L. J. McCormack

Diabetic Retinopathy, Promoter (4G/5G) Polymorphism of PAI-1 Gene, and PAI-1 Activity in Pima Indians With Type 2 Diabetes

OBJECTIVE To examine the relationship between plasma plasminogen activator inhibitor 1 (PAI-1) activity and PAI-1 gene (4G/5G) polymorphism and diabetic retinopathy in Pima Indians with type 2 diabetes. RESEARCH DESIGN AND METHODS We studied 171 Pima Indians with type 2 diabetes between the ages of 30–70 years in a population-based epidemiological survey. Plasma PAI-1 activity was measured by a spectrophotometric assay and PAI-1 4G/5G promoter genotype by the polymerase chain reaction (PCR) using allele-specific primers. Retinopathy was assessed by ophthalmoscopy after pupillary dilation and classified as any retinopathy or as nonproliferative and proliferative. RESULTS Retinopathy was present in 70 (41%) subjects, and 4 (2.3%) subjects had proliferative retinopathy. Plasma PAI-1 activity was not significantly different among subjects with and without retinopathy (17.1 ± vs. 19.7 ± 9.1 arbitrary units (AU)/ml, P = 0.09). PAI-1 activity was negatively correlated with duration of diabetes (rs = −0.18, P = 0.02). In a logistic regression analysis controlled for age, sex, BMI, and duration of diabetes, any retinopathy was significantly associated with fasting plasma glucose concentrations (P < 0.05), 2-h postload glucose (P = 0.02), and HbA1c (P = 0.008), but not with PAI-1 activity (P = 0.48). The prevalence of retinopathy in the three genotype groups differed significantly (4G/4G, 4G/5G, and 5G/5G were 44, 49, and 24%, respectively; chi χ2 = 8.22, df = 2, P = 0.016) and remained significant after controlling for age, sex, BMI, duration of diabetes, glycated hemoglobin, and urine albumin-to-creatine ratio in a logistic regression analysis. The odds ratios for retinopathy in subjects with 4G/4G and 4G/5G, compared with the 5G/5G genotype, were 2.0 and 3.1, respectively. CONCLUSIONS Although diabetic retinopathy in Pima Indians with type 2 diabetes is not associated with PAI-1 activity, subjects with the 4G/4G and 4G/5G genotype had a higher prevalence of retinopathy compared with 5G/5G PAI-1genotype. These preliminary findings indicate that in Pima Indians with type 2 diabetes, presence of the 4G allele of the PAI-1 gene was associated with a higher risk of diabetic retinopathy.

The angiotensin-I converting enzyme (ACE) gene I/D polymorphism and ACE levels in Pima Indians.

An insertion/deletion (I/D) polymorphism of the angiotensin converting enzyme (ACE) gene is associated with plasma ACE levels in white populations. The occurrence of the I/D polymorphism and relationship to ACE levels was examined in a Pima Indian group (n = 305). The frequency of the D allele was lower in Pimas than whites (0.29 v 0.52 respectively). ACE levels were significantly associated with genotype in both groups (p = 0.0001), which accounted for 6.5% of the variation in ACE levels in Pimas and 18% in whites. The association of the I/D polymorphism with ACE levels confirms the relationship across ethnic groups. The low frequency of the D allele in Pima Indians shows that ethnic differences should be accounted for when studying the ACE gene.

Promoter (4G/5G) plasminogen activator inhibitor-1 genotype in Pima Indians: relationship to plasminogen activator inhibitor-1 levels and features of the insulin resistance syndrome

Summary Elevated plasminogen activator inhibitor-1 may contribute to vascular disease in diabetes mellitus. Pima Indians have a low incidence of cardiovascular disease despite having a high prevalence of non-insulin-dependent diabetes mellitus (NIDDM) which in this population is not associated with elevated plasminogen activator inhibitor-1 activity. In Caucasians an insertion/deletion (4G/5G) polymorphism in the promoter region of the plasminogen activator inhibitor-1 gene that has been related to activity levels of its protein in plasma differentially binds repressor and enhancer elements. In 265 Pima Indians (133 diabetic, 132 non-diabetic, 129 male, 136 female, mean age 46.6, range 34–68 years) the promoter genotype frequencies were 23.0 % for 4G/4G, 49.8 % for 4G/5G and 27.2 % for 5G/5G compared to 35.4 %, 50.8 % and 13.8 % respectively (χ2 = 15.3, 2 df, p < 0.0005) previously reported in Caucasians with NIDDM. The mean plasma activity levels in the three genotypes in the Pima Indians were 18.2, 19.1 and 18.1 U/ml, respectively. Plasminogen activator inhibitor-1 activities correlated with plasma insulin (r = 0.38, p < 0.0001), body mass index (r = 0.24, p < 0.0001), and with triglyceride level (r = 0.12, p = 0.054) but there was no relationship between promoter genotype and activity. A steeper regression slope between plasminogen activator inhibitor-1 activity and triglycerides has been observed in Caucasians with the 4G/4G genotype as compared to Caucasians with the other genotypes. This was not found in the Pima population which may indicate a functional difference in this gene associated with reduced cardiovascular risk and may be involved in the lack of association of plasminogen activator inhibitor-1 levels with NIDDM in Pima Indians. [Diabetologia (1996) 39: 1512–1518]

CCAAT/enhancer binding protein α, β and δ gene variants: associations with obesity related phenotypes in the Leeds Family Study

Objective: To identify novel polymorphisms in the genes encoding the transcription factors CCAAT/enhancer binding protein α, β and δ ( CEBPA, CEBPB, CEBPD) and investigate associations between polymorphisms and obesity-related phenotypes. Methods: Denaturing high-performance liquid chromatography (HPLC) was used to screen for novel gene variants and polymorphisms were genotyped in stored DNA from participants of the Leeds Family Study (537 subjects from 89 families). Genotype and haplotype analyses were carried out in STATA and PBAT, respectively. Results: Twenty-five polymorphisms were identified; 11 in CEBPA, 12 in CEBPB and 2 in CEBPD. Several allelic variants were associated at a nominal 5% level with waist-to-hip ratio (-919G>A in CEBPA, -412G>T and 646C>T in CEBPB), leptin (1558G>A in CEBPA, -1051A>G and 1383T>- in CEBPB) and adiponectin (1382G>T and 1903G>T in CEBPB). Effects of CEBPA and CEBPB allelic variants were independent, but variants within each gene were in linkage disequilibrium. Several associations were observed between other obesity-related traits and allelic variants in CEBPA and CEBPB, but not CEBPD. Conclusion: These findings suggest that common allelic variants in CEBPA and CEBPB could influence abdominal obesity and related metabolic abnormalities associated with type 2 diabetes and cardiovascular disease in healthy White Northern European families, although results require independent confirmation.

The Effect of Number of Days in Culture and Plasminogen Activator Inhibitor-1 (PAI-1) 4G/5G Genotype on PAI-1 Antigen Release by Cultured Human Umbilical Vein Endothelial Cells

An insertion/deletion (4G/5G) polymorphism in what has been shown to be an enhancer/repressor binding site in the promoter region of the PAI-1 gene has been related to plasma PAI-1 activity. Transfection studies demonstrated increased interleukin-1 stimulated PAI-1 synthesis in cells containing the 4G sequence. To study this response in endothelial cells, first passage HUVEC from 26 umbilical cords were stimulated with interleukin-1 and tumor necrosis factor-alpha. PAI-1 antigen was measured in 24-hour conditioned medium and allele-specific PCR utilized to determine genotype at the 4G/5G locus. Analysis of covariance was used to determine whether the effect of a variable time in culture was masking a difference between genotypes. A trend towards higher PAI-1 levels with increasing time in culture was observed. The geometric mean (95% confidence interval) of the basal rate of PAI-1 release was, 4G/4G 9.7 (7.0, 13.5) ng/24 hours (n=11), 4G/5G 9.5 (6.5, 13.9) ng/24 hours (n=9), and 5G/5G 10.9 (7.8, 15.1) ng/24 hours (n=6). In cells of the same cultures, the interleukin-1 stimulated levels were 25.9 (23.1, 29.1), 27.2 (23.6, 31.3), and 23.1 (19.5, 27.3) ng/24 hours, respectively, corresponding to ratios of stimulated to basal levels of 2.68, 2.87, and 2.12. After adjustment for time in culture the basal PAI-1 release was 4G/4G 10.7, 4G/5G 9.1, and 5G/5G 9.7 ng/24 hours. For interleukin-1 stimulated release the adjusted levels were 26.3, 27.0, and 22.7 ng/24 hours, respectively. Adjusted levels in 4G/4G genotype cells were non-significantly greater than those in cells of 5G/5G genotype by a factor of 1.16 (0.95, 4.08). This study did not demonstrate a significant difference in basal or cytokine stimulated PAI-1 release from cells of different PAI-1 promoter (4G/5G) genotypes but does not exclude increased interleukin-1 stimulated PAI-1 release in the 4G/4G compared with the 5G/5G genotype.