The 2013 International Genetics of Ankylosing Spondylitis Consortium (IGAS) Immunochip study genotyped immune-related loci in 10,619 ankylosing spondylitis (AS) cases and 15,145 controls, bringing the number of independently associated genetic signals to 43 while leaving open what those signals mean biologically: most loci could not be narrowed to causal variants because linkage disequilibrium blocks were barely reduced, the mechanisms of the associated genes were unestablished, and the study's discussion could only infer — without a mechanistic account — that disease-specific loci such as HLA-B27 contribute to the organ and tissue specificity of the diseases with which they are associated. The problem is to convert these statistical associations into a causal, mechanism-based account of AS: identify the causal variant(s) and causal gene(s) at the associated loci and establish the molecular mechanism by which they alter disease risk, including whether the aminopeptidase associations act by reshaping the HLA-B27-bound peptide repertoire, by limiting peptide supply during HLA-B27 folding, or by increasing endoplasmic reticulum stress, and including a mechanistic account of why HLA-B27 predisposes to inflammation of the axial skeleton rather than the gut. Follow-up work has resolved one explicit sub-question of the source (the ERAP1-HLA-B*27 interaction has been confirmed in East Asians), but fine-mapping, functional-mechanism, and organ-specificity components remain open.

Background

Ankylosing spondylitis (AS) is a common, highly heritable inflammatory arthritis that chiefly affects the spine and sacroiliac joints and can lead to fusion of the vertebrae; it is the prototype of the spondyloarthritis family, whose members share overlapping genetics with inflammatory bowel disease (IBD) and psoriasis. The strongest known genetic risk factor is HLA-B27, an allele of a human leukocyte antigen (HLA) class I gene. HLA class I molecules are cell-surface proteins that display peptides derived from intracellular proteins to CD8+ T cells; the peptides are first trimmed in the endoplasmic reticulum (ER) by aminopeptidases such as ERAP1 and ERAP2 before being loaded onto the HLA molecule. How HLA-B27 promotes AS is unresolved; leading hypotheses include presentation of arthritogenic peptides, misfolding of the HLA-B27 heavy chain in the ER causing ER stress and the unfolded protein response, and unusual cell-surface forms of the molecule. Genome-wide association studies (GWAS) find statistical correlations between common genetic variants and disease; because neighboring variants are inherited together in linkage disequilibrium (LD), an association usually marks a whole block of correlated variants rather than the causal one, so 'fine-mapping' — narrowing the signal with denser genotyping, larger samples, and conditioning analysis — is needed before any variant can be called causal. The Immunochip was a custom genotyping array with dense coverage of immune-disease-associated regions, designed to fine-map and extend GWAS signals. The IGAS 2013 Immunochip study (Cortes et al., Nature Genetics 2013) confirmed 12 known loci, added 13 new loci (including TYK2, IL27 and IL6R in the IL-23 cytokine-signaling pathway), and identified additional associated haplotypes (inherited variant combinations) at known loci, for 43 independent signals in total; it also found associations at two regions encoding four aminopeptidase genes involved in peptide trimming before HLA class I presentation (ERAP1 and ERAP2 at the chromosome 5q15 region, together with the adjacent LNPEP whose involvement is uncertain, and NPEPPS at the chromosome 17q21 region), at G protein-coupled receptor (GPCR, a large family of cell-surface signaling receptors) loci whose functions were not well established and whose ligands (the signaling molecules that activate them) were only partly known — candidate ligands had been reported for GPR35 and GPR65, while those for GPR25 and GPR37 were not known (GPR35, GPR37, GPR65, GPR25 near KIF21B) — and at several rare variants (population frequency below 1%) in genes such as TYK2 and IL23R. The study's own discussion stated the limitations that define this problem: dense genotyping did not substantially narrow the associated region at most fine-mapped loci (less than 10% narrowing of the region was observed for four of the eight loci that were fine mapped); the key associated variants at several loci were 'not yet defined'; whether the aminopeptidase genes act through a quantitative effect on HLA class I peptide presentation, a qualitative effect on the peptide repertoire presented, or increased ER stress during HLA-B27 folding was unresolved; the interaction (epistasis, i.e. the effect of one gene depending on the genotype at another) between ERAP1 and HLA-B27 needed confirmation in East Asians, where the original analysis was underpowered; rare-variant associations required independent replication; and the observation that AS and IBD share IL-23-pathway genetics while their defining loci differ (AS associates with HLA-B27 but not NOD2, an intracellular sensor of bacterial cell-wall products; Crohn's disease associates with NOD2 but not HLA-B*27) implied that disease-specific loci contribute to the organ and tissue specificity of what is otherwise a shared immune pathway.

Problem Statement

For the ankylosing spondylitis (AS) association signals established by the IGAS 2013 Immunochip study, determine (i) the causal variant(s) and causal gene(s) at each of the 43 independently associated signals reported by the study and (ii) the molecular mechanism by which they alter AS risk. These questions all bear on one objective — converting the statistical associations into a causal, mechanism-based account of AS pathogenesis — and include: (a) fine-mapping each of these signals to its causal variant(s), resolving cases where multiple candidate genes exist within the associated haplotype (for example whether the IL23R-region signal acts through IL23R or IL12RB2); (b) for the aminopeptidase genes ERAP1, ERAP2 and NPEPPS (and LNPEP, whose involvement the source records as uncertain), establishing which molecular consequence of the disease-associated variants drives risk — qualitative reshaping of the HLA-B27-bound peptide repertoire, quantitative limitation of peptide supply during HLA-B27 folding, or increased endoplasmic reticulum stress — and identifying the key associated variants at NPEPPS; (c) establishing the functions of the associated G protein-coupled receptor loci GPR35, GPR37, GPR65 and GPR25 (KIF21B region) and their roles in autoimmune disease, which were not well established; (d) independent replication of the rare-variant associations reported in the source (including TYK2, IL23R, CARD9, LNPEP, KLKB1, RAD50, PRDM1 and DYRK4), which were tested only in a UK subset because of population-stratification risk; and (e) a mechanistic account of the organ and tissue specificity that the source's discussion could only infer — why HLA-B*27 and the AS-specific loci predispose to inflammation of the axial skeleton while the shared IL-23-pathway genetics with inflammatory bowel disease produces gut-limited disease and the Crohn's-specific locus NOD2 produces no spinal inflammation. These components (a)–(e) together constitute this objective.

References

  1. International Genetics of Ankylosing Spondylitis Consortium (IGAS), Cortes A, Hadler J, Pointon J, et al. (2013). Identification of multiple risk variants for ankylosing spondylitis through high-density genotyping of immune-related loci. Nature Genetics 45:730-738. doi:10.1038/ng.2667, https://doi.org/10.1038/ng.2667
  2. Cortes A, Pulit SL, Leo PJ, et al. (2015). Major histocompatibility complex associations of ankylosing spondylitis are complex and involve further epistasis with ERAP1. Nature Communications 6:7146. doi:10.1038/ncomms8146, https://doi.org/10.1038/ncomms8146
  3. Wang G, Kim TH, Li Z, Cortes A, Kim K, Bang SY, Leo P, Brown MA, Xu H. (2020). MHC associations of ankylosing spondylitis in East Asians are complex and involve non-HLA-B27 HLA contributions. Arthritis Research & Therapy 22:74. doi:10.1186/s13075-020-02148-5, https://doi.org/10.1186/s13075-020-02148-5
  4. Sanz-Bravo A, Campos J, Mazariegos MS, López de Castro JA. (2015). Dominant role of the ERAP1 polymorphism R528K in shaping the HLA-B27 peptidome through differential processing determined by multiple peptide residues. Arthritis & Rheumatology 67:692-701. doi:10.1002/art.38980, https://doi.org/10.1002/art.38980
  5. Kenna TJ, Lau MC, Keith P, Ciccia F, Costello ME, Bradbury L, Low PL, Agrawal N, Triolo G, Alessandro R, Robinson PC, Thomas GP, Brown MA. (2015). Disease-associated polymorphisms in ERAP1 do not alter endoplasmic reticulum stress in patients with ankylosing spondylitis. Genes and Immunity 16:35-42. doi:10.1038/gene.2014.62, https://doi.org/10.1038/gene.2014.62
  6. Zhang Z, Ciccia F, Zeng F, Guggino G, Yee K, Abdullah H, Silverberg MS, Alessandro R, Triolo G, Haroon N. (2017). Brief Report: Functional Interaction of Endoplasmic Reticulum Aminopeptidase 2 and HLA-B27 Activates the Unfolded Protein Response. Arthritis & Rheumatology 69:1009-1015. doi:10.1002/art.40033, https://doi.org/10.1002/art.40033
  7. Ellinghaus D, Jostins L, Spain SL, et al. (2016). Analysis of five chronic inflammatory diseases identifies 27 new associations and highlights disease-specific patterns at shared loci. Nature Genetics 48:510-518. doi:10.1038/ng.3528, https://doi.org/10.1038/ng.3528
  8. Robinson PC, Leo PJ, Pointon JJ, et al. (2016). Exome-wide study of ankylosing spondylitis demonstrates additional shared genetic background with inflammatory bowel disease. NPJ Genomic Medicine 1:16008. doi:10.1038/npjgenmed.2016.8, https://doi.org/10.1038/npjgenmed.2016.8
  9. Li Z, Haynes K, Pennisi DJ, Anderson LK, Song X, Thomas GP, Kenna T, Leo P, Brown MA. (2017). Epigenetic and gene expression analysis of ankylosing spondylitis-associated loci implicate immune cells and the gut in the disease pathogenesis. Genes and Immunity 18:135-143. doi:10.1038/gene.2017.11, https://doi.org/10.1038/gene.2017.11
  10. Lin R, Wu W, Chen H, Gao H, Wu X, Li G, He Q, Lu H, Sun M, Liu Z. (2022). GPR65 promotes intestinal mucosal Th1 and Th17 cell differentiation and gut inflammation through downregulating NUAK2. Clinical and Translational Medicine 12:e771. doi:10.1002/ctm2.771, https://doi.org/10.1002/ctm2.771
  11. Nagit RE, Rezus E, Cianga P. (2024). Exploring the Pathogenesis of Spondylarthritis beyond HLA-B27: A Descriptive Review. International Journal of Molecular Sciences 25:6081. doi:10.3390/ijms25116081, https://doi.org/10.3390/ijms25116081
  12. Chiñas M, Fernandez-Salinas D, Aguiar VRC, Nieto-Caballero VE, Lefton M, Nigrovic PA, Ermann J, Gutierrez-Arcelus M. (2025). Functional genomics implicates natural killer cells in the pathogenesis of ankylosing spondylitis. Human Genetics and Genomics Advances (HGG Advances) 6:100375. doi:10.1016/j.xhgg.2024.100375, https://doi.org/10.1016/j.xhgg.2024.100375
  13. Brown M, Li Z, Wu X, et al. (2025). Multi-omic insights from a multi-ancestry genome-wide meta-analysis of ankylosing spondylitis reveal novel pathways of disease susceptibility. Preprint, Research Square. doi:10.21203/rs.3.rs-6917334/v1, https://doi.org/10.21203/rs.3.rs-6917334/v1