Association of candidate genes (ALR2, RAGE, and VEGF) polymorphisms with diabetic retinopathy in type 2 diabetic patients of Khyber Pakhtunkhwa, Pakistan

To evaluate the aldose reductase (ALR2, rs759853), receptor for advanced glycation end products (RAGE, rs2070600), and vascular endothelial growth factor (VEGF, rs833061) association with diabetic retinopathy in type 2 diabetic patients of Khyber Pakhtunkhwa population. A case-control study was conducted on a total of 550 subjects consisting of 186 with diabetic retinopathy (DR) having type 2 diabetes, 180 had type 2 diabetes (T2DM), and 184 healthy controls (HC). All the samples were subjected to DNA isolation using salting-out method followed by SNP genotyping through Tetra-ARMS PCR. Chi square and Exact Fischer tests were used for alleles and genotypes distribution. Odd ratio and confidence interval values were found out by online software Medcalc Odd ratio Calculator. Multiple parameters such as random blood sugar (RBS) (p < 0.001), fasting blood sugar (FBS) (p < 0.001), HbA1c (p < 0.001), total cholesterol (p < 0.001), LDL (p < 0.001), HDL (p < 0.001), BMI (p < 0.001) and hypertension (p = 0.018) exhibited strong association with DR as compared to DM and HC. Our results displayed that the VEGF-rs833061 and RAGE- rs2070600 exhibited significant association (p < 0.05) with an increased DR risk, when compared with T2DM. In contrast, ALR2 didn’t display association with DR (p > 0.05) when compared with T2DM, but showed association (p < 0.05) when compared with HC. Statistically significant association was observed in VEGF-rs833061 and RAGE-rs2070600 with DR in type 2 diabetic patients. While, ALR2- rs759853 didn’t exhibit significant association with DR. This is the first study to report the association of candidate genes (ALR2, VEGF and RAGE) with DR in type 2 diabetes of Khyber Pakhtunkhwa population. More similar research studies are recommended with larger data sets in other ethnicities both national and international.


Introduction
Diabetic retinopathy (DR), a complication of diabetes, is a chronic condition related to hyperglycemia. DR is a progressive disorder of the retina involving pathological changes of capillaries resulting in one or more abnormal features recognized by a trained observer during diagnosis [1]. The probability of developing and progressing of DR is associated with the period of diabetes. Within the first 10 years after being diagnosed with diabetes, all patients of type 1 and more than 60% of type 2 diabetes are expected to have certain kind of retinopathy [2].
DR is categorized into non-proliferative diabetic retinopathy (NPDR) pre-proliferative diabetic retinopathy (PPDR), and proliferative diabetic retinopathy (PDR). The distinctive features of NPDR involve microaneurysm, cotton wool spots, hard exudates, venous beading, and intraretinal microvascular abnormalities. This condition either remains steady or continuously develops into PDR, which can be identified by neovascularization, pre-retinal hemorrhage, and vitreous hemorrhage [3]. In patients having DR, diabetic macular edema (DME) is the most common reason for vision loss that is identified by macula thickening or swelling due to sub-and intra-retinal build-up of fluid in the macula activated by the breakdown of the retinal blood barrier (BRB) [4]. For screening of DR, several different diagnosis methods are used, in particular slit lamp bio-microscopy, direct or indirect ophthalmoscopy, mydriatic (dilated) or non-mydriatic (nondilated) digital fundus photography [5].
In 2010, it was roughly calculated that 93 million people suffering from DR, and 28 million were affected by vision-threatening diabetic retinopathy (VTDR) globally. Diabetic retinopathy may become the prime cause of vision blindness worldwide. According to the Wisconsin Epidemiology Study of Diabetic Retinopathy (WESDR), about 3.6% of diabetes type 1 and 1.6% of type 2 diabetes individuals are legally blind. The rate of developing DR is expected to increase to 191.0 million by 2030 due to the alarming occurrence rate of diabetes mellitus [6].
Diabetic retinopathy is a complicated disorder that is strongly influenced by both genetics and environmental factors. A few biochemical pathways have been suggested to link hyperglycemia and microvascular complications. They are; polyol accumulation, oxidative stress, advanced glycation end products (AGEs), and protein kinase C (PKC). These pathways are thought to regulate the disease process by impacting cellular metabolism, growth factors, and signaling). It has been reported that, in various populations, single nucleotide polymorphism (SNP) in genes has been associated with risk for DR. Candidate genes that are widely studied for variations and association with DR in the different population are aldose reductase (ALR2), methylenetetrahydrofolate reductase (MTHFR), endothelial nitric oxide synthase (eNOS), the receptor for advanced glycation end products (RAGE), and vascular endothelial growth factor (VEGF) [7]. Studies reported that ALR2 is the first enzyme of the polyol pathway, which is also a rate-limiting enzyme of this significant metabolic pathway that's links hyperglycemia with diabetes complications. In this pathway, ALR2 converts glucose into sorbitol. In the presence of high blood glucose, sorbitol accumulates within the cell, which leads to osmotic stress [8].
Another study reported that the activation of the RAGE gene leads to cytokine secretion. Cytokines increase diabetic complications by increasing endothelial permeability and proinflammatory process [9]. Another observation that VEGF is one of the essential factors in angiogenesis and affects the vascular permeability of endothelial cells. Activation of VEGF occurs by vascular changes induced by hypoxia during diabetes mellitus and by high blood glucose levels. VEGF activation destroys the blood-retinal barrier (BRB), diabetic macular edema development, and neovascularization occur typically for PDR [10].
However, there is no such work is done so far in Khyber Pakhtunkhwa province of Pakistan, to see the association of genetic predisposition, particularly ALR2, RAGE and VEGF in DR. In this study, we have focused on finding the genetic association of candidate genes like ALR2 (rs759853), RAGE (rs2070600), and VEGF (rs833061) with DR in diabetic type 2 patients. Finding the association of the genes mentioned above in the development and pathological condition of DR in the local population will be highly significant.

Data collection, inclusion exclusion criteria, and ethical approval
This case-control study was conducted at Institute of Biotechnology and Genetic Engineering (IBGE; Health Division), The University of Agriculture, Peshawar, Pakistan. Ethical approval was taken from IBGE (IBGE Ethical Committee), and written consent was taken from all the participants while explaining the aim and objectives of the study before data collection. Patients who have a history of T2DM and or T2DM-Retinopthy were included in this study and those who have T1DM or other complications like nephropathy, cardiovascular complications and other infections (HBV, HCV) were excluded. Based on the inclusion and exclusion criteria, a total of 550 participants (186 patients of DR with confirmed T2DM, 180 type 2 diabetic patients (T2DM), and 184 healthy controls (HC)) were recruited from Lady Reading Hospital Peshawar, during 2019-2020. Retinopathy patients were examined thoroughly by an ophthalmologist with a Fundus examination and Optical Coherence Tomography (OCT) for retinopathy diagnosis. Detailed clinical history and relevant examinations, including fasting and random blood glucose, HbA1c, serum creatinine, cholesterol, triglycerides, HDL, LDL, and BMI, were obtained for each patient. Under the standard biosafety protocol and ethical consideration, 4mL of blood sample was taken via venipuncture method in an EDTA tube and stored at − 20 °C at the IBGE for DNA isolation and PCR.

DNA isolation, SNPs genotyping, and gel electrophoresis
All the samples were subjected to DNA extraction by salting out (non-enzymatic) method, and SNP genotyping was done through T-ARMS PCR, previously adopted in our lab [11,12]. Specific primers were designed through primer blast software for each SNP ( Table 1). The PCR mixture of 10 µL was prepared to consist of 5 µL master mix (Thermo scientific DreamTaq Master Mix (2X), 0.5 µL of each forward and reverse primer, 1 µL of template DNA, and 3 µL of ddH2O. The PCR conditions were; Initial denaturation 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing for 30 s at 59 °C, 58 °C, and 55.9 °C for VEGF, ALR2 and RAGE respectively, extension 72 °C for 30 s and final extension at 72 °C for 5 min. The amplified PCR products were run and confirmed with 1KB ladder (Thermo Fisher Scientific) on 1.25% agarose gel.

Statistical analysis
Statistical analysis was carried out using SPSS software version 2.0. The descriptive statistical method was used for demographics data with p < 0.05 was considered significant. For categorical data, the Chi-square test was used, while for quantitative data, a t-test was performed. The allelic and genotypic distribution was done using the chi-square and Fischer exact test. The genotypes were checked for HWE through chi-square and the exact test. Comparison of genotypes was made by obtaining Odd ratios and 95% CI value using online software named Medcalc Odd ratio Calculator. The p > 0.05 were considered nonsignificant.

General characteristics of the study subjects in various groups
A total of 550 subjects, including 186 of DR, 180 of type 2 diabetes millets (T2DM), and 184 HC, were recruited in this study. Comparison of the DR group with T2DM and HC shows a significantly higher level of RBS (p < 0.001), FBS (p < 0.001), HbA1c (p < 0.001), Total cholesterol (p < 0.001), triglycerides (0.029), LDL p < 0.001), HDL (p < 0.001), SBP (0.004), DBP (0.032), BMI (p < 0.001) ( Table 2). There was statistically no significant difference among groups in terms of gender, duration of diabetes, serum creatinine. While a statistically significant difference was observed between the age of diabetes onset (p < 0.001), age of the patients (0.043) and hypertension (0.018) among the three groups.

Genotypic and allelic frequencies confirmation
The genotypic and allelic distribution of VEGF, ALR2, and RAGE genes was conducted using TARMS-PCR. A sample gel picture shows the PCR product size confirmation run on 1.25% agarose gel (Fig. 1).

Frequencies of ALR2, VEGF and RAGE polymorphism among the DR-T2DM, and DM-HC groups
All the participants were confirmed for ALR2, VEGF and RAGE polymorphism using Tatra-ARMS-PCR protocol. Our results displayed that VEGF-rs833061 risk allele C and corresponding homozygous genotype was associated with an increased DR risk (p < 0.05), when compared with T2DM (Table 3). But VEGF risk allele and corresponding genotypes didn't show association with DR (p > 0.05) when compared with HC. In contrast, ALR2 risk allele T and associated genotypes TC and TT didn't display association with DR (p > 0.05) when compared with T2DM, but showed association (p < 0.05) when compared with HC. Interestingly, RAGE risk allele and corresponding homozygous mutant genotype displayed strong association with DR (p < 0.05) both in T2DM and HC groups (Table 3).

Discussion
The most severe microvascular complication of diabetes is DR that can be vision alarming if left unchecked. DR ranks as a common cause for impaired vision globally, mainly in grown-ups. With the worldwide prevalence of diabetes being estimated to rise to 438 million people by 2030, DR will undoubtedly pose the primary health concern in public. The exploration of DR genes has mainly been carried out using the candidate gene method. The case-control design is primarily employed and is suitable for identifying both significant and minor genes. The candidate gene method needs an adequate knowledge of the pathogenic mechanisms underlying DR. This has been acquired from many years of research in this field [13]. Several groups have confirmed the association of VEGF SNPs (− 460T/C, + 405G/C, and + 936 C/T) with increased VEGF expression, and a significant increase activity is observed in the VEGF promoter region in the presence of the + 405/− 460 polymorphisms [14]. Still, because of ethnic and regional differences and other differences, the results of previous studies have not been reliable. To find out about the VEGF (− 460T/C) association with DR risk, we carried out a case-control study of 550 subjects, containing 186 of DR, 180 of T2DM, and 184 HC. We found out that the polymorphism of the VEGF gene was significantly associated with DR risk. However, in genotypic and allelic frequencies, no significant difference was found. Previous studies showed that the C allele had greater promoter activity, about 71%, than the T allele, which might cause the VEGF gene's overexpression. Our findings were consistent with this observation. Previous studies also reported [13] a significant association of the VEGF gene (p = 0.0043) in PDR patients. Another study by Yang et al. [15] demonstrated that carrying VEGF−460T/C (rs833061) homozygous genotype was significantly associated with DR. Contrary to this observation, Szaflik et al. [16] where they performed a casecontrol association study on polish patients. No association was reported for VEGF−460T/C polymorphism with DR in the polish population. The reason for conflicting results was mainly of ethnic origin as they were Caucasians. Advanced glycation end products (AGEs) play an essential role in the development of DR, which is formed due to chronic exposure of the retina to hyperglycemia. Upon exposure of AGEs to sugars, they become non-enzymatically glycated. However, AGEs are proteins or lipids in nature. In the diabetic vasculature, production and accumulation of AGEs occur due to increased oxidative stress, prolonged hyperglycemia, and dyslipidemia in diabetes [17].
RAGE, the most important, being the cellular receptor for AGEs, partially mediate the effects of AGEs. It is ubiquitously expressed in various retinal cells. RAGE belongs to the immunoglobulin superfamily of cell surface molecules. Increased expression of RAGE in the retina associated with the AGE accumulation was observed in the animal models and humans with diabetes. In diabetic patients, up-regulation of RAGE in the retinas leads to proinflammatory and prooxidant signaling pathways [18]. Multiple intracellular signaling pathways are activated due to AGE-RAGE interaction, consequently induces the inflammatory and oxidative stress responses in vascular cells cause the transcription of target genes and oxidative stress, therefore playing an essential role in the DR progression and development [19].
In the present study, we investigated the association of RAGE-rs2070600 with DR. Results revealed a significantly increased association of RAGE-rs2070600 with DR risk; a significant difference in the genotypic and allelic frequencies was found. The AA genotype was a risk factor for DR. This suggests that the rs2070600 RAGE gene might contribute to the pathogenesis of DR. Our data was similar to the study done by Yang et al. [20]. Their results showed a significant association of rs2070600 with DR risk; a significant difference was observed in the frequency of the homozygous genotype. Similar results were reported in another study by Vanita [21] where the case-control association was done on North Indians patients. Results showed a significant association of rs2070600 with DR. A significant difference was found in the genotypic frequencies while no significant difference was found in the allelic frequencies. However, contrast results were found in the study by Ng et al. [22]. Results show no significant difference in the allelic and genotypic frequencies of RAGE-rs2070600 in Malaysian patients. This difference in results might be attributed to ethnic and regional differences.
ALR2 gene is present on location 7q35.12. In the ALR2 gene, a common polymorphism in the promoter region at nucleotide C/T at 106 positions is reported to associate with the risk of DR. Up to the present time, numerous studies tried to find the association of ALR2 C-106T polymorphism with the risk of DR. Still, the contribution of ALR2 polymorphism to DR is debated [23] as some studies support the finding while others do not. The distinct binding affinity of the gene to the transcription factor occurs due to the 2 different alleles of the polymorphism [24]. In reality, the comparison of the T allele with the C allele, the C allele was shown to have twice as high mRNA expression level [25]. This means that those with the C allele may have a higher level of aldose reductase resulting in the inefficient conversion of glucose into sorbitol. Accumulation of sorbitol in the retina is known to cause osmotic stress and, in the end, retinopathy. This is the reason why a few studies had demonstrated that the C allele was associated with an increased risk of DR [26], which is in contrast to our finding where no significant association with the DR risk was found, also higher frequency of TT genotype was found in DR group as compare to other groups. The reason and mechanism underlying our observation are not understood. However, our findings were supported by Santos et al. [27], where a case-control association study was done on euro-Brazilian type 2 diabetic patients to find an association of ALR2 C-106T polymorphism with DR. No significant difference was observed in the allelic and genotypic frequencies. Thus, no association was found. Another study by Deng et al. [28] reported no association of ALR2 C-106T with DR. However, Li et al. [29] studies showed a significantly increased association of ALR2 C-106T with DR risk. The contradiction in the results with our findings can be attributed to the ethnic and regional difference, small sample size, different sample recruitment methods, and the mechanism underlying the development and progression of DR.

Conclusion
Our data suggested a statistically significant association of VEGF-rs833061 and RAGE-rs2070600 with DR in type 2 diabetic patients. At the same time, no significant association of ALR2-rs759853 was found with DR in T2DM. Also, this is the first study to report an association of candidate genes (RAGE, VEGF, ALR2) with DR in type 2 diabetes in our region. More investigation will be required to confirm the results with large data sets, including different ethnicities both national and international.
Our study has few limitations; first, the sample size was small in the current study; a large sample size should be taken for future studies for solid clinical impact. Second, we did not perform a functional analysis to understand the mechanism behind the genetic association. Third, more indepth inspections that examine the pathogenesis of DR at the genetic level are required.