Association between sorbitol dehydrogenase gene polymorphisms and type 2 diabetic retinopathy

Available online at www.sciencedirect.com

Experimental Eye Research 86 (2008) 647e652 www.elsevier.com/locate/yexer

Association between sorbitol dehydrogenase gene polymorphisms and type 2 diabetic retinopathy Jacek P. Szaflik a, Ireneusz Majsterek b, Michal Kowalski a, Pawel Rusin b, Anna Sobczuk c, Anna I. Borucka a, Jerzy Szaflik a, Janusz Blasiak b,* a

b

Department of Ophthalmology, Medical University of Warsaw, Warsaw, Poland Department of Molecular Genetics, University of Lodz, Banacha 12/16, 90-237 Lodz, Poland c Polish Mother’s Memorial Hospital, Lodz, Poland Received 3 October 2007; accepted in revised form 8 January 2008 Available online 14 January 2008

Abstract Diabetic retinopathy (DR) may affect 98% of diabetic patients, but its aetiology is poorly understood. Besides glycaemic exposure, genetic factors likely contribute to the onset of DR. The polyol pathway, including aldose reductase and sorbitol dehydrogenase (SDH), can be activated under hyperglycaemic conditions. In our work we searched for an association between the C1214G and G888C polymorphisms of the SDH gene promoter and the occurrence and progression of type 2 DR. Two hundred and fifteen unrelated individuals with type 2 diabetes mellitus (T2DM) were divided into three groups: without DR, with non-proliferative diabetic retinopathy (NPDR) and with proliferative diabetic retinopathy (PDR). Genotypes of the C1214G (rs2055858) and G888C (rs3759890) polymorphisms of the SDH gene were determined with DNA from the peripheral blood lymphocytes of patients by restriction fragment length polymorphism and allele-specific PCR, respectively. The genotype distributions were contrasted by the c2 test and the significance of the polymorphism was assessed by multiple logistic regression producing odds ratios (ORs) and 95% confidence intervals (CIs). We found an association (OR 1.73, 95% CI 1.06e2.83) between NPDR and the G allele of the G888C polymorphism. There was no association between NPDR and the other polymorphisms of the SDH gene. No differences were found in the distributions of these polymorphisms between patients with PDR and those with NPDR. A weak association (OR 2.0, 95% CI 1.29e3.07) was found between DR and the G allele of the G888C polymorphism. Analysis of the combined genotypes (haplotypes) of both polymorphisms revealed associations between the C/GeC/G genotype and NPDR (OR 2.95, 95% CI 1.07e8.13) as well as DR in general (OR 2.91, 95% CI 1.15e7.36). The G888C polymorphism of the SDH gene may be associated with the onset of DR rather than with its progression, and its effect may be strengthened by the interaction with the C1214G polymorphism, but this association is rather weak and requires further study. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: diabetic retinopathy; genetic polymorphism; sorbitol dehydrogenase; restriction fragment length polymorphism

1. Introduction Diabetic retinopathy (DR) is the leading cause of blindness in the working-age group in the Western world (King et al., 1998). It is one of the most serious long-term complications of type 2 diabetes mellitus (T2DM). The prevalence of DR varies from 17% to 98%, depending on the duration of the

* Corresponding author. Tel.: þ48 42 6354334; fax: þ48 42 6354484. E-mail address: [email protected] (J. Blasiak). 0014-4835/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.exer.2008.01.009

diabetes (Klein and Klein, 1995). Ischaemia, macular oedema, and rapid progression of DR may cause vision loss (Lee et al., 1998). The earliest manifestations of DR are venous tortuosity and dot and blot haemorrhages with microaneurysm formation. Soft exudates reflect microinfarcts and hard exudates indicate lipid deposition in the retina. Haemorrhages in DR can be easily distinguished from those in hypertension by their depth and ability to diffuse. Visual loss in DR has two primary causes: progression from non-proliferative DR (NPDR) to the stage of microvascular proliferation called proliferative DR (PDR), in which new, pathological vessels are formed in the

648

J.P. Szaflik et al. / Experimental Eye Research 86 (2008) 647e652

retina and fibrotic scarring of the retina (Vinik, 2005), which is the culmination of PDR. The aetiology of DR is still poorly understood. A large cohort study showed that 28.8% of diabetic patients develop retinopathy in the early stages of the disease, but 22.2% are not affected by DR (Klein et al., 1984). This suggests that high glycaemic exposure is not the only factor determining the onset of retinopathy in diabetic patients. Apparently, genetic factors play a role. Therefore, these factors need to be determined to select a subgroup of diabetic patients, who should be under special ophthalmologic care, and to identify a group in the general population, who should be aware of the serious consequences of developing diabetes. The categorizations of NPDR and PDR can be used to assess the progression of DR. The most precise is the Early Treatment Diabetic Retinopathy Study Research Group scale, according to which retinopathy can be scored as: no retinopathy, non-proliferative, pre-proliferative, proliferative, and advanced proliferative retinopathy (Early Treatment Diabetic Retinopathy Study Research Group, 1991, http://clinicaltrials. gov/ct/show/NCT00000151). Again, the progression of DR seems to be unaffected by factors such as age, duration of diabetes, glycaemic haemoglobin level, and diabetic treatment. This suggests that genetic factors may contribute to the severity of DR, which can be expressed as its progression rate reflected in the classifying DR as NPDR or PDR. Thus, two issues need to be resolved: (1) which genetic factors determine the chance of developing DR in diabetic individuals and (2) the progression, expressed either as NPDR or PDR, of PDR in diabetic patients. Mutations can contribute to the disease phenotype, and are commonly accepted genetic factors that indicate occurrence and progression of a disease, especially if it occurs frequently and takes the form of gene polymorphisms. An association between the polymorphisms in the 50 -untranslated region of the vascular endothelial growth factor (VEGF ) gene and DR has been reported (Suganthalakshmi et al., 2006). This confirmed earlier suggestions about such an association (Awata et al., 2002). Polymorphisms of the insulin-like growth factor-1 (Rietveld et al., 2006), intercellular adhesion molecule 1 (ICAM-1) (Liu et al., 2006), aldose reductase (Awata et al., 2002; Rietveld et al., 2006), receptor for advanced glycation end products (RAGE) (Ramprasad et al., 2007), plasminogen activator inhibitor-1 (PAI-1) genes have also been reported as risk modifiers of DR (Funk et al., 2005; Nagi et al., 1997). Several papers reported an association between DR and polymorphisms in the aldose reductase gene (Olmos et al., 2006; dos Santos et al., 2006; Kao et al., 1999). The enzyme coded by this gene, aldose reductase (AR2, EC 1.1.1.21), converts glucose to sorbitol and represents one of two enzymes of the polyol pathway that can be activated under hyperglycaemic conditions and may therefore be considered as an important biochemical factor in the development of long-term complications of diabetes, including retinopathy (Kinoshita and Nishimura, 1988). AR2 is the first and rate-limiting enzyme of the polyol pathway. The other component of the pathway, sorbitol dehydrogenase (SDH), which converts sorbitol to

fructose, was investigated to a much lesser extent. This is probably a consequence of an apparently more direct role of AR2 than SDH in glucose metabolism, which is a hallmark of diabetes and its complications. However, Tilton et al. (1995) and Amano et al. (2003) suggested that SDH activity might make a greater contribution to the aetiology of DR than did the first step with AR2. Moreover, Amano et al. (2003) suggested that the polymorphisms of the promoter of the SDH gene could be correlated with its expression level in retinal cells in diabetes. Therefore, these polymorphisms may play a role in the pathogenesis of DR. This study was performed on a small Japanese population (97 cases and 46 controls) and a need for further research was stressed. Searching for the meaning of genetic variability in the SDH gene seems to be important in light of recent work of Hellgren et al. (2007), who showed that the Tyr110Phe mutation in the gene abolished the enzymatic activity of its product and destabilized the protein into tetramers, dimers and monomers. In our present work we investigated two polymorphisms of the sorbitol dehydrogenase gene (SDH ) promoter: a C / G transversion located at 1214 position (the C1214G polymorphism) and a G / C transversion at 888 (the G888C polymorphism). These polymorphisms, located in the promoter region, may affect the level of transcription and, consequently, the activity of the SDH enzyme. We searched for an association between genotypes of these polymorphisms and the occurrence and progression of DR in a Polish population. 2. Methods 2.1. Clinical subjects Peripheral blood samples were obtained from 215 individuals with T2DM diagnosed at the Clinical Hospital of Department of Ophthalmology, University of Warsaw, Warsaw, Poland in 2005. Of these, 82 had PDR, 72 had NPDR, and 61 individuals without DR served as controls. All subjects underwent ophthalmic examination, including best corrected visual acuity, slit lamp examination, and fundus examination using non-contact and contact fundus lenses with a slit lamp. The diagnosis or lack of DR was confirmed by fluorescein angiography (FA) in all but not 20 cases. The reasons for not performing FA were the general health status of the 16 patients and four patients’ unwillingness to have FA performed. In this group the diagnosis or lack of DR was established based on fundus examination only. The FA examinations were performed with a Topcon TRC-50I IX fundus camera with digital Image Net ver. 2.14 image system (Topcon Co., Tokyo, Japan). The clinical characteristics of patients are listed in Table 1. The study was approved by the Bioethics Committee of Medical University of Warsaw and each patient gave written informed consent. 2.2. DNA preparation Peripheral blood lymphocytes (PBLs) were isolated by centrifugation in a density gradient of Histopaque-1077 (15 min,

J.P. Szaflik et al. / Experimental Eye Research 86 (2008) 647e652

649

Table 1 Clinical characteristics of diabetes mellitus type 2 (T2DM) patients without diabetic retinopathy (DR), with non-proliferative (NPDR), or with proliferative retinopathy (PDR) (mean  SD) Patients

Number

Age (years)

Gender

Duration of T2DM (years)

Insulin uptake (number/percentage)

All T2DM without DR T2DM with NPDR T2DM with PDR

215 61 72 82

66.7  10.9 67.7  7.9 71.7  11.9 60.7  12.9

130W þ 85M 41W þ 20M 41W þ 31M 48W þ 34M

17.6  8.6 12.4  7.2 18.9  9.8 21.5  8.8

138/64% 18/30% 57/79% 63/77%

280g). The pellet containing PBLs was resuspended in TriseEDTA buffer, pH 8, to give about 1e3  105 cells/ml. Genomic DNA was extracted from PBLs by phenol/chloroform extraction and proteinase K digestion. The final samples were kept in TriseEDTA buffer, pH 8, at 20  C until use. 2.3. Determination of the SDH genotype Restriction fragment length polymorphism PCR was applied to determine the genotypes of the G888C polymorphism and allele-specific PCR the C1214G polymorphism. Each 20 ml of the PCR reaction contained 10 ng genomic DNA, 1.25 U Taq polymerase (InGen-TERPOL, Sieradz, Poland) in 1 PCR buffer (100 mM TriseHCl, pH 8.3, 500 mM KCl, 11 mM MgCl2, 0.1% gelatin), 1.5 mM MgCl2, 50 mM dNTPs, and 250 nM each primer. Thermal cycling conditions were as follows: initial denaturation step at 95  C for 1 min, 30 cycles at 95  C for 30 s and 30 s at 62  C annealing temperature, and final 30 s at 72  C. The final extension step was performed at 72  C for 5 min. The PCR was carried out in an MJ Research, INC thermal cycler, model PTC-100 (Waltham, MA, USA). The genotypes of the G888C polymorphism were determined by using sense 50 -CGCCCG GCCTCATGTCTTTT-30 and antisense 50 -TTGGGGTGGGG AATGTGAGG-30 (Institute of Biochemistry and Biophysics, Warsaw, Poland). The 307 bp PCR product was digested overnight with 5 U of the restriction enzyme BsaJI (Fermentas, Vilnius, Lithuania). The G allele was digested into 90 bp and 217 bp fragments, whereas the C variant remained intact. The genotypes of the C1214G polymorphism were determined by using the following primers: two allele-specific sense oligonucleotides (385 bp variant): 50 -TGTTGCCCA GGCTGGTGTTC-30 for C variant and 50 -TGTTGCCCAGGC TGGTGTTG-30 for G variant and two control (494 bp) primers: sense 50 -TTTGGGCAGAAACTCTG-30 and antisense 50 -ACGCAGCGTCCACGTC-30 . PCR products were separated onto a 10% polyacrylamide gel.

genotypes between groups. The odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by using a logistic regression model. The t-test (for normal distribution) or ManneWhitney test (for non-normal distribution) was used to compare each parameter between two groups. An analysis of variance test was used to identify parameters that would make significant differences between more than two groups; Scheffe’s test was then used to assess the significance of difference in each identified parameter between any two groups. STATISTICA 6.0 software (Statsoft, Tulsa, OK, USA) was used to perform analyses. 3. Results From the PCR analysis, all the patients were divided into three genotypes of the C1214G SDH gene promoter polymorphism: C/C, C/G, and G/G, as well as three genotypes: G/G, G/C, and C/C of the other polymorphism of this gene, G888C (Figs. 1 and 2). The distributions of these genotypes did not differ significantly from those expected by the Hardye Weinberg equilibrium. There was no association between DR and the genotypes and alleles of the G1214G polymorphism of the SDH gene (data not presented). A weak association (OR 2.0, 95% CI 1.29e3.07) was found between DR and the G allele of the other polymorphism, G888C, of this gene (Table 2). Another weak association (OR 1.73, 95% CI 1.06e2.83) was found between this allele and NPDR (data not presented). There

2.4. Statistical analysis The allelic frequencies were estimated by gene counting and genotypes were scored. The c2 test was used to compare the observed numbers of genotypes with those expected for a population in the HardyeWeinberg equilibrium and to test the significance of the differences of observed alleles and

Fig. 1. Genotypes of the C/G polymorphism at the 1214 bp SDH promoter region determined by the allele-specific PCR detection (ASO-PCR) and analysed by 10% polyacrylamide gel electrophoresis, stained with ethidium bromide and viewed under UV light. Lane M displays molecular weight marker; lanes 1e3 show the results of amplification with primer specific to the C allele; and lanes 4e6 with primer specific to the G allele.

J.P. Szaflik et al. / Experimental Eye Research 86 (2008) 647e652

650

Table 3 The distribution of combined genotypes of the C1214G and the G888C polymorphisms of the SDH gene and odds ratio (OR) in type 2 diabetes mellitus (T2DM) patients without or with diabetic retinopathy (DR)

Fig. 2. Genotypes of the C/G polymorphism at the 888 bp SDH promoter region determined by the PCR-based BsaJI restriction fragment length polymorphism (RFLP-PCR) and analysed by 10% polyacrylamide gel electrophoresis, stained with ethidium bromide and viewed under UV light. Lanes 1 and 2 show the results of amplification of C/G heterozygote, lanes 3 and 4 shows homozygote G/G, lane 5 shows homozygote C/C and lane M displays molecular weight marker.

was no association between NPDR and the other polymorphism of the SDH gene (data not presented). No correlation was found between the polymorphisms of the SDH gene and either PDR or NPDR (data not presented). DR correlated with the combined C/ GeC/G genotype (haplotype) of both polymorphisms (OR 2.91, 95% CI 1.15e7.36, Table 3). There was also an association between NPDR and the combined G/CeG/C genotype (OR 2.95, 95% CI 1.07e8.13, Table 4). No difference between genotype distributions was found for the combined genotypes of NPDR and PDR patients (data not presented). 4. Discussion The polyol pathway seems to be of special significance in the microvascular complications of T2DM. Therefore, genetic variability of the gene coding for two main enzymes of this pathway, aldose reductase and SDH, should be taken into account in the assessment of a patient’s susceptibility to DR. The association of DR with the polymorphism of the AR2 gene was reported for many populations, including Caucasians (Petrovic et al., 2005) Caucasian-Brazilians (Ramprasad et al., 2007), Chileans (Olmos et al., 2000), Indians (Suganthalakshmi et al., 2006), and Japanese (Awata et al., 2002). However, this association may not be linked with an ethnic group, because no correlation was found between the polymorphisms of the AR2 gene and DR in Euro-Brazilian T2DM patients (Santos et al., 2003). In those studies various polymorphisms of the AR2 gene were investigated, which suggest that the locus of the gene can be considered as a locus of susceptibility to DR independent of ethnic origin. In more general sense, those Table 2 The distribution of genotypes, frequency of alleles of the G888C polymorphism of the SDH gene and odds ratio (OR) in type 2 diabetes mellitus (T2DM) patients without or with diabetic retinopathy (DR) Genotype or allele

Number

Frequency

Number

Frequency

OR (95% CI)

C/C C/G G/G C G

17 88 38 122 164

0.12 0.61 0.26 0.43 0.56

21 31 9 73 49

0.34 0.51 0.15 0.6 0.4

0.26 1.55 2.08 0.5 2.0

DR (n ¼ 143); T2DM without DR (n ¼ 61).

(0.11e0.52) (0.84e2.82) (0.94e4.65) (0.31e0.78) (1.29e3.07)

Genotype

Number

Frequency

Number

Frequency

OR (95% CI)

C/CeC/C C/CeC/G C/CeG/G C/GeC/C C/GeC/G C/GeG/G G/GeC/C G/GeC/G G/GeG/G

2 4 11 1 36 46 2 18 17

0.01 0.03 0.08 e 0.25 0.33 0.01 0.12 0.11

3 7 10 1 6 19 1 2 7

0.05 0.12 0.18 0.02 0.11 0.34 0.02 0.03 0.12

0.26 0.2 0.38 0.4 2.91 0.96 0.8 4.0 0.96

(0.03e1.57) (0.06e0.73) (0.16e0.99) (0.01e6.45) (1.15e7.36) (0.49e1.84) (0.06e9.0) (0.9e17.9) (0.38e2.48)

‘‘e’’, Not estimated. DR (n ¼ 137); T2DM without DR (n ¼ 55).

results, along with these obtained in the present work with the SDH gene and results coming from another laboratory (Amano et al., 2003), suggest that genetic variability of the components of the polyol pathway may be important in the occurrence of DR and its progression in T2DM patients. SDH mediates the conversion of sorbitol to fructose and may play an active role in the pathogenesis of DR. Therefore, blocking of the formation of sorbitol may be a therapeutic strategy for the treatment of early phase DR (Amano et al., 2002). The works of Amano et al. (2003) and the present study indicate that this strategy can be enforced by a gene-targeting approach aimed at a particular genotype. In general, diabetic retinopathy is a multigene disease and changes in many genes may play a role in its etiopathology (Santos et al., 2003; Suganthalakshmi et al., 2006). Therefore, single nucleotide polymorphisms with relatively small odds ratios may contribute in a moderate degree to this disease. Use of a microarray technique is probably the best approach to determine changes in a set of genes along with their expression. However, to establish such a set of gene candidates for DR microarray, single nucleotide polymorphisms relevant to clinical data should be studied. The frequency of the genotypes of the G888C polymorphism ranges from 0.220 (sub-Saharan African) to 0.855

Table 4 The distribution of combined genotypes of the C1214G and the G888C polymorphisms of the SDH gene and odds ratio (OR) in type 2 diabetes mellitus (T2DM) patients without or with non-proliferative diabetic retinopathy (NPDR) Genotype

Number Frequency Number Frequency OR (95% CI)

C/CeC/C 1 C/CeC/G 3 C/CeG/G 7 C/GeC/C e C/GeC/G 17 C/GeG/G 20 G/GeC/C 2 G/GeC/G 5 G/GeG/G 9

0.01 0.05 0.11 e 0.26 0.31 0.03 0.08 0.14

3 7 10 1 6 19 1 2 7

0.05 0.12 0.18 0.02 0.11 0.34 0.02 0.03 0.12

0.27 0.34 0.55 e 2.95 0.86 1.74 2.25 1.12

‘‘e’’, Not estimated. T2DM with NPDR (n ¼ 64); T2DM without NPDR (n ¼ 55).

(0.03e2.72) (0.08e1.37) (0.19e1.56) (1.07e8.13) (0.39e1.85) (0.15-19.75) (0.42e12.06) (0.39e3.24)

J.P. Szaflik et al. / Experimental Eye Research 86 (2008) 647e652

(European) for C/C, from 0.145 (European) to 0.600 (Asian) for C/G and from near 0 (European) to 0.186 (sub-Saharan African) for G/G (http://www.ncbi.nlm.nih.gov/SNP/snp_ref. cgi?type¼rs&rs¼rs3759890). There are no frequency data on general population diversity of the other polymorphism. From the data displayed in Table 2, we can speculate that individuals carrying G/G and G/C genotypes of the G888C polymorphism may be more susceptible to DR after developing T2DM than persons with C/C. This is an interesting idea, because if true, the G888C polymorphism would be an independent genetic marker of DR, however, its association is rather weak and needs further study. Interpreting our data in a more general context and along with those indicating lack of association of this polymorphism and PDR compared with NPDR, we can speculate that the G888C polymorphism may be linked with the onset of DR, but not with its progression. This conclusion was confirmed in the analysis of combined genotypes (haplotypes) of the C1214G and G888C polymorphisms of the SDH gene in T2DM patients with or without DR (Table 3) and in patients with PDR contrasted with patients with NPDR. The results obtained are not in complete agreement with those of Amano et al. (2003). They showed that combined genotypes of the C1214G and G888C polymorphisms could be associated with DR: the frequencies of the CeG allele and homozygote of the CeG allele in DR patients tend to be higher than those in patients without DR. The difference may lie in the different ethnic origin of two populations: Japanese versus Polish, but both studies indicate that genetic variability in the SDH gene may contribute to susceptibility to DR. In both polymorphisms we studied a pyrimidine (C) is replaced with a purine (G) or vice-versa, so both polymorphisms are transversion mutations. Transversions have usually a phenotypic effect if they are located in the coding region of a gene. The polymorphisms under study are located in the 50 -untranslated region of the gene. This region contains regulatory elements of the gene and is referred to as promoter or general promoter in contrast to the core promoter, absolutely needed for the transcription to occur. In fact, little is known about the core promoter of the SDH gene. From the definition, the core promoter refers to the minimal set of elements required for the precise transcription initiation by the RNA polymerase II. The elements are usually 2 or 3 of TFIIB recognition element, the TATA element (box), the initiator, and the downstream promoter element. The promoter of the SDH gene does not contain a classical TATAA box nor CCAAT element (Iwata et al., 1995). Three Sp1 sites, which are binding sites for the AP1 transcription factor, and a repetitive sequence (CAAA)5 have been found in the 50 -UTR of the gene. Two transcription initiation sites have been identified in it. So far, there is no evidence for the presence of sequences (cis elements), which might be binding sites for transcription factors (trans elements), in so distant regions as the polymorphisms we studied. However, in eukaryotic cells, long-distance regulation in the promoter is rather common. This regulation can be cisetrans or cisecis type (or both) and can induce DNA bending, leading to bringing closer together regions

651

that otherwise are far away. Such type of interaction cannot be excluded in the regions where the C1214G and G888C polymorphisms are located. We assumed the following order of events: T2DM / T2DM/NPDR / T2DM/PDR and compared T2DM patients without any DR with subjects with T2DM/NPDR and T2DM/NPDR with both types of DR and compared T2DM patients and individuals with NPDR. Consequently, we did not contrast patients without DR with PDR patients. In summary, the C1214G and G888C polymorphisms of the SDH gene seem to be markers of the onset of DR and presumably play no role in the progression of this disease from its non-proliferative to its proliferative form. Further studies on different ethnic populations and larger groups are needed to determine the role of these polymorphisms in DR in more detail. Acknowledgements We thank Tomasz Wysocki for his technical assistance. This work was supported by the grant 505/376 from the University of Lodz (IM, JB) and by grants funded by BIOTON S.A. and PROKOM INVESTMENTS S.A. References Amano, S., Yamagishi, S., Kato, N., Inagaki, Y., Okamoto, T., Makino, M., Taniko, K., Hirooka, H., Jomori, T., Takeuchi, M., 2002. Sorbitol dehydrogenase overexpression potentiates glucose toxicity to cultured pericytes. Biochem. Biophys. Res. Commun. 299, 183e188. Amano, S., Yamagishi, S., Koda, Y., Tsuneoka, M., Soejima, M., Okamoto, T., Inagaki, Y., Yamada, K., Kimura, H., 2003. Polymorphisms of sorbitol dehydrogenase (SDH) gene and susceptibility to diabetic retinopathy. Med. Hypotheses 60, 550e551. Awata, T., Inoue, K., Kurihara, S., Ohkubo, T., Watanabe, M., Inukai, K., Inoue, I., Katayama, S., 2002. A common polymorphism in the 50 -untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes 51, 1635e1639. Early Treatment Diabetic Retinopathy Study Research Group, 1991. Fundus photographic risk factors for progression of diabetic retinopathy. Early Treatment Diabetic Retinopathy Study Report 12. Ophthalmology 98, 823e833. Funk, M., Endler, G., Exner, M., Marculescu, R., Endler, L., Abrahamian, H., Mauler, H., Grimm, A., Raith, M., Mannhalter, C., Praqer, R., Irsiqler, K., Wagner, O.F., 2005. PAI-1 4G/5G insertion/deletion promoter polymorphism and microvascular complications in type 2 diabetes mellitus. Wien. Klin. Wochenschr. 117, 707e710. Hellgren, M., Kaiser, C., de Haij, D., Norberg, A., Hoog, J.O., 2007. A hydrogen-bonding network in mammalian sorbitol dehydrogenase stabilizes the tetrameric state and is essential for the catalytic power. Cell. Mol. Life Sci. 64, 3129e3138. Iwata, T., Popescu, N.C., Zimonjic, D.B., Karlsson, C., Hoog, J.-O., Vaca, G., Rodriguez, I.R., Carper, D., 1995. Structural organization of the human sorbitol dehydrogenase gene (SORD). Genomics 26, 55e62. Kao, Y.-L., Donaghue, K., Chan, A., Knight, J., Silink, M., 1999. An aldose reductase intragenic polymorphism associated with diabetic retinopathy. Diabetes Res. Clin. Pract. 46, 155e160. King, H., Aubert, R.E., Herman, W.H., 1998. Global burden of diabetes, 1995e2025: prevalence, numerical estimates, and projections. Diabetes Care 21, 1414e1431. Kinoshita, J.H., Nishimura, C., 1988. The involvement of aldose reductase in diabetic complications. Diabetes Metab. Rev. 4, 323e337.

652

J.P. Szaflik et al. / Experimental Eye Research 86 (2008) 647e652

Klein, R., Klein, B.E., Moss, S.E., Davis, M.D., DeMets, D.L., 1984. The Wisconsin study of epidemiologic study of diabetic retinopathy. III. Prevalence and risk of diabetic retinopathy when age at diagnosis is 30 or more years. Arch. Ophthalmol. 102, 527e532. Klein, R., Klein, B.E.K., 1995. Vision disorders in diabetes. In: Harris, M.I., Cowie, C.C., Stern, M.P., et al. (Eds.), National Diabetes Data Group: Diabetes in America. National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, pp. 293e338. Publication No. 95-1468. Lee, P., Wang, C.C., Adamis, A., 1998. Ocular neovascularization: an epidemiological review. Surv. Ophthalmol. 43, 245e269. Liu, L., Yu, O., Wang, H., Zhang, S.X., Huang, C., Chen, X., 2006. Association of intercellular adhesion molecule 1 polymorphisms with retinopathy in Chinese patients with type 2 diabetes. Diabetes Med. 23, 643e648. Nagi, D.K., McCormack, L.J., Mohamed-Ali, V., Yudkin, J.S., Knowler, W.C., Grant, P.J., 1997. Diabetic retinopathy, promoter (4G/5G) polymorphism of PAI-1 gene, and PAI-1 activity in Pima Indians with type 2 diabetes. Diabetes Care 20, 1304e1309. Olmos, P., Futers, S., Acosta, A.M., Siegel, S., Maiz, A., Schiaffino, R., Morales, P., Diaz, R., Arriagada, P., Claro, J.C., Vega, R., Vollrath, V., Velasco, S., Emmerich, M., 2000. (AC)23 [Z-2] polymorphism of the aldose reductase gene and fast progression of retinopathy in Chilean type 2 diabetics. Diabetes Res. Clin. Pract. 47, 169e176. Olmos, P., Bastias, M.J., Vollrath, V., Toro, L., Trincado, A., Salinas, P., Claro, J.C., Lopez, J.M., Acosta, A.M., Miguel, J.F., Castro, J., 2006. C(106)T polymorphism of the aldose reductase gene and the progression rate of diabetic retinopathy. Diabetes Res. Clin. Pract. 74, 75e82. Petrovic, M., Peterlin, B., Hawlina, M., Petrovic, D., 2005. Aldose reductase (AC)n gene polymorphism and susceptibility to diabetic retinopathy in type 2 diabetes in Caucasians. J. Diabetes Complicat. 19, 70e73.

Ramprasad, S., Radha, V., Mathias, R.A., Majumder, P.P., Rao, M.R., Rema, M., 2007. RAGE gene promoter polymorphisms and diabetic retinopathy in a clinic-based population from South India. Eye 21, 395e401. Rietveld, I., Ikram, M.K., Vingerling, J.R., Hofman, A., Pols, H.A., Lamberts, S.W., de Jong, P.T., van Duijn, C.M., Janssen, J.A., 2006. An igf-1 gene polymorphism modifies the risk of diabetic retinopathy. Diabetes 55, 2387e2391. dos Santos, K.G., Canani, L.H., Gross, J.L., Tschiedel, B., Souto, K.E., Roisenberg, I., 2006. The 106CC genotype of the aldose reductase gene is associated with an increased risk of proliferative diabetic retinopathy in Caucasian-Brazilians with type 2 diabetes. Mol. Genet. Metabol. 88, 280e284. Santos, K.G., Tschiedel, B., Schneider, J., Souto, K., Roisenberg, I., 2003. Diabetic retinopathy in Euro-Brazilian type 2 diabetic patients: relationship with polymorphisms in the aldose reductase, the plasminogen activator innhbitor-1 and the methylenetetrahydrofolate reductase genes. Diabetes Res. Clin. Pract. 61, 133e136. Suganthalakshmi, B., Anand, R., Kim, R., Mahalakshmi, R., Karthikprash, S., Namperumalsamy, P., Sundaresan, P., 2006. Association of VEGF and eNOS gene polymorphisms in type 2 diabetic retinopathy. Mol. Vis. 12, 336e341. Tilton, R.G., Chang, K., Nyengaard, J.R., Van den Enden, M., Ido, Y., Williamson, J.R., 1995. Inhibition of sorbitol dehydrogenase. Effect on vascular and neural dysfunction in streptozocin-induced diabetic rats. Diabetes 44, 234e242. Vinik, A., 2005. The protein kinase C-b inhibitor, ruboxistaurin, for the treatment of diabetic microvascular complications. Expert Opin. Investig. Drugs 14, 1547e1559.