Influence of elevated-CRP level-related polymorphisms in non-rheumatic Caucasians on the risk of subclinical atherosclerosis and cardiovascular disease in rheumatoid arthritis

Association between elevated C-reactive protein (CRP) serum levels and subclinical atherosclerosis and cardiovascular (CV) events was described in rheumatoid arthritis (RA). CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 exert an influence on elevated CRP serum levels in non-rheumatic Caucasians. Consequently, we evaluated the potential role of these genes in the development of CV events and subclinical atherosclerosis in RA patients. Three tag CRP polymorphisms and HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 were genotyped in 2,313 Spanish patients by TaqMan. Subclinical atherosclerosis was determined in 1,298 of them by carotid ultrasonography (by assessment of carotid intima-media thickness-cIMT-and presence/absence of carotid plaques). CRP serum levels at diagnosis and at the time of carotid ultrasonography were measured in 1,662 and 1,193 patients, respectively, by immunoturbidimetry. Interestingly, a relationship between CRP and CRP serum levels at diagnosis and at the time of the carotid ultrasonography was disclosed. However, no statistically significant differences were found when CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 were evaluated according to the presence/absence of CV events, carotid plaques and cIMT after adjustment. Our results do not confirm an association between these genes and CV disease in RA.

a relationship between CRP and CRP serum levels at diagnosis and at the time of the carotid ultrasonography was disclosed. However, no statistically significant differences were found when CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 were evaluated according to the presence/absence of CV events, carotid plaques and cIMT after adjustment. Our results do not confirm an association between these genes and CV disease in RA.
Rheumatoid arthritis (RA) is a chronic disease related to an increased risk of cardiovascular (CV) mortality and high prevalence of subclinical atherosclerosis 1 . Besides traditional CV risk factors 2 and inflammation 3 , a genetic component appears to be crucial in these processes 2,4-6 .
C-reactive protein (CRP) is currently the most widely used biomarker of inflammation 7 . The production of this protein occurs almost exclusively in the liver 8 . Several studies have suggested that elevated serum concentrations of CRP are related to the development of several processes associated with atherosclerosis such as coronary heart disease 9 and stroke 10 . Accordingly, a relationship between chronic inflammation determined by CRP serum levels and the development of both subclinical atherosclerosis 11 and CV events 2 has been found in RA patients.
Genetic variation is a major determinant of CRP levels. In this sense, a meta-analysis has identified several loci associated with elevated CRP serum levels in non-rheumatic Caucasian individuals 12 . CRP gene was found to be the most significant signal related to increased CRP serum levels 12 . Other polymorphisms also exert an influence on the elevated level of serum CRP in non-rheumatic Caucasians 12 12 .
Taking together all this information, the aim of the present study was to evaluate the potential influence of CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1, polymorphisms related to elevated CRP serum levels in non-rheumatic Caucasians, on the development of CV events and subclinical atherosclerosis in RA patients. For this purpose, we took advantage of data from a large series of RA patients assessed for CV disease.

Patients and Methodology
Subjects and Study Protocol. 2 Patients fulfilled the 2010 classification criteria for RA 13 . RA patients were evaluated for CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 polymorphisms. In addition, subclinical atherosclerosis was determined in 1,298 of the RA patients recruited in the study using a carotid ultrasound (US) technique (by evaluation of carotid intima-media thickness -cIMT-and presence/absence of carotid plaques). Also, CRP serum levels at the time of RA diagnosis and at the time of carotid US study were measured in a subgroup of 1,662 and 1,193 patients, respectively, by immunoturbidimetry.
Data on epidemiological and demographical features of the patients recruited in the present work are displayed in Table 1. Traditional CV risk factors and CV events were defined previously 2,14 .
Selection of polymorphisms and genotyping. Since CRP has been identified as the most relevant signal related to elevated CRP serum levels in non-rheumatic Caucasians 12 , we performed a tagging of this gene with the aim of covering all its variability. Consequently, we genotyped 3 polymorphisms: CRP rs1417938, CRP rs1800947 and CRP rs1205.
Genotyping was performed by TaqMan predesigned assays in a 7900 HT Real-Time polymerase chain reaction system (Applied Biosystems, Foster City, CA, USA).

US evaluation.
Measurement of the cIMT values and presence/absence of carotid plaques were evaluated in 1,298 cases. Patients from Santander, Granada, Tenerife, Valencia, Ciudad Real and Madrid were evaluated by a commercially scanner, Mylab 70, Esaote (Genoa-Italy) as previously described 15 . Patients from Lugo were measured by high-resolution B-mode ultrasound, Hewlett Packard SONOS 5500 as previously reported 16 . cIMT was measured at the far wall of the right and left common carotid arteries over the proximal 15 mm-long segment. cIMT was determined as the average of three measurements in each common carotid artery. Plaque criteria in the accessible extracranial carotid tree (common carotid artery, bulb and internal carotid artery) were focal protrusion in the lumen at least cIMT > 1.5 mm, protrusion at least 50% greater than the surrounding cIMT, or arterial lumen encroaching > 0.5 mm 17 . The carotid plaques were counted in each territory and defined as no plaque, unilateral plaque or bilateral plaques 17 . Agreement between these two US methods was previously reported 18 . Experts with a high reproducibility and an excellent inter-observer reliability in the evaluation of subclinical atherosclerosis in patients with RA performed the studies.
Evaluation of CRP serum levels. CRP serum levels at the time of RA diagnosis and at the time of the carotid US study were measured in a subgroup of 1,662 and 1,193 patients, respectively, by an automated immunoturbidimetric assay using the ADVIA Chemistry system (Siemens Healthcare Diagnostics Inc.). Serum samples of RA patients containing CRP form an immune complex with the assay reagent and antibodies against CRP that precipitates, increasing the turbidity of the sample. When light is passed through the reaction solution, some light is absorbed by the precipitates, which is measured at 596/694 nm. The level of CRP is determined by comparison with a calibrator of known concentration.
The relationship between allele/haplotype frequencies and the presence/absence of CV events was tested using Cox regression adjusting for gender, follow-up time, age at RA diagnosis and traditional CV risk factors as confounder factors. Results were expressed as hazard ratios (HR) with 95% confidence interval (CI).
Differences in the allele/haplotype frequencies according to the presence/absence of carotid plaques were calculated by χ 2 or Fisher tests when necessary (expected values below 5). Strength of associations was estimated using odds ratios (OR) and 95% CI. Results were adjusted for gender, follow-up time, age at the time of US evaluation and traditional CV risk factors as confounder factors by logistic regression.
Association between allele/haplotype frequencies and data on cIMT was evaluated by unpaired t test. Results were adjusted for gender, follow-up time, age at the time of US evaluation and traditional CV risk factors as confounder factors by analysis of covariance (ANCOVA).
Results on CRP serum levels were expressed as mean ± standard deviation (SD). The relationship between allele/haplotype frequencies and CRP serum levels at RA diagnosis and at the time of the carotid US study was evaluated by unpaired t test. The statistical software used to perform all the statistical analyses was STATA 12/SE (Stata Corp., College Station, TX, USA).

Influence of CRP polymorphisms on CV events or subclinical atherosclerosis in patients with RA.
We assessed the potential influence of CRP polymorphisms (rs1417938, rs1800947 and rs1205) on the risk of CV events or subclinical atherosclerosis in RA patients.
The linkage disequilibrium (LD) pattern of these 3 CRP polymorphisms obtained by HapMap Project phase I, II and III and Haploview (v.4.2) software and measured by r 2 coefficient is displayed in Supplementary Fig. online. After adjustment for potential confounder factors, no statistically significant differences were found when each CRP polymorphism was assessed independently and according to the presence/absence of CV events, carotid plaques and cIMT values (Table 2). Similarly, after adjustment for potential confounder factors, no statistically significant differences were detected when CRP polymorphisms were tested together conforming haplotypes and according to the presence/absence of CV events, carotid plaques and cIMT values ( Table 3). HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 polymorphisms on CV events or subclinical atherosclerosis in patients with RA. In addition, we evaluated the potential relationship between HNF1A rs1183910, LEPR rs4420065, GCKR rs1260326, NLRP3 rs12239046, IL1F10 rs6734238, PPP1R3B rs9987289, ASCL1 rs10745954, HNF4A rs1800961 and SALL1 rs10521222 polymorphisms and CV events or subclinical atherosclerosis in RA patients (Table 4). Accordingly, no significant differences were obtained when RA patients were stratified according to the presence/absence of CV events after adjustment for potential confounder factors (Table 4). It was also the case when RA patients were stratified according to the presence/absence of carotid plaques and the evaluation of cIMT after adjustment for potential confounder factors (Table 4).   Table S1). Moreover, the mean ± SD of CRP serum levels at the time of the carotid US study in patients carrying the minor CRP rs1205T allele was 5.02 ± 12.81 mg/l versus 6.71 ± 17.70 mg/l in those carrying the major CRP rs1205C allele (p = 0.018) (Supplementary Table S1). Consistent with these results, the mean ± SD of CRP serum levels at RA diagnosis in patients carrying the ACC haplotype (which harbors the minor CRP rs1417938A allele) was 14.08 ± 28.28 mg/l versus 10.23 ± 19.90 mg/l in those patients carrying the TCC haplotype, the most common haplotype found in our series (p = 0.0002) ( Supplementary Table S2). Also, the mean ± SD of CRP serum levels at the time of the carotid US study in patients carrying the TCT haplotype (which harbors the minor CRP rs1205T allele) was 4.34 ± 7.75 mg/l versus 6.14 ± 16.46 mg/l in those patients carrying the TCC haplotype (p = 0.023) (Supplementary Table S2). However, no association between HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 and both CRP serum levels at RA diagnosis and at the time of the carotid US study was disclosed (Supplementary Table S3).

Discussion
Several studies have suggested that CRP has direct effects on the vessel wall promoting atherosclerosis 19 . Among them, CRP seems to stimulate the production of cellular adhesion molecules by vascular endothelial cells, facilitate the adhesion and migration of monocytes through the vessel wall, mediate the uptake of low-density lipoprotein cholesterol by macrophages and cause complement activation 19 . An association between high-grade, chronic CRP elevation and subclinical atherosclerosis in patients with RA has been reported 11 . Additionally, a higher risk of CV events in patients with RA with chronic inflammation expressed by persistently increased CRP serum levels has been found 2 .
Since CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 have been described as significant signals associated with elevated CRP serum levels in non-rheumatic Caucasians 12 , we set up a large-scale study to determine the potential influence of these 10 genetic variants on the development of atherosclerotic disease in patients with RA. Interestingly, our results disclosed a relationship between CRP gene and CRP serum levels at RA diagnosis and at the time of the carotid US study. However, we could not find an association between the elevated CRP serum level-related variants in non-rheumatic Caucasians and the presence of subclinical atherosclerosis or CV events in RA patients.The lack of evidence for the association between these polymorphisms and atherosclerosis in RA patients does not exclude the implication of CRP in the pathogenesis of this disease. This protein is produced as a part of an intricate acute phase response where several inflammatory  factors, closely interrelated, are involved. Although we could not find an association of IL6 gene polymorphisms with CV disease 20 , it is possible that complex interactions between elevated CRP serum level-related genes in non-rheumatic Caucasians and other genes implicated in the inflammation cascade may lead to up-regulation of CRP, promoting the progression of accelerated atherosclerosis in RA patients. In summary, the results obtained in the present study do not confirm association between CRP, HNF1A, LEPR, GCKR, NLRP3, IL1F10, PPP1R3B, ASCL1, HNF4A and SALL1 and CV disease in patients with RA.