Portal de Periódicos da CAPES

Lack of Evidence for Genetic Association to RUNX1 Binding Site at PSORS2 in Different German Psoriasis Cohorts

2005; Elsevier BV; Volume: 124; Issue: 1 Linguagem: Inglês

10.1111/j.0022-202x.2004.23571.x

1523-1747

Ulrike Hüffmeier, Heiko Traupe, Harald Burkhardt, Funda Schürmeier-Horst, Jesús Lascorz, Beate Böhm, Jörg Lohmann, Markward Ständer, Jörg Wendler, Reinhard Kelsch, Claudia Baumann, Wolfgang Küster, Thomas F. Wienker, André Reis,

Monoclonal and Polyclonal Antibodies Research

A DNA variant, rs734232, altering a RUNX1 binding site was recently reported as susceptibility allele at PSORS2 (17q25) in cohorts of psoriasis patients from the US. A testing of this variant in psoriasis patients from Germany did not confirm this association in 300 trios nor in two case–control studies with 281 patients with psoriasis vulgaris and 375 patients with psoriatic arthritis, respectively. These results fail to support rs734232 as a psoriasis susceptibility factor in German psoriasis patients. A DNA variant, rs734232, altering a RUNX1 binding site was recently reported as susceptibility allele at PSORS2 (17q25) in cohorts of psoriasis patients from the US. A testing of this variant in psoriasis patients from Germany did not confirm this association in 300 trios nor in two case–control studies with 281 patients with psoriasis vulgaris and 375 patients with psoriatic arthritis, respectively. These results fail to support rs734232 as a psoriasis susceptibility factor in German psoriasis patients. human leukocyte antigen single nucleotide polymorphism Psoriasis is a human leukocyte antigen (HLA)-associated inflammatory skin disease that affects about 2% of the European population. Numerous genetic studies have confirmed that HLA-C or the neighboring region harbor the yet unknown main disease-causing variant (PSORS1). Many other susceptibility loci have been reported during the last years, whereas a disease-causing variant was not proposed until recently, whenHelms et al., 2003Helms C. Cao L. Krueger J.G. et al.A putative RUNX1 binding site variant between SLC9A3R1 and NAT9 is associated with susceptibility to psoriasis.Nat Genet. 2003; 35: 349-356https://doi.org/10.1038/ng1268Crossref PubMed Scopus (261) Google Scholar described that single nucleotide polymorphism (SNP) rs734232 in a putative RUNX1 binding site was strongly associated in a family-based psoriasis cohort. Association was also present, though weaker, in a population-based cohort of European descent. Further functional in vitro studies showed loss of specific binding of nuclear extracts from a human T cell line for the associated minor A allele as compared with the major G allele. To test whether rs734232*A might be a relevant disease-causing variant in one of our cohorts, we investigated this SNP in our set of 300 trios as well as in two new cohorts: one cohort of 281 single patients with psoriasis vulgaris without joint affection, the other one comprising 375 single patients with a rheumatologist's diagnosis of psoriatic arthritis. Due to the known considerable heterogeneity of psoriasis, we stratified for an a-helix coiled-coil rod homologue (HCR)-risk haplotype that is in strong linkage disequilibrium with the HLA-Cw6 risk allele at PSORS1, as previously described byAsumalahti et al., 2003Asumalahti K. Laitinen T. Lahermo P. et al.Psoriasis susceptibility locus on 18p revealed by genome scan in Finnish families not associated with PSORS1.J Invest Dermatol. 2003; 121: 735-740https://doi.org/10.1046/j.1523-1747.2003.12483.xAbstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar. After this stratification we tested again for association to rs734232. The family-based association test was informative in 178 (59.3%) of 300 families and gave no evidence for association to rs734232 (Table I). Likewise, allele frequencies were similar in the population-based psoriasis vulgaris and psoriatic arthritis as well as control cohorts (Table II), again excluding any genetic association. Hardy–Weinberg equilibrium for rs734232 was confirmed in all cohorts.Table IAllele frequencies of rs734232 in nuclear families and results of family-based association test (FBAT)Allele frequency for allele GaAllele frequency in parents.Allele frequency for allele AaAllele frequency in parents.Number of informative familiesZp-value300 nuclear families0.5500.450178 (59.3%)1.1740.240203 families with index positive for HCR*WWCCbPSORS1 risk haplotype.0.5500.450123 (61.8%)0.0250.98097 families with index negative for HCR*WWCCbPSORS1 risk haplotype.0.5500.45055 (59.1%)2.1110.035a Allele frequency in parents.b PSORS1 risk haplotype. Open table in a new tab Table IIAllele frequencies of rs734232 in single cases and controls and results of χ2 statisticsProband groupAllele frequency for allele GAllele frequency for allele Aχ2ap-valueaCompared with genotype frequencies in controls.Psoriasis vulgaris290 (52.5%)262 (47.5%)0.0300.863Controls381 (52%)351 (48%)Psoriatic arthritis411 (56.8%)313 (43.2%)3.2670.071Psoriasis vulgaris, positive for HCR*WWCCbPSORS1 risk haplotype.195 (52.4%)177 (47.6%)0.7820.376Psoriasis vulgaris, negative for HCR*WWCCbPSORS1 risk haplotype.95 (52.8%)85 (47.2%)1.4800.224Controls, positive for HCR*WWCCbPSORS1 risk haplotype.126 (48.8%)132 (51.2%)Controls, negative for HCR*WWCCbPSORS1 risk haplotype.223 (47.4%)247 (52.6%)Psoriatic arthritis, positive for HCR*WWCCbPSORS1 risk haplotype.204 (53.7%)176 (46.3%)1.4460.229Psoriatic arthritis, negative for HCR*WWCCbPSORS1 risk haplotype.207 (60.2%)137 (39.8%)12.9110.0003a Compared with genotype frequencies in controls.b PSORS1 risk haplotype. Open table in a new tab In the set of trio index cases as well as in the patients with psoriasis vulgaris we found deviation from Hardy–Weinberg equilibrium for two HCR SNP due to significant excess of patients heterozygous for the risk alleles. In contrast, Hardy–Weinberg equilibrium was confirmed in the patient group with psoriatic arthritis. As expected, association to risk alleles of the two HCR SNPs in the 300 trios was strong: the p-value of the family-based associaton test (FBAT)-statistics for HCR-325*T was 1.0 × 10−6 whereas for HCR-2327*G the value was calculated to be 1.4 × 10−5 (196 of 207 informative families, respectively). In the family study, the index case carried the HLA-Cw6-associated haplotype in two-thirds of the families (Table I). In the case–control study, we also found strong association to the two HCR-SNP whereas in the patients with psoriasis vulgaris the p-values in the χ2 statistic were lower (1.3 × 10−14 for HCR-325*T and 1.3 × 10−8 for HCR-2327*G) than the corresponding p-values for psoriatic arthritis patients (1.3 × 10−7 for HCR-325*T and 4.5 × 10−4 for HCR-2327*G). In the case–control study, we noted also a significant difference in the number of HCR*WWCC carriers between the patient cohort with psoriasis vulgaris without arthritis and the one with psoriatic arthritis (187 subjects (66.5%) vs 198 subjects (53.0%), χ2=12.0, p=5.3 × 10−4). Next, we stratified the nuclear families and single cases for the risk haplotype HCR*WWCC associated to PSORS1. In the subgroups of nuclear families negative for the PSORS1-associated risk haplotype, this procedure resulted in a weak association to the major G allele of rs734232 (p=0.035). In the psoriatic arthritis patients negative for the PSORS1-associated risk haplotype, an association to the major G allele was observed (p=3.0 × 10−4). This association remained significant even when corrected for multiple testing (we corrected for the six tests performed) using the overly conservative Bonferroni correction (pc=1.8 × 10−3). No changes of association results to rs734232 could be observed in any of the other subgroups (Table I and Table II). The calculation of the conditional power in the nuclear families for rs734232 was done for an additive model and type I error was set to an α-level of 0.01. A genetic factor accounting for a relative risk of 1.6 for heterozygotes and for a corresponding risk of 2.8 for homozygotes could have been detected with a power of 0.93, whereas a genetic factor conferring the same relative risk for heterozygotes and a relative risk of 3.2 for homozygotes could have been detected with a power of 0.98. In the population-based study calculation of power was done under the assumption of an logarithmic additive model and a type I error rate of 0.05. The number of cases and controls has adequate power (0.8) to detect a genetic factor conferring a relative risk of 1.4 in the cohort of patients with psoriasis vulgaris and a corresponding risk of 1.35 in the one with psoriatic arthritis patients. To investigate whether rs734232, recently proposed as a disease-causing variant at PSORS2 (Helms et al., 2003Helms C. Cao L. Krueger J.G. et al.A putative RUNX1 binding site variant between SLC9A3R1 and NAT9 is associated with susceptibility to psoriasis.Nat Genet. 2003; 35: 349-356https://doi.org/10.1038/ng1268Crossref PubMed Scopus (261) Google Scholar), might be a relevant genetic factor contributing to psoriasis in patients from Germany, we tested for association in a family-based cohort and two case–control studies. We failed to detect any association to rs734232 and therefore fail to support that the putative RUNX1 binding site is a susceptibility factor in the German cohorts studied. To estimate whether our cohorts were adequate to detect this factor, we performed a calculation of conditional power in the nuclear families using the genotypes observed. Using a conservative model for an additive effect association could have been missed for a genetic factor conferring a relative risk of less than 1.6 in heterozygotes. An estimation of the power of our population-based cohorts shows that possible association could have been detected for even smaller relative risks. The discordance with data fromHelms et al., 2003Helms C. Cao L. Krueger J.G. et al.A putative RUNX1 binding site variant between SLC9A3R1 and NAT9 is associated with susceptibility to psoriasis.Nat Genet. 2003; 35: 349-356https://doi.org/10.1038/ng1268Crossref PubMed Scopus (261) Google Scholar might be explained by genetic heterogeneity that has been shown in psoriasis for example in the different linkage studies for which replication was seldom possible. PSORS1 is the major disease-causing factor in the genetics of especially the early onset form of psoriasis. We have previously reported association to HLA-Cw6 in a subgroup comprising 52 of 210 trios (Schmitt-Egenolf et al., 2001Schmitt-Egenolf M. Windemuth C. Hennies H.C. et al.Comparative association analysis reveals that corneodesmosin is more closely associated with psoriasis than HLA-Cw*0602-B*5701 in German families.Tissue Antigens. 2001; 57: 440-446https://doi.org/10.1034/j.1399-0039.2001.057005440.xCrossref PubMed Scopus (29) Google Scholar) as well as to marker tn62 adjacent to HLA-C in the full set of 210 trios (Hensen et al., 2003aHensen P. Asadullah K. Windemuth C. et al.Interleukin-10 promoter polymorphism IL10. G and familial early onset psoriasis.Br J Dermatol. 2003; 149: 381-385https://doi.org/10.1046/j.1365-2133.2003.05411.xCrossref PubMed Scopus (20) Google Scholar). We could also detect association to several regions on chromosome 19 in this cohort (Hensen et al., 2003bHensen P. Windemuth C. Huffmeier U. et al.Association scan of the novel psoriasis susceptibility region on chromosome 19: Evidence for both susceptible and protective loci.Exp Dermatol. 2003; 12: 490-496https://doi.org/10.1034/j.1600-0625.2003.00040.xCrossref PubMed Scopus (25) Google Scholar) corresponding to or adjacent to a region for which we previously reported linkage in an independent set of families (Lee et al., 2000Lee Y.A. Ruschendorf F. Windemuth C. et al.Genomewide scan in german families reveals evidence for a novel psoriasis-susceptibility locus on chromosome 19p13.Am J Hum Genet. 2000; 67: 1020-1024https://doi.org/10.1086/303075Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar). In the set of trios that we could extend to 300 nuclear families in the meantime, we found strong association to the two HCR SNP tested. Therefore we could confirm that PSORS1 is a relevant susceptibility locus in this study group. About two-thirds of trios as well as of single cases and more than half of patients with psoriatic arthritis are positive for the HCR-risk haplotype, which was used as an indirect measure of the PSORS1 risk allele. The HCR-risk haplotype HCR*WWCC as previously described byAsumalahti et al., 2003Asumalahti K. Laitinen T. Lahermo P. et al.Psoriasis susceptibility locus on 18p revealed by genome scan in Finnish families not associated with PSORS1.J Invest Dermatol. 2003; 121: 735-740https://doi.org/10.1046/j.1523-1747.2003.12483.xAbstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar is in strong linkage disequilibrium to the HLA-Cw6 allele and might therefore serve as an estimate of the PSORS1 risk allele, even when the presence of this haplotype on some non-risk haplotypes at PSORS1 is known (Chia et al., 2001Chia N.V. Stuart P. Nair R.P. et al.Variations in the HCR (Pg8) gene are unlikely to be causal for familial psoriasis.J Invest Dermatol. 2001; 116: 823-824Crossref PubMed Scopus (23) Google Scholar). Our proportion of patients with psoriasis vulgaris carrying the PSORS1-associated risk allele is in the same range as data from other studies (Enerback et al., 1997Enerback C. Martinsson T. Inerot A. Wahlstrom J. Enlund F. Yhr M. Swanbeck G. Evidence that HLA-Cw6 determines early onset of psoriasis, obtained using sequence-specific primers (PCR-SSP).Acta Derm Venereol. 1997; 77: 273-276PubMed Google Scholar; Queiro et al., 2003Queiro R. Torre J.C. Gonzalez S. Lopez-Larrea C. Tinture T. Lopez-Lagunas I. HLA antigens may influence the age of onset of psoriasis and psoriatic arthritis.J Rheumatol. 2003; 30: 505-507PubMed Google Scholar) that is suggestive of a good approximation of the used method to characterize the PSORS1 risk allele. The finding of a significant percentage of patients with psoriatic arthritis positive for the PSORS1 risk allele has been known from several studies (Murray et al., 1980Murray C. Mann D.L. Gerber L.N. Barth W. Perlmann S. Decker J.L. Nigra T.P. Histocompatibility alloantigens in psoriasis and psoriatic arthritis. Evidence for the influence of multiple genes in the major histocompatibility complex.J Clin Invest. 1980; 66: 670-675Crossref PubMed Scopus (90) Google Scholar; Beaulieu et al., 1983Beaulieu A.D. Roy R. Mathon G. et al.Psoriatic arthritis: Risk factors for patients with psoriasis – a study based on histocompatibility antigen frequencies.J Rheumatol. 1983; 10: 633-636PubMed Google Scholar; Gladman et al., 1986Gladman D.D. Anhorn K.A. Schachter R.K. Mervart H. HLA antigens in psoriatic arthritis.J Rheumatol. 1986; 13: 586-592PubMed Google Scholar). The large overlap of patients with and without arthritis carrying the PSORS1-associated HCR-risk haplotype suggests similar genetic factors in both groups. Nevertheless, the significant difference in PSORS1-associated risk allele frequency between the two patient groups might also indicate different genetic factors contributing to the arthropathy. We further investigated whether association to PSORS2 might be detected after stratification to PSORS1. Stratification of cohorts for the HLA-Cw6-associated HCR-risk haplotype indicated an association to the wild-type G allele in the subgroup of patients negative for the PSORS1 risk allele in the psoriatic arthritis group. This could mean that in this group of patients a different variant in linkage disequilibrium with the G allele is associated with the disease. Haplotype analyses are required to confirm or refute this hypothesis. No evidence for association in any of the other subgroups was found (Tables I and Table II). In summary, our results fail to support the putative disease-causing variant rs734232*A as a psoriasis susceptibility factor in psoriasis patients of German origin. Approval for the studies was obtained through the ethical committee of the Universities of Münster and Erlangen and the individuals gave their written informed consent. The investigations were conducted according to the Declaration of Helsinki principles. We investigated the formerly described cohort of 210 trios with psoriasis from Northern Germany (Hensen et al., 2003bHensen P. Windemuth C. Huffmeier U. et al.Association scan of the novel psoriasis susceptibility region on chromosome 19: Evidence for both susceptible and protective loci.Exp Dermatol. 2003; 12: 490-496https://doi.org/10.1034/j.1600-0625.2003.00040.xCrossref PubMed Scopus (25) Google Scholar) that we could expand to 300 nuclear families in the meanwhile. All index cases had an early onset form of psoriasis vulgaris with medium age of onset of 16.2±9.1 y. 53.3% of index patients were male. In 61% of trios, the father and/or the mother were also affected by psoriasis. The 281 single patients with psoriasis vulgaris were recruited through dermatology clinics at two psoriasis rehabilitation hospitals. An early onset form of psoriasis was diagnosed in all but three patients with a medium age of onset of 21.3±9.4 y. The majority of patients suffered from plaque type of psoriasis vulgaris. We excluded all patients with signs of psoriatic arthritis until the time of recruitment when medium age was 48±11.1 y. Sixty-one percent of patients were male. The 375 patients with psoriatic arthritis were recruited through four different rheumatological centers in Germany; the centers were one acute clinic, two rehabilitation hospitals and one private practice. Medium age of onset for psoriasis vulgaris was 30.1±13.0 y; 59.9% of patients were male. For 78% of the patients, the diagnosis of psoriatic arthritis was made ≥3 y prior to recruitment. Peripheral joint involvement was detectable in the majority of cases (343 or 91.4%), presenting in 85 (22.6%) patients as an oligoarthritis (affecting less than five joints) whereas polyarticular disease was recorded in 258 (68.8%) single cases. In 67% of patients with polyarthritis joint erosions were detectable by X-ray as signs of a destructive course of the disease as well as in 26% of oligoarthritis patients. Spinal involvement in form of typical spondyloarthritis and/ or sacroiliitis was observed in 72 cases (19%). The 376 controls had no psoriasis vulgaris at the time of recruitment when medium age was 32±10 y. All of them were healthy blood donors and recruited in Northern Germany. 58.7% of probands were male. We genotyped all SNP with Taqman assays on 7900HT Sequence Detection System (Applied Biosystems, Foster City, California). Sequences of primers and probes are given in the supplementary Table S1. Only SNPs with a genotyping rate of >97% were included. Taqman genotypes for all SNPs could be confirmed through direct sequencing in a set of 48 randomly chosen probands. Download .jpg (.04 MB) Help with files Table S1Sequences of primers and probes Association within the nuclear families was calculated using the family-based association test (FBAT,Laird et al., 2000Laird N.M. Horvath S. Xu X. Implementing a unified approach to family-based tests of association.Genet Epidemiol. 2000; 19: S36-S42https://doi.org/10.1002/1098-2272(2000)19:1+<::AID-GEPI6>3.3.CO;2-DCrossref PubMed Scopus (0) Google Scholar). Haplotype reconstruction for the two HCR SNPs in the trio index cases was done with the program PHASE (Stephens et al., 2001Stephens M. Smith N.J. Donnelly P. A new statistical method for haplotype reconstruction from population data.Am J Hum Genet. 2001; 68: 978-989https://doi.org/10.1086/319501Abstract Full Text Full Text PDF PubMed Scopus (6431) Google Scholar) to allow stratification for HCR haplotype. To determine significant differences in allele frequencies between the proband groups, a χ2 statistics was used. All SNPs were tested for Hardy–Weinberg-equilibrium. Haplotypes of the HCR gene in case groups were also reconstructed with PHASE (Stephens et al., 2001Stephens M. Smith N.J. Donnelly P. A new statistical method for haplotype reconstruction from population data.Am J Hum Genet. 2001; 68: 978-989https://doi.org/10.1086/319501Abstract Full Text Full Text PDF PubMed Scopus (6431) Google Scholar). The conditional power to detect association to rs734232 in the nuclear families was calculated as described by Knapp (Knapp, 1999Knapp M. A note on power approximations for the transmission/disequilibrium test.Am J Hum Genet. 1999; 64: 1177-1185Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar). To test whether our sample sizes of single cases and controls are sufficient to detect association to rs734232, we used the program QUANTO Version 0.5 (Gauderman, 2002Gauderman W.J. Sample size requirements for matched case – control studies of gene–environment interaction.Stat Med. 2002; 21: 35-50https://doi.org/10.1002/sim.973Crossref PubMed Scopus (519) Google Scholar). This work was supported by grants from the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG, Tr 228/5-4 and Re 679/10-4) and from the Interdisciplinary Centre for Clinical Research (IZKF B32) of the University of Erlangen-Nuremberg with a grant from The German Federal Ministry of Education and Research grant no. 01 KS 0002.