tyk2

TYK2 (tyrosine kinase 2)

TYK2 is a non-receptor tyrosine kinase of the Janus kinase (JAK) family, constitutively associated with cytokine receptor subunits IFNAR1 (type-I IFN), IL-12Rβ1 (IL-12/IL-23), and IL-10Rβ. It is the canonical signal transducer for type-I interferon and IL-12/IL-23 axes — two signaling arms directly relevant to the inflammaging phenotype of aging. Loss-of-function (LOF) variants in humans cause susceptibility to mycobacterial infection but confer strong protection against autoimmune disease, making TYK2 a rare example of a gene with direct human Mendelian-randomization evidence for the inflammaging-vs-host-defense tradeoff. The first-in-class allosteric TYK2 inhibitor deucravacitinib (Sotyktu, BMS) received FDA approval in September 2022 for moderate-to-severe plaque psoriasis, establishing JH2 pseudokinase-domain inhibition as a viable and highly selective pharmacological strategy.

Identity

• UniProt: P29597 (TYK2_HUMAN)
• NCBI Gene: 7297
• HGNC symbol: TYK2 (HGNC:12440)
• Ensembl: ENSG00000105397
• Chromosomal location: 19p13.2
• Mouse ortholog: Tyk2 (one-to-one ortholog)
• Protein length: 1,187 amino acids (canonical isoform)
• GenAge: Not listed as an aging-specific gene; included here due to mechanistic role in inflammaging via IFN-α/β and IL-12/IL-23 signaling

Domain architecture

TYK2 shares the canonical Janus-kinase “two-headed” (Janus) structure with JAK1, JAK2, and JAK3:

Domain	Residues (approx.)	Function
FERM	26–431	Non-catalytic; mediates constitutive association with cytokine receptor subunits (IFNAR1, IL-12Rβ1, IL-10Rβ)
SH2-like	450–529	Atypical SH2; contributes to receptor coupling
JH2 pseudokinase	589–875	Catalytically impaired; autoinhibits JH1; target of deucravacitinib
JH1 kinase	897–1176	Catalytic domain; phosphorylates receptor subunits and STAT transcription factors

The FERM domain is the primary receptor-docking module: it locks TYK2 to the box-1/box-2 cytoplasmic motifs of cognate receptor subunits, ensuring immediate kinase availability upon cytokine binding ¹.

Receptor associations and signaling context

TYK2 is constitutively paired with three receptor subunits, each anchoring a distinct cytokine axis:

Receptor subunit	Cytokine(s)	STAT activated	Biological output
IFNAR1	IFN-α, IFN-β	STAT1, STAT2	Antiviral immunity; ISG expression; MHC-I upregulation
IL-12Rβ1	IL-12, IL-23	STAT4 (IL-12); STAT3 (IL-23)	Th1 differentiation (IL-12); Th17 expansion (IL-23)
IL-10Rβ	IL-10, IL-22, IL-28/29	STAT3	Anti-inflammatory; barrier repair

Signal flow: cytokine binding → receptor dimerization → TYK2 + partner JAK (JAK1 or JAK2) transphosphorylation → receptor phosphorylation → STAT recruitment, phosphorylation, dimerization, nuclear translocation → gene expression ¹.

In aging, chronic low-level IFN-α/β signaling (driven by endogenous nucleic acids sensed via cgas-sting and Toll-like receptors) amplifies the SASP and promotes sterile inflammation — the inflammaging phenotype. TYK2-mediated IL-12/IL-23 signaling drives Th1/Th17 polarization that, when dysregulated, sustains chronic inflammatory states in aging tissues.

Key PTMs

• pY1054 / pY1055 (activation loop): autophosphorylation marks that induce full JH1 catalytic activity
• pY604 (JH2 domain): regulatory phosphorylation in the pseudokinase domain; modulates JH1 autoinhibition
• pY292 (JH2): additional phosphotyrosine in the pseudokinase domain

Deucravacitinib: JH2 allosteric mechanism

Deucravacitinib (BMS-986165; brand name Sotyktu) is an oral small molecule that binds the regulatory ATP site of the JH2 pseudokinase domain rather than the JH1 catalytic domain. This allosteric mechanism is the key pharmacological innovation ¹:

• JH2 contains a degenerate ATP-binding pocket (catalytically inactive but able to bind nucleotides and small molecules)
• Deucravacitinib locks JH2 in a conformation that maintains autoinhibitory suppression of JH1
• Because JAK1, JAK2, and JAK3 lack a competent JH2 regulatory site in the same configuration, deucravacitinib achieves 100- to 200-fold selectivity for TYK2 over JAK1/3 and ~3000-fold selectivity over JAK2 in cellular assays ¹

This contrasts with JH1-orthosteric JAK inhibitors (tofacitinib, baricitinib, upadacitinib), which compete with ATP at the catalytic domain and, while increasingly selective, retain clinically meaningful off-target JAK1/2/3 activity. The JH2-allosteric strategy predicts a more targeted immune modulation profile — selectively attenuating IFN-α/β and IL-12/IL-23 axes while leaving JAK1/2/3-dependent cytokines (e.g., IL-6, GM-CSF, EPO) less affected.

FDA approval (September 2022): deucravacitinib is approved at 6 mg once daily for adults with moderate-to-severe plaque psoriasis ¹. Primary Phase 3 POETYK PSO-1 and PSO-2 trials showed PASI 75 response rates of ~53–58% at week 16 versus ~35% for apremilast and ~9% for placebo (Armstrong et al. J Am Acad Dermatol 2023; Strober et al. J Am Acad Dermatol 2023 — primary publications not separately cited here). Armstrong et al. 2025 ² is a post-hoc analysis of apremilast non-responders who switched to deucravacitinib at week 24 (n=165 switchers from PSO-1 and PSO-2); these patients achieved PASI 75 in 46.3% (PSO-1) and 42.3% (PSO-2) by week 52, demonstrating efficacy in an apremilast-refractory population.

A 2025 network meta-analysis (Chen J et al., n = 16 RCTs, ~8,000 patients) confirmed deucravacitinib superiority over apremilast and non-inferiority versus several biologics for psoriasis, with a favorable safety signal compared to JAK1/2/3 inhibitors ³.

Aging-context rationale for druggability-tier: 1. TYK2 is FDA-approved (deucravacitinib) for an immunoinflammatory indication whose core pathobiology — dysregulated IFN-α/β and IL-12/IL-23 signaling — directly maps to the chronic-inflammation hallmark of aging. While no aging-indication clinical trial of deucravacitinib has been completed as of 2026, the underlying cytokine axes it inhibits are the same axes driving inflammaging and immunosenescence. Tier 1 is appropriate; the aging-context engagement is mechanistically grounded even without an explicit aging-indication trial.

TYK2 P1104A LOF variant: MR evidence for the inflammaging tradeoff

The rs34536443 / P1104A missense variant (exon 20; A→C substitution; Pro→Ala at position 1104 in the JH1 activation loop) is the most functionally characterized TYK2 LOF variant in humans. It impairs TYK2 catalytic activation in response to type-I IFN and IL-12 stimulation.

Dendrou et al. 2016 (Science Translational Medicine) used the P1104A variant as a Mendelian randomization instrument to resolve causal effects of TYK2 signaling amplitude on autoimmune disease risk ⁴:

• P1104A (rs34536443) homozygous carriers show >50% lower pSTAT in response to type-I IFN and ~70% less IL-12 and IL-23 signaling compared to major-allele homozygotes; the protective dosage effect is non-additive (supralinear in homozygotes), with minor allele frequency ~0.20% in the UK general population
• The variant achieves protection by reducing TYK2-mediated IFN-α/β and IL-12 responses without abolishing them (hypomorphic, not null)
• The paper established P1104A as a human pharmacogenetic “trial” of TYK2 inhibition at the constitutional level — predicting the therapeutic window for TYK2-targeted drugs

Autoimmune protection spectrum: the rs34536443 minor allele is the only SNP at the TYK2 locus that protects against 10 different autoimmune conditions (Fig. 1C) ⁴, including ankylosing spondylitis, Crohn’s disease, multiple sclerosis, psoriasis, ulcerative colitis, rheumatoid arthritis, SLE, and others (full list in paper Table S1). This breadth is consistent with TYK2’s role as a shared transducer across IFN-α/β and IL-12/IL-23 — the two axes most implicated in chronic autoimmune inflammation.

The host-defense tradeoff. The price of constitutively reduced TYK2 signaling is impaired pathogen clearance:

• Kreins et al. 2015 described patients with complete TYK2 deficiency — recurrent mycobacterial infections (including BCG and Mycobacterium tuberculosis), viral infections (including severe herpes and Epstein-Barr), and susceptibility to Salmonella, but notably absent hyper-IgE syndrome (unlike human TYK2 null mutations previously described) ⁵.
• Mechanistically: TYK2 deficiency impairs IFN-α/β-driven antiviral immunity AND IL-12-dependent IFN-γ production from NK/T cells — the primary anti-mycobacterial effector axis.

Implications for aging pharmacology. The P1104A haplotype is functionally equivalent to a moderate (not complete) TYK2 inhibitor sustained constitutively over a lifetime. Its protection against autoimmune disease without causing overt mycobacterial susceptibility in heterozygous carriers suggests there is a therapeutic window in which partial TYK2 inhibition suppresses inflammaging-relevant signaling while preserving adequate host defense. This is the scientific basis for the anticipated investigation of TYK2 inhibitors in aging-associated inflammatory conditions (e.g., frailty, cardiovascular inflammation, neuroinflammation in Alzheimer’s), though no Phase 2/3 aging-indication trial of deucravacitinib has been reported as of 2026. needs-human-replication

Pathway membership

• jak-stat-pathway — JAK family member; TYK2 is the primary transducer for IFNAR1 and IL-12Rβ1 complexes within the broader JAK-STAT signaling network
• type-i-interferon-signaling — TYK2 phosphorylates IFNAR1 and STAT1/STAT2 in the type-I IFN production and response arms; constitutive FERM-domain association with IFNAR1 is required for IFN-α/β signaling competence

Aging-context evidence summary

Claim	Evidence type	Human?	Note
TYK2 LOF (P1104A / rs34536443) protective vs autoimmune disease	MR / human genetics	yes	Dendrou 2016; 10 conditions; MAF ~0.20% UK
Complete TYK2 LOF → mycobacterial susceptibility	Human case series	yes	Kreins 2015; n=10 patients
Deucravacitinib suppresses IFN-α/β + IL-12/IL-23 signaling	Phase 3 RCT	yes	POETYK PSO; Armstrong 2025
Chronic IFN-α/β drives inflammaging SASP	in-vitro / mouse	partial	cGAS-STING pathway; see cgas-sting
TYK2 as a druggable inflammaging target	None (aging indication)	no	Mechanistic extrapolation only needs-human-replication

Key interactors

• IFNAR1 — constitutive FERM-domain binding; required for type-I IFN signaling
• IL-12Rβ1 — constitutive FERM-domain binding; required for IL-12 and IL-23 signaling
• IL-10Rβ — constitutive FERM-domain binding; required for IL-10/IL-22 signaling
• JAK1 — transphosphorylation partner at the IFNAR complex (JAK1 on IFNAR2 / TYK2 on IFNAR1)
• JAK2 — transphosphorylation partner at the IL-12Rβ2/IL-12Rβ1 complex
• STAT1, STAT2, STAT4, STAT3 — direct phosphorylation substrates; activated depending on receptor complex

Limitations and knowledge gaps

• No aging-indication trial. As of 2026, deucravacitinib has not been tested in a Phase 2 or Phase 3 trial for any aging phenotype (frailty, inflammaging biomarkers, cardiovascular inflammation, neuroinflammation). The aging-context rationale is mechanistically strong but empirically untested. needs-human-replication
• GTEx aging correlation not populated. TYK2 tissue-by-age expression correlation (Spearman ρ across GTEx tissues) not yet extracted. Populate per sops/finding-tissue-expression.md. unsourced
• Mouse aging phenotype unclear. Tyk2 knockout mice have been generated and show impaired IFN + IL-12 responses, but whether they show accelerated or attenuated aging phenotypes has not been systematically studied. needs-human-replication
• IL-23-specific inhibition tradeoff. Selective IL-23p19 antibodies (guselkumab, risankizumab) target the same IL-23 cytokine axis as TYK2 without affecting type-I IFN. Whether JH2-allosteric TYK2 inhibition provides incremental benefit vs IL-23-selective biologics for aging-context inflammation is unstudied. needs-replication
• Infection risk at aging doses. The mycobacterial susceptibility signal from complete TYK2 deficiency (Kreins 2015) is most severe in the null state. Whether chronic partial TYK2 inhibition in older, immunosenescent individuals creates a qualitatively different infection-risk profile compared to younger trial populations is unknown. long-term-unknown
• Pseudokinase domain conformational pharmacology. The precise structural mechanism by which deucravacitinib-bound JH2 maintains JH1 autoinhibition at the atomic level has been modeled computationally but full crystallographic characterization with the intact FERM-SH2-JH2-JH1 full-length protein remains an open area. no-mechanism

Footnotes

Footnotes

1. doi:10.1016/j.phrs.2022.106642 · Roskoski R Jr · Pharmacological Research 2023 · review · cited 70+ times (per archive) · model: biochemical/pharmacological review of deucravacitinib JH2 mechanism and FDA-approval data; citation percentile 100th in archive ↩ ↩² ↩³ ↩⁴ ↩⁵

2. doi:10.1007/s13555-025-01606-9 · Armstrong AW et al. · Dermatology and Therapy 2025 · post-hoc analysis of Phase 3 POETYK PSO-1 (n=54) and PSO-2 (n=111) apremilast non-responders switched to deucravacitinib at week 24 · model: adults with moderate-to-severe plaque psoriasis; primary POETYK PSO-1/PSO-2 week-16 efficacy data (PASI 75 ~53–58%) are from Armstrong et al. J Am Acad Dermatol 2023 and Strober et al. J Am Acad Dermatol 2023; OA (gold) ↩

3. doi:10.1007/s10067-025-07597-4 · Chen J et al. · Clinical Rheumatology 2025 · systematic-review + meta-analysis · n=~16 RCTs (~8,000 patients) · comparative efficacy deucravacitinib vs biologics/apremilast; not OA locally ↩

4. doi:10.1126/scitranslmed.aag1974 · Dendrou CA et al. · Science Translational Medicine 2016 · mendelian-randomization / human genetics · model: human cohorts (multiple autoimmune GWAS datasets); n=large-scale GWAS; P1104A protective haplotype; 241 citations per archive; citation percentile 100th ↩ ↩²

5. doi:10.1084/jem.20140280 · Kreins AY et al. · Journal of Experimental Medicine 2015 · observational · n=10 patients with complete TYK2 deficiency · model: human primary immunodeficiency case series; 349 citations per archive; citation percentile 100th; not OA locally — no-fulltext-access ↩