SLC2A9 rs1014290 Polymorphism is Associated with Prediabetes and Type 2 Diabetes

Purpose To investigate the association of the A/G rs1014290 polymorphism in SLC2A9 with type 2 diabetes (T2DM) and prediabetes mellitus (pre-DM). Patients and Methods. We enrolled 1058 patients who attended the Hebei General Hospital, Shijiazhuang, Hebei Province, China. The patients underwent general testing and oral glucose tolerance tests and were divided into three groups: 352 patients newly diagnosed with T2DM, 358 patients with pre-DM, and 348 healthy controls. The single nucleotide polymorphism (SNP) was detected by ligase detection reactions. The χ2 test, one-way ANOVA, and binary logistic regression analysis were used to analyze the results. Results In the T2DM group, the GG genotype frequency at the rs1014290 locus was significantly lower (14.8%) than it was in the healthy controls. Furthermore, the GG genotype group was associated with a reduced risk of T2DM in unadjusted and confounder-adjusted models compared with the risk in the AA genotype group. The G allele in the SLC2A9 rs1014290 locus decreased susceptibility to T2DM. In the pre-DM group, the GG and AG genotype groups had no significant correlation with the risk of pre-DM in any of the models. In the T2DM group, the uric acid level was significantly lower in the GG genotype group. In the T2DM and pre-DM groups, the HOMA-β levels were significantly higher in the GA (P  < 0.001) and GG (P  < 0.001) genotype groups than it was in the AA genotype group, and HOMA-IR was significantly lower in the GA (P  < 0.001) and GG (P  < 0.001) genotype groups than it was in the AA genotype group. Conclusion The A/G (rs1014290) SNP in SLC2A9 is closely related to the occurrence and development of diabetes.


Introduction
Type 2 diabetes (T2DM) is a complex disease characterized by insulin resistance and a relative lack of insulin. Its etiology is related to heredity and the environment. Te association between genetic polymorphisms and type 2 diabetes mellitus has been confrmed by many studies. Te study of Aka et al. in 2021 shows that KCNJ11rs5219, SLC30A8 rs13266634, and HHEXrs1111875 polymorphisms are associated with T2DM [1]. Te prevalence of T2DM and prediabetes (pre-DM) in the Chinese adult population were reported to be 11.6% and 50.1%, respectively [2]. Pre-DM is an early warning signal of T2DM. If early detection and early intervention do not occur, pre-DM will progress to T2DM, leading to delays in diagnosis and treatment.
Glucose transporter 9 (GLUT9), encoded by the solute carrier SLC2A9, can transport glucose, fructose, and/or uric acid [3,4]. Te expression and functions of GLUT9 are tissue-selective. Te abundance of GLUT9 is signifcantly increased in the liver and kidney tissue of T2DM mice [5]. In pancreatic beta cells, GLUT9 was found to be related to insulin secretion after glucose stimulation [6]. GLUT9 is also a high-throughput uric acid transporter [7][8][9][10], and serum uric acid is associated with insulin resistance and is considered an independent risk factor for T2DM [11][12][13][14]. Despite these fndings, the relationship between GLUT9 and diabetes is still unclear, and the specifc mechanism needs to be further explored. Terefore, we investigated whether the SLC2A9 rs1014290 polymorphism was associated with T2DM and pre-DM.
Many studies have shown that SLC2A9 has >60 variant sites [15]. Tese studies explored mainly the correlation between single nucleotide polymorphisms (SNPs) in SLC2A9 and hyperuricemia and gout [16,17], but a few of them focused on T2DM. One study in a Taiwan Han population showed that SNP rs1014290 in the third intron of SLC2A9 was associated with T2DM [18]. Te wild type of this gene locus is AA, the homozygous mutant is GG, and the heterozygous mutant is AG. Te GG genotype of rs1014290 can reduce the risk of T2DM, and therefore the correlation between SLC2A9 SNPs and glucose metabolism disorders is of interest. Most of the current studies have focused on the relationship between SLC2A9 SNPs and T2DM and not on pre-DM, especially in the Han population. In this study, descendants of Han nationality in Hebei Province, China, with T2DM and pre-DM were selected as the study population. Te correlation between rs1014290 and T2DM and pre-DM was investigated by identifying rs1014290 variants.

Study Subjects.
We enrolled 1058 patients who were outpatients or inpatients at the Endocrinology Department of Hebei General Hospital, Shijiazhuang City, Hebei Province, China, from January to December 2018 in this study. Among them, there were 358 patients with pre-DM (165 males and 193 females), 352 patients with T2DM (159 males and 193 females) who were controlled by diet and exercise alone and did not receive diabetes medication, and 348 healthy individuals (145 males and 203 females) in the physical examination who were matched by age and sex. We use the 1999 World Health Organization's diagnostic and classifcation criteria for diabetes to determine [19]: pre-DM, 6.1 mmol/L ≤ fasting blood glucose <7.0 mmol/L, 7.8 mmol/ L ≤ 2 h postprandial blood glucose (measured after taking 75 g glucose orally) < 11.1 mmol/L; T2DM: fasting blood glucose ≥7.0 mmol/L or 2 h postprandial blood glucose (measured after taking 75 g glucose orally) ≥ 11.1 mmol/L. Collect general information such as age, gender, body mass index, and blood pressure. Te study obtained the consent of all subjects and was approved by the Ethics Committee of Hebei General Hospital. Subjects voluntarily signed an informed consent form.

Biochemical Analysis.
We use an automatic biochemical analyzer to detect blood glucose at each time of the glucose tolerance test, uric acid (UA), and blood lipids (including TC, TG, LDL, and HDL) (Hitachi 7600-110, Japan). Detection of Glycated Hemoglobin (HbA1c) by Chromatographic Analysis (ADAMS A1c HA-8180 analyzer, Japan). Determination of serum insulin levels by electrochemiluminescence (Roche Cobas e601, Germany). Te degree of insulin resistance was assessed by calculating beta cell function (HOMA-β) and the insulin resistance index (HOMA-IR), using the HOMA2 calculator version 2.2.3 (Oxford University Diabetes Trial Unit).

Statistical Analysis.
Te data were analyzed using SPSS 22.0 (IBM Corp., Armonk, NY, USA). Quantitative data conforming to a normal distribution were analyzed by one-way ANOVA. Te Pearson's χ 2 test was applied to verify whether each genotype conforms to Hardy-Weinberg equilibrium. Te comparison of genotype and allele frequencies between groups was performed using a simple χ 2 test, and the risk of T2DM and pre-DM among the genotype groups was assessed by binary logistic regression analysis. All P values were two-tailed, and P values <0.05 were considered signifcant.

Comparison of the General Data between the Tree Groups.
Systolic blood pressure and BMI in the T2DM and pre-DM groups were signifcantly higher than those in the healthy control group (P < 0.05), whereas UA, TC, TG, LDL, HbA1c, FPG, and PBG levels were signifcantly higher than those in the control group, and the diferences were statistically signifcant (P < 0.05). HOMA-IR in the T2DM group was higher than that in the control group, whereas there was no signifcant diference in HOMA-IR between the pre-DM and control groups. Te levels of HDL, fasting serum insulin, and HOMA-β in the T2DM and pre-DM groups were signifcantly decreased compared with their levels in the control group (P < 0.001). Comparisons of the general data and biochemical indexes for the three groups are provided in Table 1.

Comparison of the Genetic Correlation of the SLC2A9 A/G (rs1014290) SNP in the Tree Groups.
Te expression of the three genotypes of the SLC2A9 rs1014290 SNP conforms to Hardy-Weinberg equilibrium. In the pre-DM group, the frequencies of the GA and GG genotypes were 50.3% and 17.6%, respectively, and the frequency of the G allele (42.7%) was lower than its frequency in the healthy controls, but the diference was not statistically signifcant. In the T2DM group, the frequencies of the GA and GG genotypes were 48.9% and 14.8%, respectively, and the frequency of the G allele (39.2%) was also signifcantly lower than its frequency in the controls. Analysis of the SLC2A9 A/G comparisons, dominant ((GG + AG) vs. AA), additive (GG vs. AA), recessive (GG vs. (AG + AA)), and codominant ((AA + GG) vs. AG), showed that the recessive and additive models were both signifcantly correlated with T2DM (P � 0.05, P � 0.021). In the pre-DM group, no statistically signifcant diference was found for the dominant, additive, recessive, and codominant comparisons (P � 0.520, P � 0.316, P � 0.342, and P � 0.880 respectively). Te results are shown in Table 2.
In the T2DM group, the GG genotype was associated with a reduced risk of T2DM in both the unadjusted and adjusted models. In the model that was not adjusted for confounding factors, the odds ratio (OR) was 0.595 (95% confdence index (CI) 0.383-0.925, P � 0.021). In model 1, which was adjusted for age, sex, and BMI, the OR was 0.375 (95% CI 0.197-0.646, P � 0.001). In model 2, which was adjusted for age, sex, BMI, and UA, the OR was 0.278 (95% CI 0.078-0.990, P � 0.048). In model 3, which was adjusted for age, sex, BMI, UA, TC, TG, HDL, and LDL, the OR was 0.248 (95% CI 0.063-0.966, P � 0.045). In the T2DM group, the association of the AG and AA genotypes with reduced risk was only signifcantly diferent in model 1 (OR � 0.484, 95% CI 0.265-0.884, P � 0.018) ( Table 3).
In the pre-DM group, the GG and AG genotypes had no signifcant correlation with the risk of pre-DM in the unadjusted and adjusted models (Table 4).

Correlation of SLC2A9 rs1014290 with Diferent Clinicopathological Characteristics of T2DM and Pre-DM Groups.
In the T2DM group, the levels of UA, FPG, PBG, and HbA1c in the GG genotype group were signifcantly decreased compared with their levels in the GA and AA genotype groups. Te levels of TC and TG in the GG genotype group were signifcantly decreased compared with their levels in the AA genotype group. In the pre-DM group, the FPG level in the GG genotype group was signifcantly decreased compared with its levels in the AA and GA genotype groups, whereas the UA, TG, TC, and HbA1c levels were not signifcantly diferent among the three genotype groups. In both the T2DM and pre-DM groups, the HOMA-β levels in the GA (P < 0.001) and GG (P < 0.001) genotype groups were signifcantly higher than those in the AA genotype group. However, HOMA-IR was signifcantly lower in the GA (P < 0.001) and GG (P < 0.001) genotype groups compared with its level in the AA genotype group (Tables 5  and 6).

Discussion
Te results of this study show that the serum UA level in the T2DM and pre-DM groups was signifcantly higher than it was in the healthy control group. Signifcant increases in TC, TG, LDL, HbA1c, FPG, and PBG levels were also found in the T2DM and pre-DM groups compared with their levels in the control group. Te number of patients with the GG genotype was signifcantly lower in the T2DM group than it was in the control group. Te risk of developing T2DM was signifcantly reduced for the GG genotype compared with the risk for the AA genotype, and this association persisted after adjusting for confounders of UA. Te serum UA, TG, TC, FPG, PBG, and HbA1c levels in the GG genotype group were signifcantly decreased compared with their levels in the AA and GA genotype groups. Tese fndings indicate the GG genotype was signifcantly associated with lower blood UA levels and a lower risk of T2DM compared with the other two genotypes. GLUT9 [20] was once classifed as a glucose and/or fructose transporter and, although its transport activity is very low, GLUT9 has a large efect on glucose metabolism [7,21,22]. GLUT9 is upregulated in the liver and kidney tissues of diabetic mice and afects glucoseinduced insulin secretion in pancreatic β cells [6,[23][24][25]. In two of these studies [6,25], small interfering RNA knockdown of GLUT9 in MIN6 and INS cells resulted in reduced cellular ATP levels that correlated with reductions in glucose-stimulated insulin secretion. Tese studies confrmed that GLUT9 was expressed in murine and human β-cells and suggested that GLUT9 may participate in glucose sensing in β-cells.
In previous studies, we also showed that GLUT9 was upregulated in the renal tissues of hyper-lipid-induced insulin resistant mice, along with elevated serum UA, which can be ameliorated by resveratrol intervention [26].
More recently, GLUT9 has been identifed as a UA transporter, as well as being a glucose and/or fructose transporter [7,23,27,28]. Increased UA levels were found to be associated with the incidence of T2DM, and every 60 μmol/L increase of UA was associated with a 1.17 times an increase in T2DM incidence [14,28]. Te incidence of T2DM in patients with hyperuricemia is about 35% [11]. UA afects the risk of diabetes by afecting glucose metabolism and insulin secretion of pancreatic β-cells [29,30]. Other studies have shown that UA was closely correlated with FPG, fasting serum insulin, and PBG, and was positively proportional to HOMA-IR [31,32]. Te occurrence of diabetes and hyperuricemia is closely related to eating habits, but genetic background may be the internal source of the occurrence and development of T2DM and per-DM.
SLC2A9 encodes GLUT9 [33], and SNPs in SLC2A9 are not only associated with UA levels, but also with pancreatic β-cell function and diabetes [31,[34][35][36]. Tis association is infuenced by region, ethnicity, and sex [7]. SNP rs1014290 is an A/G variant located in the third intron of SLC2A9 [37,38]. Previous studies have shown that the SLC2A9 rs1014290 SNP was associated with the risk of diabetes, and that the GG genotype played a protective role in the occurrence of T2DM [18,33,39]; however, another study concluded that this protective role was not related to the level of blood UA [40]. Only a few studies have investigated the association between the rs1014290 SNP and T2DM and pre-DM, and no relevant studies have been conducted in northern China. In this study, a northern Chinese cohort was divided into T2DM and pre-DM groups. In the T2DM group, the GG genotype and G allele frequencies were signifcantly lower than they were in the healthy control group. In the pre-DM group, there were no signifcant diferences in the GA and GG genotype and G allele frequencies compared with those in the control group. However, the FPG level was signifcantly decreased and the HOMA-β level was signifcantly higher in the GG genotype group than they were in the AA genotype group, and HOMA-IR was signifcantly lower than it was in the AA genotype group. Te binary regression analysis showed that the risk of developing T2DM was signifcantly reduced in GG genotype group compared with the risk in the AA genotype group. Studies have shown that the G allele in GLUT9 rs1014290 locus reduced susceptibility to T2DM.
Tis study has a number of limitations. (1) To avoid the infuence of diferent regions on the results, all participants were from northern China. Te pathogenesis of diabetes is complex and is known to be afected by heredity, environment. And dietary habits, but the role of environmental factors was not considered in this study. To confrm the efects of the SLC2A9 A/G (rs1014290) SNP on diabetes, large-scale clinical data are needed. (2) GLUT9 expression was not measured directly in this study, so the efects of the A/G (rs1014290) SNP on GLUT9 expression and its impact on the onset and progression of pre-DM and T2DM needs further study. (3) SLC2A9 has multiple SNP loci, but in this study only a single site rs1014290 in intron 3 was considered. Given the complexity and conjunction of gene regulation, further studies that include more SNP loci are needed to guide the diagnosis and treatment of clinical T2DM and pre-DM.

Conclusion
Te SLC2A9 A/G (rs1014290) SNP is closely related to the occurrence and development of pre-DM and T2DM. Te G allele in the SLC2A9 rs1014290 locus reduced the susceptibility to T2DM. Our fndings need to be replicated in a larger group of patients with diferent ethnic backgrounds to confrm this relationship.

Data Availability
Te data can be obtained by e-mail request to the corresponding author.

Conflicts of Interest
Te authors declare that there are no conficts of interest.