Adiponectin gene single-nucleotide polymorphisms in patients with type 2 diabetes mellitus and nonalcoholic fatty liver disease

Background. It is generally believed that environmental and genetic factors interact with the formation of nonalcoholic fatty liver disease (NAFLD) phenotype and determine its progression. Both NAFLD and type 2 diabetes (T2D) are heterogeneous diseases with common pathogenic pathways. Adiponectin is an adipokine, which increases the sensitivity of hepatocytes and muscle to insulin, modulates energy homeostasis, glucose/lipid metabolism, and inflammatory response. A number of significant adiponectin gene polymorphisms are known in this area. The purpose of the study was to evaluate the possible association between two adiponectin gene (ADIPOQ) variants, +276 G/T (rs1501299) and –11391 G/A (rs17300539), and susceptibility to NAFLD in T2D patients of Ukrainian population. Materials and methods. Case-control study included a total of 155 persons with T2D (males/females: 77/78, age 54.55 ± 0.73 years, T2D duration 6.66 ± 0.49 years, body mass index 32.20 ± 0.43 kg/m2, waist/hip circumference 0.98 ± 0.01 m, HbA1c 7.26 ± 0.11 %) for biochemical characteristics (lipid profile, non-esterified fatty acids (NEFA), insulin, total adiponectin, etc.), including 90 T2D patients with NAFLD, 245 — with rs1501299 genotyping, 155 — with rs17300539 genotyping, and 51 sex and age-matched control subjects. The +276 G/T and –11391 G/A were determined by polymerase chain reaction — restriction fragment length polymorphism method with endonucleases Mva1269I (BsmI) and MspI (HpaII). Insulin resistance (IR) was assessed using homeostasis model assessment (HOMA) algorithm and as adipose IR (Adipo-IR, NEFAxinsulin). Unpaired Student’s t test, χ2 test and Spearman’s rank order were used. To predict the probabilities of genetic risk in NAFLD, the odds ratio (OR) and 95% confidence interval (CI) were calculated. Results. T2D patients were characterized by overweight and obesity, which were more significant in the presence of NAFLD (p < 0.01). It was accompanied by an increase in НОМА-IR (p < 0.05) and triglycerides (p < 0.001) levels. We found that Adipo-IR was higher in patients with T2D as compared to the controls (p < 0.001), and this index was significantly increased in T2D patients with NAFLD in contrast to obesity-matched persons without NAFLD (190.18 ± 22.15 vs 133.32 ± 13.58 mmol/L·pmol/L, p < 0.02), with negative correlation between Adipo-IR and adiponectin level in T2D patients with NAFLD only (rs = –0.350, p = 0.021). Stratification of nonNAFLD patients by +276G/T genotype suggests the prevalence of GTand TT-genotypes. Thus, the rs1501299 G-allele increased the risk of NAFLD in comparison with T-allele (OR = 4.44, 95% CI = 2.89–6.81, p < 0.05). We also found a significant difference in the frequency of –11391G/A between T2D and control groups, but not between the patients with and without NAFLD. We observed that the haplotype of GT/GG had been more common in T2D with NAFLD, and twice less often detected in patients without hepatic disease (33 and 16.49 %, respectively, p < 0.05). Conclusions. We can recommend Adipo-IR index as a predictive marker for the NAFLD development and the indicator for therapy success in T2D patients. We established new genetic markers (rs1501299 G-allele, rs17300539 and rs1501299 GG/GG and GT/GG haplotypes, respectively) for the risk of NAFLD development in T2D patients.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is closely related with visceral obesity and insulin resistance (IR) [1], which are early and powerful determinants of type 2 diabetes mellitus (T2D).Although NAFLD is not one of the defining criteria for metabolic (IR) syndrome, it is a common hepatic manifestation [2].On the other hand, L. Amedeo et al. (2015) are thought, that the conventional paradigm of NAFLD representing the "hepatic manifestation of the metabolic syndrome" is outdated [1].However, NAFLD is usually associated with obesity or IR and thus can be regarded as sign of the metabolic syndrome.It should be looked for in patients with obesity, diabetes or lipid disturbances [3].Dyslipidemia, IR, increased production of proinflammatory cytokines, low adiponectin, high PAI-1 levels, hypertension, and hyperglycemia are main factors that lead to NAFLD, further aggravate the course of NAFLD, and accelerate the progress of atherosclerosis and the development of cardiovascular disease (CVD) [4].The high prevalence of T2D, about 2 % of the world's population [5], the specificity of clinical manifestations and the life quality of these patients make the research of this disease and its complications especially urgent.On the other hand, obesity, hyperglycemia, T2D and hypertriglyceridemia are known as risk factors for the NAFLD development [6][7][8].It is generally believed that environmental and genetic factors interact to produce NAFLD phenotype and determine its progression [9].Both NAFLD and T2D are heterogeneous diseases.Common pathways involved in the pathogenesis of NAFLD and T2D include IR, atherogenic dyslipidemia, subclinical inflammation, oxidative stress, CVD, chronic kidney disease, and obesity [10].
Recent studies have pointed role of adipose tissue as an active endocrine organ that can secrete a number of specific adipokines (including chemokines, cytokines and hormones) which can affect energy balance and endocrine mechanisms responsible for regulating appetite, lipid metabolism, blood pressure, insulin sensitivity, and systemic inflammation.An important role in this system plays adiponectin [5].Adiponectin is an adipose tissuespecific plasma protein produced in growing adipocytes, prevents the formation of IR syndrome, increases the sensitivity of hepatocytes and muscle to insulin, possesses a protective effect on the vascular wall, modulates insulin sensitivity, energy homeostasis, glucose and lipid metabolism, and anti-inflammatory responses in the vascular system [11,12].However, as soon as fat tissue increases in volume, adiponectin concentration is reduced [13], and the low level of this hormone is combined with the further development of obesity-related disturbances [14].
Adiponectin shows protective properties in alcoholic and nonalcoholic fatty liver disease [15].In liver, adiponectin attenuates IR by increasing of insulin sensitivity [16].At present time, however, the molecular mechanisms underlying the development and the progression of NAFLD are poorly understood.To clarify the nature of the link between the gene and metabolic state of the organism, ADIPOQ SNPs are investigated, among others [17].Adiponectin is encoded by ARM1 (ADIPOQ, ACDC) gene, located on the long arm of chromosome 3 locus at 3q27 [18,19], which covers 16 kb and contains three exons and two introns.The results of genome wide study of the locus 3q27 identified its relationship with susceptibility to diabetes [20,21]: it is associated with T2D and metabolic syndrome in Japanese [21], USA [22], French (Caucasoids) [19] and Italian [23] populations.We have previously identified association of adiponectin gene (ADIPOQ) SNP +276 G/T and paraoxonase-1 gene (PON-1) SNP Q192R with the risk of T2D development in Ukrainian population [24,25].
The purpose of the present study was to evaluate the possible association between two adiponectin gene variants, +276 G/T (rs1501299) and -11391 G/A (rs17300539), and susceptibility to NAFLD in type 2 diabetic subjects of Ukrainian population.

Materials and methods
Case-control study included a total of 155 patients with a diagnosis of T2D (M/F: 77/78, age 54.55 ± 0.73, T2D duration 6.66 ± 0.49 yrs, body mass index (BMI) 32.20 ± 0.43 kg/m 2 , waist-to-hip ratio (WHR) 0.98 ± 0.01, HbA 1c 7.26 ± 0.11 %) including 90 T2Ds with NAFLD and 51 sex and age-match control subjects (C) for biochemical characteristics, 245 T2Ds and 103 C for rs1501299 genotyping, 155 T2Ds and 51 C for rs17300539 genotyping.The data were collected through a standard questionnaire.All patients were interviewed regarding a full medical history that included age, sex, occupation, duration of diabetes, mode and duration of treatment, presence of any associated illness, surgical history, personal history of smoking/alcohol/drug abuse, dietary habit and family history of diabetes.Controls were individuals with no clinically significant abnormal physical findings.None of the controls had any personal history of diabetes at the time of blood donation, which was ascertained with a questionnaire completed by each healthy volunteer.All cases and controls signed an informed consent for clinical, biochemical and genetic studies and the protocols were approved by the institutional review board of SI "V.Danilevsky Institute of Endocrine Pathology Problems of NAMS of Ukraine".
Cases and controls from study were all Caucasoids and residents of Kharkiv region (Ukraine).The cases were clinically and biochemically confirmed as T2D.The diagnosis NAFLD was verified in accordance with the recommendations of the American Gastroenterological Association (AGA) and the American Association for the Study of liver disease based on the clinical course of the disease, lipid and carbohydrate metabolism, activity of alaninaminotransferase (ALT), aspartataminotransferase (AST), ratio ALT/AST and sonographic examination [26].Nutritional status was assessed by measuring weight, height and abdominal circumference using well established techniques.
The genotype and allele frequencies of ADIPOQ SNP +276G/Т and -11391G/A are shown in Table 1.For SNP +276G/T 155 patients with T2D without NAFLD and 90 T2D patients with NAFLD were genotyped.Stratification of non-NAFLD patients by +276G/T genotype suggests the prevalence of heterozygotes GT and homozygotes TT compared with GG homozygotes.Thus, the rs1501299 G-allele increased the risk of NAFLD in comparison with T-allele (OR = 4.44, 95% CI = 2.89-6.81,p < 0.05).
The frequencies of alleles among healthy persons for polymorphic variants ADIPOQ +276G/T were p G = 0.45, p T = 0.55.The frequencies of alleles for polymorphic variants ADIPOQ +276G/T among T2D patients without NAFLD were p G = 0.42, p T = 0.58; among the patients with T2D complicated NAFLD were p G = 0.41, p T = 0.59.No statistically significant differences in allele frequencies between studied groups were found (p > 0.05).
We also found that there was a significant difference in frequency of ADIPOQ SNP -11391G/A between T2D patients and control subjects, but not between the patients with and without NAFLD.
The ADIPOQ SNP -11391G/A in controls followed Hardy-Weinberg equilibrium (HWE) (χ 2 = 0.078, p = 0.962) but in case of +276G/Т, it was out of HWE (χ 2 = 17.114, p < 0.001).The most prevalent genotype for the +276G/Т SNP was GT in both the non-NAFLD T2D patients (45.2 %) and T2D individuals with NAFLD (41.1 %).The most prevalent genotype of the -11391G/A SNP was GG in both the T2D patients with NAFLD (75.6 %) and without NAFLD (74.5 %).As shown in Table 1, genotype frequencies of +276G/Т Том 13, № 4, 2017 Міжнародний ендокринологічний Передова стаття < 0.005).The genotype frequencies also significantly different between NAFLD group and control subjects (p < 0.01).We evaluated also the frequencies of ADIPOQ -11391G/A alleles in Kharkiv population and the possibility of their further use as a prognostic marker for the risk of T2D complicated by NAFLD.Frequencies of ADIPOQ -11391G/A alleles for healthy individuals were: p G = 0.911, p A = 0.088.The frequencies of ADIPOQ -11391G/A alleles for total T2Ds were p G = 0.758, p A = 0.134.No significant differences in allele frequencies between studied groups were found (p > 0.05).The distribution of genotypes in the case and control groups corresponds to the HWE.Frequencies of ADIPOQ -11391G/A alleles for T2Ds with NAFLD were p G = 0.863, p A = 0,137.No statistically significant differences of the allele frequencies between the control group and T2Ds with NAFLD were found (p > 0.05).Also, there were any significant differences of allele frequencies between T2Ds with or without NAFLD (p G = 0.873, p A = 0.127).Thus, the initial data shown that homo-or heterozygous carriers of allele A (AA + GA) have increased risk of T2D (OR = 1.45, 95% CI 0.60-4.57),and the absence of this allele (GG-genotype) reduces the risk (OR = 0.69, 95% CI 0.29-1.22)as compared to the average for the whole population.
We did also haplotype analysis in order to get more accurate assessment about the influence of the two ADIPOQ gene polymorphisms.
As a result of a haplotypic analysis of the 2 polymorphic variants of the ADIPOQ gene in patients with T2D and NAFLD (table 2), the tendency to increase the frequency of the haplotype GT/GG was shown in comparison with the sample of patients without the liver disease, but the difference was not statistically significant (p > 0.05).It was found that the haplotype of GT/GG was more common in patients with T2D complicated by NAFLD and was twice less often detected in patients without hepatic disease (33 and 16.49 %, respectively, p < 0.05).
According to the results of our study, there was no significant difference in adiponecin levels in blood and frequency distribution of the genotypes of the polymorphic loci G276T and G11391A of the adiponectin gene in patients with T2D with and without NAFLD (p > 0.05).But in the analysis of haplotypes at these polymorphic loci, it was found that the levels of adiponectin in the group of patients with T2D and NAFLD are numerically or, in some cases statistically, lower than in the group without NAFLD (table 3).

Передова
NAFLD who lower serum total adiponectin levels without NAFLD who carried the GG/GG-or GG/AG-genotype.Moreover, in patients with T2D without NAFLD, there was no significant difference in plasma adiponectin levels between patients with different genotypes (p > 0.05).

Discussion
At present time, T2D is considered as a multifactorial, polygenic disease.The level of family risk incompatible with any of the hypotheses of monogenic inheritance, without the additional assumption of incomplete penetrance of the putative gene, which from a formally genetic point of view coincides with the polygenic hypothesis, indicates the polygenic nature.A multifactorial model of inheritance suggests that the manifestation of a disease is determined by the interaction of multiple environmental and genetic factors.By genetic factors, it means certain alleles of a number of polymorphic genes involved in the development of T2D, which in clinical practice are called alleles predisposing to the development of T2D.The study of hereditary predisposition to multifactorial diseases is extremely important for their diagnosis and selection of optimal therapy.Thus, of great practical value is the study of polymorphic markers in candidate genes, whose products are involved in the pathogenesis of a multifactorial disease.In this study, we investigated the possible association between ADIPOQ candidate gene polymorphisms of and NAFLD in the group of patients with T2D in Kharkiv population.It was recently shown that the level of fat in the liver correlated with such indicators of total fat as BMI and body fat percent.Furthermore, there is clear correlation between the liver fat accumulation of and WHR [31].In addition, large population studies have proved an association of nonalcoholic steatohepatitis with increased BMI.It was shown that 91 % of individuals with a BMI over 30 kg/m 2 had signs of steatosis by ultrasound data [32].
Adipose tissue IR is present in the majority of patients with NAFLD, whether they are obese or not, and adipose tissue lipolysis provides approximately 60 % of the fatty acids used for hepatic triglyceride synthesis [33,34].
Even though hepatic IR has been documented in T2D patients using homeostasis model assessment of IR (HOMA-IR) as a measure, there is insufficient data on adipose IR (Adipo-IR) and its relationship with the dysregulation of adipokines in T2D and NAFLD.We first found, that Adipo-IR index was significantly increased in T2D patients with NAFLD as compared to obesitymatched non-NAFLD diabetic subject with pronounced negative correlation between Adipo-IR and adiponectin level in T2D patients with NAFLD only.We found also, that the minor allele of SNP +276G/Т, but not SNP -11391G/A, had the substantial effect on adipose resistance to insulin in T2D patients with NAFLD.Therefore, we can recommend Adipo-IR index as a predictive marker for the NAFLD development and the indicator for monitoring therapy success in T2D patients.
Hipoadiponectinemia considered as unfavorable prognostic sign because adiponectin has angioprotective properties [35].In NAFLD patients, adiponectin levels are generally decreased [36].Decreased plasma adiponectin levels were observed in obese and T2D patients, as well as in the ob/ob mouse line (mice with congenital obesity and hyperglycaemia) [37,38].In our previous study hipoadiponectinemia was found in all patients with T2D without modulating influence of the NAFLD [39].This trend is confirmed in the current work, in both diabetic groups, with or without NAFLD, adiponectin le vels are significantly lower (p < 0.001) than in control subjects.Our results shown, in contrast to previous studies in the other countries [40,41], that levels of total adiponectin in circulation were characterized by no gender difference in T2D patients.Moreover, there was no effect of NAFLD on adiponectin levels in T2D patients likely due to the fact that adiponectin is synthesized exclusively by immature adipocytes and depends largely on insulin, glycemia and NEFA levels, which were not statistically different in patients with T2D with or without NAFLD (p > 0,05).On the other hand, we found that total adiponectin levels in patients with T2D and NAFLD who carried the GG/GG-and GG/AG-haplotypes of ADIPOQ SNPs +276G/T and -11391G/A were significantly lower than in T2D patients without NAFLD who carried the GG/GG-or GG/AG-genotype.
Low levels of adiponectin can promote the development of IR by blocking the phosphorylation of insulin receptor, increasing the flow of fatty acids in the liver and slowing their oxidation, thus, improving liver glucose, TG, and LDL production.So far, there are no comprehensive information about genetic determination of fat accumulation in the whole body and in the liver in humans.It is believed that gene polymorphism that exists in patients with NAFLD, associated with a large number of substances involved in the metabolism of lipids and carbohydrates in the liver.
The distribution of genotypes in the control group for ADIPOQ SNP +276G/T, which at some assumptions can be considered as a sample of the population, the ratio deviates from HWE toward high heterozygosity.The share of heterozygotes +276G/T at 1.53 times higher (p < 0.001) than the theoretically expected value at equilibrium [42].Our data correspond in part to the стаття /Leading Articlе/ but among females, the GG geno was reported to be significantly more prevalent in patients with NAFLD [43].
Genetic variation in ADIPOQ -11391G/A was associated with plasma levels of adiponectin, obesity and resistance to insulin, also tested the association between insulin resistance and -11391G and BMI, suggest that the impact of the variant adiponectin gene polymorphism on glucose homeostasis may depend on the body fat [44].Due to reduced activity and low circulating levels of adiponectin associated with an increased risk of serious diseases such as T2D and coronary heart disease, experimental data suggest that minor haplotype (by promoter polymorphisms) affects the expression of adiponectin in vivo to such measures that might be unfavorable [45].SNPs in the promoter and intron 2 regions of ADIPOQ gene play a functional role in adiponectin regulation [46].
G. Dolley et al. (2008) did not found the influence of ADIPOQ gene polymorphisms -11391G/A and -11377C/G on anthropometric indicator associations separately, but haplotype was associated with WHR.If there was a genotype -11391A and 11377C, WHR was significantly higher compared to other haplotypes.Also AC/AC-haplotype carriers had higher levels of adiponectin than the carriers of GG/GG-haplotype (p < 0.0001) [47].
Our data shown that homo-or heterozygous allele A (AA + GA) increases the risk of T2D (OR = 1.45, 95% CI 0.60-4.57),and the absence in the genotype the A-allele reduced the risk (OR = 0.69, 95% CI 0.29-1.22)compared to the average for the whole population.However, the odds ratio was not statistically significant, perhaps due to the limited amount of samples, which did not allow to achieve the minimum capacity criteria which justifies further studies with increased samples.
Thus, verification of SNPs of adiponectin gene and characteristics of its association with the level of the hormone in the circulation for patients with T2D complicated by NAFLD allow improved preventive therapy algorithm.
NAFLD is a complex disease, which may be the results of multiple loci commonly affected mutations [48,49].Previous studies have shown that polymorphisms of adiponectin gene were associated with NAFLD [50].At the same time, in our study the combination of GG/GGand GT/GG-haplotypes (33 % respectively) in patients with NAFLD was more frequent.Some combinations of alleles of different loci (haplotypes) can give a highly adapted phenotype, a combination of other alleles give phenotypes poorly adapted.Such deviations from equilibrium may mean that carriers of some genotypes are likely to have a selective advantage, possibly GT/AA have reduced adaptivity, and their absence may be due to the action of selection or the small number of samples.

Conclusions
Our results demonstrate for the first time, that Adipo-IR index was significantly increased in Ukrainian T2D patients with NAFLD as compared to obesity-matched non-NAFLD diabetic subject with pronounced negative correlation between Adipo-IR and adiponectin level in T2D patients with NAFLD only.We grounded the prospects of using the Adipo-IR index for monitoring of T2D therapy and associated liver disease development.Against the background of the research the adiponectin gene polymorphic variant haplotype association with NAFLD in T2Ds in the Kharkiv region population it have been established new markers (rs1501299 G-allele, rs17300539 and rs1501299 GG-/GG-and GT/GGhaplotypes, respectively) of an increased risk of NAFLD development in T2D patients.Thus, the contribution of various factors to the development of the NAFLD is significantly different depending on the population.Further identification of genetic markers for NAFLD risk in patients with T2D will allow to understand the main pathological mechanisms of the diseases, and to select the respectively treatment.