Vascular endothelial growth factor (VEGF)-related polymorphisms rs10738760 and rs6921438 are not risk factors for proliferative diabetic retinopathy (PDR) in patients with type 2 diabetes mellitus (T2DM)

  • Jana Sajovic Institute of Histology and Embryology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
  • Ines Cilenšek Institute of Histology and Embryology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
  • Sara Mankoč Institute of Histology and Embryology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
  • Špela Tajnšek Department of Animal Science, Biotechnical Faculty, University of Ljubljana, Domžale, Slovenia
  • Tanja Kunej Department of Animal Science, Biotechnical Faculty, University of Ljubljana, Domžale, Slovenia
  • Daniel Petrovič Institute of Histology and Embryology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
  • Mojca Globočnik Petrovič University Medical Centre, University Eye Clinic, Ljubljana, Slovenia Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
Keywords: Vascular endothelial growth factor, VEGF, SNP, single nucleotide polymorphism, proliferative diabetic retinopathy, PDR, T2DM, diabetes, VEGFR-2

Abstract

Vascular endothelial growth factor (VEGF) is an important regulator of angiogenesis and has been investigated as a candidate gene in a number of conditions, including diabetes and its microvascular complications (e.g., retinopathy and nephropathy). Several VEGF-related polymorphisms have been shown to contribute to nearly half of the variability in circulating VEGF levels in healthy individuals. Our aim was to assess the association between VEGF-related rs10738760 and rs6921438 polymorphisms and proliferative diabetic retinopathy (PDR) in Slovenian patients with type 2 diabetes mellitus (T2DM). We also investigated the effect of these polymorphisms on VEGF receptor 2 (VEGFR-2) expression in fibrovascular membranes (FVMs) from patients with PDR. This case-control study enrolled 505 unrelated patients with T2DM: 143 diabetic patients with PDR as a study group, and 362 patients with T2DM of >10 years duration and with no clinical signs of PDR as a control group. Patient clinical and laboratory data were obtained from their medical records. rs10738760 and rs6921438 polymorphisms were genotyped using TaqMan SNP Genotyping assay. VEGFR-2 expression was assessed by immunohistochemistry in 20 FVMs from patients with PDR, and numerical areal density of VEGFR-2-positive cells was calculated. The occurrence of PDR was 1.7 times higher in diabetic patients carrying GA genotype of rs6921438 compared to patients with GG genotype, with a borderline statistical significance (OR = 1.7, 95% CI = 1.00 – 2.86, p = 0.05). In addition, A allele of rs6921438 was associated with increased VEGFR-2 expression in FVMs from PDR patients. However, we observed no association between AA genotype of rs6921438 nor between rs10738760 variants and PDR, indicating that the two polymorphisms are not genetic risk factors for PDR.

Downloads

Download data is not yet available.

Author Biographies

Špela Tajnšek, Department of Animal Science, Biotechnical Faculty, University of Ljubljana, Domžale, Slovenia
Department of Animal Science
Tanja Kunej, Department of Animal Science, Biotechnical Faculty, University of Ljubljana, Domžale, Slovenia
Department of Animal Science
Vascular endothelial growth factor (VEGF)-related polymorphisms rs10738760 and rs6921438 are not risk factors for proliferative diabetic retinopathy (PDR) in patients with type 2 diabetes mellitus (T2DM)
Published
2019-02-12
How to Cite
1.
Sajovic J, Cilenšek I, Mankoč S, Tajnšek Špela, Kunej T, Petrovič D, Globočnik Petrovič M. Vascular endothelial growth factor (VEGF)-related polymorphisms rs10738760 and rs6921438 are not risk factors for proliferative diabetic retinopathy (PDR) in patients with type 2 diabetes mellitus (T2DM). Bosn J of Basic Med Sci [Internet]. 2019Feb.12 [cited 2020Sep.25];19(1):94-100. Available from: https://www.bjbms.org/ojs/index.php/bjbms/article/view/3519
Section
Translational and Clinical Research

INTRODUCTION

Diabetic retinopathy (DR) is a common microvascular complication of diabetes and is the leading cause of blindness in adults aged 20–74 years in developed countries [1-5]. It occurs in more than 60% of patients with type 2 diabetes mellitus (T2DM) and in almost all patients with type 1 diabetes mellitus (T1DM) within the first decade of diabetes onset [6-8]. DR may progress through several stages, including early non-proliferative diabetic retinopathy (NPDR), moderate NPDR, severe NPDR and finally advanced, proliferative, DR (PDR) [9,10]. The main characteristics of NPDR are microaneurysms, hemorrhage, cotton wool spots and hard exudates (lipid deposits). The distinctive features of PDR include increased vascular permeability, tissue ischemia and neovascularization that leads to fibrovascular changes, vitreoretinal traction and retinal detachment, eventually resulting in blindness [3,5,11].

It is accepted that both genetic and environmental factors influence the severity and progression of DR [12,13]. As an important regulator of angiogenesis, vascular endothelial growth factor (VEGF) has been investigated as a candidate gene in a large number of conditions including cardiovascular, inflammatory diseases, diabetes, DR, diabetic nephropathy (DN) and cancer [14]. Several studies have shown that VEGF is involved in the progression of DR, leading to the proliferative stage of the disease [15]. Under physiological conditions VEGF is expressed at low levels, whereas it is markedly upregulated in the vitreous and aqueous humours of the eyes from patients with PDR [16-19]. VEGF is a potent angiogenic and vascular permeability factor (VPF); in addition to its involvement in neovascularization process in PDR, VEGF has been implicated in the development of early DR, as well as in breakdown of the blood-retina barrier in diabetic macular edema [14,20]. Hypoxia is the major stimulus for VEGF production in retinal areas with reduced perfusion. The mechanisms involve increased VEGF transcription, mRNA stability, mRNA translation as well as increased production of the chaperone oxygen-regulated protein 150 (ORP150). The increase in VEGF gene expression is predominantly mediated by hypoxia-inducible factor-1 (HIF-1) [12].

The human VEGF gene is located on chromosome 6p21.3 and consists of 8 exons separated by 7 introns [21,22]. Single nucleotide polymorphisms (SNPs) of or related to the VEGF gene have been associated with VEGF expression/serum levels [23-27]. SNPs located in the VEGF gene, such as rs833061 and rs699947 (both located in the promoter), rs2010963 (in the 5′ untranslated region [5′ UTR]) and rs3025039 (in the 3′ UTR) have been best studied in the context of diabetes microvascular complications [26]. On the other hand, only a few studies investigated the association of diabetes and diabetic complications with SNPs related to but located outside the VEGF gene. A genome-wide association (GWA) study showed that the polymorphism rs10738760 (located on chromosome 9p24.2, between the VLDLR and KCNV2 genes) and rs6921438 (on 6p21.1 chromosome 171 kb downstream of the VEGFA and close to the MRPL14 gene) together with rs4416670 (also located on 6p21.1) and rs6993770 (on 8q23.1 within the ZFPM2 gene) explain nearly half of the variability in circulating VEGF levels [27].

In otherwise healthy individuals, the rs6921438 polymorphism was shown to interact with hypertension resulting in decreased VEGF levels [28] as well as to contribute to decreased high-density lipoprotein-cholesterol (HDL-C) and increased low-density lipoprotein cholesterol (LDL-C) levels [29], which may have a negative effect on the cardiovascular system. In addition, an association of the rs10738760 polymorphism with increased risk of metabolic syndrome, higher VEGF and lower HDL levels was demonstrated in patients with metabolic syndrome [30]. Bonnefond et al. [31] investigated the effect of rs6921438 and rs10738760 on the risk of T2DM, nephropathy and retinopathy in Danish and/or French populations. Overall they reported no significant association between the two VEGF-related SNPs and the risk of T2DM, DN or DR [31]. Similarly, Terzić et al. [32] showed that rs6921438 and rs10738760 were not associated with DR in Slovenian patients with T2DM.

In this study, we aimed to assess the association between the rs6921438 and rs10738760 polymorphisms and PDR in Slovenian patients with T2DM. Moreover, we investigated the effect of these SNPs on VEGF receptor 2 (VEGFR-2) expression in fibrovascular membranes (FVMs) from patients with PDR.

MATERIALS AND METHODS

Patients

This case-control study enrolled 505 unrelated Slovenian (white) patients with T2DM who had a defined ophthalmologic diagnosis. The participants were recruited from the University Eye Clinics Ljubljana and from the internal medicine clinics (cardiology, diabetic, and endocrinology clinics) of the University Clinical Center in Ljubljana and Maribor. The patients were classified as having type 2 diabetes according to the 2003 American Diabetes Association criteria for the diagnosis and classification of T2DM [33]. For a dilated fundus examination, the pupils were dilated with 2.5% tropicamide and phenylephrine and a senior ophthalmologist (M.G.P.) performed the examination using a slit lamp biomicroscope with non-contact lens. The result was electronically documented by a fundus camera with a 50-degree angle (Topcon-TRC 40-IX, Tokyo, Japan). DR staging was determined according to the Early Treatment Diabetic Retinopathy Study (ETDRS) diabetic retinopathy severity scale [34].

The study group consisted of 143/505 patients with T2DM and PDR (evidence of new blood vessel formation at the optic disc or elsewhere in the retina and/or fibrous proliferation with or without vitreous hemorrhage). The control group consisted of 362/505 patients with T2DM of more than 10 years duration who had NPDR or had no clinical signs of DR. Demographic, clinical and laboratory parameters such as age, gender, diabetes duration, insulin therapy, systolic blood pressure, diastolic blood pressure, body mass index (BMI), history of hypertension, cigarette smoking, total cholesterol, LDL-C, HDL-C, triglycerides and glycated hemoglobin (HbA1c), were retrieved from the patients’ medical records.

To avoid the confounding effect of impaired kidney function, patients with overt nephropathy were excluded from the study. The study was approved by the National Medical Ethics Committee. After informed consent was obtained, a detailed interview with patients was performed.

Genotyping

DNA was isolated from peripheral blood using the QIAamp DNA Blood Mini Kit (Qiagen GmbH, Hilden, Germany). The rs10738760 and rs6921438 polymorphisms were genotyped using the TaqMan SNP Genotyping assay (Applied Biosystems, Foster City, CA, USA), according to the manufacturer protocol. Genotyping was done for cases and multiple controls (homozygous sample or one allele, heterozygous sample and homozygous sample for another allele and negative control).

Immunohistochemistry

FVMs were obtained from 20 patients with T2DM and PDR during pars plana vitrectomy, as described previously [17]. Formalin-fixed paraffin-embedded (FFPE) tissue sections of FVMs were used for hematoxylin-eosin (HE) staining [18]. Consecutive 5-µm tissue sections were cut from each paraffin block and then mounted and dried on glass slides. Tissues were deparaffinized and dehydrated in graded alcohol solutions. Detection of VEGFR-2-positive cells was performed with the NovoLink Max Polymer Detection System (Leica Biosystems Newcastle Ltd, United Kingdom) following the manufacturer’s instructions. The slides were incubated with anti-VEGFR-2 monoclonal antibodies (diluted 1:100, ThermoFisher, USA) overnight at 4°C. Placental tissue was used as a positive and tonsils as a negative control. The cells were defined as VEGFR-2-positive/negative. The area with VEGFR-2-positive cells was manually marked and numerical areal density of VEGFR-2-positive cells was calculated (the number of positive cells per mm2) [17].

Statistical analysis

Allelic discrimination analysis was conducted using StepOne Software version 2.2 (Applied Biosystems, Foster City, CA, USA). Statistics was performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp., Armonk, NY). Categorical variables were expressed as the number and percentage of patients. Continuous variables were expressed as mean ± standard deviation (SD) and compared by unpaired Student’s t-test. Discrete variables were compared using Chi-square test. All variables showing significant differences in the univariate analysis (p < 0.05) were analyzed by logistic regression analysis. A value of p < 0.05 was considered statistically significant. The deviation from Hardy –Weinberg equilibrium (HWE) was assessed by Fisher’s exact test (http://ihg.gsf.de/) [35].

RESULTS

Clinical and laboratory characteristics of study group (patients with T2DM and PDR) and control group (patients with T2DM without PDR) are presented in Table 1. The two groups of T2DM patients were well matched for age, systolic and diastolic blood pressure, HbA1c, total cholesterol and triglyceride levels. A statistically significant difference between T2DM patients with and without PDR was observed in diabetes duration (p < 0.001), insulin therapy (p < 0.001), LDL-C (p = 0.001), HDL-C levels (p = 0.001), and BMI (p < 0.001). Diabetic patients with PDR had a longer duration of type 2 diabetes, higher LDL-C, lower HDL-C levels, and higher prevalence of insulin therapy compared to controls. BMI was significantly higher in controls. Moreover, hypertension and cigarette smoking were more frequent in the group of diabetics with PDR compared to control group.

TABLE 1: Clinical and laboratory characteristics of study group (patients with T2DM and PDR) and control group (patients with T2DM without PDR)

Table 2 shows the genotype and allele frequencies of rs10738760 and rs6921438 polymorphisms. The two SNPs conformed to HWE in both groups (diabetic patients with and without PDR); i.e., no deviations from HWE were observed (p > 0.05).

TABLE 2: Distribution of rs10738760 and rs6921438 genotypes and alleles in study group (patients with T2DM and PDR) and control group (patients with T2DM without PDR)

The allele frequencies of rs10738760 and rs6921438 in our cohort were consistent with the 1000 Genomes Project data of the European population. We observed no significant difference in the genotype or allele frequencies of rs10738760 and rs6921438 polymorphisms between diabetics with PDR and diabetic patients without PDR (Table 2).

We then performed a logistic regression analysis using a co-dominant model of inheritance to evaluate whether rs10738760 and rs6921438 polymorphisms were independently associated with PDR after adjusting for duration of diabetes, insulin therapy, BMI, LDL-C and HDL-C (Table 3). The logistic regression analysis showed a tendency for increased risk of PDR in patients with the GA genotype of rs6921438, with a borderline statistical significance (OR = 1.70, 95% CI = 1.00 – 2.86, p = 0.05) compared to patients with the GG genotype. Namely, the GA genotype of rs6921438 was found to modify the susceptibility to PDR, since the occurrence of PDR was 1.70 times higher in diabetic patients with GA genotype compared to the reference group (diabetics with GG genotype). However, the AA genotype of rs6921438 did not achieve a statistically significant association with PDR. Additionally, no association between rs10738760 variants and PDR was observed.

TABLE 3: Association between rs10738760 and rs6921438 polymorphisms and the risk for PDR

In FVMs from diabetic patients with PDR a significantly higher numerical areal density of VEGFR-2-positive cells (Figure 1) was found in patients with the A allele of rs6921438 (AA+AG genotypes) compared to the homozygotes for wild type G allele (105 ± 31/mm2 vs. 65 ± 20/mm2, respectively; p < 0.001). On the contrary, there was no significant difference in numerical areal density of VEGFR-2-positive cells between rs10738760 genotypes (patients with AA+AG genotypes vs. patients with GG genotype: 75 ± 25/mm2 vs. 67 ± 28/mm2; p < 0.001).

FIGURE 1: VEGF receptor 2-positive cells in fibrovascular membranes from patients with PDR (anti-VEGF receptor 2 monoclonal antibody, ×63 magnification). PDR: Proliferative diabetic retinopathy; VEGF: Vascular endothelial growth factor.

DISCUSSION

Previously, it was suggested that rs10738760 and rs6921438 polymorphisms affect the levels of VEGF in serum, i.e., in healthy individuals, these SNPs contributed to nearly half of the variability in circulating VEGF levels [27]. As a potent angiogenic and vascular permeability factor, VEGF has been implicated in the development of diabetic microvascular complications. In the present study, we investigated the association between rs10738760 and rs6921438 polymorphisms and PDR in Slovenian patients with T2DM. The logistic regression analysis showed a tendency of rs6921438 GA genotype for an increased risk of PDR (p = 0.05). In addition, the A allele of rs6921438 was found to affect the levels of VEGFR-2 in FVMs of diabetic patients with PDR. On the contrary, we did not observe an association between rs10738760 variants and PDR in our group of patients with T2DM.

Bonnefond et al. [31] investigated the contribution of rs6921438 and rs10738760 polymorphisms to T2DM and microvascular complications (DR and DN) in several case-control studies including French and Danish populations. Although in the French population the G allele of rs6921438 (increases circulating VEGF levels in the general population) was associated with increased T2DM risk, this was not confirmed in the Danish population. Also, rs10738760 SNP was not associated with T2DM in neither of those populations. Similarly, they did not observe significant effects of rs6921438 or rs10738760 on diabetic microvascular complications in patients with T2DM [31]. In line with these and our findings are also the results of another study on Slovenian patients with T2DM [32], which reported no significant association between DR and rs6921438 and rs10738760 polymorphisms. Overall, the results from our and the previous studies indicate a more complex and perhaps indirect relationship between VEGF and T2DM and its microvascular complications, as suggested by Bonnefond et al. [31]. In addition to genetic factors including polymorphisms in the VEGF and other related genes [36-39], environmental factors such as diabetes duration, poor glycemic control, and hypertension also affect the development of PDR.

VEGF is considered as a possible molecular target/biomarker for PDR since the heritability of circulating VEGF levels is very high (60–80% heritability) [27] and markedly higher VEGF levels have been observed in diabetic patients with PDR [12,37]. In addition, inhibition of VEGF-mediated retinal neovascularization and retinal vascular permeability using anti-VEGF agents or inhibitory VEGF isoforms has a potential as a therapeutic strategy in diabetic retinopathy [40-43].

Finally, our results suggested that the expression of VEGFR-2 is affected by the presence of VEGF-related polymorphisms, since the numerical areal density of VEGFR-2-positive cells was significantly higher in FVMs from PDR patients carrying the A allele of rs6921438 (AA+AG genotypes) compared to PDR patients with the GG genotype (wild type).

There are several limitations to our study. First, we enrolled a relatively small number of patients and a larger sample size may be necessary to demonstrate significant association. Second, we only assessed the association between two VEGF-related polymorphisms and PDR. Therefore, we cannot exclude the effect of other VEGF or VEGF-related polymorphisms on the development of PDR. Third, we did not evaluate the levels of VEGF in the vitreous fluid or serum of diabetic patients with PDR, and it would be informative to see if rs6921438 and/or rs10738760 polymorphism affect VEGF expression in PDR. In the general population, the minor A allele of rs6921438 was associated with decreased and the G allele with increased levels of serum VEGF [27].

Overall, our results suggest that VEGF-related rs10738760 and rs6921438 polymorphisms do not contribute to genetic susceptibility to PDR, despite the effect of rs6921438 on VEGFR-2 expression. Due to the aforementioned limitations, further studies with a larger sample size should be carried out to clarify the role of VEGF-related SNPs in PDR.

Acknowledgements

ACKNOWLEDGMENTS

The authors thank Mrs. Polona Sajovic and Mrs. Petra Nussdorfer for the preparation of histological slides, and Mrs. Visam Bajt, BA for revising the English.

DECLARATION OF INTERESTS

The authors declare no conflict of interests.

REFERENCES

  1. , , (). Ocular complications of diabetes mellitus. World J Diabetes. http://doi.org/10.4239/wjd.v6.i1.92
  2. (). Diabetic retinopathy:Treatment and prevention. Diab Vasc Dis Res. https://doi.org/10.3132/dvdr.2007.051
  3. , , , , , (). Diabetic retinopathy in a multi-ethnic cohort in the United States. Am J Ophthalmol. http://doi.org/10.1016/j.ajo.2005.08.063
  4. , , (). Important causes of visual impairment in the world today. JAMA. http://dx.doi.org/10.1001/jama.290.15.2057
  5. , , , (). Diabetic retinopathy and angiogenesis. Curr Diabetes Rev. https://doi.org/10.2174/157339909787314149
  6. , , (). Epidemiology of proliferative diabetic retinopathy. Diabetes care. https://doi.org/10.2337/diacare.15.12.1875
  7. (). The natural history of active retinopathy of prematurity. Ophthalmology. https://doi.org/10.1016/S0161-6420(84)34120-3
  8. , (). Growth factors in proliferative diabetic retinopathy. Exp Diabesity Res. http://doi.org/10.1155/EDR.2003.287
  9. , , (). Diabetic retinopathy. N Engl J Med. https://doi.org/10.1056/NEJMra1005073
  10. , , (). Diabetic retinopathy. Lancet. https://doi.org/10.1016/S0140-6736(09)62124-3
  11. , , (). Biochemical and molecular mechanisms of diabetic retinopathy. Curr Sci. https://www.jstor.org/stable/24108174
  12. , , , , , (). Vascular endothelial growth factor in eye disease. Prog Retin Eye Res. http://doi.org/10.1016/j.preteyeres.2008.05.001
  13. , , (). The role of genetics in susceptibility to diabetic retinopathy. Int Ophthalmol Clin. http://doi.org/10.1097/IIO.0b013e31819fd5d7
  14. , (). The biology of vascular endothelial growth factor. Endocr Rev. https://doi.org/10.1210/edrv.18.1.0287
  15. , (). Vascular endothelial growth factor and diabetes:The agonist versus antagonist paradox. Diabetes. https://doi.org/10.2337/diabetes.48.10.1899
  16. , (). The vascular endothelial growth factor (VEGF) family:Angiogenic factors in health and disease. Genome Biol. http://doi.org/10.1186/gb-2005-6-2-209
  17. , , , , (). Epiretinal membrane inflammatory cell density might reflect the activity of proliferative diabetic retinopathy. Invest Ophthalmol Vis Sci. http://dx.doi.org/10.1167/iovs.13-13634
  18. , , , , , (). Detection of vascular endothelial growth factor messenger RNA and vascular endothelial growth factor-like activity in proliferative diabetic retinopathy. Arch Ophthalmol. http://dx.doi.org/10.1001/archopht.1994.01090230090028
  19. , , , , , (). Increased vascular endothelial growth factor levels in the vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol. https://doi.org/10.1016/S0002-9394(14)75794-0
  20. , , , , , (). VEGF-initiated blood-retinal barrier breakdown in early diabetes. Invest Ophthalmol Vis Sci.
  21. , , , , , (). The role of VEGF and KDR polymorphisms in moyamoya disease and collateral revascularization. PLoS One. https://doi.org/10.1371/journal.pone.0047158
  22. , , , (). Assignment of the vascular endothelial growth factor gene to human chromosome 6p21.3. Circulation. https://doi.org/10.1161/01.CIR.93.8.1493
  23. , , , , (). Polymorphisms of the vascular endothelial growth factor and susceptibility to diabetic microvascular complications in patients with type 1 diabetes mellitus. J Diabetes Complications. https://doi.org/10.1016/S1056-8727(02)00181-2
  24. , , , , , (). Association of vascular endothelial growth factor polymorphisms with nonproliferative and proliferative diabetic retinopathy. J Clin Endocrinol Metab. https://doi.org/10.1210/jc.2009-2719
  25. , , , , , (). Local and genetic determinants of vascular endothelial growth factor expression in advanced proliferative diabetic retinopathy. Mol Vis.
  26. , , , (). Update on genetics and diabetic retinopathy. Clin Ophthalmol. http://doi.org/10.2147/OPTH.S94508
  27. , , , , , (). Identification of cis- and trans-acting genetic variants explaining up to half the variation in circulating vascular endothelial growth factor levels. Circ Res. http://doi.org/10.1161/CIRCRESAHA.111.243790
  28. , , , , , (). Angiogenesis related genes NOS3, CD14, MMP3, and IL4R are associated to VEGF gene expression and circulating levels in healthy adults. BMC Med Genet. http://doi.org/10.1186/s12881-015-0234-6
  29. , , , , , (). A common variant highly associated with plasma VEGFA levels also contributes to the variation of both LDL-C and HDL-C. J Lipid Res. http://doi.org/10.1194/jlr.P030551
  30. , , , , , (). Association of a vascular endothelial growth factor genetic variant with serum VEGF level in subjects with metabolic syndrome. Gene. https://doi.org/10.1016/j.gene.2016.10.034
  31. , , , , , (). What is the contribution of two genetic variants regulating VEGF levels to type 2 diabetes risk and to microvascular complications?. PLoS One. http://doi.org/10.1371/journal.pone.0055921
  32. , , , , (). Vascular endothelial growth factor (VEGF)-related single nucleotide polymorphisms rs1073∸and rs6921438 are not associated with diabetic retinopathy (DR) in Slovenian patients with type 2 diabetes mellitus (T2DM). Bosn J Basic Med Sci. https://doi.org/10.17305/bjbms.2017.2068
  33. (). Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. https://doi.org/10.2337/diacare.26.2007.S5
  34. (). Grading diabetic retinopathy from stereoscopic color fundus photographs –An extension of the modified Airlie House classification. ETDRS report number 10. Ophthalmology. https://doi.org/10.1016/S0161-6420(13)38012-9
  35. , (). Testing for Hardy-Weinberg equilibrium in small samples. Biometrics. https://doi.org/10.2307/2529370
  36. , , , , , (). A common polymorphism in the 5'- untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes. https://doi.org/10.2337/diabetes.51.5.1635
  37. , , , , (). A hemochromatosis-causing mutation C282Y is a risk factor for proliferative diabetic retinopathy in Caucasians with type 2 diabetes. J Hum Genet. http://dx.doi.org/10.1007/s10038-003-0094-3
  38. , , (). Biomarkers in diabetic retinopathy and the therapeutic implications. Mediators Inflamm. https://doi.org/10.1155/2013/193604
  39. , , , , (). Pathophysiology of diabetic retinopathy. ISRN Ophthalmol. http://doi.org/10.1155/2013/343560
  40. , , , (). Neurotrophic effects of vascular endothelial growth factor on organotypic cortical explants and primary cortical neurons. J Neurosci. https://doi.org/10.1523/JNEUROSCI.23-35-11036.2003
  41. , (). New roles for VEGF in nervous tissue –Beyond blood vessels. Exp Neurol. https://doi.org/10.1016/j.expneurol.2004.01.022
  42. , , , , (). VEGF165b, an endogenous C-terminal splice variant of VEGF, inhibits retinal neovascularization in mice. Mol Vis.
  43. , , , , , (). Suppression of retinal neovascularization in vivo by inhibition of vascular endothelial growth factor (VEGF) using soluble VEGF-receptor chimeric proteins. Proc Natl Acad Sci USA. https://doi.org/10.1073/pnas.92.23.10457