Polymorphisms of lipid metabolism enzyme-coding genes in patients with diabetic dyslipidemia.

OBJECTIVE: The polymorphisms/mutations of genes encoding proteins and enzymes involved in lipoprotein metabolism play important roles in the development of diabetic dyslipidemia. The aim of our study was to investigate the effects of LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), PON1 (rs662), and MNSOD (rs4880) gene polymorphisms on lipid metabolism and diabetic dyslipidemia.
METHODS: This case-control study included 217 patients with diabetic dyslipidemia and 212 healthy age- and gender-matched individuals. Genomic DNA isolation was performed from blood samples, and genotype analysis was performed using melting curve analysis on a LightCycler® 480 Instrument. The chi-square test was used to compare genotype distribution and allele frequencies between the groups.
RESULTS: Significant associations were observed between LPL (rs320) (p<0.001), LIPC (rs2070895) (p<0.001), SCARB1 (rs5888) (p<0.001), LCAT (rs2292318) (p<0.001), CETP (rs708272) (p<0.001), ADIPOQ (rs1501299) (p=0.01), RETN (rs3745367) (p<0.001), and MNSOD (rs4880) (p<0.001) polymorphisms and diabetic dyslipidemia. However, no association was observed between PON1 (rs662) polymorphisms and diabetic dyslipidemia (p=0.611).
CONCLUSION: LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), and MNSOD (rs4880) polymorphisms play an important role in basic molecular metabolism in diabetic dyslipidemia. Therefore, these polymorphisms may be used as a predictive marker for diabetic dyslipidemia in high-risk patients.

Introduction

Diabetes mellitus is a metabolic disorder with a complex etiology resulting from the disturbance of insulin secretion/action and is characterized by carbohydrate, fat, and protein metabolism dysfunction (1). Dyslipidemia occurs in conjunction with insulin resistance and is considered to be one of the most important risk factors for cardiovascular (CV) disease (2). Quantitative and qualitative changes in the properties of lipoproteins, the degradation of lipoprotein metabolism, genetic predispositions, and environmental factors are the main etiologic mechanisms of diabetic dyslipidemia. Recent studies have shown that the polymorphisms or mutations of genes that encode proteins and enzymes involved in lipoprotein metabolism may play an important role in the development of diabetic dyslipidemia (3–5). Therefore, determining genetic profiles associated with diabetic dyslipidemia is gaining importance in terms of reducing the risk of micro- and macrovascular complications.

Roles of enzymes and enzyme-coding candidate genes in cholesterol metabolism

Lipoprotein lipase (LPL) is a key enzyme in lipoprotein metabolism. It hydrolyzes triglycerides from very-low-density lipoproteins and separates lipoproteins from the circulation (6, 7). This enzyme is coded by the LPL gene, and several genetic variants of LPL have been associated with plasma lipoprotein levels (7, 8). Lecithin–cholesterol acyltransferase (LCAT) catalyzes the esterification of free cholesterol in the blood (9). Both common and rare LCAT gene mutations/polymorphisms have been found to affect high-density lipoprotein (HDL) levels (6). Adiponectin is a cytokine that inhibits adipose-specific expression in the blood (10). Resistin, encoded by the RETN gene, plays a key role in glucose homeostasis; RETN has been described as one of the candidate genes for type 2 diabetes mellitus (11, 12). Serum paraoxonase/arylesterase 1 (PON1) is an enzyme that protects low-density lipoproteins from oxidation and is responsible for the antioxidant properties of HDL (13). Cholesteryl ester transfer protein (CETP) is a hydrophobic glycoprotein that regulates the exchange of triglycerides by transferring esterified cholesterol from HDL to apo-B-containing particles (14). CETP also plays an important role in the regulation of HDL cholesterol levels (6, 14). CETP gene polymorphisms have been associated with HDL cholesterol concentrations in many reports (14, 15). Scavenger receptor class B member 1 (SRB1), encoded by the SCARB1 gene, is an HDL receptor involved in reverse cholesterol transport (16). Hepatic lipase (LIPC) is an enzyme that regulates the metabolism of low-density lipoprotein (LDL), intermediate-density lipoprotein (IDL), and HDL particles and plays a crucial role in the selective uptake of cholesterol esters from HDL (17). Recent studies have reported that polymorphisms of the promoter region of the LIPC gene influence the regulation of insulin sensitivity and type 2 diabetes development (17, 18). Oxidative stress also plays an important role in the pathogenesis of insulin resistance. Increased reactive oxygen species (ROS) levels lead to insulin resistance in type 2 diabetes (19). Manganese-dependent superoxide dismutase (MnSOD, SOD2) is a member of the most important family of superoxide dismutases and may prevent damage from ROS-induced hyperglycemia, oxidative stress, and ionizing radiation. It has been reported that the polymorphism of the MNSOD gene leads to differences in protein function (20).

The aim of this study was to identify the genotype distributions of enzyme-coding genes involved in the cholesterol pathway and the development of diabetic dyslipidemia in a group of patients from a single outpatient diabetes clinic in Turkey. These enzyme-coding genes included LPL, LIPC, SCARB1, LCAT, CETP, ADIPOQ, RETN, PON1, and MNSOD.

Methods

Study population

Our case-control study group initially consisted of 250 patients with diabetic dyslipidemia who were consecutively admitted to the Endocrinology and Metabolic Disease outpatient clinic of Ege University School of Medicine between 2007 and 2011. This study also included 225 age- and gender-matched healthy individuals. Control subjects were recruited from healthy volunteers whose routine health checkup was within normal limits. The diagnosis of diabetes was based on the presence of fasting blood glucose (FBG) levels of >126 mg/dL and/or postprandial glucose levels of >200 mg/dL and hemoglobin A1c (HbA1c) levels of >6.5%. The presence of diabetic dyslipidemia was defined as triglyceride levels of >200 mg/dL and/or HDL levels of <45 mg/dL. During the analysis, 13 control subjects were excluded due to elevated FBG levels (>126 mg/dL) indicating diabetes mellitus and 33 patients were excluded from the study group due to normal FBG levels. The remaining 429 subjects (217 diabetics and 212 healthy subjects) constituted the study population. The study protocol was approved by the local Ethics Committee, and written informed consent was obtained from all subjects.

The clinical characteristics of the groups were retrospectively obtained from patient charts. Biochemical analysis included HbA1c, total cholesterol, triglyceride, HDL cholesterol, LDL cholesterol, and FBG levels measured by an Olympus AU2700 automated analyzer (Toshiba, Tokyo, Japan). All participants’ peripheral venous blood samples were collected into EDTA-containing tubes and were stored at –20°C until genomic DNA isolation was performed.

Genotyping

Genomic DNA was extracted from peripheral blood leukocytes using a MagNA Pure LC instrument and MagNA Pure LC DNA Isolation Kit (Roche Applied Science, Mannheim, Germany). The analysis of LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), PON1 (rs662), and MNSOD (rs4880) gene polymorphisms was performed using specific primers and a simple probe mix with a LightCycler-FastStart DNA Master Hybridization Probes Kit (Roche Applied Science). All experiments were performed in a LightCyclerTM 480 Instrument (Roche Applied Science) according to the protocol provided by the manufacturer (TIB MOLBIOL, Berlin, Germany). All polymorphisms were detected in the F2 channel. Alleles were identified by the specific melting temperature of the resulting amplicons, as individuals with two copies of the wild-type genotype show a single melting peak, individuals with heterozygous alleles show two melting peaks, and individuals with two copies of the polymorphic genotype show a single melting peak for all polymorphisms. Among the patients with diabetic dyslipidemia, errors occurred in the analysis of LPL (rs320) in 1 patient, in the analysis of LIPC (rs2070895) and ADIPOQ (rs1501299) in 2 patients, and in the analysis of PON1 (rs662) in 3 patients; therefore, these 5 patients were not included in the final statistical analysis.

Biochemical analyses were compared between the patients with diabetic dyslipidemia and the control group according to the genotypic status of LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), PON1 (rs662), and MNSOD (rs4880) gene polymorphisms.

Statistical analysis

All statistical analyses were performed using SPSS for Windows version 18.0 (SPSS, Chicago, IL, USA). Data are presented as percentages for discrete variables and as mean ± standard deviation (SD) for continuous variables. Comparisons between the groups were made using Student’s t-test, and discrete variables were compared by chi-square analysis. Distribution analyses of the variables were performed using the Kolmogorov–Smirnov test. A p value of <0.05 (two-sided) was regarded to be statistically significant.

Results

This study consisted of 217 patients with diabetic dyslipidemia (133 females and 84 males) and 212 unrelated healthy individuals (122 females and 90 males). The comparison of clinical characteristics and biochemical parameters between the groups is shown in Table 1. FBG, total cholesterol, triglyceride, and LDL cholesterol levels were significantly higher in the diabetic dyslipidemia group than in the control group (p<0.001), while HDL cholesterol levels were significantly lower in the diabetic dyslipidemia group. Within our patient group, those with diabetic dyslipidemia were treated with oral antidiabetics, insulin, and lipid-lowering drugs (Table 1).

Table 1

Characteristics and biochemical parameters of patients with diabetic dyslipidemia and control subjects

	Patients with diabetic dyslipidemia Mean±SD n=217	Control subjects Mean±SD n=212	P
Gender, female/male	133/84	122/90	–
Age, years	53.2±9.8	52.8±9.2	–
Fasting glucose, mg/dL	161.9±72.5	87.7±11.6	<0.001
Total cholesterol, mg/dL	219.3±61.8	178.2±20.7	<0.001
Triglycerides, mg/dL	231.6±255.8	109.5±25.9	<0.001
HDL cholesterol, mg/dL	48.5±13.6	63±8.9	<0.001
LDL cholesterol, mg/dL	131.6±42.3	117±17.9
Lipid-lowering drugs, n, %	68 (31.3%)	–
Oral antidiabetics, n, %	20 (9.2%)	–
Insulin, n, %	29 (13.4%)	–

HDL - high-density lipoprotein; LDL - low-density lipoprotein

LPL (rs320) TT, TG and GG genotype distributions were 10.1%, 38%, and 51.9% in the diabetic dyslipidemia group and 69.8%, 25.5%, and 4.7% in the control group, respectively. Patients with diabetic dyslipidemia had a higher frequency of the polymorphic GG genotype of LPL (rs320) than control subjects (p<0.001) (Table 2).

Table 2

Genotype distributions and allele frequencies of LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), PON1 (rs662), and MNSOD (rs4880) gene polymorphisms in the study groups

Gene/SNP	Genotype/haplotype	Patients with diabetic dyslipidemia n (%)	Control subjects n (%)	P
LPL (rs320) T495G	Wild type (TT)	22 (10.1)	148 (69.8)	<0.001
	Heterozygote (TG)	82 (38)	54 (25.5)
	Polymorphic (GG)	112 (51.9)	10 (4.7)
	T	126	350
	G	164	74
LIPC (rs2070895) G250A	Wild type (GG)	1 (0.5)	156 (73.6)	<0.001
	Heterozygote (GA)	46 (21.4)	48 (21.4)
	Polymorphic (AA)	168 (78.1)	8 (3.8)
	G	48	360
	A	382	64
SCARB1 (rs5888) C1050T	Wild type (CC)	67 (30.9)	166 (79.6)	<0.001
	Heterozygote (CT)	85 (39.2)	40 (18.9)
	Polymorphic (TT)	65 (30.0)	6 (2.8)
	C	219	372
	T	215	52
LCAT (rs2292318) C511T	Wild type (CC)	0 (0)	150 (69.1)	<0.001
	Heterozygote (CT)	38 (17.9)	67 (30.9)
	Polymorphic (TT)	174 (82.1)	0 (0)
	C	38	367
	T	67	386
CETP (rs708272) G279A	Wild type (GG)	68 (31.3)	150 (70.8)	<0.001
	Heterozygote (GA)	112 (51.6)	53 (25)
	Polymorphic (AA)	37 (17.1)	9 (4.2)
	G	248	353
	A	186	71
ADIPOQ (rs1501299) G276T	Wild type (GG)	130 (60.5)	156 (73.6)	=0.01
	Heterozygote (GT)	69 (32.1)	49 (23.1)
	Polymorphic (TT)	16 (7.4)	7 (3.3)
	G	329	361
	T	101	63
RETN (rs3745367) G299A	Wild type (GG)	42 (19.4)	122 (57.5)	<0.001
	Heterozygote (GA)	101 (46.5)	75 (35.4)
	Polymorphic (AA)	74 (34.1)	15 (7.1)
	G	185	319
	A	249	105
PON1 (rs3745367) Q192R	Wild type (AA)	110 (51.4)	119 (56.1)	0.611
	Heterozygote (AG)	87 (40.7)	77 (36.3)
	Polymorphic (GG)	17 (7.9)	16 (7.5)
	A	307	315
	G	121	109
MNSOD (rs4880) Ala16Val	Wild type (CC)	101 (46.5)	164 (77.4)	<0.001
	Heterozygote (CT)	75 (34.6)	41 (19.3)
	Polymorphic (TT)	41 (18.9)	7 (3.3)
	C	277	369
	T	157	55

ADIPOQ - adiponectin; CETP - cholesterol ester transfer protein; LCAT - lecithin–cholesterol acyltransferase; LIPC - hepatic lipase; LPL - lipoprotein lipase; MNSOD - manganese-dependent superoxide dismutase; PON1 - paraoxonase; RETN - resistin; SCARB1 - scavenger receptor class B member 1.

Statistical analysis was performed using SPSS 18.0. The chi-square test was used to compare the genotype distribution and allele frequencies between the groups. The level of significance was accepted as P<0.05.

GG, GA and AA genotype distributions of the LIPC (rs2070895) gene were 0.5%, 21.4%, and 78.1% in the diabetic dyslipidemia group and 73.6%, 21.4%, and 3.8% in the control group, respectively. Patients with diabetic dyslipidemia showed a higher frequency of the polymorphic AA genotype of LIPC (rs2070895) than control subjects (p<0.001) (Table 2).

The frequencies of the CC, CT, and TT SCARB1 (rs5888) genotypes were 30.9%, 39.2%, and 30% in the diabetic dyslipidemia group and 79.6%, 18.9%, and 2.8% in the control group, respectively. Patients with diabetic dyslipidemia had higher frequencies of the SCARB1 (rs5888) CT and TT genotypes than control subjects (p<0.001) (Table 2).

LCAT (rs2292318) CC, CT, and TT genotypes were found in 0%, 17.9%, and 82.1% of patients with diabetic dyslipidemia and in 69.1%, 30.9%, and 0% of the control subjects, respectively. Genotype analysis revealed that patients with diabetic dyslipidemia showed a higher frequency of the polymorphic TT genotype of LCAT (rs2292318) than control subjects (p<0.001), while control subjects had a higher frequency of the CC genotype than patients with diabetic dyslipidemia (p<0.001) (Table 2).

For the CETP (rs708272) gene, the frequencies of the GG, GA and AA genotypes were 31.3%, 51.6%, and 17.1% in the diabetic dyslipidemia group and 70.8%, 25%, and 4.2% in the control group, respectively. The frequency of the AA genotype was significantly higher in patients with diabetic dyslipidemia than in control subjects (p<0.001) (Table 2).

The GG, GT, and TT genotypes of ADIPOQ (rs1501299) were detected at frequencies of 60.5%, 32.1%, and 7.4% in the diabetic dyslipidemia group and 73.6%, 23.1%, and 3.3% in the control group, respectively. The TT genotype was significantly more common in the diabetic dyslipidemia group (p=0.01) (Table 2).

The distributions of the GG, GA, and AA genotypes of RETN (rs3745367) were 19.4%, 46.5%, and 34.1% in the diabetic dyslipidemia group and 57.5%, 35.4%, and 7.1% in the control group, respectively. The frequency of the polymorphic AA genotype was significantly higher among patients with diabetic dyslipidemia than among control subjects (p<0.001) (Table 2).

For PON1 (rs662), frequencies of the AA, AG, and GG genotypes were 51.4, 40.7%, and 7.9% in the diabetic dyslipidemia group and 56.1%, 36.3%, and 7.5% in the control group, respectively. No significant differences emerged between patients with diabetic dyslipidemia and control subjects (p=0.611) (Table 2).

The distributions of the MNSOD (rs4880) CC, CT, and TT genotypes were 46.5%, 34.6%, and 18.9% in patients with diabetic dyslipidemia patients and 77.4%, 19.3%, and 3.3% in control subjects, respectively. The differences between patients with diabetic dyslipidemia and control subjects were statistically significant (p<0.001) (Table 2).

In the control group, CETP (rs708272) and PON1 (rs662) gene polymorphisms showed a negative correlation with triglycerides (r=–0.137, p=0.047) and HDL cholesterol (r=–0.158, p=0.022) levels, respectively, whereas no significant correlation was observed between LPL (rs320), LCAT (rs2292318), ADIPOQ (rs1501299), RETN (rs3745367), LIPC (rs2070895), SCARB1 (rs5888), or MNSOD (rs4880) gene polymorphisms and lipid parameters including total cholesterol, triglyceride, HDL cholesterol, and LDL cholesterol levels (p>0.005). Furthermore, no significant association was observed between these gene polymorphisms and HbA1c or FBG levels in control subjects.

In the diabetic dyslipidemia group, SCARB1 (rs5888) and ADIPOQ (rs1501299) gene polymorphisms showed a weak positive correlation with FBG (r=0.172, p=0.011) and HDL cholesterol (r=0.176, p=0.010) levels, respectively. However, no significant association was found between LPL (rs320), LIPC (rs2070895), LCAT (rs2292318), CETP (rs708272), RETN (rs3745367), PON1 (rs662), or MNSOD (rs4880) gene polymorphisms and lipid parameters, plasma glucose levels, or HbA1c levels (p>0.005).

Discussion

To the best of our knowledge, no previous studies have reported the effects of LPL (rs320), LCAT (rs2292318), and PON1 (rs662) gene polymorphisms in patients with diabetic dyslipidemia. Our study is the first comprehensive study to provide evidence on possible associations between LPL (rs320), LIPC (rs2070895), SCARB1 (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), and MNSOD (rs4880) gene polymorphisms and diabetic dyslipidemia in Turkish adults.

According to our results, heterozygote and polymorphic genotype rates in patients with diabetic dyslipidemia were found to be 38% TG and 51.9% GG for LPL, 21.4% GA and 78.1% AA for LIPC, 39.2% CT and 30% TT for SCARB1, 17.9% CT and 82.1% TT for LCAT, 51.6% GT and 17.1% TT for CETP, 32.1% GT and 7.4% TT for ADIPOQ, 46.5% GA and 34.1% AA for RETN, and 34.6% CT and 18.9% TT for MNSOD.

LPL, LIPC, SCARB1, LCAT, and CETP proteins play important roles in the endogenous and exogenous pathways of lipoprotein metabolism. Genetic and environmental factors such as obesity, dyslipidemia, and blood glucose and blood insulin levels are defined as risk factors for the development of type 2 diabetes. Therefore, the determination of the relationship between LPL T495G gene polymorphisms and risk factors associated with diabetes mellitus is becoming more important in understanding the etiology of type 2 diabetes. In our study, the frequency of heterozygous and polymorphic genotypes of the T495G polymorphism was observed in 38% and 51.9% of patients with diabetic dyslipidemia, respectively. Daoud et al. (7) determined TG and GG genotype frequencies to be 35.8% and 19%, respectively, in Saudi Arabian patients with coronary artery disease (CAD). They also found lower TG and GG frequencies than those in our population in their subgroups of patients with CAD having diabetes (53.04% and 28.15%, respectively) and dyslipidemia (45.88% and 24.35%, respectively). The systemic overexpression of the LPL gene may initiate a tissue-specific insulin signaling cascade, causing insulin resistance in skeletal muscles as well as other metabolic tissues and eventually leading to the development of type 2 diabetes mellitus. LPL activity is also sensitive to environmental factors such as hormonal regulation. We believe that the LPL GG genotype has utility as a biomarker for type 2 diabetes. Qi et al. (8) reported that even in patients with a normal lipid profile, the GG genotype of LPL T495G was significantly associated with an increased risk of developing type 2 diabetes. Individuals with the GG genotype have higher pre-heparin LPL levels and lower triglycerides levels than those with the TT genotype. Ariza et al. (21) reported that T495G LPL gene variants have a lowering effect on triglyceride levels (p<0.005) and that this polymorphism has a protective effect against the development of hypertrigly-ceridemia (p=0.042). In summary, the LPL T495G polymorphism may regulate the magnitude of dyslipidemia, but its effects on lipid metabolism are not yet clearly understood.

HDL cholesterol levels are associated with the LCAT enzyme due to the important physiological function of LCAT; however, only a few large studies have determined a significant relationship between LCAT gene polymorphisms and HDL cholesterol levels (9, 22). In our study, the CT and TT genotypes were detected in 17.9% and 82.1% of patients with diabetic dyslipidemia, respectively. Ghanei et al. (9) reported that LCAT activity is significantly decreased in patients with type 2 diabetes mellitus. The LCAT rs2292318 polymorphism may result in a decreased enzyme activity of LCAT. As there are few studies on rs2292318, its effect on LCAT enzyme activity is unknown. Paré et al. (22) showed a significant relationship between the TT genotype of the LCAT rs2292318 polymorphism (7.6%) and increased HDL cholesterol levels in patients with CAD. In recent studies, 5 LCAT polymorphisms (Gly230Arg, P143L, rs4986970, rs5922, and rs2292318) have been associated with HDL cholesterol levels, though the findings were not entirely consistent with each other (6). To determine the effects of the LCAT rs2292318 genetic polymorphism on HDL cholesterol levels and enzyme activity, rare and frequent genetic variations of this polymorphism should be investigated in genome-wide association studies.

To date, the direct effects of the SCARB1 C1050T gene polymorphism could not be fully explained by in vivo studies. Constantineau et al. (16) have investigated the effects of the C1051T polymorphism on SRB1 protein expression and function by an in vitro study. They concluded that the C1050T variant of SCARB1 affects the secondary structure and protein translation of SRB1, leading to reduced protein expression and function. Roberts et al. (23) determined an association between high HDL cholesterol levels and C1050T polymorphism in women (<50 years old), but any association between lipid levels and the C1050T polymorphism in men was not examined. We determined a positive correlation between FBG and the SCARB1 rs5888 variant in patients with diabetic dyslipidemia. However, no association was found between other lipid parameters and the C1050T polymorphism in the diabetic dyslipidemia group. Stanislovaitiene et al. (24) reported that SCARB1 TT genotype frequency (9.4%) was significantly lower in the oldest male myocardial infarction group than in the control group (22.3%). McCarthy et al. (25) also reported that the rs5888 variants of SCARB1 were associated with insulin resistance (p=0.0003), particularly in women. In our study, we revealed the SCARB1 C1050T polymorphism to be a risk factor for the development of diabetic dyslipidemia, regardless of gender.

CETP plays an important role in the regulation of HDL cholesterol concentrations. Carlquist et al. (15) reported that the frequencies of the GG, GA and AA genotypes of CETP rs708272 were 32.9%, 50.3%, and 16.8%, respectively, in patients with CV diseases. We also found very similar genotypic frequencies (31.3% GG, 51.6% GA, and 17.1% AA). However, Dixit et al. (14) demonstrated that the prevalence of the AA mutant genotype of CETP rs708272 was higher in patients with type 2 diabetes than in those without diabetes, whereas no association was observed between the rs708272 polymorphism and type 2 diabetes mellitus. They also found that the heterozygous genotype of CETP increased the risk of type 2 diabetes-induced hypertension (p=0.028), while the wild-type genotype had a protective effect against disease development (p=0.038) (14). Population studies in Singapore (China, Malaysia, and Indies) demonstrated that the AA polymorphic genotype was associated with higher HDL cholesterol concentrations (26). As our study, also Indian patients have a higher frequency of the A allele, they have low HDL cholesterol levels than Malaysian and Chinese individuals. Our results also support of the results obtained by Yılmaz et al. (27) who indicated that polymorphic gene carriers in the control group show a negative correlation between CETP (rs708272) gene polymorphism and triglyceride concentrations. Furthermore, Ağırbaşlı et al. (28) determined an association between the CETP (rs708272) polymorphism and low HDL cholesterol phenotype in children. However, this result was not as strong as that in adults. Contrary to the findings of Yılmaz et al. (27), we did not find any association between the CETP rs708272 polymorphism and biochemical parameters such as triglyceride or HDL cholesterol levels in our diabetic dyslipidemia patient group. A recent study demonstrated that a CETP rs708272 gene variant is associated with low HDL cholesterol phenotype in Turkish children (28, 29). As our patient population was on lipid-lowering drugs, insulin, and oral antidiabetics, the effects of polymorphisms on lipid parameters and glucose might be masked by drug treatments. We also suspect that these discrepancies in the results of various studies in the same population are attributable to differences between selected patient groups’ baseline characteristics and gene–gene and gene–environment interactions. Similar to our study, Yılmaz et al. (27) also demonstrated that the CETP rs708272 polymorphism was associated with an increased risk of developing type 2 diabetes mellitus in Turkish patients.

LIPC plays an important role in the metabolism of LDL, IDL, and HDL particles. The common polymorphisms of the LIPC gene change the LIPC enzymatic activity by approximately 20–30% of the individual variation (18). Population studies have indicated that lipid and lipoprotein abnormalities are messengers for developing type 2 diabetes and that LIPC plays a central role in insulin resistance in dyslipidemia. Regarding the LIPC (rs2070895) G250A polymorphism, AA (78.1%) genotype carriers significantly outnumbered GA (21.4%) and GG (0.5%) genotype carriers among our patients with diabetic dyslipidemia. Although LIPC levels were not measured, we can assume that polymorphic genotypes (AA) increase LIPC levels by causing the accumulation of visceral fat (17).

Adiponectin, resistin, PON1, and MnSOD proteins have crucial roles in glucose homeostasis. To date, studies have not fully elucidated how the ADIPOQ (rs1501299) G276T polymorphism affects insulin sensitivity and type 2 diabetes development. Şenol et al. (30) found the frequencies of G276T genotypes in patients with CAD to be 65.5% GG, 29.1% GT, and 5.5% TT in Turkish patients, which is comparable to the results of our study. Cnop et al. (31) and Tschritter et al. (32) demonstrated in large studies that plasma adiponectin concentrations are positively correlated with HDL cholesterol levels and negatively correlated with triglyceride levels. In our patient group, we found a positive correlation between ADIPOQ G276T polymorphisms and HDL concentrations. Therefore, adiponectin may have a direct role in HDL catabolism by promoting HDL assembly in the liver. Various studies have indicated that adiponectin levels are affected by multigenic control, particularly genetic variability in the ADIPOQ gene, contributing to the regulation of the protein (33, 34). Hara et al. (35) and other previous studies have shown that GG ADIPOQ gene variants had a greater effect on adiponectin levels, leading to insulin resistance and the progression of glucose intolerance to type 2 diabetes (10, 36). Contrary to these results, Ramya et al. (37) and Tsai et al. (38) demonstrated that TT genotypes were significantly associated with type 2 diabetes in Indian and Taiwan Chinese Han populations, respectively. Tu et al. (39) reported that ADIPOQ polymorphisms confer genetic susceptibility for type 2 diabetes in East Asians. ADIPOQ gene polymorphisms (rs2241766 and rs1501299) were also associated with an increased risk of developing type 2 diabetes mellitus in the Chinese Han population in a meta-analysis (38). Consistent with our results, various studies have concluded that the TT genotype plays an important role in the pathogenesis of type 2 diabetes.

In recent years, resistin was defined as a new potential signaling molecule proposed to link obesity and diabetes mellitus (11). Our results revealed that the RETN G299A polymorphism AA genotype (34.1%) was significantly associated with an increased risk of diabetic dyslipidemia compared to the control group. Ma et al. (12) concluded that obesity and the AA genotype were associated with an increased risk of type 2 diabetes. In contrast, Fehmann et al. (40) and Ochi et al. (41) did not detect any association between the RETN G299A polymorphism and type 2 diabetes mellitus. Furthermore, rs3745367 and rs1423096 RETN variants were significantly associated with resistin levels in a genome-wide association study conducted by Chung et al. (10). However, rs1423096 variants in the downstream region of RETN seem to be more associated with an increased risk of type 2 diabetes than the rs3745367 variant (42). These discrepancies could be attributed to ethnic differences between the selected populations.

PON1 is an HDL-associated antioxidant enzyme that prevents LDL oxidation (43). According to the genotype analysis of PON1 Q192R polymorphisms, the diabetic dyslipidemia and control groups were 40.7% and 36.3% heterozygous and 7.9% and 7.5% polymorphic, respectively. The R allele frequencies of the PON1 Q192R polymorphism widely vary between distinct ethnic groups; rates in Italy, Sardinia, Ethiopia, Benin, and Ecuador were reported to be 31.3%, 24.8%, 40.8%, 61.2%, and 78.9%, respectively (44). Ghanei et al. (9) reported that the QQ (AA) homozygote and R (G) alleles were associated with myocardial infarction in dyslipidemia cases, whereas only the R (G) allele was associated with myocardial infarction in patients with diabetes (p<0.05). Alegria-Torres et al. (45) identified the Q192R polymorphism of the PON1 gene as a risk factor for insulin resistance and metabolic syndrome in Mexican children. The study determined that for this population, the R allele of the PON1 Q192R polymorphism was not an independent risk factor for the development of myocardial infarction (p<0.005) (46). Gamboa et al. (13), Pérez-Herrera et al. (47), Rojas-García et al. (48), and other studies in Chinese, Afro-American, and other Mexican populations demonstrated that individuals with the PON1 192RR genotype had lower levels of HDL cholesterol than 192QQ genotype carriers. These results are similar to those in our control group. Hence, these data suggest that PON1 polymorphisms affect HDL cholesterol levels Although we did not find significant differences in the PON1 Q192R genotype distribution and allele frequencies between patients and control subjects, in the control group, we observed a negative correlation between the Q192R polymorphism and HDL cholesterol levels. This result may support the protective effect of PON1 Q192R gene polymorphisms on HDL cholesterol levels in healthy individuals.

Increased levels of ROS and endogenous antioxidants are related to various diseases such as Parkinson’s, Alzheimer’s, CV disease, pulmonary disease, diabetes, ocular disorders, and cancer, as well as aging and radiation damage (49). Nakanishi et al. (19) investigated the effects of the MNSOD Ala16Val polymorphism on the development of type 2 diabetes mellitus. Of 523 nondiabetic cases, only 65 developed type 2 diabetes during the 9.9-year-follow-up period. However, the number of homozygous TT genotype carriers (n=55) was higher than that of CC genotype carriers (n=10) in patients with type 2 diabetes mellitus. Therefore, they concluded that the MNSOD Ala16Val polymorphism plays an important role in the development of diabetes (p=0.041) (19). Li et al. (20) also demonstrated that MNSOD rs4880 gene variants confer an increased risk of type 2 diabetes mellitus to the Chinese Han population, whereas Nomiyama et al. (50) reported that Ala16Val allele frequency differences were not statistically significant between diabetic and nondiabetic patients in the Japanese population. Mitochondrial oxidative stress plays a key role in beta cell dysfunction, insulin resistance, and glucose tolerance, which leads to diabetes mellitus (20). Our data suggest that MNSOD Ala16Val polymorphisms impacts oxidative metabolism and increases the risk of diabetic dyslipidemia.

Study limitations

The main limitation of our study was the lack of whole-gene sequencing for LPL, LIPC, SCARB1, LCAT, CETP, ADIPOQ, RETN, and MNSOD genes. However, all studied SNPs were well-known mutations associated with either dyslipidemia and/or CV disease.

Conclusion

LPL (rs320), LIPC (rs2070895), SCARB (rs5888), LCAT (rs2292318), CETP (rs708272), ADIPOQ (rs1501299), RETN (rs3745367), and MNSOD (rs4880) gene polymorphisms can be used as markers for the early detection of diabetic dyslipidemia. We believe that further studies are necessary to determine the effects of other enzyme-coding gene polymorphisms and interactions on plasma lipid concentrations. Although the exact molecular mechanisms of diabetic dyslipidemia are not yet fully understood, our study results will guide further investigations to elaborate the development of diabetic dyslipidemia.