Polymorphisms in TIE2 and ANGPT-1 genes are associated with protection against diabetic retinopathy in a Brazilian population

ABSTRACT

Objective:

The objective of this study was to investigate the association between SNPs in the TIE2 and ANGPT-1 genes and diabetic retinopathy (DR).

Subjects and methods:

This study comprised 603 patients with type 2 diabetes mellitus (T2DM) and DR (cases) and 388 patients with T2DM for more than 10 years and without DR (controls). The TIE2 rs639225 (A/G) and rs638203 (A/G) SNPs and the ANGPT-1 rs4324901 (G/T) and rs2507800 (T/A) SNPs were genotyped by real-time PCR using TaqMan MGB probes.

Results:

The G/G genotype of the rs639225/TIE2, the G/G genotype of the rs638203/TIE2 and the T allele of the rs4324901/ANGPT-1 SNPs were associated with protection against DR after adjustment for age, glycated hemoglobin, gender, and presence of hypertension (P = 0.042, P = 0.003, and P = 0.028, respectively). No association was found between the rs2507800/ANGPT-1 SNP and DR.

Conclusion:

We demonstrated, for the first time, the association of TIE2 rs638203 and rsrs939225 SNPs and ANGPT-1 rs4324901 SNP with protection against DR in a Brazilian population.

Keywords
ANGPT-1 gene; TIE2 gene; polymorphism; type 2 diabetes mellitus; diabetic retinopathy

INTRODUCTION

Diabetic retinopathy (DR) is one of the commonest complications of diabetes mellitus (DM) and is characterized by damage to small blood vessels of the retina (¹). DR is a neurovascular disorder and the leading cause of legal blindness in working-age adults, affecting over 93 million people worldwide (², ³). Clinically, DR is classified into two main stages: nonproliferative diabetic retinopathy (NPDR) and proliferative diabetic retinopathy (PDR). NPDR is the first stage of DR, characterized by increased vascular permeability, capillary occlusion, microaneurysms, bleeding, and hard exudates (⁴). In contrast, PDR is a progressive stage of DR characterized by neovascularization (⁴).

Vascular growth and angiogenesis are involved in the pathogenesis of this diabetic complication. Angiopoietins (ANGPTs) are key regulators of these processes. ANGPT-1 is involved in vascular maturation, endothelial cell (EC) survival, EC interactions with other cells, and vascular permeability (⁵). In contrast, ANGPT-2 is an ANGPT-1 antagonist and inhibits endothelial quiescence, neutralizing the vascular maintenance activities of ANGPT-1 in cases where endothelial remodeling is necessary, such as during inflammation and angiogenesis (⁶, ⁷).

ANGPT-1 and ANGPT-2 signaling occur through transmembrane tyrosine kinase receptor (TIE2) (⁶-⁸), which is expressed in both vascular endothelial tissue and non-EC, and has a vital role in vascular stability (⁹-¹³). ANGPT-1 activation by TIE2 generally leads to protective effects on cells, such as cell migration, adhesion, and survival. ANGPT-2 acts as a partial agonist of TIE2, where its high concentrations lead to a competitive inhibition of ANGPT-1 signaling through TIE2 (¹⁴).

In this context, studies have reported the involvement of ANGPT-1 and TIE2 in the pathogenesis of chronic complications of DM (¹⁵). Khalaf and cols. (¹⁶) reported that serum ANGPT-1 levels were higher in patients with NPDR compared to patients with DM but without DR, while no differences were found between the PDR group and control patients with DM. Serum TIE2 levels did not differ between patients with and without DR (¹⁶). Another study showed that ANGPT-1 and ANGPT-2 levels were increased in vitreous samples of patients with PDR compared to control patients without DM (¹⁷). Moreover, Jeansson and cols. (¹⁸) demonstrated that diabetic mice with Angpt-1 deletion and streptozotocin-induced diabetes developed more severe diabetic kidney disease (DKD) compared to diabetic mice with normal Angpt-1 expression. Taking this background into consideration, single nucleotide polymorphisms (SNPs) that could influence the expression of these genes may be associated with susceptibility to chronic complications of DM, including DR. Therefore, we investigated, for the first time, the association of rs639225 (A/G) and rs638203 (A/G) SNPs in the TIE2 gene and rs4324901 (G/T) and rs2507800 (T/A) SNPs in the ANGPT-1 gene with DR.

SUBJECTS AND METHODS

Study participants

Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) and Strengthening the Reporting of Genetic Association Studies (STREGA) guidelines were used to design this case-control study (¹⁹, ²⁰). The sample consisted of 991 unrelated patients with type 2 DM (T2DM) selected from Hospital de Clínicas de Porto Alegre and Grupo Hospitalar Conceição (Porto Alegre, Rio Grande do Sul, Brazil) between 2002 and 2013, as previously described (²¹).

T2DM was diagnosed according to American Diabetes Association criteria (²²). The diagnosis of DR was made by an experienced ophthalmologist using fundoscopy through dilated pupils. DR was classified as “absent DR” (no fundus abnormalities), ‘NPDR’ (microaneurysms, hemorrhage, and hard exudates), or “PDR” (newly formed blood vessels and/or growth of fibrous tissue into the vitreous cavity). DR classification considered the most severely affected eye, according to the scale developed by the Global Diabetic Retinopathy Group (²³). Cases were defined by the presence of DR (NPDR or PDR). Controls were defined by the absence of this complication and known DM for at least 10 years.

A standard questionnaire was used to collect information about age, age at T2DM diagnosis, T2DM duration, and pharmaceutical treatment, and all patients underwent physical and laboratory evaluations as previously described (²⁴). Serum and plasma samples were taken after 12 h of fasting for laboratory analyses (²⁴). Glycated hemoglobin (HbA1c) measurements were performed by different methods and the results were traceable to the Diabetes Control and Complications Trial (DCCT) method by off-line calibration or through conversion formulae (²⁵). Creatinine was measured by the Jaffe reaction; total plasma cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides were analyzed by enzymatic methods, and albuminuria was verified by immunoturbidimetry (Sera-Pak Immuno Microalbumin kit, Bayer, Tarrytown, NY, USA; mean intra- and inter-assay coefficients of variance of 4.5% and 11%, respectively) (²⁶).

DKD was diagnosed based on the Kidney Disease Improving Global Outcomes (KDIGO) guidelines (²⁷), using both urinary albumin excretion (UAE) levels and estimated glomerular filtration rate (eGFR); the latter was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (²⁸). Ethnic groups were defined based on self-classification.

The study protocol was approved by the Ethic Committee in Research from Hospital de Clínicas de Porto Alegre (CAAE number: 97844918.1.0000.5327), and all participants provided assent and written informed consent prior to inclusion in the study.

Genotyping

DNA was extracted from peripheral blood leucocytes using a standardized salting-out procedure (²⁹). The rs639225 (A/G) (Assay ID = C__1305224_30) and rs638203 (A/G) (Assay ID = C__8775841_10) SNPs in the TIE2 gene, as well as the rs4324901 (G/T) (Assay ID = C__26472342_10) and rs2507800 (T/A) (Assay ID = C__1252396_10) SNPs in the ANGPT-1 gene were genotyped using specific Human TaqMan SNP Genotyping Assays 40x (Thermo Fisher Scientific, Foster City, CA, USA). Real-time PCR reactions were performed in 384-well plates, in a volume of 5 µL, using 2 ng of DNA, TaqPath ProAmp 1 X Mastermix (Thermo Fischer Scientific), and TaqMan SNP Genotyping Assay 1 X. Plates were placed in a real-time PCR thermal cycler (ViiA7 Real-Time PCR System; Thermo Fisher Scientific) and heated for 10 min at 95 °C, followed by 50 cycles of 95 °C for 15 s and 62 °C for 1 min.

Statistical analyses

Allele frequencies were determined by gene counting, and departures from Hardy–Weinberg equilibrium (HWE) were verified using the x² test. Allele and genotype frequencies were compared between groups using x² tests. Genotypes were also compared between groups under additive, recessive, and dominant inheritance models, categorized as suggested by a previous publication (³⁰). We also examined the widely used measures of linkage disequilibrium (LD), Lewontin's D’ (|D’|), and r² between all pairs of biallelic loci (³¹). Haplotypes constructed with the combination of the two TIE2 and two ANGPT-1 SNPs and their frequencies were inferred using PHASE 2.1 software, which implements a Bayesian statistical method (³²).

Clinical and laboratory characteristics were compared between groups of patients categorized according to the different SNP genotypes using an unpaired Student's t-test or a x² test, as appropriate. Normal distribution of quantitative variables was checked using Kolmogorov-Smirnov and Shapiro–Wilk tests. Variables with normal distribution are shown as means ± standard deviation (SD). Variables with skewed distribution were log-transformed before the analyses and are shown as medians (25th-75th percentile values). Categorical variables are shown as n (%). Multivariate logistic regression analyses were done to evaluate the independent association of SNPs with DR, adjusting for possible confounding factors. Variables with significant associations with DR in the univariate analysis or with a biologically relevant association with this complication were chosen for inclusion in the multivariate model. DM duration was not included as an independent variable in these analyses since the DM control group was selected based on this characteristic. Statistical analyses were performed using SPSS 18.0 software (SPSS, Chicago, IL), and P values < 0.05 were considered significant.

Sample sizes were calculated using the OpenEpi website (www.openepi.com). Since no previous study investigated the association of the SNPs of interest in TIE2 and ANGPT-1 genes with DR, data from studies that evaluated the association of these SNPs with other diseases were used (mean minor allele frequency = 0.30 and odds ratio [OR] = 0.4 or 1.4) (33-35). Therefore, the calculated sample size was estimated at 403 controls and 504 cases to find an OR = 1.4, with 80% power; or 113 controls and 142 cases to find an OR = 0.4, with 80% power.

RESULTS

Sample description

Table 1 describes the clinical and laboratorial characteristics of patients with T2DM for more than 10 years without DR (controls) and patients with T2DM and DR (cases). As expected, mean HbA1c and UAE levels, as well as the prevalence of arterial hypertension and DKD, were higher in patients with T2DM and DR compared to controls (all P values < 0.050). Moreover, the frequency of males and non-white subjects and LDL cholesterol levels were increased in the case group compared to controls (all P values < 0.050). Mean age, BMI, and eGFR levels were lower in cases compared to controls (all P values < 0.050).

Table 1
Clinical and laboratory characteristics of patients with DM (controls) and patients with DM and DR (cases)

Characteristics	Patients with T2DM		P *
	Without DR n = 388	With DR n = 603
Age (years)	69.4 ± 10.5	65.8 ± 10.5	0.0001
Gender (% male)	37.4	51.6	0.0001
BMI (kg/m²)	29.0 ± 5.1	28.2 ± 5.2	0.017
Ethnicity (% non-white)	18.4	25.2	0.017
HbA1c (%)	7.4 ± 1.8	7.8 ± 2.1	0.005
Hypertension (%)	82.3	91.2	0.0001
Duration of diabetes (years)	20.6 ± 8.8	20.6 ± 9.6	0.935
Cholesterol total (mg/dL)	194.9 ± 49.6	198.3 ± 51.3	0.340
Triglycerides (mg/dL)	148.0 (103.5-209.0)	149.0 (102.0-215.0)	0.818
HDL cholesterol (mg/dL)	46.5 ± 12.6	45.3 ± 14.0	0.163
LDL cholesterol (mg/dL)	112.7 ± 43.5	119.2 ± 46.4	0.045
eGFR (mL/min per 1.73 m2)	74.0 (51.0-90.0)	53.0 (16.0-76.0)	0.0001
UAE (mg/g)	8.3 (3.1-36.6)	53.1 (8.6-369.6)	0.0001
Diabetic kidney disease (%)	48.4	75.7	0.0001

• Variables are shown as means ± standard deviations (SD), median (25th-75th percentiles), or %.
• ^*
  P value was computed using Student's t-tests or x² tests, as appropriate. T2DM: type 2 diabetes mellitus; BMI: body mass index; DR: diabetic retinopathy; HDL: high-density lipoprotein; LDL: low-density lipoprotein; eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; T2DM: type 2 diabetes mellitus; UAE: urinary albumin excretion.

Association between SNPs in TIE2 and ANGPT-1 genes and DR

Table 2 shows genotype and allele frequencies of rs639225 and rs638203 in TIE2 gene and of rs4324901 and rs2507800 in ANGPT-1 gene between patients with T2DM without DR (controls) and patients with T2DM and DR (cases). The allele and genotype frequencies of the TIE2 rs639225 SNP did not differ significantly between groups (P = 0.107 and P = 0.211, respectively). However, after adjusting for age, HbA1c, gender, and presence of hypertension, the rs639225 G/G genotype was associated with protection against DR (OR = 0.642, 95% CI 0.419-0.984; P = 0.042). This association was also found when considering the additive inheritance model (OR = 0.643, 95% CI 0.419-0.986; P = 0.043).

Table 2
Genotype and allele frequencies of SNPs in the TIE2 and ANGPT-1 genes in patients with T2DM, with and without DR

	Patients with DM (controls)	Patients with DR (cases)	Unadjusted P*	Adjusted OR (95% CI) /† P
rs639225 – TIE2	368	547
Genotype
A/A	119 (32.3)	193 (35.3)		1
A/G	157 (42.7)	244 (44.6)		0.886 (0.622-1.264)/0.505
G/G	92 (25.0)	110 (20.1)	0.211	0.642 (0.419-0.984)/0.042
Allele
A	0.54	0.58
G	0.46	0.42	0.107
Recessive model
A/A + A/G	276 (75.0)	437 (79.9)		1
G/G	92 (25.0)	110 (20.1)	0.095	0.687 (0.471-1.001)/0.051
Additive model
A/A	119 (56.4)	193 (63.7)		1
G/G	92 (43.6)	110 (36.3)	0.115	0.643 (0.419-0.986)/0.043
Dominant model
A/A	119 (32.3)	193 (35.3)		1
A/G + G/G	249 (67.7)	354 (64.7)	0.395	0.798 (0.575-1.108)/0.178
rs638203 – TIE2	368	562
Genotype
A/A	118 (32.1)	197 (35.1)		1
A/G	156 (42.4)	260 (46.3)		0.921 (0.645-1.313)/0.648
G/G	94 (25.5)	105 (18.7)	0.046	0.523 (0.340-0.804)/0.003
Allele
A	0.53	0.58
G	0.47	0.42	0.040
Recessive model
A/A + A/G	274 (74.5)	457 (81.3)		1
G/G	94 (25.5)	105 (18.7)	0.016	0.548 (0.376-0.800)/0.002
Additive model
A/A	118 (55.7)	197 (65.2)		1
G/G	94 (44.3)	105 (34.8)	0.036	0.527 (0.343-0.812)/0.004
Dominant model
A/A	118 (32.1)	197 (35.1)		1
A/G + G/G	250 (67.9)	365 (64.9)	0.384	0.770 (0.554-1.070)/0.119
rs4324901 – ANGPT-1	354	532
Genotype
G/G	152 (42.9)	271 (50.9)		1
G/T	149 (42.1)	194 (36.5)		0.724 (0.514-1.019)/0.064
T/T	53 (15.0)	67 (12.6)	0.064	0.640 (0396-1.034)/0.068
Allele
G	0.64	0.69
T	0.36	0.31	0.026
Recessive model
G/G + G/T	301 (85.0)	465 (87.4)		1
T/T	53(15.0)	67(12.6)	0.361	0.742 (0.473-1.166)/0.196
Additive model
G/G	152 (74.1)	271 (80.2)		1
T/T	53 (25.9)	67 (19.8)	0.125	0.651 (0.402-1.053)/0.080
Dominant model
G/G	152 (42.9)	271 (50.9)		1
G/T + T/T	202 (57.1)	261 (49.1)	0.023	0.701 (0.510-0.963)/0.028
rs2507800 – ANGPT-1	356	547
Genotype
T/T	154 (43.3)	246 (45.0)		1
T/A	142 (39.9)	221 (40.4)		0.1030 (0.733-1.448)/0.865
A/A	60 (16.9)	80 (14.6)	0.656	0.832 (0.523-1.323)/0.437
Allele
T	0.63	0.65
A	0.37	0.35	0.421
Recessive model
T/T + T/A	296 (83.1)	467 (85.4)		1
A/A	60 (16.9)	80 (14.6)	0.418	0.820 (0.531-1.267)/0.371
Additive model
T/T	154 (72.0)	246 (75.5)		1
A/A	60 (28.0)	80 (24.5)	0.420	0.835 (0.523-1.333)/0.449
Dominant model
T/T	154 (43.3)	246 (45.0)		1
T/A + A/A	202 (56.7)	301 (55.0)	0.661	0.972 (0.709-1.333)/0.861
Haplotype ANGPT-1	331	494
0-2 mutated alleles	262 (79.2)	418 (84.6)		1
3-4 mutated alleles	69 (20.8)	76 (15.4)	0.054	0.602 (0.392-0.925)/0.021

• Data are shown as number (%) or proportion.
• ^*
  P-values were calculated using x² tests.
• ^†
  P-values and OR (95% CI) obtained using logistic regression analyses adjusting for age, gender, HbA1c, and presence of hypertension. DM: diabetes mellitus; DR: diabetic retinopathy; OR: odds ratio; CI: confidence interval.

Frequencies of the TIE2 rs638203 G allele and G/G genotype were higher in controls compared to cases (P = 0.040 and P = 0.046, respectively). After adjustment for the same covariates described above, the G/G genotype was associated with protection against DR (OR = 0.523, 95% CI 0.340-0.804; P = 0.003). This association remained significant when considering recessive (OR = 0.548, 95% CI 0.376-0.800; P = 0.002) and additive (OR = 0.527, 95% CI 0.343-0.812; P = 0.004) inheritance models (Table 2).

The frequency of the T allele of the rs4324901 SNP in ANGPT-1 was higher in controls compared to cases (36% vs. 31%; P = 0.026). Genotype frequency of this SNP did not differ significantly between groups (P = 0.064). When considering the dominant model, the T allele was significantly associated with protection against DR (P = 0.023), which was maintained after adjustment for age, HbA1c, gender, and presence of hypertension (OR = 0.701, 95% CI 0.510-0.963; P = 0.028). No association was found when considering the additive and recessive models (Table 2).

The frequency of the A allele of the ANGPT-1 rs2507800 SNP was 37% in controls and 35% in cases with DR (P = 0.421). Genotype frequencies of this SNP did not differ significantly between groups (P = 0.656) or when considering different inheritance models (all P values > 0.050). Furthermore, adjustments for age, HbA1c, gender, and presence of hypertension did not change the lack of association between the rs2507800 SNP and DR (Table 2).

Moreover, when we compared control patients to patients with PDR, the TIE2 rs639225 G/G and TIE2 rs638203 G/G genotypes conferred protection against PDR (OR = 0.573; 95% CI 0.377-0.872; P = 0.012 and OR = 0.512; 95% CI 0.335-0.783; P = 0.003, respectively, for the recessive model). No association was found between the rs4324901 and rs2807800 SNPs in ANGPT-1 and PDR (P = 0.279 and P = 0.338, respectively). Additionally, when comparing control patients to patients with NPDR, the presence of the T allele of the ANGPT-1 rs4324901 SNP was associated with protection against NPDR (OR = 0.698, 95% CI 0.512-0.953; P = 0.029). No association was found between the ANGPT-1 rs2807800 SNP (P = 0.915) and the TIE2 rs639225 and rs638203 SNPs (P = 0.780 and P = 0.462, respectively) and NPDR.

Haplotype distributions

The TIE2 rs639225 SNP is in almost complete LD with the TIE2 rs638203 SNP (|D’| = 0.971, r² = 0.905) in our population. For this reason, we did not proceed with the haplotype analysis between these two SNPs. In contrast, the ANGPT-1 rs4324901 SNP presented a weak LD with the ANGPT-1 rs2507800 SNP (|D’| = 0.531, r² = 0.258) in our population. Four haplotypes comprising the ANGPT-1 rs4324901 and rs2507800 SNPs were inferred in both groups: T/G (51.8%), T/T (11.6%), A/G (14.7%), and A/T (21.9%). Their distributions did not differ significantly between groups (P = 0.080). However, after comparing patients carrying 0 to 2 minor alleles to patients with 3 or 4 minor alleles in the haplotypes (Table 2), the presence of ≥ 3 minor alleles in the ANGPT-1 haplotypes seemed to be associated with protection against DR (20.8% in controls vs. 15.4% in cases; P = 0.054). This difference was significant after adjustment for age, HbA1c, gender, and presence of hypertension (OR = 0.602 95% CI 0.392-0.925; P = 0.021; Table 2).

Association between SNPs in the TIE2 and ANGPT-1 genes and DKD

In order to evaluate whether the analyzed SNPs could also be associated with DKD in our population, we categorized our sample according to the presence of this complication: 349 patients without DKD vs. 761 patients with this complication. Frequencies of the TIE2 rs639225, ANGPT-1 rs4324901, and ANGPT-1 rs2507800 SNPs did not differ between patients with or without DKD (^{Supplementary Table 1}). However, the frequency of the TIE2 rs638203 G/G genotype was lower in patients with T2DM and DKD compared to patients without this complication (19.3% vs. 24.9%, OR = 0.721, 95% CI 0.533-0.975; P = 0.040 for the recessive model), but this association was not independent of DR presence, age, gender, HbA1c, and hypertension in the logistic regression analysis (OR = 0.839, 95% CI 0.574-1.226; P = 0.365).

DISCUSSION

The ANGPT family has been reported as being involved in DR pathogenesis as a mediator of the permeability of the blood-retinal barrier and a regulator of pericyte function, angiogenesis, and apoptosis (¹⁵, ³⁶). Thus, considering that ANGPT plays an important role in DR, the present study investigated the association of four SNPs in the TIE2 and ANGPT-1 genes with DR in patients with T2DM. Our results show, for the first time, an association between the TIE2 rs638203 G/G and TIE2 rs639225 G/G genotypes and protection against DR. Moreover, the presence of the T allele of the ANGPT-1 rs4324901 SNP and the presence of ≥ 3 minor alleles in the ANGPT-1 haplotypes conferred protection against DR.

TIE2, also known as TEK, comprises immunoglobulin-like domains, epidermal growth factor-like domains, and fibronectin type III domains (¹⁵). After activation, TIE2 shows strong kinase activity and becomes phosphorylated on several cytoplasmic tyrosine residues, resulting in downstream activation of some pathways, such as PI3-kinase/protein kinase B (AKT) and extracellular signal regulated kinase (ERK) pathways (³⁷). The activation of these pathways inhibits de novo blood vessel growth and vascular hyperpermeability (³⁷), key processes in DR pathogenesis.

To our knowledge, no other study so far has investigated the frequencies of these two TIE2 SNPs in patients with DR. The rs638203 and rs639225 SNPs were previously associated with risk of vascular malformations (³⁸), and the rs639225 SNP was associated with a baseline peritoneal transport property (³⁴). Few studies have reported TIE2 levels in patients with DM and DR (¹⁶, ³⁹). Khalaf and cols. (¹⁶) reported that serum TIE2 levels were similar between patients with T2DM and PDR, NPDR, or without this complication. In accordance, another study found no difference in serum TIE2 levels in patients with T2DM with and without DR (³⁹).

Although these results suggest that serum TIE2 levels are not associated with DR, TIE2 activation has been investigated as a treatment or prevention strategy of DR, since TIE2 activation increased EC survival and adhesion, as well as cell-cell junction integrity, thereby stabilizing the vasculature (⁴⁰, ⁴¹). Thus, polymorphisms in TIE2 gene associated with DR protection, such as the rs638203 SNP, could be involved with better TIE2 activation and, consequently, better function and stabilization of the vasculature, which is important in DR pathogenesis. However, to date, no study has evaluated the functional impact of the TIE2 rs638203 SNP; hence, functional studies are necessary to better understand the involvement of TIE2 SNPs in DR.

Regarding the ANGPT-1 SNPs, we demonstrated an association of the presence of the T allele of the rs4324901 SNP and the presence of ≥ 3 minor alleles in ANGPT-1 haplotypes with protection against DR. ANGPT-1 seems to be involved in DR pathogenesis since ANGPT-1 is reported to help delay diabetic complications owing to the restoration of microvascular function maintenance of quiescence in some adult stem cells [reviewed in (⁴²)]. Accordingly, ANGPT-1 treatment protected retinal pericytes against apoptosis after 48 h of culture with tumor necrosis factor (TNF) or high glucose (³⁶). Moreover, ANGPT-1 seems to improve the activation and migration of retinal pericytes during the establishment of new retinal vessels (³⁶). Joussen and cols. (⁴³) demonstrated that Angpt-1 protects the retinal vasculature of patients with DM against leukocyte-mediated EC injury and death and suppresses diabetic blood-retinal barrier breakdown and vascular endothelial growth factor (Vegf) and intercellular adhesion molecule 1 (Icam-1) expression in mice.

Expression levels of ANGPT-1 were also investigated in patients with DM (16,39,44). Serum ANGPT-1 levels in T2DM patients without DR and those of patients with T2DM and PDR were similar; however, patients with NPDR had higher ANGPT-1 levels compared to PDR and T2DM control groups (³⁹). Furthermore, Khalaf and cols. (¹⁶) observed that serum ANGPT-1 levels were higher in patients with NPDR vs. DM control patients with DM without this complication, but no difference was found between patients with PDR and controls. These authors suggested that the increased expression of ANGPT-1 in patients with DR occurs early in the development of this complication as a compensatory mechanism to help cellular repair and preserve the integrity of ECs (¹⁶). Accordingly, when we stratified patients according to DR severity, the ANGPT-1 rs4324901 T allele was associated with protection against NPDR but not PDR. Thus, we hypothesized that the T allele might increase ANGPT-1 expression, thus conferring protection against the development of NDPR. Functional studies are also needed to confirm this hypothesis.

ANGPT-1 expression seems to be altered by the rs2507800 A/A genotype. Chen and cols. (⁴⁴) demonstrated that the A allele of the rs2507800 SNP suppressed ANGPT-1 translation by facilitating miR-211 binding, which was not observed for the T allele. Accordingly, individuals carrying the T/T genotype had higher plasma ANGPT-1 levels than those with the A allele (⁴⁴). Even though the rs2507800 SNP seems to alter ANGPT-1 levels, we did not demonstrate an individual association between this SNP and DR in the present study. Of note, no previous case-control study evaluated the association of ANGPT-1 SNPs with DR.

The present study has a few limitations. First, we cannot rule out the possibility of population stratification bias when analyzing our samples. Both minor alleles of the TIE2 SNPs did not differ significantly between white and non-white patients (all P values > 0.700). Regarding ANGPT-1 SNPs, although their frequencies differed between white and non-white participants (P = 0.0001), when analyzing white and non-white patients separately, the T allele of the rs4324901 SNP remained increased in controls without DR in comparison to patients with DR in both ethnic groups. Second, we cannot fully exclude the possibility of a type II error when analyzing associations between the rs2507800 SNP in ANGPT-1 and DR. Although we had more than 80% power (α = 0.050) to detect an OR ≤ 0.4 for DR protection and the sample sizes were in agreement to those calculated, we cannot exclude the possibility that this SNP would be associated with DR with a lower OR.

In conclusion, we demonstrated, for the first time, an association of rs638203 and rs639225 SNPs in TIE2 and of rs4324901 SNP in ANGPT-1 with protection against DR in a Brazilian population. These results are biologically plausible considering the involvement of TIE2 and ANGPT-1 in key pathways related to DR pathogenesis. No functional study has been published on the impact of these SNPs on TIE2 and ANGPT-1 expression, thus analyses are still required to clarify how they influence the expression of their respective genes as well as DR pathogenesis. Nevertheless, the present data contribute to the identification of new genetic markers of DR protection. Additional studies are needed to confirm the associations of these SNPs with DR in other populations and also in patients with T1DM.

Acknowledgments:

this study was partially supported by grants from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundo de Incentivo à Pesquisa e Eventos (Fipe) at Hospital de Clínicas de Porto Alegre (grant number: 2018-0471), Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (Fapergs), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes) (Finance Code 001), and Post-Graduation Program in Medical Sciences: Endocrinology - Federal University of Rio Grande do Sul. D.C. is recipient of scholarships from CNPq, while C.D. and F.M.P. are recipients of scholarships from CAPES.

REFERENCES

• ¹
  Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract. 2019;157:107843.
• ²
  Lee R, Wong TY, Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond). 2015;2:17.
• ³
  Yau JW, Rogers SL, Kawasaki R, Lamoureux EL, Kowalski JW, Bek T, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care. 2012;35(3):556-64.
• ⁴
  Wang W, Lo ACY. Diabetic Retinopathy: Pathophysiology and Treatments. Int J Mol Sci. 2018;19(6).
• ⁵
  Isidori AM, Venneri MA, Fiore D. Angiopoietin-1 and Angiopoietin-2 in metabolic disorders: therapeutic strategies to restore the highs and lows of angiogenesis in diabetes. J Endocrinol Invest. 2016;39(11):1235-46.
• ⁶
  Moss A. The angiopoietin: Tie 2 interaction: a potential target for future therapies in human vascular disease. Cytokine Growth Factor Rev. 2013;24(6):579-92.
• ⁷
  van Meurs M, Kumpers P, Ligtenberg JJ, Meertens JH, Molema G, Zijlstra JG. Bench-to-bedside review: Angiopoietin signalling in critical illness – a future target? Crit Care. 2009;13(2):207.
• ⁸
  Huang H, Bhat A, Woodnutt G, Lappe R. Targeting the ANGPT-TIE2 pathway in malignancy. Nature reviews Cancer. 2010;10(8):575-85.
• ⁹
  Brkovic A, Pelletier M, Girard D, Sirois MG. Angiopoietin chemotactic activities on neutrophils are regulated by PI-3K activation. J Leukoc Biol. 2007;81(4):1093-101.
• ¹⁰
  Kosacka J, Figiel M, Engele J, Hilbig H, Majewski M, Spanel-Borowski K. Angiopoietin-1 promotes neurite outgrowth from dorsal root ganglion cells positive for Tie-2 receptor. Cell Tissue Res. 2005;320(1):11-9.
• ¹¹
  Nakayama T, Hatachi G, Wen CY, Yoshizaki A, Yamazumi K, Niino D, et al. Expression and significance of Tie-1 and Tie-2 receptors, and angiopoietins-1, 2 and 4 in colorectal adenocarcinoma: Immunohistochemical analysis and correlation with clinicopathological factors. World J Gastroenterol. 2005;11(7):964-9.
• ¹²
  Voskas D, Jones N, Van Slyke P, Sturk C, Chang W, Haninec A, et al. A cyclosporine-sensitive psoriasis-like disease produced in Tie2 transgenic mice. Am J Pathol. 2005;166(3):843-55.
• ¹³
  Uchida T, Nakashima M, Hirota Y, Miyazaki Y, Tsukazaki T, Shindo H. Immunohistochemical localisation of protein tyrosine kinase receptors Tie-1 and Tie-2 in synovial tissue of rheumatoid arthritis: correlation with angiogenesis and synovial proliferation. Ann Rheum Dis. 2000;59(8):607-14.
• ¹⁴
  Yuan HT, Khankin EV, Karumanchi SA, Parikh SM. Angiopoietin 2 is a partial agonist/antagonist of Tie2 signaling in the endothelium. Mol Cell Biol. 2009;29(8):2011-22.
• ¹⁵
  Whitehead M, Osborne A, Widdowson PS, Yu-Wai-Man P, Martin KR. Angiopoietins in Diabetic Retinopathy: Current Understanding and Therapeutic Potential. J Diabetes Res. 2019;2019:5140521.
• ¹⁶
  Khalaf N, Helmy H, Labib H, Fahmy I, El Hamid MA, Moemen L. Role of Angiopoietins and Tie-2 in Diabetic Retinopathy. Electron Physician. 2017;9(8):5031-5.
• ¹⁷
  Yu Y, Zhang J, Zhu R, Zhao R, Chen J, Jin J, et al. The Profile of Angiogenic Factors in Vitreous Humor of the Patients with Proliferative Diabetic Retinopathy. Curr Mol Med. 2017;17(4):280-6.
• ¹⁸
  Jeansson M, Gawlik A, Anderson G, Li C, Kerjaschki D, Henkelman M, et al. Angiopoietin-1 is essential in mouse vasculature during development and in response to injury. J Clin Invest. 2011;121(6):2278-89.
• ¹⁹
  von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-9.
• ²⁰
  Little J, Higgins JP, Ioannidis JP, Moher D, Gagnon F, von Elm E, et al. STrengthening the REporting of Genetic Association Studies (STREGA)--an extension of the STROBE statement. Genet Epidemiol. 2009;33(7):581-98.
• ²¹
  Canani LH, Capp C, Ng DP, Choo SG, Maia AL, Nabinger GB, et al. The fatty acid-binding protein-2 A54T polymorphism is associated with renal disease in patients with type 2 diabetes. Diabetes. 2005;54(11):3326-30.
• ²²
  American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2019. Diabetes Care. 2019;42(Suppl 1):S13-28.
• ²³
  Wilkinson CP, Ferris FL 3rd, Klein RE, Lee PP, Agardh CD, Davis M, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology. 2003;110(9):1677-82.
• ²⁴
  de Souza BM, Michels M, Sortica DA, Boucas AP, Rheinheimer J, Buffon MP, et al. Polymorphisms of the UCP2 Gene Are Associated with Glomerular Filtration Rate in Type 2 Diabetic Patients and with Decreased UCP2 Gene Expression in Human Kidney. PLoS One. 2015;10(7):e0132938.
• ²⁵
  Camargo JL, Zelmanovitz T, Paggi A, Friedman R, Gross JL. Accuracy of conversion formulae for estimation of glycohaemoglobin. Scand J Clin Lab Invest. 1998;58(6):521-8.
• ²⁶
  Zelmanovitz T, Gross JL, Oliveira JR, Paggi A, Tatsch M, Azevedo MJ. The receiver operating characteristics curve in the evaluation of a random urine specimen as a screening test for diabetic nephropathy. Diabetes Care. 1997;20(4):516-9.
• ²⁷
  Andrassy KM. Comments on ‘KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease’. Kidney Int. 2013;84(3):622-3.
• ²⁸
  Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-12.
• ²⁹
  Lahiri DK, Nurnberger JI. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res. 1991;19(19):5444.
• ³⁰
  Zintzaras E, Lau J. Synthesis of genetic association studies for pertinent gene-disease associations requires appropriate methodological and statistical approaches. J Clin Epidemiol. 2008;61(7):634-45.
• ³¹
  Hedrick PW. Gametic disequilibrium measures: proceed with caution. Genetics. 1987;117(2):331-41.
• ³²
  Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet. 2001;68(4):978-89.
• ³³
  Andraweera PH, Dekker GA, Thompson SD, North RA, McCowan LM, Roberts CT, et al. A functional variant in ANGPT1 and the risk of pregnancies with hypertensive disorders and small-for-gestational-age infants. Mol Hum Reprod. 2012;18(6):325-32.
• ³⁴
  Ding L, Shao X, Cao L, Fang W, Yan H, Huang J, et al. Possible role of IL-6 and TIE2 gene polymorphisms in predicting the initial high transport status in patients with peritoneal dialysis: an observational study. BMJ Open. 2016;6(10):e012967.
• ³⁵
  He L, Dang L, Zhou J, Bai J, Li YZ. Association of angiopoietin-1, angiopoietin-2 and caspase-5 polymorphisms with psoriasis vulgaris. Clin Exp Dermatol. 2015;40(5):556-63.
• ³⁶
  Cai J, Kehoe O, Smith GM, Hykin P, Boulton ME. The angiopoietin/Tie-2 system regulates pericyte survival and recruitment in diabetic retinopathy. Invest Ophthalmol Vis Sci. 2008;49(5):2163-71.
• ³⁷
  Thurston G, Daly C. The complex role of angiopoietin-2 in the angiopoietin-tie signaling pathway. Cold Spring Harb Perspect Med. 2012;2(9):a006550.
• ³⁸
  Zheng Q, Du J, Zhang Z, Xu J, Fu L, Cao Y, et al. Association study between of Tie2/angiopoietin-2 and VEGF/KDR pathway gene polymorphisms and vascular malformations. Gene. 2013;523(2):195-8.
• ³⁹
  You QY, Zhuge FY, Zhu QQ, Si XW. Effects of laser photocoagulation on serum angiopoietin-1, angiopoietin-2, angiopoietin-1/angiopoietin-2 ratio, and soluble angiopoietin receptor Tie-2 levels in type 2 diabetic patients with proliferative diabetic retinopathy. Int J Ophthalmol. 2014;7(4):648-53.
• ⁴⁰
  Rubsam A, Parikh S, Fort PE. Role of Inflammation in Diabetic Retinopathy. Int J Mol Sci. 2018;19(4).
• ⁴¹
  Campochiaro PA, Peters KG. Targeting Tie2 for Treatment of Diabetic Retinopathy and Diabetic Macular Edema. Curr Diab Rep. 2016;16(12):126.
• ⁴²
  Koh GY. Orchestral actions of angiopoietin-1 in vascular regeneration. Trends Mol Med. 2013;19(1):31-9.
• ⁴³
  Joussen AM, Poulaki V, Tsujikawa A, Qin W, Qaum T, Xu Q, et al. Suppression of diabetic retinopathy with angiopoietin-1. Am J Pathol. 2002;160(5):1683-93.
• ⁴⁴
  Chen J, Yang T, Yu H, Sun K, Shi Y, Song W, et al. A functional variant in the 3’-UTR of angiopoietin-1 might reduce stroke risk by interfering with the binding efficiency of microRNA 211. Hum Mol Genet. 2010;19(12):2524-33.

Supplementary Table 1 Genotype and allele frequencies of SNPs in TIE2 and ANGPT-1 genes in T2DM patients with and without DKD

T2DM patients without DKD DKD patients Unadjusted P* Adjusted OR (95% IC) /† P rs639225 – TIE2 337 742 Genotype A/A 114 (33.9) 263 (35.4) 0.224 1 A/G 140 (41.5) 331 (44.7) 1.081 (0.756-1.546)/0.669 G/G 83 (24.6) 148 (19.9) 0.976 (0.633-1.504)/0.912 Allele A 0.55 0.58 0.186 - G 0.45 0.42 Recessive model A/A + A/G 254 (75.4) 594 (80.1) 0.097 1 G/G 83 (24.6) 148 (19.9) 0.935 (0.63-1.373)/0.730 Additive model A/A 114 (57.9) 263 (64.0) 0.172 1 G/G 83 (42.1) 148 (36.0) 0.939 (0.609-1.446)/0.774 Dominant model A/A 114 (33.8) 263 (35.4) 0.655 1 A/G + G/G 223 (66.2) 479 (64.6) 1.046 (0.752-1.454)/0.791 rs638203 – TIE2 349 761 Genotype A/A 119 (34.1) 264 (34.7) 0.088 1 A/G 143 (41.0) 350 (46.0) 1.159 (0.814-1.650)/0.412 G/G 87 (24.9) 147 (19.3) 0.912 (0.595-1.398)/0.672 Allele A 0.55 0.58 0.185 - G 0.45 0.42 Recessive model A/A + A/G 262 (75.1) 614 (80.7) 0.040 1 G/G 87 (24.9) 147 (19.3) 0.839 (0.574-1.226)/0.365 Additive model A/A 119 (57.8) 264 (64.2) 0.141 1 G/G 87 (42.2) 147 (35.8) 0.868 (0.566-1.332)/0.518 Dominant model A/A 119 (34.1) 264 (34.7) 0.900 1 A/G + G/G 230 (65.9) 497 (65.3) 1.073 (0.775-1.487)/0.671 rs4324901 – ANGPT-1 331 726 Genotype G/G 151 (45.6) 353 (48.6) 0.645 1 G/T 133 (40.2) 279 (38.5) 0.897 (0.634-1.268)/0.539 T/T 47 (14.2) 94 (12.9) 1.289 (0.790-2.103)/0.309 Allele G 0.66 0.68 0.115 - T 0.34 0.32 Recessive model G/G + G/T 284 (85.8) 632 (87.1) 0.647 1 T/T 47 (14.2) 94 (12.9) 1.355 (0.853-2.152)/0.198 Additive model G/G 151 (76.3) 353 (79.0) 0.506 1 T/T 47 (23.7) 94 (21.0) 1.328 (0.807-2.184)/0.264 Dominant model G/G 151 (45.6) 353 (48.6) 0.401 1 G/T + T/T 180 (54.4) 373 (51.4) 0.990 (0.718-1.364)/0.951 rs2507800 – ANGPT-1 335 739 Genotype T/T 150 (44.8) 313 (42.4) 0.270 1 T/A 129 (38.5) 321 (43.4) 1.256 (0.888-1.776)/0.197 A/A 56 (16.7) 105 (14.2) 1.231 (0.765-1.981)/0.391 Allele T 0.64 0.64 0.976 - A 0.36 0.36 Recessive model T/T + T/A 279 (83.3) 634 (85.8) 0.330 1 A/A 56 (16.7) 105 (14.2) 1.103 (0.706-1.723)/0.666 Additive model T/T 150 (72.8) 313 (74.9) 0.648 1 A/A 56 (27.2) 105 (25.1) 1.269 (0.780-2.063)/0.337 Dominant model T/T 150 (44.8) 313 (42.4) 0.499 1 T/A + A/A 185 (55.2) 426 (57.6) 1.249 (0.907-1.722)/0.174 Data are shown as number (%) or proportion. *
†

	T2DM patients without DKD	DKD patients	Unadjusted P*	Adjusted OR (95% IC) /† P
rs639225 – TIE2	337	742
Genotype
A/A	114 (33.9)	263 (35.4)	0.224	1
A/G	140 (41.5)	331 (44.7)		1.081 (0.756-1.546)/0.669
G/G	83 (24.6)	148 (19.9)		0.976 (0.633-1.504)/0.912
Allele
A	0.55	0.58	0.186	-
G	0.45	0.42
Recessive model
A/A + A/G	254 (75.4)	594 (80.1)	0.097	1
G/G	83 (24.6)	148 (19.9)		0.935 (0.63-1.373)/0.730
Additive model
A/A	114 (57.9)	263 (64.0)	0.172	1
G/G	83 (42.1)	148 (36.0)		0.939 (0.609-1.446)/0.774
Dominant model
A/A	114 (33.8)	263 (35.4)	0.655	1
A/G + G/G	223 (66.2)	479 (64.6)		1.046 (0.752-1.454)/0.791
rs638203 – TIE2	349	761
Genotype
A/A	119 (34.1)	264 (34.7)	0.088	1
A/G	143 (41.0)	350 (46.0)		1.159 (0.814-1.650)/0.412
G/G	87 (24.9)	147 (19.3)		0.912 (0.595-1.398)/0.672
Allele
A	0.55	0.58	0.185	-
G	0.45	0.42
Recessive model
A/A + A/G	262 (75.1)	614 (80.7)	0.040	1
G/G	87 (24.9)	147 (19.3)		0.839 (0.574-1.226)/0.365
Additive model
A/A	119 (57.8)	264 (64.2)	0.141	1
G/G	87 (42.2)	147 (35.8)		0.868 (0.566-1.332)/0.518
Dominant model
A/A	119 (34.1)	264 (34.7)	0.900	1
A/G + G/G	230 (65.9)	497 (65.3)		1.073 (0.775-1.487)/0.671
rs4324901 – ANGPT-1	331	726
Genotype
G/G	151 (45.6)	353 (48.6)	0.645	1
G/T	133 (40.2)	279 (38.5)		0.897 (0.634-1.268)/0.539
T/T	47 (14.2)	94 (12.9)		1.289 (0.790-2.103)/0.309
Allele
G	0.66	0.68	0.115	-
T	0.34	0.32
Recessive model
G/G + G/T	284 (85.8)	632 (87.1)	0.647	1
T/T	47 (14.2)	94 (12.9)		1.355 (0.853-2.152)/0.198
Additive model
G/G	151 (76.3)	353 (79.0)	0.506	1
T/T	47 (23.7)	94 (21.0)		1.328 (0.807-2.184)/0.264
Dominant model
G/G	151 (45.6)	353 (48.6)	0.401	1
G/T + T/T	180 (54.4)	373 (51.4)		0.990 (0.718-1.364)/0.951
rs2507800 – ANGPT-1	335	739
Genotype
T/T	150 (44.8)	313 (42.4)	0.270	1
T/A	129 (38.5)	321 (43.4)		1.256 (0.888-1.776)/0.197
A/A	56 (16.7)	105 (14.2)		1.231 (0.765-1.981)/0.391
Allele
T	0.64	0.64	0.976	-
A	0.36	0.36
Recessive model
T/T + T/A	279 (83.3)	634 (85.8)	0.330	1
A/A	56 (16.7)	105 (14.2)		1.103 (0.706-1.723)/0.666
Additive model
T/T	150 (72.8)	313 (74.9)	0.648	1
A/A	56 (27.2)	105 (25.1)		1.269 (0.780-2.063)/0.337
Dominant model
T/T	150 (44.8)	313 (42.4)	0.499	1
T/A + A/A	185 (55.2)	426 (57.6)		1.249 (0.907-1.722)/0.174

• Data are shown as number (%) or proportion.
• ^*
  P-values were calculated using x² tests.
• ^†
  P-values and OR (95% CI) obtained using logistic regression analyses adjusting for age, gender, HbA1c, presence of hypertension and diabetic retinopathy.

Publication Dates

• Publication in this collection
  05 June 2023
• Date of issue
  2023

History

• Received
  04 Apr 2022
• Accepted
  06 Dec 2022