Abstract
Objective. To investigate whether a functional integrin alpha M (ITGAM) variant is involved in susceptibility to and clinical manifestations of giant cell arteritis (GCA).
Methods. A Spanish cohort of 437 white patients with biopsy-proven GCA and 1388 healthy controls were genotyped using the TaqMan allele discrimination technology.
Results. No association was observed between ITGAM rs1143679 and GCA (p = 0.80, OR 0.97). Similarly, subphenotype analyses did not yield significant differences between the case subgroups and the control set or between GCA patients with or without the main specific features of GCA.
Conclusion. Our results suggest that the ITGAM rs1143679 variant does not play an important role in the pathophysiology of GCA.
Integrin alpha M (ITGAM) encodes the alpha subunit of the αMß2-integrin. A nonsynonymous single-nucleotide polymorphism (SNP) at exon 3 of the ITGAM gene, rs1143679, produces a functional modification in the protein by changing the 77th amino acid residue arginine to histidine (R77H). This genetic variant has been reported to be one of the signals most highly associated with systemic lupus erythematosus (SLE), and the most likely causal polymorphism within the ITGAM region in several populations of different ethnicity1,2. There is evidence suggesting that ITGAM rs1143679 is also involved in the pathogenesis of systemic sclerosis (SSc)3.
Giant cell arteritis (GCA) is a complex polygenic autoimmune disease characterized by inflammatory lesions of blood vessels, mainly involving medium- and large-size arteries. It represents the most common vasculitis in elderly individuals from Western countries4.
Accumulation of polyclonal T lymphocytes in the arterial lesions has been observed in patients with GCA, and it was hypothesized that integrin-type adhesion receptors may play a key role in this process5. Our objective was to determine whether the ITGAM functional polymorphism rs1143679 is implicated in the genetic susceptibility and specific clinical features of GCA.
MATERIALS AND METHODS
Study population
A Spanish cohort of 437 white patients with GCA and 1388 unrelated healthy controls recruited in the same geographic areas and matched by age, sex, and ethnicity was analyzed. All patients had a positive temporal artery biopsy6 and fulfilled the 1990 American College of Rheumatology classification criteria for GCA7. Table 1 shows their main clinical characteristics, as defined6,8.
Written informed consent from subjects and approval of the local ethical committees were obtained.
Genotyping methods
DNA was extracted from peripheral blood cells using standard procedures. All participants were genotyped for the ITGAM variant rs1143679 using the TaqMan allele discrimination assay technology (assay ID: C___2847895_1_) in a 7900HT Fast Real-time polymerase chain reaction system, both from Applied Biosystems (Foster City, CA, USA).
Statistical analysis
The overall statistical power of the analysis was 70% to detect associations, with OR = 1.3 at the 5% significance level, according to Power Calculator for Genetic Studies 2006 software (Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA; Website: http://www.sph.umich.edu/csg/abecasis/CaTS/). Plink (v1.07; Center for Human Genetic Research, Massachusetts General Hospital, Boston, MA, USA; Website: http://pngu.mgh.harvard.edu/purcell/plink/) was used to perform 2×2 contingency tables and chi-square test and/or Fisher’s exact test. OR and 95% CI were obtained according to Woolf’s method. P values < 0.05 were considered statistically significant.
RESULTS
No divergence from Hardy-Weinberg equilibrium was observed (p < 0.05) in controls or cases, and control allele frequencies were similar to those reported in the white European population9.
Table 2 shows the genotype and minor allele frequencies of the control cohort and the different case sets. No significant differences were detected between the allele frequencies of the patients with biopsy-proven GCA and the control group (p = 0.80, OR 0.97). To test for a possible gender-specific association, the female patients with GCA (n = 288) were compared with female controls (n = 825). No association of this polymorphism was observed in women (p = 0.65, OR 1.06).
GCA patients were further subdivided into those positive for polymyalgia rheumatica (PMR), visual ischemic manifestations (VIM), severe ischemic manifestations (SIM), and irreversible occlusive disease (IOD). However, analysis of data stratified according to the different phenotypes of clinical expression of this vasculitis revealed no significant heterogeneity between the GCA subgroups and the control population (PMR-positive vs controls, p = 0.71, OR 0.95; VIM-positive vs controls, p = 0.58, OR 0.89; SIM-positive vs controls, p = 0.75, OR 0.95; IOD-positive vs controls, p = 0.61, OR 0.89). Similar negative results were also observed when GCA patients with and without specific clinical disease features were compared to one another (Table 3). Finally, no significant deviation in genotype frequencies was evident in any of the comparisons above (data not shown).
DISCUSSION
Shared immunological pathways have been proposed to underlie different autoimmune conditions10. ITGAM rs1143679 has been associated with other autoimmune diseases, including SLE and SSc1,2,3. Previous studies in SLE predicted that the R77H substitution caused by this nonsynonymous SNP alter the structure and function of this integrin, and may contribute to endothelial injury and impairment of immune complex clearance in patients with SLE1,11,12. Because integrin molecules have been proposed to be crucial in the vasculopathy of GCA5, ITGAM represented a good susceptibility candidate locus for investigation. The power of this study and the differences in the effect magnitude between SSc and SLE make it difficult to draw definitive conclusions. However, considering the well-defined clinical cohort we analyzed, and that our statistical analysis had enough power to detect a possible moderate signal, it is unlikely that this gene may play a relevant role in GCA.
Recent studies in rheumatoid arthritis (RA) also described a lack of association with rs114367913. It could be speculated that common pathways involving the αMß2-integrin were restricted to specific subgroups of patients with RA and/or GCA that share more similarity with SLE and SSc. For instance, variation in the IRF5 gene has been associated with a subgroup of RA cases (not clearly delimited to date) with a pattern of association very similar to that described in patients with SLE14; no association of IRF5 rs2004640 and CGGGG insertion/deletion gene polymorphisms in the susceptibility to and clinical expression of GCA was described15. It should be noted that, despite not achieving statistical significance, all the OR that we observed were in the same direction; this might suggest a hypothetical role of ITGAM in a specific subgroup of patients that could not be categorized using the current definitions to describe the clinical features of GCA. However, the direction of the OR was the opposite of those reported for SLE and SSc1,2,3 and thus the previous assumption does not seem feasible. Another possibility could be that this gene is only weakly involved in predisposition to GCA, and the lack of high power of our study (e.g., > 90% to detect associations with OR = 1.2) did not make it possible to observe statistical significance. Further studies are needed to clarify whether the non-synonymous change caused by ITGAM rs1143679 is a key event in the development of autoimmunity or if it is only associated with some SLE-related manifestations.
Acknowledgment
The authors thank Sofía Vargas, Sonia García, and Gema Robledo for their excellent technical assistance, and all the patients and controls for kindly providing their essential collaboration. We thank Banco Nacional de ADN (University of Salamanca, Spain) for supplying part of the control material.
Footnotes
-
F.D. Carmona and A. Serrano share first authorship, and M.A. González-Gay and J. Martín share senior authorship of this report.
-
Supported by Fondo de Investigaciones Sanitarias, grants PI06-0024 and PS09/00748 (Spain), and by RETICS Program RD08/0075 (RIER) from Instituto de Salud Carlos III (ISCIII). F.D. Carmona was supported by Consejo Superior de Investigaciones Científicas (CSIC) through the JAE-DOC program.
- Accepted for publication July 21, 2011.