Polymorphism R25P in the gene encoding transforming growth factor-beta (TGF-beta1) is a newly identified risk factor for proliferative diabetic retinopathy

• Published in: American journal of medical genetics Vol. 109; no. 4; p. 278
• Main Authors: Beránek, Michal, Kanková, Katerina, Benes, Petr, Izakovicová-Hollá, Lydie, Znojil, Vladimír, Hájek, Dobroslav, Vlková, Eva, Vácha, Jirí
• Format: Journal Article
• Language: English
• Published: United States 15.05.2002
• Subjects: Aged; Alleles; Amino Acid Substitution; Diabetes Mellitus, Type 2 - complications; Diabetes Mellitus, Type 2 - genetics; Diabetic Retinopathy - etiology; Diabetic Retinopathy - genetics; DNA - chemistry; DNA - genetics; DNA Mutational Analysis; Female; Gene Frequency; Genotype; Haplotypes; Humans; Infant, Newborn; Male; Middle Aged; Polymerase Chain Reaction; Polymorphism, Genetic; Polymorphism, Single-Stranded Conformational; Risk Factors; Transforming Growth Factor beta - genetics
• ISSN: 0148-7299
• DOI: 10.1002/ajmg.10372

Associations of the genetic polymorphisms in the promoter region and the signal peptide sequence of the transforming growth factor-beta (TGF-beta1) gene with proliferative diabetic retinopathy (PDR) in patients with non-insulin-dependent diabetes mellitus (NIDDM) were studied. A total of 245 Caucasian subjects comprised the two groups: NIDDM patients with PDR (n = 73) and NIDDM patients without PDR (n = 172). Allele frequencies of common TGF-beta1 polymorphisms (at positions -988C/A, -800G/A, -509C/T, +869T/C (L10P), and +915G/C (R25P)) were determined by PCR-based methodology. All polymorphisms were in strong linkage disequilibrium (P < 10(-2)). Significantly higher frequencies of both the L allele and the R allele of the signal sequence polymorphisms in PDR subjects were found (after a correction for multiple comparisons, P(corr) < 10(-2) and P(corr) < 10(-4), respectively). Calculated odds ratios (ORs) for the LL and RR genotypes were 2.89 (95% confidence interval (CI), 1.6-5.1) and 19.73 (95% CI, 2.6-146.8), respectively. No significant differences between groups were found for the -800G/A and -509C/T polymorphisms. The -988A allele was not represented in our sample. Multiple logistic regression identified age, diabetes duration, and R25P polymorphism as significant predictors (P = 0.002, P = 0.000003, and P = 0.007, respectively). The frequencies of genotype combinations of the -800G/A, -509C/T, L10P, and R25P TGF-beta(1) polymorphisms were significantly different between the PDR and non-PDR groups (chi(2) = 37.83, df = 20, P < 10(-2)). The frequency of haplotype consisting of majority alleles was found significantly associated with PDR (P < 0.03). The presented data indicate that the R25P polymorphisms in the TGF-beta1 gene could be regarded as a strong genetic risk factor for PDR.