Genetic Variants of SIRT1 Gene Promoter in Type 2 Diabetes

Type 2 diabetes (T2D) is a highly heterogeneous and polygenic disease. To date, genetic causes and underlying mechanisms for T2D remain unclear. SIRT1, one member of highly conserved NAD-dependent class III deacetylases, has been implicated in many human diseases. Accumulating evidence indicates that SIRT1 is involved in insulin resistance and impaired pancreatic β-cell function, the two hallmarks of T2D. Thus, we speculated that altered SIRT1 levels, resulting from the genetic variants within its regulatory region of SIRT1 gene, may contribute to the T2D development. In this study, the SIRT1 gene promoter was genetically analyzed in T2D patients (n = 218) and healthy controls (n = 358). A total of 20 genetic variants, including 7 single-nucleotide polymorphisms (SNPs), were identified. Five heterozygous genetic variants (g.4114-15InsA, g.4801G > A, g.4816G > C, g.4934G > T, and g.4963_64Ins17bp) and one SNP (g.4198A > C (rs35706870)) were identified in T2D patients, but in none of the controls. The frequencies of two SNPs (g.4540A > G (rs3740051) (OR: 1.75, 95% CI: 1.24–2.47, P < 0.001 in dominant genetic model) and g.4821G > T (rs35995735)) (OR: 3.58, 95% CI: 1.94–6.60, P < 0.001 in dominant genetic model) were significantly higher in T2D patients. Further association and haplotype analyses confirmed that these two SNPs were strongly linked, contributing to the T2D (OR: 1.442, 95% CI: 1.080–1.927, P < 0.05). Moreover, most of the genetic variants identified in T2D were disease-specific. Taken together, the genetic variants within SIRT1 gene promoter might contribute to the T2D development by altering SIRT1 levels. Underlying molecular mechanism needs to be further explored.


Introduction
Type 2 diabetes (T2D) is a highly heterogeneous and polygenic disease. Dysfunction of pancreatic β cells and insulin resistance in tissues are involved in the pathophysiology of T2D [1,2]. Candidate gene association, linkage, and genome-wide association studies have identifed large number of genetic loci and gene variants for T2D. A highly polygenic architecture of T2D has been established, which is dominated by common alleles with small and cumulative efects [3]. Although a small proportion of T2D cases can be explained, genetic causes and underlying molecular mechanisms of T2D remain largely unknown [4,5]. Rare and low-frequency genetic variants that modulate β-cell mass and function may account for the missing inheritance for T2D [6].
Sirtuins are NAD-dependent protein deacetylases and broadly regulate many cellular processes, including cell fate determination, DNA damage repair, cellular protection, calorie restriction, and energy metabolism. Sirtuins have been implicated in age-related diseases, such as cancer, diabetes, and cardiovascular and neurodegenerative diseases [7][8][9][10]. Tere are seven members in the mammalian sirtuin family, SIRT1-7. SIRT1 gene is highly expressed in metabolically active tissues, including the liver, muscle, adipose tissue, heart, pancreas, and brain. SIRT1 regulates glucose and lipid metabolism, mitochondrial biogenesis, stress responses, infammation, autophagy, circadian rhythms, and chromatin silencing [11]. In addition, SIRT1 is involved in the epigenetic regulation in the diferentiation of the human stem cells [12,13].
SIRT1 has been involved in insulin resistance and impaired β-cell function, which are the hallmarks of T2D [14]. In experimental animals, SIRT1 regulates insulin secretion and protects pancreatic β-cells against toxic stresses [15,16]. In mouse pancreatic beta cells, loss of SIRT1 leads to impaired glucose sensing and insulin secretion [17]. SIRT1 ameliorates insulin resistance by repressing protein tyrosine phosphatase 1B, a major negative regulator of insulin action [18]. In a T2D rat model, SIRT1 regulates glucose homeostasis and insulin sensitivity [19]. SIRT1 improves insulin sensitivity in skeletal muscle and liver [20,21]. In cultured 3T3-L1 adipocytes and human adipose tissues, SIRT1 functions as a suppressor of infammation, which is strongly associated with insulin resistance [22,23]. Terefore, SIRT1 plays an important role in the T2D development.
Te human SIRT1 gene has been mapped to chromosome 10q21.3 [24]. Te expression of the SIRT1 gene is strictly controlled at transcription level. Hypermethylated in cancer 1 (HIC1), a transcriptional repressor, directly binds the SIRT1 gene promoter and represses its transcription [25]. P53 has been shown to upregulate SIRT1 gene expression by binding to a P53-binding element [26]. E2F1, a cell cycle and apoptosis regulator, induces the expression of the SIRT1 gene [27]. Proinfammatory cytokine interferon gamma IFN-c represses SIRT1 gene expression [28]. In human pancreatic islet cells, SIRT1 is induced by gammaaminobutyric acid (GABA), which protects pancreatic beta cells against apoptosis [29]. SIRT1 gene is also regulated by extracellular-signal-regulated kinase 5 in leukemic Jurkat T cells [30].
Dysregulation of gene expression has been implicated in human diseases [31]. Variations in SIRT1 gene expression levels have been associated with obesity and T2D [32][33][34]. SIRT1 gene expression in circulating peripheral blood mononuclear cells is signifcantly associated with abdominal visceral fat accumulation [35]. In human adipose tissue, SIRT1 mRNA expression is signifcantly associated with energy expenditure and insulin sensitivity [36]. Terefore, we postulated that altered SIRT1 gene expression levels, caused by the genetic variants within its regulatory regions, may contribute to the T2D development. Identifcation and subsequent functional analysis of genetic variants in SIRT1 gene associated with T2D may provide a basis for manipulating SIRT1 gene expression with genetic approaches or pharmaceutical chemicals as potential therapies for T2D patients. In the present study, the promoter region of the SIRT1 gene was genetically analyzed in cohorts of T2D patients and controls. University, Jining, Shandong, China. T2D patients were diagnosed according to the American Diabetes Association guideline (2014) with fasting plasma glucose >7.0 mmol/L, 2-hour plasma glucose level >11.1 mmol/L, and glycated hemoglobin A1c >6.5%. Subjects with type 1 diabetes and other metabolic or endocrinological diseases were excluded from this study. Te healthy controls (n � 358, mean age: 52.76 years), including 206 males and 152 females, were recruited from Physical Examination Center in the same hospital. Subjects with family history of T2D were excluded. Tis study was approved by the Human Ethics Committee of Afliated Hospital of Jining Medical University. Informed consent was obtained from all participants. According to the power calculations for genetic association studies, more than 200 cases were included to eliminate the bias in diferent genetic models in this study [37][38][39].

Statistical Analysis.
Distributions of genetic variants were compared between T2D patients and controls using SPSS v13.0. Te frequency of single-nucleotide polymorphisms (SNPs) in T2D and control groups was tested for deviation from Hardy-Weinberg equilibrium (HWE) by Fisher's test. Pearson chi-squared test was performed to evaluate the signifcant diferences on allele and genotype frequencies between T2D patients and controls. Te statistical power was generally set as 80% for determining the sample size that may yield the acceptable probability estimates. Odds ratio (OR) values and 95% confdence intervals (CIs) were measured using unconditional logistic regression analysis. Te associations in diferent genetic models (codominant, dominant, over-dominant, and recessive) were analyzed with web-based software SNPStats. Linkage disequilibrium (LD) analysis and haplotype associations were conducted with Haploview software package (version 4.2) and SHEsis software platform. P < 0.05 was considered statistically signifcant.
Distributions of A/A, A/G, and G/G genotypes in SNP rs3740051 were 51.8%, 43.1%, and 5.0% in the T2D patient group and 65.4%, 31.8%, and 2.8% in the control group, respectively. Tere were signifcant associations between genotype frequency and distribution with the T2D patient group in codominant, dominant, and over-dominant models (P � 0.005, P < 0.001, and P � 0.006). Te G allele frequency of rs3740051 was higher in the T2D patient group (26.6%) than in the control group (18.7%) (P � 0.002).
Distributions of G/G, G/T, and T/T genotypes in SNP rs35995735 were 84.9%, 15.1%, and 0.0% in the T2D patient group and 95.3%, 4.5%, and 0.2% in the control group, respectively. Tere were signifcant associations between genotype frequency and distribution with the T2D patient group in codominant, dominant, and over-dominant models (P < 0.001, P < 0.001, and P < 0.001). Te T allele frequency of rs35995735 was higher in the T2D patient group (7.6%) than in the control group (2.5%) (P < 0.001).
In addition, distributions of A/A, A/G, and G/G genotypes in SNP rs3740053 were 52.8%, 41.7%, and 5.5% in the T2D patient group and 61.7%, 34.4%, and 3.9% in the control group, respectively. Tere was a signifcant association between genotype frequency and distribution with the T2D patient group in the dominant model (P � 0.34). Te G allele frequency of rs3740053 was higher in the T2D patient group (26.4%) than in the control group (21.1%) (P � 0.039).

Associations between Haplotypes and T2D Risk.
To further analyze the association between haplotypes and T2D, we characterized the linkage disequilibrium (LD) of the SIRT1 gene promoter SNPs in T2D patients and controls. D values and R2 values were examined with Haploview (version 4.2) and SHEsis ( Figure 2). Two SNPs (rs35995735 and rs3740053) had no linkage, and all other SNPs showed strong linkage. Furthermore, R2 values showed a strong linkage between SNPs rs3740051 and rs3740053 as well as between SNPs rs932658 and rs2394443. Tese results further confrmed that SNPs rs3740051 and rs35995735 were associated with T2D.
Te haplotypes of the fve SNPs (rs3740051, rs932658, rs35995735, rs3740053, and rs2394443) and their frequencies in T2D patients and controls are shown in Table 3. Te haplotypes G-A-G-G-G and A-A-T-A-G were associated with T2D (P < 0.05 and P < 0.001, respectively). Te haplotype A-C-G-A-C provided protection from T2D (P<0.001). Te most common haplotype A-A-G-A-G was not associated with T2D (P > 0.05).

Discussion
SIRT1 gene mutations have been reported in type 1 diabetes [43]. Genetic variations in SIRT1 gene have been related to the risk for obesity [44][45][46]. In a Dutch population, SIRT1 gene SNPs are associated with prenatal famine exosure to infuence the T2D risk [47]. In Pima Indians, an upstream variant (NC_000010.10: g.69635204T > A, rs10509291) and International Journal of Endocrinology 3 an intron variant (NC_000010.10: g.69651125A > G, rs7896005) in SIRT1 gene are associated with reduced insulin secretion and increased risk for T2D [48]. In this study, we analyzed the proximal promoter region of the SIRT1 gene and found six genetic variants in 7.3% (16/218) of T2D patients. Te frequencies of two SNPs (g.4540A > G (rs3740051) and g.4821G > T (rs35995735)) were signifcantly higher in T2D patients compared to controls. Further genotypes analysis indicated that these two SNPs had strong linkage and were signifcantly associated with T2D in codominant, dominant, and over-dominant models. Collectively, these genetic variants and SNPs may abolish, create, or modify the binding sites for transcription factors within the SIRT1 gene promoter, which then alter SIRT1 levels, contributing to the T2D development. Many downstream targets of SIRT1 have been identifed, including forkhead-box transcription factors (FOXOs), peroxisome proliferator-activated receptor c (PPARc), PPARc-coactivator 1α (PGC-1α), myogenic diferentiation 1, p53, and autophagy-related proteins. SIRT1 has been involved in insulin signaling by regulating insulin receptor substrate 2 and FOXO3 [49,50]. In adipocytes, SIRT1 increases adiponectin gene expression in maintaining energy homeostasis [51]. In transgenic mice, adiponectin improves insulin sensitivity and acts against infammation [52]. SIRT1 deacetylates FOXO1 in liver, adipose tissue, and pancreatic β-cells and protects β-cells against oxidative stress [53]. SIRT1 is involved in regulating infammatory responses, gluconeogenesis, and levels of reactive oxygen species, which contribute to the insulin resistance [54,55]. SIRT1 forms a complex with FOXA2 to regulate pancreas duodenum homeobox 1 (PDX1) gene, which is essential for pancreas development and β-cell formation [56]. In addition, SIRT1 induces autophagy by deacetylating autophagy-related (ATG) proteins, such as ATG5, ATG7, and LC3 (microtubule-associated protein 1 light chain 3 alpha) [57]. Te crosstalk between SIRT1 and autophagy has been implicated in obesity and T2D [58]. Terefore, changed SIRT1 levels may afect pancreatic β-cell functions, insulin signaling, infammation, autophagy, and other processes, contributing to the T2D development.
Genetic variants in SIRT1 gene have been associated with many human diseases. T2D and coronary artery disease are closely linked. A number of genetic loci have been identifed and shared in both diseases [59]. In patients with coronary artery disease, SIRT1 gene expression levels are signifcantly  International Journal of Endocrinology decreased [60]. SIRT1 gene SNPs have been associated with SIRT1 levels in patients with cardiovascular diseases [61]. In previous studies, we have identifed several genetic variants within the SIRT1 gene promoter in AMI patients, including g.4816G > C and g.4934G > T [40]. In this study, these two genetic variants were also found in T2D patients, providing further evidence that T2D and AMI shared common molecular mechanisms.
Tis study has limitations. SIRT1 gene expression was not measured directly with clinical samples due to lack of sample availability. Moreover, the efects of the SNPs (g.4540A > G (rs3740051) and g.4821G > T (rs35995735)) on SIRT1 gene expression need further study, as have been by previous studies [62,63]. In addition, the impact of the SNPs on the onset and progression of T2D needs to be investigated.  International Journal of Endocrinology      International Journal of Endocrinology In conclusion, we genetically analyzed the promoter region of SIRT1 gene in T2D patients and controls. Te genetic variants identifed in T2D patients may contribute to the T2D development by changing SIRT1 levels. As natural and pharmacological compounds have been identifed for regulating SIRT1 gene expression, pharmacological targeting of SIRT1 gene genetic variants may emerge as a novel therapy for T2D patients.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon reasonable request.

Ethical Approval
Tis study was approved by the Human Ethics Committee of Afliated Hospital of Jining Medical University.

Consent
Te participants provided written informed consent.

Conflicts of Interest
Te authors declare that they have no conficts of interest.

Authors' Contributions
SP, JZ, and BY were responsible for the conception and design of the study. SP, ZZ, and YZ conducted the experiments. SP and JZ performed the statistical analysis. SP and ZZ wrote the frst draft of the manuscript. JZ and BY revised the draft of the manuscript. All authors have read and approved the submitted version.