rs547154 — C2 IVS10

C2 IVS10 — The H7 Haplotype and Complement-Driven Eye Disease

The C2 gene encodes complement component 2, a serine protease that is indispensable for activating the classical complement pathway¹¹ classical complement pathway
the arm of innate immunity triggered by antibody-antigen complexes and certain pathogens. rs547154, also called IVS10 (intron variant 10), sits in an intron — a non-coding stretch of DNA that is spliced out before the protein is made. Yet this quiet intronic change is one of the most robustly replicated protective variants in ocular genetics, because it tags a specific chromosomal block that fundamentally alters how the complement system responds in ageing retinal tissue.

The Mechanism

rs547154 itself (G>T on the plus strand) does not change any amino acid in the C2 protein. Its power derives from haplotype context: it is inherited in extremely tight linkage disequilibrium (r²≈0.92–0.96) with the CFB R32Q missense variant (rs4151667)²² CFB R32Q missense variant (rs4151667)
the functional variant in complement factor B that reduces C2/CFB complex activity. Together, rs547154-T and CFB R32Q-A constitute the H7 protective haplotype — a segment of the HLA class III region on chromosome 6p21.3 that is inherited as a unit and reduces the efficiency of the classical complement pathway.

The classical pathway begins when C1q binds antibody-coated targets. C1 then cleaves C4 and C2, generating the C3 convertase (C4b2a) that splits C3 into C3b (opsonin) and C3a (anaphylatoxin). The R32Q substitution in CFB produces a less stable C3 convertase with reduced catalytic activity, blunting the entire downstream cascade. Less C3 cleavage means fewer complement fragments accumulate in the retinal pigment epithelium, reducing chronic sub-retinal inflammation and drusen formation — the hallmarks of early AMD.

rs547154 itself may additionally influence C2 expression through intronic regulatory elements, though this has not been fully characterised. The strong LD with CFB R32Q means separating their independent contributions requires conditional analyses that go beyond most published studies.

The Evidence

The H7 haplotype was first described in Gold et al. 2006 in Nature Genetics³³ Gold et al. 2006 in Nature Genetics
case-control study of ~900 AMD cases and ~400 controls: the haplotype carried an odds ratio of 0.45 (95% CI 0.33–0.61) for AMD risk, independent of age, CFH Y402H, and ARMS2/HTRA1. The combined model of C2/CFB haplotype plus CFH variants correctly classified 74% of AMD cases and 56% of controls — a remarkable predictive accuracy for a complex disease.

The individual rs547154 signal has since been replicated across populations and study designs. A HuGE meta-analysis of 19 studies⁴⁴ HuGE meta-analysis of 19 studies
Thakkinstian et al. 2012 found an odds ratio of 0.47 (95% CI 0.39–0.57) for the T allele, translating to a 2–6% absolute risk reduction in Caucasian cohorts. A separate meta-analysis of 15 studies covering 8,905 subjects⁵⁵ meta-analysis of 15 studies covering 8,905 subjects
Sun et al. 2012 found the dominant model OR of 0.40 (95% CI 0.29–0.55), meaning T carriers have roughly 60% of the AMD risk compared to GG individuals. The Lu et al. 2018 mega-analysis⁶⁶ Lu et al. 2018 mega-analysis
53,774 AMD cases and 56,973 controls across 53 studies confirmed the heterozygote OR of 0.52 (95% CI 0.43–0.62), with the strongest protection in Caucasians.

Protection extends beyond classic AMD. A meta-analysis of 4,076 subjects including polypoidal choroidal vasculopathy (PCV) cases⁷⁷ 4,076 subjects including polypoidal choroidal vasculopathy (PCV) cases
Chen et al. 2015 found the T allele equally protective against PCV (OR 0.64, p<0.0001), with no statistical difference between AMD and PCV subgroups. The effect replicates in Japanese populations⁸⁸ Japanese populations
Mori et al. 2012 (AMD OR 0.47, PCV OR 0.53, both p<0.01) and in Indian populations⁹⁹ Indian populations
Kaur et al. 2010 (protective haplotype OR 0.10, p=5.4×10⁻¹¹), demonstrating a cross-ethnic biological signal.

The variant is notably absent from AMD associations in some smaller populations (e.g. a Greek cohort of 260 subjects), likely because of low statistical power at the low T-allele frequency.

Practical Implications

For T allele carriers, this variant reduces complement-mediated retinal inflammation independent of other AMD risk loci. However, it does not confer immunity: AMD can still develop, particularly in those who smoke, carry CFH Y402H risk genotypes, or have other AMD risk factors. The protection is graded — one T allele gives roughly 50% risk reduction; two T alleles (rare) give maximum protection from this locus.

For GG homozygotes — about 83% of Europeans — this protective mechanism is absent, meaning AMD risk from this locus is unmodified. GG individuals should focus on other modifiable risk factors: smoking cessation, dietary carotenoids, omega-3 fatty acids, and regular ophthalmological surveillance.

Interactions

rs547154 acts in tight concert with CFB rs4151667 (R32Q) as the H7 haplotype. This is a distinct signal from the H10 haplotype tagged by C2 E318D (rs9332739) + CFB L9H (rs1270942). Both haplotypes reduce complement activation but through different structural perturbations in the C2/CFB complex and are inherited independently, so an individual can carry one, both, or neither.

The H7 haplotype protection is independent of and additive with the CFH Y402H (rs1061170) signal — the two variants operate at different nodes of the complement cascade (CFH regulates the C3 convertase at the alternative pathway level; C2/CFB modulates classical pathway C3 convertase assembly). C3 R102G (rs2230199) adds a third independent AMD risk locus in the same pathway. Individuals who carry the H7 haplotype and lack CFH risk alleles have the strongest overall complement protection; those who carry CFH risk alleles but also carry the H7 haplotype experience partial but meaningful buffering of their AMD risk.