs34536443 to proxy TYK2 inhibition across 339,197 UK Biobank participants and 260,405 FinnGen participants confirmed associations with hypothyroidism, psoriasis, SLE, and RA.

The pharmacological validation of this finding is striking: deucravacitinib¹⁰¹⁰ deucravacitinib (Sotyktu), FDA-approved in 2022 for moderate-to-severe psoriasis, was explicitly designed to mimic the P1104A allele by targeting the TYK2 pseudokinase domain allosterically. Early-phase SLE trials have shown promising results, with additional indications under investigation. The drug's existence as a validated pharmaceutical target that copies a common natural variant is among the strongest evidence that this variant's effects are real and clinically meaningful.

Practical Implications

The C allele (protective) at rs34536443 is predominantly a European variant: about 4.2% of Europeans carry at least one copy, compared to <1% in East Asians and ~0.9% in Africans. Most people worldwide carry GG (the reference genotype) and have standard TYK2 signaling. CG heterozygotes have partially attenuated signaling and roughly 22-24% reduced autoimmune disease odds per allele. CC homozygotes — carrying two copies of the protective variant — are rare (less than 0.2% of Europeans) but show the strongest protection and closest pharmacological equivalence to deucravacitinib's mechanism.

Interactions

TYK2 sits downstream of multiple cytokine receptors and its effects compound with other immune regulatory variants. rs2476601 (PTPN22 R620W), rs3087243 (CTLA4), and rs6920220 (TNFAIP3 6q23) each modulate T cell and B cell activation thresholds in overlapping autoimmune disease networks. Carriers of P1104A who also carry protective alleles at PTPN22 or CTLA4 likely have additive protection — but these interactions have not been formally modeled in compound heterozygosity studies. The P1104A effect on cancer immune surveillance also may interact with PTPN22 and HLA variants that affect NK and CD8 T cell licensing.