rs4349859 — HLA-B HLA-B27 proxy

HLA-B27 Status — The Dominant Genetic Risk Factor for Ankylosing Spondylitis

HLA-B27¹¹ HLA-B27
Human Leukocyte Antigen B*27 — a class I MHC molecule encoded in the highly polymorphic HLA region of chromosome 6, present in approximately 8% of Europeans and 60–90% of ankylosing spondylitis patients worldwide represents the single strongest genetic association with any common disease known to medicine — individuals carrying this allele face a roughly 60-fold increased risk for ankylosing spondylitis (AS) compared to non-carriers. Direct HLA typing is complex and expensive, but rs4349859 — a common intronic SNP located 41 kb centromeric of the HLA-B gene and 5.4 kb telomeric of MICA — serves as a highly accurate genetic proxy for European HLA-B27 subtypes, making it possible to infer HLA-B27 status directly from standard genotyping arrays.

The A allele of rs4349859 arose on a chromosome bearing HLA-B27 in an ancestor shared by nearly all European and some Asian HLA-B27 carriers. This ancient founder event means the A allele travels with HLA-B27 in strong linkage disequilibrium²² linkage disequilibrium
a statistical association between alleles at nearby loci that persists across generations because recombination has not yet broken apart the ancestral haplotype, allowing the SNP to stand in for direct HLA-B27 typing in research and clinical screening applications.

The Mechanism

rs4349859 is an intronic variant in MICA-AS1 (MICA antisense RNA 1), a non-coding RNA gene located between HLA-B and MICA in the MHC class I region. The SNP itself has no known functional consequence — it does not change any protein or alter gene expression. Its clinical relevance derives entirely from the haplotype it marks: the A allele at rs4349859 is a reliable indicator that the individual carries an HLA-B27 allele³³ HLA-B27 allele
most commonly HLA-B*27:05 in Europeans, the primary AS-associated subtype; also tags B*2702, B*2708, and B*2709 on the same chromosome.

HLA-B27 itself causes disease through several proposed mechanisms. The arthritogenic peptide hypothesis holds that HLA-B27 presents self-peptides (or microbial peptides with structural similarity to self) to CD8+ T cells, triggering an autoimmune cascade directed at joint tissues. Supporting this model, variants in ERAP1 — the enzyme that trims peptides before they are loaded onto HLA-B27 — affect AS risk exclusively in HLA-B27-positive individuals, directly implicating the peptide-HLA-B27 interaction in disease pathogenesis. A complementary hypothesis involves the spontaneous misfolding of HLA-B27 heavy chains into homodimers, which activate NK cells and innate lymphocytes independently of peptide presentation.

The Evidence

rs4349859 as HLA-B27 proxy. In the landmark epistasis study by Evans et al.⁴⁴ Evans et al.
Interaction between ERAP1 and HLA-B27 in ankylosing spondylitis implicates peptide handling in the mechanism for HLA-B27 in disease susceptibility. Nature Genetics, 2011, rs4349859 was used as the HLA-B27 proxy across a discovery cohort of 1,787 British and Australian cases plus 4,800 controls and a replication cohort of 2,111 cases and 4,483 controls. The SNP achieved 98.0% sensitivity and 99.0% specificity for HLA-B27 in the European ancestry subset (538 cases and 741 controls with known HLA-B27 status). The dominant model (A allele present = HLA-B27 positive) was used throughout, with the study confirming that ERAP1 variants have no detectable effect on AS risk in rs4349859 GG individuals (presumed HLA-B27-negative), but powerfully modify risk in A-allele carriers.

Population prevalence. A comprehensive New Zealand population study of 1,220 Caucasian controls found rs4349859 genotype frequencies of GG: 90.7%, AG: 8.8%, AA: 0.4%, corresponding to an A-allele frequency of ~4.7%⁵⁵ corresponding to an A-allele frequency of ~4.7%, consistent with the approximately 8–9% HLA-B27 prevalence in Northern European populations (heterozygotes carry one A allele; both A alleles in homozygotes are rarer). Concordance with direct serological HLA-B27 typing was 98.7–100% in European-ancestry individuals. The SNP tags all major European AS-associated subtypes (B*2702, B*2705, B*2708) but does not tag African (B*2703) or most Asian subtypes (B*2704, B*2706, B*2707), limiting its utility to European-ancestry individuals.

HLA-B27 and AS risk. The overall odds ratio for AS given HLA-B27 positivity is approximately 60, the strongest common-variant disease association documented in human genetics. Despite this extraordinary OR, only 1–2% of HLA-B27-positive individuals develop AS in their lifetime⁶⁶ Despite this extraordinary OR, only 1–2% of HLA-B27-positive individuals develop AS in their lifetime
population attributable risk is nonetheless substantial because B27 prevalence is 8% in Europeans and 60-90% of AS cases carry B27. Risk rises to approximately 20% if a first-degree relative has AS. HLA-B27 accounts for approximately 25% of AS heritability, with 116 additional genetic loci contributing ~30% combined.

Practical Implications

Knowing your rs4349859 genotype provides a genetic proxy for HLA-B27 status without the need for traditional HLA typing. For GG individuals (no A allele), the probability of carrying HLA-B27 is below 2%; residual risk comes from rare B27-positive individuals whose HLA-B27 subtype is not tagged by this SNP. For AG individuals (one A allele), HLA-B27 positivity is overwhelmingly likely — approximately 90% of A-allele carriers in European populations are HLA-B27 positive. For AA individuals (two A alleles), HLA-B27 positivity is essentially certain, and these individuals likely carry two HLA-B27 alleles.

The clinical significance of HLA-B27 positivity depends heavily on context. In isolation, carrying HLA-B27 (A allele at rs4349859) means a modestly elevated lifetime risk of AS (~1–2% absolute) that rises substantially with additional genetic risk (ERAP1 variants) or family history. The most important implication is that inflammatory back pain symptoms in an HLA-B27-positive individual should be evaluated rapidly for axial spondyloarthritis, since early treatment with NSAIDs has disease-modifying effects on radiographic progression.

Interactions

rs4349859 × ERAP1 variants (rs26653, rs30187, rs10050860): Epistasis in ankylosing spondylitis. This is the defining interaction for the entire ERAP1 locus. Evans et al. 2011 demonstrated unequivocally that all ERAP1 AS associations are conditional on HLA-B27 status⁷⁷ Evans et al. 2011 demonstrated unequivocally that all ERAP1 AS associations are conditional on HLA-B27 status
ERAP1 SNPs show zero association with AS in HLA-B27-negative individuals (GG at rs4349859), while in HLA-B27-positive individuals (AG or AA at rs4349859), ERAP1 risk alleles confer 3–4-fold differences in AS risk. The combined interaction p-value was 7.3 × 10⁻⁶. This epistasis is mechanistically coherent: ERAP1 trims peptides that are subsequently loaded onto HLA-B27; HLA-B27 status determines whether the trimmed peptide repertoire reaches a disease-triggering threshold.

rs4349859 × rs12191877 (HLA-C*06:02 proxy): Co-occurring immune susceptibility. HLA-C*06:02 is the primary genetic risk factor for psoriasis. Individuals carrying both HLA-B27 (rs4349859 A allele) and HLA-C*06:02 (rs12191877 A allele) are at risk for both AS and psoriasis and face elevated risk for psoriatic arthritis, a spondyloarthritis subtype overlapping both conditions. ERAP1 variants interact with both HLA alleles, albeit through partially distinct mechanisms and peptide repertoires.