Elevated C-reactive protein levels and ARMS2/HTRA1 gene variants in subjects without age-related macular degeneration

PURPOSE To investigate the association between the serum high sensitivity C-reactive protein (hs-CRP) levels and variants in age-related maculopathy susceptibility 2 (ARMS2)/HtrA serine peptidase 1 (HTRA1) genes in normal subjects with no evidence of age-related macular degeneration (AMD). METHODS After clinical evaluation, information related to medical and social history was collected from 476 Japanese individuals (age range 17-89 years) along with blood samples. These subjects were medical checkup participants recruited at Nagoya University Hospital with no macular disease, as confirmed by fundus photographs. Serum hs-CRP levels were measured using a highly sensitive latex aggregation immunoassay. The genotypes of three polymorphisms in the ARMS2/HTRA1 locus, i.e., *372_815del443ins54 (del/ins), rs10490924, and rs11200638 were determined using direct sequencing and/or PCR-based assays. After the haplotype was constructed and analyzed, the associations between hs-CRP levels and representative del/ins genotypes were studied with and without adjustment for potential confounding factors. RESULTS All three polymorphisms in the ARMS2/HTRA1 region were in almost complete linkage disequilibrium. Haplotype analyses showed the existence of only two common haplotypes, together comprising 98.9%. Regression analyses showed that the level of hs-CRP was positively correlated with increasing age. This age-dependent increase of hs-CRP levels was greatest in those with homozygous del/ins alleles and lowest in those with homozygous wild-type alleles, which was significant assuming an additive model for gene-dosage association (univariate analyses: p=0.016, multivariate analyses including smoking status, past medical history, and BMI: p=0.043). Consequently, the level of hs-CRP was greatest in those with homozygous del/ins alleles and lowest in those with homozygous wild-type alleles when subjects older than 60 were analyzed. This was significant assuming an additive model for gene-dosage association (univariate analyses: p=0.032). CONCLUSIONS An age-dependent elevation of serum hs-CRP levels may be accelerated in normal subjects with one or two risk alleles in the ARMS2/HTRA1 locus compared to those with homozygous wild-type alleles. The results of the current study show that the as-yet undetermined function of variants in the ARMS2/HTRA1 locus might be linked to inflammation, possibly contributing to the development of neovascular AMD.