Association of CTG repeat polymorphism in carnosine dipeptidase 1 (CNDP1) gene with diabetic nephropathy in north Indians.

BACKGROUND &
OBJECTIVES: CNDP1 gene, present on chromosome 18q22.3-23, encodes carnosinase, the rate-limiting enzyme in hydrolysis of carnosine to ß-alanine and L-histidine. Linkage of CTG trinucleotide (leucine) repeat polymorphism in CNDP1 gene with diabetic nephropathy has been observed in several populations. However, this association is conflicting and population-dependent. We investigated this association in type 2 diabetes mellitus (T2DM) patients with and without nephropathy in north India.
METHODS: A total of 564 individuals [199 T2DM without nephropathy (DM), 185 T2DM with nephropathy (DN) and 180 healthy individuals (HC)] were enrolled. CNDP1 CTG repeat analysis was done by direct sequencing of a 377 base pair fragment in exon 2.
RESULTS: The most frequent leucine (L) repeats were 5L-5L, 6L-5L and 6L-6L. 5L-5L genotype frequency was reduced in DN (24.3%) as compared to DM (34.7%, P=0.035) and HC (38.4%, P=0.005). Similarly, 5L allele frequency was lower in DN (46.8%) as compared to DM (57.3%, P=0.004) and HC (60.5%, P<0.001). The genotype and allelic frequencies were similar in DM and HC groups. No gender specific difference was observed in the genotype or allelic frequencies between groups. INTERPRETATION &
CONCLUSIONS: Compared to healthy individuals and those with diabetes but no kidney disease, patients with diabetic nephropathy exhibited lower frequencies of 5L-5L genotype and 5L allele of CNDP1 gene, suggesting that this allele might confer protection against development of kidney disease in this population.

Diabetic nephropathy (DN), a microvascular complication of diabetes, is the leading cause of end-stage renal disease (ESRD) worldwide. About 35 per cent of type 2 diabetes mellitus (T2DM) patients develop DN1. A population based study showed DN to be the commonest (44%) cause of ESRD in India2. Although the precise mechanisms leading to DN remain unclear, genetic susceptibility is considered to be an important factor in its development and/or progression3.

Several candidate genes have been evaluated for their role in DN, with conflicting results. These include ACE, AGT, APOE, CCR5, ELMO1, SLC12A3 and many more4. Carnosinase, encoded by the gene CNDP1 (carnosine dipeptidase 1) has also been shown to have a role in the genesis of DN5. This enzyme hydrolyzes carnosine into β-alanine and histidine, inhibits the advanced glycation end products (AGE) formation6 and acts as an inhibitor of angiotensin converting enzyme (ACE)7. Apart from these characteristics, carnosinasealso possesses antioxidant properties8 and inhibits hyperglycaemia induced extracellular matrix accumulation5.

Vardali et al9 discovered a susceptibility locus for DN at locus 18q22.3-q23 in Turkish families8. Later, an association of CTG trinucleotide repeat polymorphism in second exon of the CNDP1 gene at 18q with nephropathy in both type 1 and 2 diabetes was reported5. Homozygosity for 5 leucine repeats (5L) was associated with reduced risk of DN5. This finding was replicated in a cohort of European American individuals with T2DM10. In contrast, other studies have failed to show such an association. Examples include Japanese with type 2 diabetes11, type 1 diabetes in Caucasians12, and Scandinavians and African Americans with T2DM1314. One study showed a gender specific association of 5L-5L CNDP1 in Caucasians with type 2 diabetes15. An association between this polymorphism and microvascular complications of diabetes seems to have biological plausibility, since the number of CTG repeats is associated with serum carnosinase concentrations5. Moreover, CNDP1 is differentially expressed in kidney of animal models of DM16.

With the contradictory results in various ethnic populations, the association of 5L-5L homozygous repeat in CNDP1 with DN remains uncertain. There are no studies on CNDP1-leucine repeat polymorphism in Indian population. We, therefore, investigated the association between the CTG trinucleotide (leucine) repeats in CNDP1 gene with DN in T2DM patient with and without DN in north India.

Material & Methods

This study was carried out at the Nehru Hospital, Postgraduate Institute of Medical Education and Research, Chandigarh during 2011-2012. Adult individuals aged between 18 and 70 yr, including 199 subjects patients with T2DM but no kidney disease (DM) and 185 T2DM patients with nephropathy (DN) were recruited consecutively from Endocrinology and Nephrology Clinics. A control population comprising 180 healthy adults was also included. Healthy individuals were voluntary kidney donors and relatives of the patients. The Institute Ethics Committee approved the study, and all participants provided written informed consent.

All patients and controls underwent a detailed clinical examination including ocular fundus examination for diabetic retinopathy; T2DM was diagnosed on the basis of World Health Organization guidelines17. Those with diabetic retinopathy and sustained albuminuria (>150 mg/day) for ≥3 months in the absence of any other cause were considered to have DN. Diabetics with duration of disease >5 years, normal blood pressure, and urinary albumin excretion <150 mg/day formed the group of diabetes without nephropathy (DM).

Genotype analysis: Peripheral blood (5 ml) was collected from each patient after overnight fast. Buffy coat was separated using Ficoll-histopaque density gradient method18, and genomic DNA extracted from the buffy coat using QIAamp DNA mini kit (Qiagen GmbH, Hilden, Germany) according to manufacturer's instructions. The trinucleotide CTG repeat polymorphism in exon 2 of CNDP1 gene was analyzed by direct sequencing of 377 base pair covering exon2 using forward primer 5’-GTGTTTGGGGAAGGACGTAG-3’ and reverse primer 5’-TCTCACCCAAATACCAAAGG-3’19 on ABI 3730XL DNA analyzer (Applied Biosystems, Darmstadt, Germany).

Statistical analysis: Continuous and nominal clinical variables were compared with Student's t test and χ2 test (with Yates correction where appropriate). Mann-Whitney U test was used for parameters with skewed distribution. One-way ANOVA was used for comparison of more than two groups. Hardy-Weinberg equilibrium (HWE) was tested for each population. Allelic and genotypic associations of CTG repeat were evaluated by Pearson's χ2 test and odds ratio (OR) and 95% confidence intervals (CI). Multivariate logistic regression analysis was performed to find association of DN with 5L-5L genotype and other parameters. Statistical analysis was performed using SPSS v16.0 (Chicago, IL, USA).

Results

Table I shows the characteristics of study participants in different groups. There was no difference in age of the DN, DM and HC individuals. DN group had more males than DM (70 vs 55%). The duration of diabetes and prevalence of hypertension were more in DN group compared to DM. Systolic and diastolic blood pressure were higher in DN group as compared to DM and HC (P<0.001 for both), whereas BMI was higher in DM group (P<0.01). As expected, serum creatinine, total cholesterol, low-density lipoprotein cholesterol, triglyceride and HbA1c were higher in DN group compared to DM and HC groups (Table I). DN was also accompanied more frequently by diabetic neuropathy (P<0.01) and retinopathy (P<0.001) compared to DM.

Table I

Baseline parameters of the study groups

Tables II and III show the frequency of leucine repeats and comparison of genotype and allelic frequencies between groups. All populations were in the Hardy-Weinberg equilibrium (P=0.297, χ2 =1.09 for HC, P=0.284, χ2 =1.14 for DM and P=0.178, χ2 =1.87 for DN). 5L-5L, 5L-6L and 6L-6L were most common repeats found in this study. Only three other repeats, one 4L-7L and two 5L-7L repeats were noted in healthy individuals. These three were excluded from further analyses. The Figure shows representative DNA sequence chromatograph for 5 and 6 CTG repeats.

Table II

Frequency of leucine repeat allele and genotype in study groups

Table III

Association of CNDP1 CTG repeat genotype and allele with diabetic nephropathy

Figure

Representative DNA sequence chromatograph showing (A) 5 -5 CTG repeats, (B) 6-6 CTG repeats, and (C) 5-6 CTG repeats. 5-5, 5-6 and 6-6 CTG repeats are highlighted and underlined.

Frequency of 5L-5L genotype was lower in DN group (24.3%) compared to DM (34.7%, P=0.035) and HC (38.4%, P=0.005) groups. Similarly, the 5L allele frequency was also reduced in patients with DN (46.8%) compared to DM (57.3%, P=0.004) and HC (60.5%, P<0.001) (Table II). Further, to assess whether the susceptibility for DN is independent of susceptibility for T2DM, the DM group was compared with HC, and similar frequencies of the 5L-5L genotypes and 5L alleles were found in the two groups (Table III).

Next, the sex-specific frequency of 5L-5L homozygous in DN, DM and HC groups was compared. Amongst women, 5L-5L genotype was noted in 28.5, 33.7 and 39.4 per cent in DN, DM and HC groups, respectively. The frequency of males with 5L-5L homozygosity was 22.5 per cent in DN, 31.8 per cent in DM and 33.3 per cent in HC group. There was no sex-specific difference in the frequencies.

Multivariate logistic regression analysis was performed to assess the association of nephropathy with genotype and various clinical parameters. Apart from CNDP1 5L-5L genotype, serum creatinine, HbA1c, diastolic blood pressure, and duration of diabetes were found to be independently associated with DN (Table IV).

Table IV

Multivariate logistic regression analysis of factors related to diabetic nephropathy

Discussion

Our study showed that amongst north Indian patients with diabetic nephropathy, the frequency of the 5L-5L genotype in CNDP1 gene was reduced in comparison to those with T2DM but no kidney disease and healthy non-diabetic individuals. These findings suggest the possible protective effect of this polymorphism for development of kidney disease in T2DM patients.

This finding fulfils the criterion of biological plausibility, as 5L repeats are associated with carnosinase activity. Upon transfecting COS cells with CNDP1 construct containing 4-8 CTG repeats, cells with higher number of CTG repeats showed increased carnosinase in the supernatant20. Individuals homozygous for the 5L allele (also called Mannheim allele) exhibit reduced serum carnosinase concentrations compared to those with higher number of CTG repeats15. Carnosinase hydrolyses the dipeptide carnosine, hence lower level of this enzyme is likely to result in higher renal carnosine concentrations. Carnosine has been shown to have multiple potentially renoprotective effects21. In a cell culture experiment, Janssen et al5 showed that carnosine blocked glucose-induced extracellular matrix accumulation by renal cells. Moreover, epithelial cells in glomeruli and renal tubules express CNDP1 mRNA and carnosinase protein22.

CNDP1 polymorphism has been studied in the context of diabetic kidney disease in various ethnicities, with conflicting results. Some studies51015 showed an association of D18S880 microsatellite marker with DN whereas others11121314,[ 23 failed to find any such association. A meta-analysis confirmed the association of 5L-5L genotype with DN24.

The 5L-5L homozygous genotype was encountered with reduced frequency in South Asian Surinamese compared to White Dutch and African Surinamese22. The frequency of various genotypes in DM, DN and HC populations in this study was similar to those reported in the Caucasians510. The findings of this study were also consistent with the finding of reduced risk for developing diabetic renal disease amongst European-Americans with 5L-5L homozygous repeat polymorphism in the CNDP1 gene10. In another study, lower 5L-5L frequency was not associated with DN and in the follow up study 5L-5L was associated with higher risk of ESRD beyond six years23. They also showed an association of 5L-5L genotype with sex specific risk of mortality, with the risk being higher in women23. In another study, 5L-5L genotype was associated with DN in sex specific manner, with protection seen only in women22. We did not find any gender-specific association of 5L-5L genotype with DN.

Our study had some limitations. The number of patients in different groups was small, and only albuminuric patients were included in the DN group. The diagnosis of diabetic nephropathy was made on clinical grounds alone. In absence of kidney biopsy, it was not possible to exclude non-diabetic lesions with certainty. The strength of the study was a well selected homogenous population reducing phenotype errors and bias. The study included well matched controls of the same ethnic background which reduced the risk of population stratification.

In conclusion, our findings show an association of CTG repeat polymorphism in the CNDP1 gene with DN in north Indian population; with reduced frequency of 5L-5L repeat genotype and 5L allele in DN.