Polymorphism at the C-reactive protein locus influences gene expression and predisposes to systemic lupus erythematosus

Abstract

Relative deficiency of pentraxin proteins is implicated in the pathogenesis of systemic lupus erythematosus. The C-reactive protein (CRP) response is defective in patients with acute flares of disease, and mice with targeted deletions of the serum amyloid P component gene (Sap) develop a lupus-like illness. In humans, the genes for CRP (CRP) and SAP (APCS) map to 1q23.2 within an interval linked with SLE. We have investigated the candidate genes CRP and APCS in two cohorts totalling 586 UK simplex SLE families. The inheritance of an intronic dinucleotide repeat and seven single nucleotide polymorphisms in the CRP and APCS genes was examined by application of family-based tests of association and linkage. Basal levels of CRP were influenced independently by two polymorphisms at the CRP locus, CRP 2 and CRP 4. Furthermore, the latter polymorphism was linked/associated with SLE and antinuclear autoantibody production. Thus, the polymorphism associated with reduced basal CRP was also associated with the development of SLE. These data support the hypothesis that defective disposal of potentially immunogenic material is a contributory factor in lupus pathogenesis. The identification of polymorphisms that determine basal CRP levels has implications in ischaemic heart disease, where CRP level is an important predictor of risk.

Figure 1

The structure and orientation of the pentraxin genes. The pentraxin genes, CRP and APCS are paralogues sharing approximately 60% nucleotide identity. They both consist of two exons separated by a single intron. The first exon encodes the leader peptide and first two amino acids of the mature peptide and the second exon encodes the remaining 202 (SAP) and 204 (CRP) amino acids. The orientation of these genes in centromeric (Cen) and telomeric (Tel) directions is marked by arrows. Exons are represented by boxes, and introns and intergenic regions by lines. Filled boxes denote translated regions; open boxes untranslated regions. The region boxed in grey represents a portion of the gene downstream from the limit of the 3′-UTR, but which contains a possible polyadenylation signal. Arrows mark the locations of SNPs within the CRP and APCS genes. The GT dinucleotide repeat located within the intron of CRP is represented by CRP (GT)_n.

Figure 2

LD across the pentraxin locus. The results of linkage disequilibrium mapping using GENEHUNTER 2 are presented in this matrix. The values for D′ between each SNP are presented in each box. Black boxes denote strong LD (D′ > 0.8), dark grey boxes intermediate LD (D′ 0.7–0.8), pale grey boxes denote weak LD (D′ 0.5–0.7) and white boxes very weak or negligible LD (0.2 > D′ < 0.5).

Figure 3

The association of CRP haplotype with basal serum CRP levels. The left-hand part of the figure shows the five common CRP haplotypes with corresponding allelic composition at the five markers across the CRP locus. To the right of this the mean serum basal CRP levels are shown for each haplotype. Significant differences (P < 0.001) are indicated by double asterisks.

Figure 4

(A) CRP 4 genotype influences CRP levels. This bar chart illustrates the effect on CRP 4 genotype on serum CRP levels in the combined analysis of Caucasian probands and siblings. Median CRP levels in the heterozygous and 2/2 homozygous groups were both significantly lower than in 1/1 homozygotes (P < 0.0001, double asterisk). (B) The interaction of CRP 2 and CRP 4 genotype on basal CRP. In this analysis Caucasian probands and siblings were divided by their CRP 2 and CRP 4 genotypes. The results are graphed on this bar chart. Individuals homozygous for the major (1) allele at box loci are represented by the white bar, individuals homozygous for 1 at CRP 2 but heterozygous at CRP 4 by a pale grey bar. The dark grey bar represents double heterozygotes and the black bar, those individuals homozygous for 2 at CRP 4, and heterozygous at CRP 2. There were too few double homozygotes for meaningful analysis.