Interferon-γ and interferon-γ receptor 1 and 2 gene polymorphisms and restenosis following coronary stenting

Atherosclerosis 182 (2005) 145–151

Interferon-␥ and interferon-␥ receptor 1 and 2 gene polymorphisms and restenosis following coronary stenting K. Tiroch ∗ , N. von Beckerath, W. Koch, J. Lengdobler, A. Joost, A. Sch¨omig, A. Kastrati Deutsches Herzzentrum M¨unchen and 1. Medizinische Klinik rechts der Isar, Technische Universit¨at M¨unchen, M¨unchen, Germany Received 2 August 2004; received in revised form 29 January 2005; accepted 2 February 2005 Available online 24 February 2005

Abstract In a recent study, analysis of gene expression in atherectomy specimens derived from restenotic coronary lesions revealed 223 differentially expressed genes. Thirty-seven of these genes indicated activation of interferon- (IFN-) ␥ signaling in neointimal smooth muscle cells. Moreover, genetic disruption of IFN-␥ signaling in a mouse model of restenosis significantly reduced the vascular proliferative response. Thus, IFN-␥ is assumed to play an important role in the control of tissue proliferation during neointima formation. We hypothesized that genetic variants of IFN-␥ and its receptor subunits are involved in upregulation of IFN-␥ related genes in neointimal tissue of patients that develop in-stent restenosis. Polymorphisms in the genes encoding for IFN-␥ (IFNG T874A) and its receptors 1 (IFNGR1 C-56T) and 2 (IFNGR2 A839G) were tested for their association with restenosis. IFNG T874A, IFNGR1 C-56T and IFNGR2 A839G genotypes were determined in a consecutive series of patients (n = 2591) that had been treated with coronary stents. Follow-up angiography 6 months after stent implantation was performed in 76.8% of the patients. Genotyping was performed with PCR-based methods. IFNG T874A, IFNGR1 C-56T and IFNGR2 A839G genotypes were not associated with the incidence of angiographic and clinical restenosis (P > 0.23). Moreover, there was no association between IFNG, IFNGR1 and IFNGR2 genotypes and the combined incidence of death form any cause and non-fatal myocardial infarction during the first 12 months following the intervention (P > 0.61). Thus, this study does not support a clinically relevant role of the studied polymorphisms in the processes leading to in-stent restenosis. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Interferon-␥; Receptor; Restenosis; Genetics

1. Introduction Percutaneous intracoronary intervention has become the mainstay of therapy for coronary artery disease but restenosis continues to have a major impact on long-term outcome. Formation of restenosis that is secondary to neointimal proliferation is strongly mediated and controlled by inflammatory mechanisms and inflammatory cytokines [1–3]. Interferon(IFN-) ␥ and its receptor, consisting of two subunits acting together after heterodimerisation, are the starting line for a complex interaction of molecules resulting in a macrophage-rich inflammatory reaction [4]. In non-infectious diseases most of the effects of IFN-␥ are disease promoting [4]. ∗

Corresponding author. Tel.: +49 89 1218 4011; fax: +49 89 1218 4593. E-mail address: [email protected] (K. Tiroch).

0021-9150/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2005.02.003

In a previous study from our institution, analysis of the expression of 2435 genes in atherectomy specimens derived from restenotic lesions following coronary stent implantation revealed 223 differentially expressed genes compared to specimens from muscular arteries of the gastrointestinal tract and from coronary arteries of patients who underwent cardiac transplantation [5]. Thirty-seven of these genes indicated activation of IFN-␥ signaling in neointimal smooth muscle cells (SMC). In cultured SMC’s, IFN-␥ inhibits apoptosis. Genetic disruption of IFN-␥ signaling in a mouse model of restenosis significantly reduces the vascular proliferative response [5]. In two elegant studies, it was shown that interferon-␥ promotes atherosclerosis in animal models [6,7]. Thus, besides its involvement in atherosclerosis [6,7] IFN-␥ is suspected to play a pivotal role in the formation of in-stent restenosis.

146

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

The genes coding for IFN-␥ (IFNG, chromosomal localization 12q14.1.) and its structurally homologous receptor subunits 1 (IFNGR1, 6q23.2.) and 2 (IFNGR2, member of the cytokine receptor-II family, 21q22.11.) are polymorphic. The IFNG intron 1 T874A polymorphism lies within a NF␬B binding site with specific binding of NF␬B to the allelic sequence containing a T at position 874 [8]. Moreover, the 874T allele was in complete linkage disequilibrium with the 12 CA allele of the adjacent intron 1 CA repeat polymorphism. The 12 CA allele was previously shown to be associated with increased production of IFN-␥ in vitro [9]. The IFNGR1 C-56T polymorphism is located in the promoter region of IFNGR1 and the T allele is associated with higher susceptibility and lower morbidity of Helicobacter pylori infection and increased levels of anti-H. pylori antibodies. This suggests a shift of the Th1/Th2 balance of T helper cells towards Th2 resulting from a reduced IFN␥ receptor 1 function [10]. The A839G polymorphism of IFNGR2 [11] that is part of a gene cluster harboring other IFN receptor genes and the gene encoding for the interleukin(IL-) 10 receptor [12] has not been functionally characterized but leads to a change in the amino acid sequence (Gln64Arg). We hypothesized that variants in the genes encoding for IFN-␥ and its receptor subunits are involved in upregulation of IFN-␥ related genes in neointimal tissue of patients that develop in-stent restenosis. In this context, we tested the three above-mentioned polymorphisms in the genes of IFN-␥ and its receptor subunits 1 and 2 for association with restenosis and adverse clinical events following coronary stent implantation.

2. Methods 2.1. Patients The significance of the studied polymorphisms was evaluated in a cohort study that comprised 2591 consecutive Caucasian patients with symptomatic coronary artery disease who underwent coronary angiography and stent implantation at Deutsches Herzzentrum M¨unchen and 1. Medizinische Klinik rechts der Isar, Technische Universit¨at M¨unchen. Coronary stent implantation was performed as previously described [13,14]. Postprocedural therapy consisted of aspirin (100 mg twice daily, indefinitely) and ticlopidine (250 mg twice daily for 4 weeks) or clopidogrel (75 mg once daily for 4 weeks). Patients deemed at higher risk for stent thrombosis received additional therapy with abciximab given as a bolus injection during stent insertion procedure and as a 12-h continuous infusion thereafter. The decision to give abciximab was left to the operator’s discretion. Follow-up angiography was routinely scheduled at 6 months poststenting or whenever the patient complained of anginal symptoms and was available in 76% of the patients. Creatine kinase levels and ECG changes were assessed systematically over 48 h after

the procedure. Clinical events were monitored throughout a 1-year period following the intervention. The data regarding cardiovascular risk factors were obtained during the actual hospitalization or from the patient’s chart. The study was approved by the local Ethics Committee and informed genetic consent was obtained from all patients. 2.2. Definitions The diagnosis of myocardial infarction (MI) was based on the presence of new pathological Q waves or a value of creatine kinase or its MB isoenzyme at least three times the upper limit. Patients with acute MI were those who were admitted for the treatment of acute MI (time period from onset of pain to treatment <72 h) with percutaneous coronary intervention. Systemic arterial hypertension was defined as systolic blood pressure of 140 mmHg or greater and/or a diastolic blood pressure of 90 mmHg or greater at least on two separate occasions. Diabetes mellitus was defined in the presence of an active treatment with insulin or an oral antidiabetic agent; for patients on dietary treatment, documentation of an abnormal fasting blood glucose or glucose tolerance test based on the World Health Organization criteria [15] was required for establishing this diagnosis. Persons reporting regular smoking in the prior 6 months were considered as current smokers. Hypercholesterolemia was defined as documented total cholesterol ≥240 mg/dl. 2.3. Sequencing of the IFNG locus The IFNG locus was sequenced in 22 randomly selected patients from our institution. For this purpose 25 primers were designed with an interval of 250–350 base pairs (bp) spanning a region of 6650 bp including the four exons and the introns in between. The sequence of both DNA strands (forward and reverse) was determined with the BigDye® Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Pat. No. 4336776) and an ABI Prism 3100 Automated Sequencer (Applied Biosystems, Weiterstadt, Germany). Sequences were aligned with the published sequence (GenBank accession no. AF375790) and detected polymorphic sites were compared with those available in NCBI’s dbSNP. 2.4. Genotyping Genomic DNA was extracted from 200 ␮l of peripheral blood using commercially available kits (Qiagen, Hilden, Germany and Roche Molecular Biochemicals, Mannheim, Germany). Genotype analysis was based on PCR and involved fluorogenic oligonucleotide probes (TaqMan). TaqMan assays were performed with the ABI Prism Sequence Detection System 7700 (Applied Biosystems, Darmstadt, Germany) [16]. Primers and probes were selected with the help of previously reported sequences (GenBank accession no. AF375790 for IFNG, AY594694 for IFNGR1 and NM005534 for IFNGR2) surrounding the polymorphic sites

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

and the “Primer Express” software (Applied Biosystems). Of each probe pair, one probe was labeled with the fluorescent dye 6-carboxy-fluorescein (FAM) and the other with the fluorescent dye VIC (Applied Biosystems, patent pending). Two primers (forward and reverse) and two probes (FAM and VIC labeled) were used for each gene (IFNG, IFNGR1 and IFNGR2) with the following sequences: IFNG forward 5 GCT GTT ATA ATT ATA GCT GTC ATA ATA ATA TTC AGAC, IFNG reverse 5 -AGG GTA TTA TTA TAC GAG CTT TAA AAG ATA GTT C, IFNG probe FAM 5 -FAM-A ATC AAA TCT CAC ACA CAC, IFNG probe VIC 5 -VICATC AAA TCA CAC ACA CAC; IFNGR1 forward 5 -ACG GTC GCT GGC TCC AA, IFNGR1 reverse 5 -CGG TGA CGG AAG TGA CGT AAG, IFNGR1 probe FAM 5 -FAMCAG CCC AGC GCT GCC CTC, IFNGR1 probe VIC 5 VIC-ACC AGC CCA GCA CTG CCC TC; IFNGR2 forward 5 -GGT CCT GAG TTG GGA GCC A, IFNGR2 reverse 5 GGA TCC AAC AGA AAT ACC GGC, IFNGR2 probe FAM 5 -FAM-AGGC CTG TTG TCT ACC AAG TGC AGT TT, IFNGR2 probe VIC 5 -VIC-GGC CTG TTG TCT ACC GAG TGC AGT TT. DNA was amplified during 40 cycles of denaturation at 95 ◦ C for 15 s and primer annealing as well as extension at 65 ◦ C for 1 min (60 ◦ C for 1 min for IFNGR1). The DNA from two healthy volunteers served as standards for the TaqMan analysis for each polymorphism. These initial genotypes were determined by screening of 50 healthy volunteers through restriction enzymes analysis—HaeII for IFNGR1 and MunI for IFNGR2—and sequencing for IFNG polymorphism with the ABI PRISM 3100 Genetic Analyzer (Applied Biosystems) capillary sequencing system and the BigDye® Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Pat. No. 4336776) using the same primers as for the subsequent TaqMan assay. TaqMan assays were read by two independent operators and every fifth assay was repeated as a further control. 2.5. Angiographic assessment Quantitative angiographic analysis was performed off-line using the automated edge detection system CMS (Medis Medical Imaging Systems, Nuenen, The Netherlands) assessing matched views of the target lesions. The operators were not aware of the patients’ respective genotypes. The lesion morphology was assessed according to the modified American Heart Association/American College of Cardiology and classified as type A, B1, B2 or C. Lesions of types B2 and C were considered complex lesions [22]. Lesion length, reference diameter, minimal lumen diameter (MLD), diameter stenosis, and diameter of the maximally inflated balloon during the intervention were measured and assessed for each patient. Angiographic parameters were recorded before and immediately following the intervention, as well as at follow-up angiography. Late lumen loss was calculated as the difference between the final poststenting MLD and the MLD measured at follow-up angiography.

147

2.6. Study end points The primary end point of the study was restenosis that was evaluated angiographically and clinically. Angiographic restenosis was defined as a diameter stenosis of ≥50% at 6-month follow-up angiography. Clinical restenosis was defined as the necessity for target vessel revascularization (TVR) with percutaneous transluminal coronary angioplasty or aortocoronary bypass grafting due to symptoms or signs of ischemia during the first year following stenting. Secondary end point was the combined incidence of death from any cause and non-fatal MI during a 1-year follow-up period. 2.7. Statistical analysis Continuous variables were expressed as mean ± S.D. and were compared by means of analysis of variance. Discrete variables were expressed as counts or percentages and compared with the χ2 or Fisher’s exact test, as appropriate. Statistical analyses were performed using S-Plus software (Mathsoft Inc., Seattle, Washington). Consistency with Hardy–Weinberg equilibrium was tested with Pearson’s χ2 test. A P-value < 0.05 was considered statistically significant.

3. Results First, we sequenced the IFNG locus (including all four exons and the complete sequence of all introns) in 22 randomly selected patients from our institution. We identified eight SNPs, seven already available in NCBI’s dbSNP. The genotype distribution of these eight SNPs is displayed in Table 1. None of the eight detected polymorphism did affect the amino acid sequence of the gene product or was located in or close to a splice site at an exon/intron border. Therefore, we decided to analyze the functionally characterized IFNG intron 2 T874A polymorphism in our association study. The genotype distribution for IFNG T874A polymorphism was 23.1% TT, 46.8% TA and 30.1% AA, for IFNGR1 Table 1 The eight detected polymorphisms by sequencing the IFNG locus for 22 patients with the respective genotype distributions Polymorphic site

Common allele homozygotes

Heterozygotes

Rare allele homozygotes

rs2069707/Pos. −892 rs2430561/Pos. T874A rs1861494/Pos. 1984 rs1861493/Pos. 2197 rs2069716/Pos. 2578 rs2069733/Pos. 3161 rs2069718/Pos. 3232 Pos. 4150

18 CC 6 AA 12 AA 12 TT 20 AA 14 GG 10 CC 10 CC

4 CG 12 AT 8 AG 8 TC 2 AG 7 G10 CT 12 CT

0 GG 4 TT 2 GG 2 CC 0 GG 1 -2 TT 0 TT

G- and - - denote hetero- or homozygous carriage the deletion, respectively. The polymorphisms are displayed from 5 to 3 with the rs# number and position from the starting codon in the IFNG sequence in GenBank accession no. AF375790.

148

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

Table 2 Baseline characteristics of the patients according to IFNG, IFNGR1 and IFNGR2 genotypes IFNG T874A

Age (years) Women Arterial hypertension Diabetes Current smoker Hyper-cholesterolemia Acute MI Unstable angina Previous MI Previous CABG

IFNGR1 C-56T

IFNGR2 A839G

TT (n = 599)

TA (n = 1213)

AA (n = 779)

CC (n = 450)

CT (n = 1260)

TT (n = 881)

AA (n = 1998)

AG (n = 548)

GG (n = 45)

65 ± 11 23.9 73.6 19.4 25.7 76.5 23.0 31.9 28.4 13.7

65 ± 11 23.1 75.1 21.4 22.8 75.8 20.9 33.6 32.2 12.0

66 ± 12 27.7 77.4 24.1 22.0 76.1 20.2 32.9 31.1 11.3

64 ± 12 24.9 74.9 20.2 25.3 75.6 23.6 30.7 31.3 13.1

66 ± 11 25.6 75.1 22.8 23.0 76.7 20.6 34.0 31.5 11.8

65 ± 12 23.2 76.3 21.1 22.4 75.5 20.8 32.7 30.2 12.3

65 ± 11 24.4 74.8 22.6 23.5 75.8 22.0 33.0 31.2 11.9

66 ± 12 26.1 78.3 19.0 22.8 77.9 18.1 33.4 31.0 13.9

66 ± 9 20.0 71.1 17.8 15.6 64.4 20.0 28.9 24.4 6.7

28.9 31.1 39.7

29.9 30.6 39.3

30.9 31.8 36.7

28.1 29.9 41.7

29.5 30.9 39.3

29.5 30.9 39.1

27.7 29.4 42.7

26.7 28.9 44.4

Number of diseased vessels One vessel 28.4 Two vessels 29.6 Three vessels 41.6

P, not significant for all comparisons. Age is mean ± S.D.; other variables are percentage of patients. CABG, coronary artery bypass grafting; MI, myocardial infarction.

C-56T 17.4% CC, 48.6% CT and 34.0% TT, and for IFNGR2 A839G polymorphism 77.1% AA, 21.2% AG and 1.7% GG, respectively. The genotype distribution of both receptor subunit gene SNPs was consistent with Hardy–Weinberg equilibrium (P = 0.99 and 0.30 for IFNGR1 and IFNGR2 polymorphisms respectively) whereas this was not true for IFNG T874A polymorphism (P = 0.003). The baseline characteristics of the patients displayed in Table 2 were evenly distributed between the IFNG T874A, IFNGR1 C-56T and IFNGR2 A839G genotype groups. The same was true for lesion and procedural characteristics at the time of the intervention (Table 3). The lowest P-value that we yielded was 0.0509 for the rate of diabetes regarding the IFNG T874A genotypes. Since none of the analyses yielded P values < 0.05 it was not necessary to perform correction for multiple testing. IFNG T874A, IFNGR1 C-56T and IFNGR2 A839G polymorphisms did not have an appreciable influence on the occurrence of early thrombotic events. In the nine genotype groups the combined rate of death, non-fatal MI and TVR was between 4 and 6.6% (P > 0.12). Follow-up angiography 6 months after stent implantation was performed in 76.8% of the patients. The incidence of angiographic restenosis was not significantly different between genotype groups of the studied polymorphisms—ranging from 25.5 to 27% for IFNG T874A (P = 0.80), from 24.9 to 27.8% for IFNGR1 C-56T (P = 0.43) and from 15.6 to 28.3% for IFNGR2 A839G (P = 0.24). Other quantitative measures of angiographic restenosis and the incidence of clinical restenosis (P > 0.23) were also comparable within the different genotype groups (Table 4). The combined incidence of death and non-fatal MI during the first 12 months following the intervention ranged from 7.3 to 11.1% in the IFNG, IFNGR1 and IFNGR2 genotype groups (P > 0.61). The TVR rates were 16.9% (TT), 15.7% (TA) and 15.5% (AA) for IFNG T874A

(P = 0.77). For IFNGR1 the incidence of TVR was 17.3% (CC), 14.9% (CT) and 16.7% (TT) (P = 0.37) and for IFNGR2 this incidence was 16.2% (AA), 15.9% (AG) and 6.7% (GG) (P = 0.23).

4. Discussion The main result of this study is that the assessed polymorphisms in the genes encoding IFN-␥ and its two receptor subunits are of no detectable relevance for the development of restenosis and other adverse events following coronary stenting. The lack of statistical difference in restenosis rate between the different genotypes was independent of the presence or absence of a history of myocardial infarction or aortocoronary bypass surgery. A random false negative result is quite improbable given the large series of patients. The low incidence of restenosis in the group of patients carrying the IFNGR2 A839G GG genotype is probably a chance finding related to the small number of patients carrying this genotype. Moreover, the data does not suggest a gene-dose effect of the G allele. Overall, the reported genotype distributions agree with the genotype distributions described previously for the IFNG and IFNGR1 polymorphisms in Caucasians [10,17], and for the IFNGR2 polymorphism in Japanese [11]. The genotype distribution of the IFNG T874A polymorphism deviated from Hardy–Weinberg equilibrium. Incorrect genotyping can be excluded as a cause for this deviation. The standards for the TaqMan assay were derived from direct sequencing, the assays were read by two independent operators and every fifth assay was repeated as a further control (see methods). Thus, the deviation of IFNG genotypes from Hardy–Weinberg equilibrium may either be a chance finding or this SNP may actually be related to coronary artery disease for which the studied patients were treated.

Table 3 Lesion and procedural characteristics at the time of intervention IFNG T874A

IFNGR1 C-56T

IFNGR2 A838G

TA (n = 1213)

AA (n = 779)

CC (n = 450)

CT (n = 1260)

TT (n = 881)

AA (n = 1998)

AG (n = 548)

GG (n = 45)

Stented coronary vessel Left main (%) LAD (%) LCx (%) RCA (%) Venous bypass graft (%)

1.7 43.7 18.0 30.9 5.7

2.6 43.1 18.2 29.6 6.5

1.9 43.2 19.3 29.9 5.7

2.2 45.1 18.2 29.1 5.4

2.1 42.9 18.7 30.6 5.7

2.3 42.9 18.3 29.6 6.9

2.0 44.0 18.5 29.7 5.7

2.1 40.5 18.4 31.0 8.0

6.7 46.6 17.8 28.9 0.0

Chronic occlusion (%) Complex lesions (%) Lesion length (mm) Reference diameter (mm) Initial MLD (mm) Stented segment length (mm) Final MLD (mm) Therapy with abciximab (%)

6.8 79.3 13.9 ± 7.9 3.04 ± 0.52 0.69 ± 0.57 22.9 ± 11.5 2.94 ± 0.52 26.0

6.6 79.6 13.7 ± 7.9 3.02 ± 0.56 0.74 ± 0.61 22.6 ± 11.7 2.93 ± 0.54 23.0

7.8 82.4 14.4 ± 8.2 3.00 ± 0.56 0.69 ± 0.57 23.4 ± 11.9 2.91 ± 0.55 23.0

6.2 78.0 14.1 ± 8.0 2.99 ± 0.56 0.70 ± 0.61 23.6 ± 11.8 2.89 ± 0.55 21.6

6.5 81.0 13.8 ± 7.7 3.02 ± 0.55 0.71 ± 0.58 22.8 ± 11.5 2.94 ± 0.54 24.6

7.5 80.8 14.1 ± 8.6 3.02 ± 0.51 0.71 ± 0.56 23.2 ± 12.2 2.92 ± 0.50 23.8

6.7 80.0 13.9 ± 7.7 3.01 ± 0.55 0.71 ± 0.59 22.8 ± 11.5 2.92 ± 0.54 24.6

6.4 81.9 14.5 ± 8.9 3.03 ± 0.56 0.72 ± 0.57 23.4 ± 12.8 2.95 ± 0.52 21.4

15.6 80.0 12.3 ± 7.0 3.01 ± 0.44 0.71 ± 0.63 22.8 ± 9.4 3.01 ± 0.42 20.0

P, not significant for all comparisons. LAD, left anterior descending coronary artery; LCx, left circumflex coronary artery; MLD, minimal lumen diameter; MLD, minimal lumen diameter; RCA, right coronary artery.

Table 4 Results of 6-month follow-up angiography according to the IFNG, IFNGR1 and IFNGR2 genotypes IFNG T874A

Minimal lumen diameter (mm) Late lumen loss (mm) Angiographic restenosis rate (%)

IFNGR1 C-56T

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

TT (n = 599)

IFNGR2 A839G

TT (n = 452)

TA (n = 917)

AA (n = 596)

CC (n = 335)

CT (n = 507)

TT (n = 680)

AA (n = 1516)

AG (n = 417)

GG (n = 32)

1.86 ± 0.85 1.12 ± 0.75 26.5

1.86 ± 0.85 1.07 ± 0.74 25.5

1.83 ± 0.85 1.11 ± 0.74 27.0

1.77 ± 0.86 1.10 ± 0.74 26.6

1.85 ± 0.85 1.09 ± 0.75 24.9

1.86 ± 0.83 1.10 ± 0.75 27.8

1.84 ± 0.84 1.09 ± 0.73 25.9

1.87 ± 0.87 1.12 ± 0.79 28.3

2.23 ± 0.80 0.85 ± 0.57 15.6

P, not significant for all comparisons.

149

150

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

Previous results of our group favor a role for genetic determinants in the processes leading to in-stent restenosis. First, angiographic restenosis showed a bimodal distribution with one peak at 30% lumen narrowing and another peak at 70% lumen narrowing suggesting two groups of patients with different proneness for restenosis [19]. Second, an analysis of patients with multiple stenting revealed that the likelihood to develop restenosis is more than twice as high in the presence of other already restenosed lesions [20]. Attempts have been made to identify genetic variants that are related to formation of restenosis [21–25]. This study is unique in the way that the selection of the candidate genes was based on an extensive analysis of differentially expressed genes in neointimal atherectomy specimens that was carried out at our institution [5]. As controls in that study served specimens from muscular arteries of the gastrointestinal tract and from coronary arteries of patients who underwent cardiac transplantation. This previous gene expression study revealed that of the 223 differentially regulated genes 37 were related to IFN-␥ signaling including target genes of the IFN-␥ pathway like CD40, TRAIL and thrombospondin-1. Since the study by Zohlnh¨ofer et al. [5] showed extensive alteration of the entire interferon-␥ signaling cascade in neointimal smooth muscle cells we assumed that genes comprising the starting point of this signaling cascade are the most promising candidate genes. Therefore, we chose to restrict the genetic analysis to the gene of interferon␥ (IFNG) and its receptor heterodimer (IFNGR1, IFNGR2). Of note, the gene expression profiling revealed that an active IFN-␥ receptor containing both chains was mainly expressed in neointimal vascular smooth muscle cells [5]. Well-described polymorphisms in the genes encoding for IFN-␥ and its two receptor subunits were chosen for this investigation. IFNG T874A polymorphism was shown to influence the rate of IFN-␥ production [8]. Recently, it was shown in a combined case–control and family study that the IFNG 874T allele that is associated with increased IFN-␥ production protects from tuberculosis [17]. A higher susceptibility and lower morbidity of H. pylori infection in the presence of high levels of anti-H. pylori antibodies was observed for the IFNGR1 C-56T polymorphism T allele [10]. This suggests a reduced IFNGR1 function with a shift towards an increased Th2 response. The more common IFNGR2 allele (839A/64Gln) has been shown to be associated with the risk to develop systemic lupus erythematosus [11]. The relevance of alterations in the genes of IFN-␥ and its two receptor subunits is also underlined by the severe consequences of IFN-␥ receptor 1 deficiency resulting from IFNGR1 mutations [26]. Considering the actual distributions of the genotypes and the actual overall angiographic restenosis rate (the primary endpoint of the study), this study had a 83, 80 and 50% power for detecting a 8% difference between the genotype with the highest and that with the lowest restenosis rate, for each of the three polymorphisms investigated here—IFNG T874A, IFNG C-56T and IFNGR2 A839G—respectively. The lack of influence of the polymorphisms studied in the

present investigation is probably related to the complexity of cytokine interactions in the processes leading to the formation of restenosis. One limitation of this study deserves particular consideration. Gene variants encoding for proteins further downstream the interferon-␥ signaling pathway that may also be involved in the formation of neointima were not assessed in this study. Moreover, we cannot exclude that other polymorphisms/haplotypes of the studied genes are relevant to restenosis formation.

Acknowledgement The authors thank Angela Ehrenhaft, Marianne Eichinger, and Wolfgang Latz for skilful technical assistance. This study was supported by a research grant from the Deutsches Herzzentrum, Klinik an der Technischen Universit¨at, Munich, Germany (10-02).

References [1] Farb A, Sangiorgi G, Carter AJ, et al. Pathology of acute and chronic coronary stenting in humans. Circulation 1999;99:44–52. [2] Farb A, Weber DK, Kolodgie FD, Burke AP, Virmani R. Morphological predictors of restenosis after coronary stenting in humans. Circulation 2002;105:2974–80. [3] Welt FG, Rogers C. Inflammation and restenosis in the stent era. Arterioscler Thromb Vasc Biol 2002;22:1769–76. [4] Billiau A. Interferon-gamma: biology and role in pathogenesis. Adv Immunol 1996;62:61–130. [5] Zohlnh¨ofer D, Richter T, Neumann F, et al. Transcriptome analysis reveals a role of interferon-gamma in human neointima formation. Mol Cell 2001;7:1059–69. [6] Gupta S, Pablo AM, Jiang X, Wang N, Tall AR, Schindler C. IFNgamma potentiates atherosclerosis in ApoE knock-out mice. J Clin Invest 1997;99:2752–61. [7] Tellides G, Tereb DA, Kirkiles-Smith NC, et al. Interferongamma elicits arteriosclerosis in the absence of leukocytes. Nature 2000;403:207–11. [8] Pravica V, Perrey C, Stevens A, Lee JH, Hutchinson IV. A single nucleotide polymorphism in the first intron of the human IFNgamma gene: absolute correlation with a polymorphic CA microsatellite marker of high IFN-gamma production. Hum Immunol 2000;61:863–6. [9] Pravica V, Asderakis A, Perrey C, Hajeer A, Sinnott PJ, Hutchinson IV. In vitro production of IFN-gamma correlates with CA repeat polymorphism in the human IFN-gamma gene. Eur J Immunogenet 1999;26:1–3. [10] Thye T, Burchard GD, Nilius M, M¨uller-Myhsok B, Horstmann RD. Genomewide linkage analysis identifies polymorphism in the human interferon-␥ receptor affecting Helicobacter pylori infection. Am J Hum Genet 2003;72:448–53. [11] Nakashima H, Inoue H, Akahoshi M, et al. The combination of polymorphisms within interferon-gamma receptor 1 and receptor 2 associated with the risk of systemic lupus erythematosus. FEBS Lett 1999;453:187–90. [12] Reboul J, Gardiner K, Monneron D, Uze G, Lutfalla G. Comparative genomic analysis of the interferon/interleukin-10 receptor gene cluster. Genome Res 1999;9:242–50. [13] Sch¨omig A, Kastrati A, Mudra H, et al. Four-year experience with Palmaz-Schatz stenting in coronary angioplasty complicated

K. Tiroch et al. / Atherosclerosis 182 (2005) 145–151

[14]

[15] [16] [17]

[19]

[20]

by dissection with threatened or present vessel closure. Circulation 1994;90:2716–24. Sch¨omig A, Neumann FJ, Kastrati A, et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary artery stents. N Engl J Med 1996;334:1084– 9. World Health Organisation Study Group. Diabetes mellitus. WHO Tech Rep Ser 1985;727:1–104. Livak KJ. Allelic discrimination using fluorogenic probes and the 5 nuclease assay. Genet Anal 1999;14:143–9. Rossouw M, Nel HJ, Cooke GS, van Helden PD, Hoal EG. Association between tuberculosis and a polymorphic NFkappaB binding site in the interferon gamma gene. Lancet 2003;361:1871– 2. Sch¨omig A, Kastrati A, Elezi S, et al. Bimodal distribution of angiographic measures of restenosis six months after coronary stent placement. Circulation 1997;96:3880–7. Kastrati A, Sch¨omig A, Elezi S, Sch¨uhlen H, Wilhelm M, Dirschinger J. Interlesion dependence of the risk for restenosis in patients with coronary stent placement in multiple lesions. Circulation 1998;97:2396–401.

151

[21] Kastrati A, Sch¨omig A, Seyfarth M, et al. PlA polymorphism of platelet glycoprotein IIIa and risk of restenosis after coronary stent placement. Circulation 1999;99:1005–10. [22] Kastrati A, Koch W, Berger PB, et al. Protective role against restenosis from an interleukin-1 receptor antagonist gene polymorphism in patients treated with coronary stenting. J Am Coll Cardiol 2000;36:2168–73. [23] Amant C, Bauters C, Bodart JC, et al. D allele of the angiotensin Iconverting enzyme is a major risk factor for restenosis after coronary stenting. Circulation 1997;96:56–60. [24] Ribichini F, Steffenino G, Dellavalle A, et al. Plasma activity and insertion/deletion polymorphism of angiotensin I-converting enzyme: a major risk factor and a marker of risk for coronary stent restenosis. Circulation 1998;97:147–54. [25] Koch W, Kastrati A, Mehilli J, B¨ottiger C, von Beckerath N, Sch¨omig A. Insertion/deletion polymorphism of the angiotensin Iconverting enzyme gene is not associated with restenosis after coronary stent placement. Circulation 2000;102:197–202. [26] Jouanguy E, Dupuis S, Pallier A, et al. In a novel form of IFNgamma receptor 1 deficiency, cell surface receptors fail to bind IFNgamma. J Clin Invest 2000;105:1429–36.