- Email: [email protected]
Association of VEGF-1499C→T polymorphism with diabetic nephropathy in type 1 diabetes mellitus
Journal of Diabetes and Its Complications 21 (2007) 242 – 245
Association of VEGF-1499CYT polymorphism with diabetic nephropathy in type 1 diabetes mellitus Amy-Jayne McKnighta,4, A. Peter Maxwella, Chris C. Pattersonb, Hugh R. Bradyc, David A. Savagea a
Nephrology Research Group, Queen’s University of Belfast, Belfast, UK Department of Epidemiology and Public Health, Queen’s University of Belfast, Belfast, UK c Dublin Molecular Medicine Centre, Conway Institute, Dublin, Ireland
b
Received 21 February 2006; accepted 8 May 2006
Abstract Vascular endothelial growth factor (VEGF) is reported to be implicated in the development of diabetic nephropathy. We performed a casecontrol study to determine if VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC single-nucleotide polymorphisms (SNPs) in the VEGF gene are associated with predisposition to diabetic nephropathy in type 1 diabetes. Genomic DNA was obtained from Irish type 1 diabetic individuals with nephropathy (cases, n=242) and those without nephropathy (controls, n=301), in addition to 400 healthy control samples. These samples were genotyped for the three SNPs using TaqMan or Pyrosequencing technology. Chi-squared analyses revealed no significant differences in genotype or allele frequencies in cases versus controls for VEGF-2578CYA (genotype, P=.58; allele, P=.52) and VEGF-635GYC (genotype, P=.58; allele, P=.33). However, a positive association with diabetic nephropathy was observed for the VEGF1499T allele in the Northern Ireland population ( Pb.001) and subsequently replicated in a separate population from the Republic of Ireland ( Pb.001; combined, Pb.001). Carriage of the VEGF-1499T allele was associated with a twofold excess risk of developing diabetic nephropathy (OR=2.24, 95% CI=1.50–3.36, Pb.0001). No significant differences were found between the healthy control population and the type 1 diabetic population. Our results suggest that the VEGF-1499T allele, or an allele in linkage disequilibrium with this allele, is associated with susceptibility to diabetic nephropathy in the Irish population. D 2007 Elsevier Inc. All rights reserved. Keywords: Association; Nephropathy; SNP; Type 1 diabetes; VEGF
1. Introduction Diabetic nephropathy is a microvascular complication of diabetes characterized by persistent proteinuria, decreased glomerular filtration rate, and increased blood pressure. Diabetic kidney disease is the leading cause of end-stage renal failure in the Western world and is associated with an increased premature mortality due to cardiovascular
4 Corresponding author. Nephrology Research Group, Queen’s University of Belfast, c/o Regional Genetics Centre, Level A, Tower Block, Belfast City Hospital, Lisburn Road, Belfast BT9 7AB, Ireland. Tel.: +353 2890 329241x2558; fax: +353 2890 236911. E-mail address: [email protected] (A.-J. McKnight). 1056-8727/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2006.05.005
disease. Although poor glycemic control, hypertension, and prolonged duration of diabetes are recognized risk factors for diabetic nephropathy, there is also evidence for genetic susceptibility to renal disease in a subset of type 1 diabetic patients. The genetic variation contributing to the development of diabetic nephropathy remains to be identified. Vascular endothelial growth factor (VEGF) is implicated in the pathogenesis of diabetic nephropathy, as well as other diabetic complications including retinopathy, neuropathy, and cardiovascular disease (Duh & Aiello, 1999). Marked up-regulation of VEGF gene expression has been observed in glomerular podocytes in the early phase of diabetic kidney disease, and urinary VEGF excretion is significantly increased in accordance with the degree of proteinuria for
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245 Table 1 Clinical characteristics of cases (n=242) and controls (n=301) Age at diagnosis of diabetes (years) Duration of diabetes (years)a HbA1C (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
243
1499CYT, and VEGF-635GYC SNPs for association with diabetic nephropathy in the Irish population.
Cases
Controls
17.6F9.0 29.1F6.2 9.4F2.5 150.1F22.6 86.5F11.4
15.9F8.4 30.5F6.2 8.6F1.2 126.9F16.6 76.1F7.3
Data are expressed as meanFS.D. a Duration of diabetes was calculated from the date of diagnosis of diabetes to date of recruitment.
more advanced nephropathy (Cha et al., 2000). Increased VEGF mRNA and protein levels have been observed in models of diabetic kidney disease (Schrijvers, Flyvbjerg, De Vriese, 2003). Furthermore, antibodies directed against VEGF have been shown to improve early renal dysfunction in experimental diabetes (De Vriese et al., 2001). The VEGF gene is clearly an excellent candidate for diabetic nephropathy. VEGF is located on chromosome 6p21.3 and comprises eight exons (Vincenti, Cassano, Rocchi, & Persico, 1996). To date, at least 15 gene variants have been reported in the promoter region and 3 in the 5V untranslated region of the VEGF gene (Brogan et al., 1999; Watson, Webb, Bottomley, & Brenchley, 2000). VEGF1499CYT (rs833061) and VEGF-635GYC (rs2010963) single-nucleotide polymorphisms (SNPs) occur at the highest minor allele frequency, are correlated with altered transcriptional activity, and are associated with several disease phenotypes (Awata et al., 2002; Stevens, Soden, Brenchley, Ralph, & Ray, 2003; Watson et al., 2000). The VEGF-2578CYA SNP (rs699947) is in strong linkage disequilibrium with an 18-base insertion, VEGF2548_2549insTCCCACTCTTCCCACAGG (ss38346984), which is associated with increased transcriptional activity (Schrijvers et al., 2003; Yang, Cross, Ollerenshaw, Millward, & Demaine, 2003). Previous studies have described associations with VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC SNPs and renal disease (Watson et al., 2000; Yang et al., 2003; Summers, 2002a, 2002b). In the present report, we have assessed VEGF-2578CYA, VEGF-
2. Methods Ethical approval was obtained from the appropriate research ethics committees, and written informed consent was obtained from subjects prior to conducting this study. All patients in the case-control cohort (n=543) were Caucasian, with parents and grandparents born in Ireland, who were diagnosed with type 1 diabetes mellitus before 31 years of age and who required insulin from diagnosis. Subjects were recruited from two centers, one in Northern Ireland (NI) and the other in the Republic of Ireland (ROI). Patients with nephropathy (cases, n=242; NI, 153; ROI, 89) had diabetes for at least 10 years before the onset of proteinuria (N0.5 g/24 h) and diabetic retinopathy. Patients without nephropathy (controls, n=301; NI, 184, ROI, 117) had diabetes for at least 15 years with no evidence of microalbuminuria on repeated testing, were not prescribed antihypertensive medication, and had no evidence of nondiabetic renal disease. Healthy controls (n=400) were selected from the Young Hearts collection (Boreham, Savage, Primrose, Cran, & Strain, 1993). Genotyping was performed for VEGF-1499CYT and VEGF-635GYC SNPs by Pyrosequencing using a PSQ 96HS instrument (Biotage, Uppsala, Sweden). DNA samples were genotyped for the VEGF-2578AYC SNP by TaqMan technology using a 7700 Sequence Detection System (Applied Biosystems, Warrington, UK). Reactions were performed according to manufacturer’s instructions, and details of primers and probes are available from the authors. DNA samples representing all three genotypes (confirmed by direct sequencing) for each SNP and no DNA template controls were included in each genotyping run. All genotype calls were independently double scored to minimize errors, and no mismatches were observed. Genotype and allele frequencies were compared using the v 2 test; the level of statistical significance was set at 5%. A stratified analysis was used to combine case-control results from the two centers. The extent of linkage
Table 2 Distributions of genotype and allele frequencies for the VEGF-1499CYT SNP in cases and controls; data are expressed as n (%) NI Cases (n=153)
ROI Controls (n=184)
Cases (n=89)
Combined Controls (n=117)
Cases (n=242)
Controls (n=301)
Genotype CC CT TT
34 (22.2) 66 (43.1) 53 (34.6)
74 (40.2) 84 (45.7) 26 (14.1) Pb.001
18 (20.2) 38 (42.7) 33 (37.1)
40 (34.2) 56 (47.9) 21 (17.9) P=.004
52 (21.5) 104 (43.0) 86 (35.5)
114 (37.9) 140 (46.5) 47 (15.6) Pb.001
Allele C T
134 (43.8) 172 (56.2)
232 (63.0) 136 (37.0) Pb.001
74 (41.6) 104 (58.4)
136 (58.1) 98 (41.9) Pb.001
208 (43.0) 276 (57.0)
368 (61.1) 234 (38.9) Pb.001
244
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245
disequilibrium between pairs of SNPs was quantified using Lewontin’s DVvalue.
3. Results Table 1 lists the clinical characteristics of case and control groups. There were a total of 38.3% and 59.5% females in cases and controls, respectively. The mean duration of diabetes was at least 25 years for both groups. Not surprisingly, average blood pressure was higher in cases versus controls, despite the use of antihypertensive drug treatment in the cases. Genotype distributions were in Hardy–Weinberg equilibrium for all three SNPs in case, control, and healthy populations ( PN.05). The VEGF-1499T allele was significantly increased in cases compared with controls in the NI population (56.2% vs. 37.0%, Pb.001; Table 2). This finding was replicated in a separate population from the ROI (58.4% vs. 41.9%, Pb.001; Table 2). Analysis of the combined data confirmed the association between the VEGF-1499T allele and diabetic nephropathy ( Pb.001; Table 2). Carriage rates for the VEGF-1499T allele (proportion of individuals with VEGF-1499T allele) in cases and controls for the combined population revealed a twofold increase in risk of developing nephropathy (OR=2.24, 95% CI=1.50–3.36, Pb.0001). In contrast, no significant differences were observed in genotype and allele frequencies for VEGF-2578CYA (cases 58 CC, 122 CA, 62 AA giving a C allele frequency of 49.2%; controls 73 CC, 162 CA, 66 AA giving a C allele frequency of 51.2%) or VEGF-635GYC (cases 60 GG, 135 GC, 47 CC giving a G allele frequency of 52.7%; controls 85 GG, 165 GC, 51 CC giving a G allele frequency of 55.6%). No significant differences were observed in genotype and allele frequencies between the healthy control groups and those with type 1 diabetes (genotypes: VEGF-2578CYA, P=.4; VEGF-1499CYT, P=.3; VEGF-635GYC, P=.6). Although |DV| values were not particularly large (|DV|=0.73 for VEGF-2578CYA and VEGF-1499CYT, |DV|=0.50 for VEGF-2578CYA and VEGF-635GYC, and (|DV|=0.27 for VEGF-1499CYT and VEGF-635GYC), all were highly statistically significant ( Pb.001). Allele frequencies for all three SNPs were similar to those reported in other Caucasian populations (Brogan et al., 1999).
4. Discussion In the present report, three putatively functional SNPs, VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC, were genotyped for association with diabetic nephropathy in 242 cases, 301 type 1 diabetic controls, and 400 healthy controls from Ireland. We observed a significant increase in the VEGF-1499T allele in type 1 diabetic patients with
nephropathy (n=242) compared with nonnephropathic controls (n=301). Carriage of this allele conferred a twofold increase in risk for nephropathy in diabetic patients in the combined Irish population. Our findings are consistent with a recent UK study in which the VEGF-1499T allele was found to have a significantly higher frequency in diabetic patients with microalbuminuria or proteinuria compared with normoalbuminuric controls (Summers, 2002a, 2002b). In contrast, Ray et al. (2004) failed to replicate this association in their study that comprised type 1 and type 2 diabetic patients with nephropathy defined by the presence of microalbuminuria. Interestingly, the VEGF-1499T allele has also been found to be significantly associated with other diseases (Hsieh et al., 2004; Ku et al., 2005). We used much larger sample sizes compared with others, and our patient groups had longer duration of type 1 diabetes compared with the cohort described by Ray et al. who employed patients groups comprising both type 1 and type 2 diabetes with duration of diabetes just over 20 years. We also employed rigorous phenotypic criteria for inclusion of cases and controls, excluding those individuals with microalbuminuria from case or control groups. Our initial finding of a positive association of the VEGF-1499T allele with diabetic nephropathy in the NI population was also replicated in the ROI population and remained positive in the combined population. In addition, genotyping was performed in this study by Pyrosequencing, which generates a sequence output, thus providing confidence in assigning genotypes. We failed to find an association between VEGF-635GYC and diabetic nephropathy, despite the earlier report that this SNP has been reported associated with end-stage renal disease (Summers, 2002a). However, our results are consistent with those of Ray et al. (2004) who also found no association between VEGF-635GYC and diabetic nephropathy in type 1 and type 2 diabetes. Recently, the VEGF2548_2549insTCCCACTCTTCCCACAGG has been reported by Yang et al. (2003) to be associated with diabetic nephropathy in type 1 diabetes in a UK population. As this insertion is in strong linkage disequilibrium with the VEGF2578CYA SNP, our results suggest differences with those reported by Yang et al. Carriage of the VEGF-2578A allele was more common in our control group (76%), but our study still had 85% power to detect an OR of 2.0 associated with this allele. We compared genotype and allele frequencies for a healthy control group (n=400) to a group with type 1 diabetes (combined cases and controls, n=543). No significant differences were observed between these groups, indicating that these SNPs do not play a major role in type 1 diabetes in our population. In conclusion, our results suggest that the VEGF-1499T allele, or an allele in linkage disequilibrium with this SNP, is associated with diabetic nephropathy in the Irish population. This finding will need to be replicated in larger casecontrol collections such as the Juvenile Diabetes Research
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245
Foundation Genetics of Kidneys in Diabetes (GoKinD) casecontrol collection. Acknowledgments This work was funded by the Research and Development Office Northern Ireland and the Health Research Board of Ireland. AJ McK is supported by a Northern Ireland Kidney Research Fund Postdoctoral Fellowship. References Awata, T., Inoue, K., Kurihara, S., Ohkubo, T., Watanabe, M., Inukai, K., Inoue, I., & Katayama, S. (2002). A common polymorphism is the 5V untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes, 51 (5), 1635 – 1639. Boreham, C., Savage, J. M., Primrose, D., Cran, G., & Strain, J. (1993). Coronary risk factors in school-children. Archives of Disease in Childhood, 68, 182 – 186. Brogan, I. J., Khan, N., Isaac, K., Hutchinson, J. A., Pravica, V., & Hutchinson, I. V. (1999). Novel polymorphisms in the promoter and 5V UTR regions of the human vascular endothelial growth factor gene. Human Immunology, 60 (12), 1245 – 1249. Cha, D. R., Kim, N. H., Yoon, J. W., Jo, S. K., Cho, W. Y., Kim, H. K., Won, N. H. (2000). Role of vascular endothelial growth factor in diabetic nephropathy. Kidney International. Supplement, 77, S104 – S112. De Vriese, A. S., Tilton, R. G., Elger, M., Stephan, C. C., Kriz, W., & Lameire, N. H. (2001). Antibodies against vascular endothelial growth factor improve early renal dysfunction in experimental diabetes. Journal of the American Society of Nephrology, 12 (5), 993 – 1000.
245
Duh, E., & Aiello, L. P. (1999). Vascular endothelial growth factor and diabetes: The agonist versus antagonist paradox. Diabetes, 48, 1899 – 1906. Hsieh, Y. Y., Chang, C. C., Tsai, F. J., Yeh, L. S., Lin, C. C., & Peng, C. T. (2004). T allele for VEGF gene-460 polymorphism at the 5V-untranslated region: Association with a higher susceptibility to endometriosis. Journal of Reproductive Medicine, 49 (6), 468 – 472. Ku, K. T., Wan, L., Peng, H. C., Tsai, M. H., Tsai, C. H., & Tsai, F. J. (2005). Vascular endothelial growth factor gene-460 C/T polymorphism is a biomarker for oral cancer. Oral Oncology, 41 (5), 497 – 502. Ray, D., Mishra, M., Ralph, S., Read, I., Davies, R., & Brenchley, P. (2004). Association of the VEGF gene with proliferative diabetic retinopathy but not proteinuria in diabetes. Diabetes, 53 (3), 861 – 864. Schrijvers, B. F., Flyvbjerg, A., & De Vriese, A. S. (2004). The role of vascular endothelial growth factor (VEGF) in renal pathophysiology. Kidney International, 65 (6), 2003 – 2017. Stevens, A., Soden, J., Brenchley, P. E., Ralph, S., & Ray, D. W. (2003). Haplotype analysis of the polymorphic human vascular endothelial growth factor gene promoter. Cancer Research, 63 (4), 812 – 816. Summers, A. (2002a). VEGF polymorphisms and diabetic nephropathy [Abstract]. Nephrology Dialysis Transplantation, 17 (Suppl. 1), O80. Summers, A. (2002b). VEGF polymorphisms and diabetic nephropathy [Abstract]. Journal of the American Society of Nephrology, 13, 248A. Vincenti, V., Cassano, C., Rocchi, M., & Persico, G. (1996). Assignment of the vascular endothelial growth factor gene to human chromosome 6p21.3. Circulation, 93 (8), 1493 – 1495. Watson, C. J., Webb, N. J., Bottomley, M. J., & Brenchley, P. E. (2000). Identification of polymorphisms within the vascular endothelial growth factor (VEGF) gene: Correlation with variation in VEGF protein production. Cytokine, 12 (8), 1232 – 1235. Yang, B., Cross, D. F., Ollerenshaw, M., Millward, B. A., & Demaine, A. G. (2003). Polymorphisms of the vascular endothelial growth factor and susceptibility to diabetic microvascular complications in patients with type 1 diabetes mellitus. Journal of Diabetes and its Complications, 17 (1), 1 – 6.
Association of VEGF-1499CYT polymorphism with diabetic nephropathy in type 1 diabetes mellitus Amy-Jayne McKnighta,4, A. Peter Maxwella, Chris C. Pattersonb, Hugh R. Bradyc, David A. Savagea a
Nephrology Research Group, Queen’s University of Belfast, Belfast, UK Department of Epidemiology and Public Health, Queen’s University of Belfast, Belfast, UK c Dublin Molecular Medicine Centre, Conway Institute, Dublin, Ireland
b
Received 21 February 2006; accepted 8 May 2006
Abstract Vascular endothelial growth factor (VEGF) is reported to be implicated in the development of diabetic nephropathy. We performed a casecontrol study to determine if VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC single-nucleotide polymorphisms (SNPs) in the VEGF gene are associated with predisposition to diabetic nephropathy in type 1 diabetes. Genomic DNA was obtained from Irish type 1 diabetic individuals with nephropathy (cases, n=242) and those without nephropathy (controls, n=301), in addition to 400 healthy control samples. These samples were genotyped for the three SNPs using TaqMan or Pyrosequencing technology. Chi-squared analyses revealed no significant differences in genotype or allele frequencies in cases versus controls for VEGF-2578CYA (genotype, P=.58; allele, P=.52) and VEGF-635GYC (genotype, P=.58; allele, P=.33). However, a positive association with diabetic nephropathy was observed for the VEGF1499T allele in the Northern Ireland population ( Pb.001) and subsequently replicated in a separate population from the Republic of Ireland ( Pb.001; combined, Pb.001). Carriage of the VEGF-1499T allele was associated with a twofold excess risk of developing diabetic nephropathy (OR=2.24, 95% CI=1.50–3.36, Pb.0001). No significant differences were found between the healthy control population and the type 1 diabetic population. Our results suggest that the VEGF-1499T allele, or an allele in linkage disequilibrium with this allele, is associated with susceptibility to diabetic nephropathy in the Irish population. D 2007 Elsevier Inc. All rights reserved. Keywords: Association; Nephropathy; SNP; Type 1 diabetes; VEGF
1. Introduction Diabetic nephropathy is a microvascular complication of diabetes characterized by persistent proteinuria, decreased glomerular filtration rate, and increased blood pressure. Diabetic kidney disease is the leading cause of end-stage renal failure in the Western world and is associated with an increased premature mortality due to cardiovascular
4 Corresponding author. Nephrology Research Group, Queen’s University of Belfast, c/o Regional Genetics Centre, Level A, Tower Block, Belfast City Hospital, Lisburn Road, Belfast BT9 7AB, Ireland. Tel.: +353 2890 329241x2558; fax: +353 2890 236911. E-mail address: [email protected] (A.-J. McKnight). 1056-8727/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2006.05.005
disease. Although poor glycemic control, hypertension, and prolonged duration of diabetes are recognized risk factors for diabetic nephropathy, there is also evidence for genetic susceptibility to renal disease in a subset of type 1 diabetic patients. The genetic variation contributing to the development of diabetic nephropathy remains to be identified. Vascular endothelial growth factor (VEGF) is implicated in the pathogenesis of diabetic nephropathy, as well as other diabetic complications including retinopathy, neuropathy, and cardiovascular disease (Duh & Aiello, 1999). Marked up-regulation of VEGF gene expression has been observed in glomerular podocytes in the early phase of diabetic kidney disease, and urinary VEGF excretion is significantly increased in accordance with the degree of proteinuria for
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245 Table 1 Clinical characteristics of cases (n=242) and controls (n=301) Age at diagnosis of diabetes (years) Duration of diabetes (years)a HbA1C (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
243
1499CYT, and VEGF-635GYC SNPs for association with diabetic nephropathy in the Irish population.
Cases
Controls
17.6F9.0 29.1F6.2 9.4F2.5 150.1F22.6 86.5F11.4
15.9F8.4 30.5F6.2 8.6F1.2 126.9F16.6 76.1F7.3
Data are expressed as meanFS.D. a Duration of diabetes was calculated from the date of diagnosis of diabetes to date of recruitment.
more advanced nephropathy (Cha et al., 2000). Increased VEGF mRNA and protein levels have been observed in models of diabetic kidney disease (Schrijvers, Flyvbjerg, De Vriese, 2003). Furthermore, antibodies directed against VEGF have been shown to improve early renal dysfunction in experimental diabetes (De Vriese et al., 2001). The VEGF gene is clearly an excellent candidate for diabetic nephropathy. VEGF is located on chromosome 6p21.3 and comprises eight exons (Vincenti, Cassano, Rocchi, & Persico, 1996). To date, at least 15 gene variants have been reported in the promoter region and 3 in the 5V untranslated region of the VEGF gene (Brogan et al., 1999; Watson, Webb, Bottomley, & Brenchley, 2000). VEGF1499CYT (rs833061) and VEGF-635GYC (rs2010963) single-nucleotide polymorphisms (SNPs) occur at the highest minor allele frequency, are correlated with altered transcriptional activity, and are associated with several disease phenotypes (Awata et al., 2002; Stevens, Soden, Brenchley, Ralph, & Ray, 2003; Watson et al., 2000). The VEGF-2578CYA SNP (rs699947) is in strong linkage disequilibrium with an 18-base insertion, VEGF2548_2549insTCCCACTCTTCCCACAGG (ss38346984), which is associated with increased transcriptional activity (Schrijvers et al., 2003; Yang, Cross, Ollerenshaw, Millward, & Demaine, 2003). Previous studies have described associations with VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC SNPs and renal disease (Watson et al., 2000; Yang et al., 2003; Summers, 2002a, 2002b). In the present report, we have assessed VEGF-2578CYA, VEGF-
2. Methods Ethical approval was obtained from the appropriate research ethics committees, and written informed consent was obtained from subjects prior to conducting this study. All patients in the case-control cohort (n=543) were Caucasian, with parents and grandparents born in Ireland, who were diagnosed with type 1 diabetes mellitus before 31 years of age and who required insulin from diagnosis. Subjects were recruited from two centers, one in Northern Ireland (NI) and the other in the Republic of Ireland (ROI). Patients with nephropathy (cases, n=242; NI, 153; ROI, 89) had diabetes for at least 10 years before the onset of proteinuria (N0.5 g/24 h) and diabetic retinopathy. Patients without nephropathy (controls, n=301; NI, 184, ROI, 117) had diabetes for at least 15 years with no evidence of microalbuminuria on repeated testing, were not prescribed antihypertensive medication, and had no evidence of nondiabetic renal disease. Healthy controls (n=400) were selected from the Young Hearts collection (Boreham, Savage, Primrose, Cran, & Strain, 1993). Genotyping was performed for VEGF-1499CYT and VEGF-635GYC SNPs by Pyrosequencing using a PSQ 96HS instrument (Biotage, Uppsala, Sweden). DNA samples were genotyped for the VEGF-2578AYC SNP by TaqMan technology using a 7700 Sequence Detection System (Applied Biosystems, Warrington, UK). Reactions were performed according to manufacturer’s instructions, and details of primers and probes are available from the authors. DNA samples representing all three genotypes (confirmed by direct sequencing) for each SNP and no DNA template controls were included in each genotyping run. All genotype calls were independently double scored to minimize errors, and no mismatches were observed. Genotype and allele frequencies were compared using the v 2 test; the level of statistical significance was set at 5%. A stratified analysis was used to combine case-control results from the two centers. The extent of linkage
Table 2 Distributions of genotype and allele frequencies for the VEGF-1499CYT SNP in cases and controls; data are expressed as n (%) NI Cases (n=153)
ROI Controls (n=184)
Cases (n=89)
Combined Controls (n=117)
Cases (n=242)
Controls (n=301)
Genotype CC CT TT
34 (22.2) 66 (43.1) 53 (34.6)
74 (40.2) 84 (45.7) 26 (14.1) Pb.001
18 (20.2) 38 (42.7) 33 (37.1)
40 (34.2) 56 (47.9) 21 (17.9) P=.004
52 (21.5) 104 (43.0) 86 (35.5)
114 (37.9) 140 (46.5) 47 (15.6) Pb.001
Allele C T
134 (43.8) 172 (56.2)
232 (63.0) 136 (37.0) Pb.001
74 (41.6) 104 (58.4)
136 (58.1) 98 (41.9) Pb.001
208 (43.0) 276 (57.0)
368 (61.1) 234 (38.9) Pb.001
244
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245
disequilibrium between pairs of SNPs was quantified using Lewontin’s DVvalue.
3. Results Table 1 lists the clinical characteristics of case and control groups. There were a total of 38.3% and 59.5% females in cases and controls, respectively. The mean duration of diabetes was at least 25 years for both groups. Not surprisingly, average blood pressure was higher in cases versus controls, despite the use of antihypertensive drug treatment in the cases. Genotype distributions were in Hardy–Weinberg equilibrium for all three SNPs in case, control, and healthy populations ( PN.05). The VEGF-1499T allele was significantly increased in cases compared with controls in the NI population (56.2% vs. 37.0%, Pb.001; Table 2). This finding was replicated in a separate population from the ROI (58.4% vs. 41.9%, Pb.001; Table 2). Analysis of the combined data confirmed the association between the VEGF-1499T allele and diabetic nephropathy ( Pb.001; Table 2). Carriage rates for the VEGF-1499T allele (proportion of individuals with VEGF-1499T allele) in cases and controls for the combined population revealed a twofold increase in risk of developing nephropathy (OR=2.24, 95% CI=1.50–3.36, Pb.0001). In contrast, no significant differences were observed in genotype and allele frequencies for VEGF-2578CYA (cases 58 CC, 122 CA, 62 AA giving a C allele frequency of 49.2%; controls 73 CC, 162 CA, 66 AA giving a C allele frequency of 51.2%) or VEGF-635GYC (cases 60 GG, 135 GC, 47 CC giving a G allele frequency of 52.7%; controls 85 GG, 165 GC, 51 CC giving a G allele frequency of 55.6%). No significant differences were observed in genotype and allele frequencies between the healthy control groups and those with type 1 diabetes (genotypes: VEGF-2578CYA, P=.4; VEGF-1499CYT, P=.3; VEGF-635GYC, P=.6). Although |DV| values were not particularly large (|DV|=0.73 for VEGF-2578CYA and VEGF-1499CYT, |DV|=0.50 for VEGF-2578CYA and VEGF-635GYC, and (|DV|=0.27 for VEGF-1499CYT and VEGF-635GYC), all were highly statistically significant ( Pb.001). Allele frequencies for all three SNPs were similar to those reported in other Caucasian populations (Brogan et al., 1999).
4. Discussion In the present report, three putatively functional SNPs, VEGF-2578CYA, VEGF-1499CYT, and VEGF-635GYC, were genotyped for association with diabetic nephropathy in 242 cases, 301 type 1 diabetic controls, and 400 healthy controls from Ireland. We observed a significant increase in the VEGF-1499T allele in type 1 diabetic patients with
nephropathy (n=242) compared with nonnephropathic controls (n=301). Carriage of this allele conferred a twofold increase in risk for nephropathy in diabetic patients in the combined Irish population. Our findings are consistent with a recent UK study in which the VEGF-1499T allele was found to have a significantly higher frequency in diabetic patients with microalbuminuria or proteinuria compared with normoalbuminuric controls (Summers, 2002a, 2002b). In contrast, Ray et al. (2004) failed to replicate this association in their study that comprised type 1 and type 2 diabetic patients with nephropathy defined by the presence of microalbuminuria. Interestingly, the VEGF-1499T allele has also been found to be significantly associated with other diseases (Hsieh et al., 2004; Ku et al., 2005). We used much larger sample sizes compared with others, and our patient groups had longer duration of type 1 diabetes compared with the cohort described by Ray et al. who employed patients groups comprising both type 1 and type 2 diabetes with duration of diabetes just over 20 years. We also employed rigorous phenotypic criteria for inclusion of cases and controls, excluding those individuals with microalbuminuria from case or control groups. Our initial finding of a positive association of the VEGF-1499T allele with diabetic nephropathy in the NI population was also replicated in the ROI population and remained positive in the combined population. In addition, genotyping was performed in this study by Pyrosequencing, which generates a sequence output, thus providing confidence in assigning genotypes. We failed to find an association between VEGF-635GYC and diabetic nephropathy, despite the earlier report that this SNP has been reported associated with end-stage renal disease (Summers, 2002a). However, our results are consistent with those of Ray et al. (2004) who also found no association between VEGF-635GYC and diabetic nephropathy in type 1 and type 2 diabetes. Recently, the VEGF2548_2549insTCCCACTCTTCCCACAGG has been reported by Yang et al. (2003) to be associated with diabetic nephropathy in type 1 diabetes in a UK population. As this insertion is in strong linkage disequilibrium with the VEGF2578CYA SNP, our results suggest differences with those reported by Yang et al. Carriage of the VEGF-2578A allele was more common in our control group (76%), but our study still had 85% power to detect an OR of 2.0 associated with this allele. We compared genotype and allele frequencies for a healthy control group (n=400) to a group with type 1 diabetes (combined cases and controls, n=543). No significant differences were observed between these groups, indicating that these SNPs do not play a major role in type 1 diabetes in our population. In conclusion, our results suggest that the VEGF-1499T allele, or an allele in linkage disequilibrium with this SNP, is associated with diabetic nephropathy in the Irish population. This finding will need to be replicated in larger casecontrol collections such as the Juvenile Diabetes Research
A.-J. McKnight et al. / Journal of Diabetes and Its Complications 21 (2007) 242 – 245
Foundation Genetics of Kidneys in Diabetes (GoKinD) casecontrol collection. Acknowledgments This work was funded by the Research and Development Office Northern Ireland and the Health Research Board of Ireland. AJ McK is supported by a Northern Ireland Kidney Research Fund Postdoctoral Fellowship. References Awata, T., Inoue, K., Kurihara, S., Ohkubo, T., Watanabe, M., Inukai, K., Inoue, I., & Katayama, S. (2002). A common polymorphism is the 5V untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes, 51 (5), 1635 – 1639. Boreham, C., Savage, J. M., Primrose, D., Cran, G., & Strain, J. (1993). Coronary risk factors in school-children. Archives of Disease in Childhood, 68, 182 – 186. Brogan, I. J., Khan, N., Isaac, K., Hutchinson, J. A., Pravica, V., & Hutchinson, I. V. (1999). Novel polymorphisms in the promoter and 5V UTR regions of the human vascular endothelial growth factor gene. Human Immunology, 60 (12), 1245 – 1249. Cha, D. R., Kim, N. H., Yoon, J. W., Jo, S. K., Cho, W. Y., Kim, H. K., Won, N. H. (2000). Role of vascular endothelial growth factor in diabetic nephropathy. Kidney International. Supplement, 77, S104 – S112. De Vriese, A. S., Tilton, R. G., Elger, M., Stephan, C. C., Kriz, W., & Lameire, N. H. (2001). Antibodies against vascular endothelial growth factor improve early renal dysfunction in experimental diabetes. Journal of the American Society of Nephrology, 12 (5), 993 – 1000.
245
Duh, E., & Aiello, L. P. (1999). Vascular endothelial growth factor and diabetes: The agonist versus antagonist paradox. Diabetes, 48, 1899 – 1906. Hsieh, Y. Y., Chang, C. C., Tsai, F. J., Yeh, L. S., Lin, C. C., & Peng, C. T. (2004). T allele for VEGF gene-460 polymorphism at the 5V-untranslated region: Association with a higher susceptibility to endometriosis. Journal of Reproductive Medicine, 49 (6), 468 – 472. Ku, K. T., Wan, L., Peng, H. C., Tsai, M. H., Tsai, C. H., & Tsai, F. J. (2005). Vascular endothelial growth factor gene-460 C/T polymorphism is a biomarker for oral cancer. Oral Oncology, 41 (5), 497 – 502. Ray, D., Mishra, M., Ralph, S., Read, I., Davies, R., & Brenchley, P. (2004). Association of the VEGF gene with proliferative diabetic retinopathy but not proteinuria in diabetes. Diabetes, 53 (3), 861 – 864. Schrijvers, B. F., Flyvbjerg, A., & De Vriese, A. S. (2004). The role of vascular endothelial growth factor (VEGF) in renal pathophysiology. Kidney International, 65 (6), 2003 – 2017. Stevens, A., Soden, J., Brenchley, P. E., Ralph, S., & Ray, D. W. (2003). Haplotype analysis of the polymorphic human vascular endothelial growth factor gene promoter. Cancer Research, 63 (4), 812 – 816. Summers, A. (2002a). VEGF polymorphisms and diabetic nephropathy [Abstract]. Nephrology Dialysis Transplantation, 17 (Suppl. 1), O80. Summers, A. (2002b). VEGF polymorphisms and diabetic nephropathy [Abstract]. Journal of the American Society of Nephrology, 13, 248A. Vincenti, V., Cassano, C., Rocchi, M., & Persico, G. (1996). Assignment of the vascular endothelial growth factor gene to human chromosome 6p21.3. Circulation, 93 (8), 1493 – 1495. Watson, C. J., Webb, N. J., Bottomley, M. J., & Brenchley, P. E. (2000). Identification of polymorphisms within the vascular endothelial growth factor (VEGF) gene: Correlation with variation in VEGF protein production. Cytokine, 12 (8), 1232 – 1235. Yang, B., Cross, D. F., Ollerenshaw, M., Millward, B. A., & Demaine, A. G. (2003). Polymorphisms of the vascular endothelial growth factor and susceptibility to diabetic microvascular complications in patients with type 1 diabetes mellitus. Journal of Diabetes and its Complications, 17 (1), 1 – 6.