A two-stage genome-wide association study (GWAS) in Han Chinese (Lin et al. 2012) identified two new ankylosing spondylitis (AS) susceptibility loci — between EDIL3 and HAPLN1 at 5q14.3 (rs4552569) and within ANO6 at 12q12 (rs17095830) — and raised a follow-up agenda: identify the causal variant(s) and gene(s) behind these tag-SNP associations, settle whether the suggestive 6q21 signal (rs13210693) is real, characterize the causes of the weaker effects seen in Han Chinese at loci such as ERAP1 and IL12B relative to Europeans, and determine whether the implicated cartilage/bone-related genes are functionally involved in the pathological new bone formation that characterizes the disease. The non-HLA-B27 MHC component of this question has been largely resolved (HLA-C15, HLA-B40, HLA-DQB104 and HLA-B amino-acid positions 70/97 identified in East Asians), but the causal variants and genes at 5q14.3 and 12q12 remain unidentified, replication of the two tag SNPs is inconsistent, 6q21 has had only one underpowered follow-up, the causes of cross-ancestry heterogeneity remain unexplained, and no functional study establishes how these genes contribute to new bone formation. The problem remains open.

Background

Ankylosing spondylitis (AS) is a chronic inflammatory rheumatic disease that primarily affects the sacroiliac joints and the spine; its hallmark is pathological new bone formation (syndesmophytes) that can fuse the spine, and structural damage is scored radiographically with instruments such as the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS). AS is highly heritable (twin estimates around 90%), and its strongest genetic risk factor is HLA-B*27, an allele of a class I gene in the major histocompatibility complex (MHC), the gene-dense immune-recognition region on chromosome 6. Other established non-MHC risk genes include ERAP1, which encodes an endoplasmic reticulum aminopeptidase that trims peptides before their presentation by MHC class I molecules, and IL12B, which encodes the IL-12p40 subunit shared by the cytokines IL-12 and IL-23.

Genome-wide association studies (GWAS) test hundreds of thousands to millions of single-nucleotide polymorphisms (SNPs) for statistical association with disease. A GWAS hit is usually a tag SNP: because nearby variants are co-inherited in blocks called linkage disequilibrium (LD), the associated SNP often merely marks a causal variant that travels with it. Identifying the causal variant and gene requires fine-mapping (denser genotyping or sequencing plus statistical methods that shrink the candidate set, for example Bayesian credible sets) and functional assays. The conventional threshold for genome-wide significance is P < 5 × 10⁻⁸, and effects are reported as odds ratios (OR), which for most non-MHC AS loci are modest (1.1–1.3). Because LD structure and allele frequencies differ between populations, a tag SNP discovered in one ancestry may not track the same causal variant — or show the same effect size — in another.

European-ancestry GWAS had established non-MHC AS loci at ERAP1, IL23R, 2p15, 21q22, IL12B and others, and shown that ERAP1's effect exists only in HLA-B27-positive individuals (epistasis, i.e. a statistical interaction between two variants). Lin et al. (2012) reported the first large GWAS in Han Chinese (1,837 cases and 4,231 controls in discovery; 2,100 cases and 3,496 controls in validation). They found two new loci: rs4552569 between EDIL3 (EGF-like repeats and discoidin I-like domains 3, an adhesion protein) and HAPLN1 (hyaluronan and proteoglycan link protein 1, a stabilizer of the cartilage extracellular matrix) at 5q14.3, and rs17095830 within ANO6 (anoctamin 6, a calcium-activated phospholipid scramblase with roles in membrane remodeling and mineralization) at 12q12. Notably, these genes relate to bone formation and cartilage development, which resonates with AS's pathological new bone formation. The study confirmed the MHC association (top SNP rs13202464, which the authors showed mainly represents the risk of HLA-B27 subtypes B2704, B2705 and B2715 in Chinese) and the European 2p15 locus, while the suggestive 6q21 signal (rs13210693, OR ≈ 1.15, P = 9.31 × 10⁻⁷ in the combined scan) did not reach genome-wide significance. Signals at European-discovered loci such as ERAP1 were much weaker in Han Chinese than in Europeans, raising the question of ancestry heterogeneity. The paper's discussion thus posed a follow-up agenda: which genes and variants are causal at the new loci; what the additional non-HLA-B27 MHC risk variation is; whether 6q21 is real; what causes the cross-ancestry effect differences; and whether the cartilage/bone-related genes are functionally involved in the cartilage development and pathological new bone formation that characterize the disease. The MHC-identity part of this agenda was subsequently answered — in Europeans by Cortes et al. (2015), who fine-mapped the MHC to specific HLA alleles, amino-acid positions and ERAP1 epistasis, and in East Asians by Kim et al. (2015, HLA-C amino-acid variant in Koreans) and Wang et al. (2020), who used Immunochip genotyping with HLA imputation in 1,637 Chinese/Taiwanese/Korean cases and 1,589 controls to identify HLA-C15 (OR 7.62, P = 9.30 × 10⁻¹⁹), HLA-B40 and HLA-DQB1*04 as additional risk alleles, with HLA-B amino-acid positions 70 and 97 driving the associations — so the remaining open problem is the causal, mechanistic and ancestry-heterogeneity resolution of the non-MHC discoveries.

Problem Statement

For ankylosing spondylitis (AS) in Han Chinese and East Asian populations, Lin et al. (2012, Nature Genetics, doi:10.1038/ng.1005) raised the following open questions. The core target is the causal resolution of the two novel loci — identifying the causal variant(s) and gene(s) behind the tag-SNP associations (item 1), together with whether the implicated cartilage/bone-formation genes are functionally involved in the disease (item 4). The status of the 6q21 signal (item 2) and the causes of cross-ancestry effect differences (item 3) are related open questions arising from the same study's reported data.

  1. Causal resolution of the two novel loci. Identify the causal variant(s) and gene(s) — HAPLN1, EDIL3, ANO6, or other nearby genes — mediating the tag-SNP associations at 5q14.3 (rs4552569) and 12q12 (rs17095830).
  2. Status of the 6q21 signal (related open question). Determine whether rs13210693, or the 6q21 region more generally, is a genuine AS susceptibility locus, with cohorts powered to detect an effect of the originally reported size (OR ≈ 1.15).
  3. Ancestry heterogeneity (related open question). Characterize the extent and causes of the differences in effect-size estimates and allele frequencies between Han Chinese and European populations at AS loci such as ERAP1 and IL12B.
  4. Functional and mechanistic consequences. Establish whether the cartilage/bone-related susceptibility genes (e.g., HAPLN1, EDIL3, ANO6) are functionally involved in the cartilage development and pathological new bone formation that characterize AS.

The fifth component of the original question — identification of the additional non-HLA-B*27 MHC risk variants indicated by conditional analysis — has been substantially resolved by later work (Kim et al. 2015, doi:10.1186/s13075-015-0855-3; Cortes et al. 2015, doi:10.1038/ncomms8146; Wang et al. 2020, doi:10.1186/s13075-020-02148-5); the functional characterization of those HLA alleles remains open.

References

  1. Lin Z, Bei J-X, Shen M, Li Q, Liao Z, Zhang Y, Lv Q, Wei Q, Low H-Q, Guo Y-M, Cao S, Yang M, Hu Z, Xu M, Wang X, Wei Y, Li L, Li C, Li T, Huang J, Pan Y, Jin O, Wu Y, Wu J, Guo Z, He P, Hu S, Wu H, Song H, Zhan F, Liu S, Gao G, Liu Z, Li Y, Xiao C, Li J, Ye Z, He W, Liu D, Shen L, Huang A, Wu H, Tao Y, Pan X, Yu B, Tai ES, Zeng Y-X, Ren EC, Shen Y, Liu J, Gu J. A genome-wide association study in Han Chinese identifies new susceptibility loci for ankylosing spondylitis. Nature Genetics 44(1):73-77 (2012). doi:10.1038/ng.1005. https://doi.org/10.1038/ng.1005
  2. Zhai J, Rong J, Li Q, Gu J. Immunogenetic study in Chinese population with ankylosing spondylitis: are there specific genes recently disclosed? Clinical and Developmental Immunology 2013:419357 (2013). doi:10.1155/2013/419357. https://doi.org/10.1155/2013/419357
  3. Wei JC-C, Hsu Y-W, Hung K-S, Wong R-H, Huang C-H, Liu Y-T, Guo Y-C, Ikegawa S, Chang W-C. Association study of polymorphisms rs4552569 and rs17095830 and the risk of ankylosing spondylitis in a Taiwanese population. PLoS ONE 8(1):e52801 (2013). doi:10.1371/journal.pone.0052801. https://doi.org/10.1371/journal.pone.0052801
  4. Wen Y-F, Wei J-C, Hsu Y-W, Chiou H-Y, Wong H-S, Wong R-H, Ikegawa S, Chang W-C. rs10865331 associated with susceptibility and disease severity of ankylosing spondylitis in a Taiwanese population. PLoS ONE 9(9):e104525 (2014). doi:10.1371/journal.pone.0104525. https://doi.org/10.1371/journal.pone.0104525
  5. Kim K, Bang SY, Lee S, Lee HS, Shim SC, Kang YM, Suh CH, Sun C, Nath SK, Bae SC, Kim T-H. An HLA-C amino-acid variant in addition to HLA-B*27 confers risk for ankylosing spondylitis in the Korean population. Arthritis Research & Therapy 17:342 (2015). doi:10.1186/s13075-015-0855-3. https://doi.org/10.1186/s13075-015-0855-3
  6. Cortes A, Pulit SL, Leo PJ, Pointon JJ, Robinson PC, Weisman MH, Ward M, Gensler LS, Zhou X, Garchon HJ, Chiocchia G, Nossent J, Lie BA, Førre Ø, Tuomilehto J, Laiho K, Bradbury LA, Elewaut D, Burgos-Vargas R, Stebbings S, Appleton L, Farrah C, Lau J, Haroon N, Mulero J, Blanco FJ, Gonzalez-Gay MA, Lopez-Larrea C, Bowness P, Gaffney K, Gaston H, Gladman DD, Rahman P, Maksymowych WP, Crusius JB, van der Horst-Bruinsma IE, Valle-Oñate R, Romero-Sánchez C, Hansen IM, Pimentel-Santos FM, Inman RD, Martin J, Breban M, Wordsworth BP, Reveille JD, Evans DM, de Bakker PI, Brown MA. Major histocompatibility complex associations of ankylosing spondylitis are complex and involve further epistasis with ERAP1. Nature Communications 6:7146 (2015). doi:10.1038/ncomms8146. https://doi.org/10.1038/ncomms8146
  7. Wang Q, Yang Y, Lv J, Lin Q, Wang L, Fang Z. Association study of single nucleotide polymorphisms rs4552569/rs17095830 with ankylosing spondylitis in a Chinese population. Open Rheumatology Journal 10:1-7 (2016). doi:10.2174/1874312901610010001. https://doi.org/10.2174/1874312901610010001
  8. Layh-Schmitt G, Lu S, Navid F, Brooks SR, Lazowick E, Davis KM, Montagna C, Gadina M, Colbert RA. Generation and differentiation of induced pluripotent stem cells reveal ankylosing spondylitis risk gene expression in bone progenitors. Clinical Rheumatology 36(1):143-154 (2017). doi:10.1007/s10067-016-3469-5. https://doi.org/10.1007/s10067-016-3469-5
  9. Li Z, Brown MA. Progress of genome-wide association studies of ankylosing spondylitis. Clinical and Translational Immunology 6(12):e163 (2017). doi:10.1038/cti.2017.49. https://doi.org/10.1038/cti.2017.49
  10. Gao S, Wang H, Zhou C, Guan M, Li X, Dai J, Zou Y, Fang H. Association of rs4552569 and rs17095830 single-nucleotide polymorphisms with susceptibility to ankylosing spondylitis in east Asian population: a meta-analysis. Journal of Genetics 97(4):825-833 (2018). doi:10.1007/s12041-018-0966-0. https://doi.org/10.1007/s12041-018-0966-0
  11. Wang G, Kim T-H, Li Z, Cortes A, Kim K, Bang S-Y, Leo P, Brown MA, Xu H. MHC associations of ankylosing spondylitis in East Asians are complex and involve non-HLA-B27 HLA contributions. Arthritis Research & Therapy 22(1):74 (2020). doi:10.1186/s13075-020-02148-5. https://doi.org/10.1186/s13075-020-02148-5
  12. Zheng X, Li Q, Li X, Zhang Y, Wu X, Wei Q, Cao S, Yang M, Lin Z, Liao Z, Qi J, Lv Q, Wang L, Li Y, Irwanto A, Cheng C-Y, Chai X, Khor CC, Heng CK, Koh W-P, Yuan J-M, Bei J, Zhang F, Zhang X, Zeng Y, Shen Y, Liu J, Gu J. Analysis of 47 non-MHC ankylosing spondylitis susceptibility loci regarding associated variants across Whites and Han Chinese. Journal of Rheumatology 47(5):674-681 (2020). doi:10.3899/jrheum.190184. https://doi.org/10.3899/jrheum.190184
  13. Ko C-L, Lin W-Z, Lee M-T-M, Chang Y-T, Lin H-C, Wu Y-S, Lin J-F, Pan K-T, Chang Y-C, Lee K-H, Lee Y-L, Hsieh T-T, Huang J-H, Wang C-H, Yang S-S, Chen H-C, Chu C-M. Genome-wide association study reveals ethnicity-specific SNPs associated with ankylosing spondylitis in the Taiwanese population. Journal of Translational Medicine 20(1):589 (2022). doi:10.1186/s12967-022-03701-3. https://doi.org/10.1186/s12967-022-03701-3
  14. Nam B, Jo S, Bang SY, Park Y, Shin JH, Park YS, Lee S, Joo KB, Kim T-H. Clinical and genetic factors associated with radiographic damage in patients with ankylosing spondylitis. Annals of the Rheumatic Diseases 82(4):527-532 (2023). doi:10.1136/ard-2022-222796. https://doi.org/10.1136/ard-2022-222796
  15. Lian Z, Luo W, Liu J, Wang J, Chai W, Wang Y, Sethi S, Ma X. Analysis of ANO6, HAPLN1, and EDIL3 polymorphisms in patients with ankylosing spondylitis in a Chinese Han population: a case-control study. Genetic Testing and Molecular Biomarkers 28(10):385-392 (2024). doi:10.1089/gtmb.2023.0569. https://doi.org/10.1089/gtmb.2023.0569
  16. Brown M, Li Z, Wu X, Harvey N, Garrido-Mesa J, Wang X, Xu Z, Wang G, Marzo-Ortega H, McGonagle D, Morgan A, Akkoc N, Artin GK, Macfarlane G, Jones G, Bradbury L, Leo P, Zimmerman K, Duncan E, Brown J, Merriman T, Stebbings S, Mahmoudi M, Jamshidi A, Farhadi E, Haroon N, Inman R, Breban M, Gensler L, Ward M, Wordsworth B, Weisman M, Evans D, Kenna T, Kim T-H, Reveille J, Xu H. Multi-omic insights from a multi-ancestry genome-wide meta-analysis of ankylosing spondylitis reveal novel pathways of disease susceptibility. Research Square preprint rs-6917334 (2025). doi:10.21203/rs.3.rs-6917334/v1. https://doi.org/10.21203/rs.3.rs-6917334/v1
  17. Yang M, Wen X, Wu X, Zhang Y, Gu J. Association of ANO6 gene polymorphism and expression with ankylosing spondylitis in a Chinese Han population. Reumatologia 64(3):192-199 (2026). doi:10.5114/reum/213485. https://doi.org/10.5114/reum/213485