Association of the Pro12Ala, C161T polymorphisms in the PPARγ2 gene with obesity in Han nationality of Yunnan Plateau in Southwest China

Obesity is considered a major public health problem in the world. At present, the association of single nucleotide polymorphisms (SNPS) Pro12Ala and C161-T in peroxisome proliferator-activated receptor gamma 2 (PPARγ2) gene with obesity is still controversial. The aim of this study is to evaluate the relationship between the Pro12Ala and C161-T polymorphisms of PPARγ2 gene and obesity among Han nationality of Yunnan Plateau in Southwest China. The genotypes of 284 extremely overweight patients, 475 obese patients and 759 normal controls from Yongsheng County of Yunnan Province (China) were analyzed. The Pro12Ala and C161-T genotypes of PPARγ2 were detected by SNaPshot genotyping assay. The data were analyzed statistically.

The evidence for assessing the risk of PPARγ2 Pro12Ala and C161-T polymorphisms in obesity is currently controversial and needs to be reassessed. There is no related studies been reported in cold and high altitude regions. Therefore, the purpose of this study was to evaluate the association between the PPARγ2 Pro12Ala, the C161-T gene polymorphisms and the susceptibility to obesity in adults of Han nationality in the Yunnan Plateau of Southwest China.

Study population
The present case-control study is part of The China Multi-Ethnic Cohort, Yunnan region, whose baseline survey was carried out between May 2018 and September 2019 (23). A total of 1518 subjects were recruited in our study, including 284 extremely overweight subjects (26.9 ≤ BMI < 28), 475 obese subjects (BMI ≥ 28) and 759 normal controls subjects (18.5 ≤ BMI < 24). All participants, aged 30-79 years on the day of the investigation, were from the Chinese Han population in the Yongsheng County of Yunnan Province. We matched the normal weight group (named as control 1) to the extremely overweight group, the normal weight group (named as control 2) to the obese group according to age, gender and town by 1 : 1 ratio. Prior to the survey, all of the participants signed an informed consent form. This study protocol was approved by the Ethics Review Board of Kunming Medical University.
The relevant data about socio-demographic and lifestyle factors of the subjects were collected face to face by questionnaire. Weight, Height, Body Mass Index (BMI), Waist Circumference (WC) and Hip Circumference (HC) were measured by standard methods. Waist-to-hip ratio (WHR) was calculated as a ratio between waist and hips circumference. BMI was calculated as weight in kilograms divided by squared height in meters (kg/m 2 ). Height in light clothing was determined using a standard steel strip stadiometer to the nearest 0.1 cm. Weight without shoes was measured by digital electronic scale to the nearest 0.1 kg. All subjects were investigated for their systolic blood pressure (SBP) and diastolic blood pressure (DBP) to the nearest 2 mmHg by a trained nurse in a sitting position using an appropriately sized cuff and a standard mercury sphygmomanometer. In addition, after overnight fasting, blood samples for evaluating triglyceride (TG), total serum cholesterol (TC), high-density lipoprotein (HDL), lowdensity lipoprotein (LDL), and fasting blood glucose were taken from the anterior cubital vein in the morning. Biochemical markers such as TG, TC and fasting blood glucose were tested by Kunming Jinyu Medical Inspection Institute Co., Ltd. Genotyping DNA is extracted from a blood sample using the AxyPreP Blood Genomic DNA Small Volume Kit, according to the manufacturer's instructions. PCR ampli cation using a perfect match upstream primer [F: 5'-ACGGATTGATC TTTTGCTAGATAGA-3' (Pro12Ala), F: 5'-ACAACCTGCTACAAGCCCTG-3' (C161-T)] and a downstream primer [R:5'-ACATAAATGCCCCCACGTCC-3' (Pro12Ala), R:5'-GAAATGTTGGC AGTGGCTCA-3'(C161-T)] was used to create a BstUI restriction site in the PCR product. The quality of the extracted genomic DNA was determined by agarose gel electrophoresis and the content of DNA was determined by spectrophotometry. All SNP genotyping tests were performed using SNaPshot genotyping test. The PCR products of some samples were randomly selected and sent to Kunming Shuoqing Biotechnology Co., Ltd for genotype sequencing to further verify whether the typing of our SNaPshot method is correct.

Statistical analysis
Data of normal distribution were evaluated by t test and analysis of variance (ANOVA). Parameters of skew distribution were analyzed using the Wilcoxon Rank Sum test. The categorical data were analyzed by χ 2 test. Hardy-Weinberg equilibrium was tested by a goodness-of-t χ 2 test. Conditional Logistic regression was used to adjust the covariates to evaluate the odds ratio (OR) and its 95% con dence interval (CI). The signi cance criterion for all tests was set at P < 0.05. All data were analyzed using SPSS 23 statistical software.

Results
The subjects in this study were all from the natural population cohort living on the plateau area of Yunnan Province. A total of 10544 subjects were included in the baseline survey, and the obesity rate was 4.99% (526/10544), and the overweight rate was 23.93% (2523/10544). In our study, the distribution of genotype frequency was in line with Hardy-Weinberg balance test. We analyzed polymorphisms of the PPARγ2 gene Pro12Ala and C161-T in sample population and their association with the risk of obesity. Additionally, the values of fasting glucose were signi cantly higher in obesity in comparison to control 2 (P < 0.001). However, fasting glucose values were similar between extreme overweight and control 1 (P = 0.052).  Table 2 describes the genotype and allele distributions of Pro12Ala, C161-T and the two combined genotypes compared between the case group and the matched-group, respectively. The results showed that in the Pro12Ala polymorphism, the extremely overweight and the control 1 were found three genotypes. These genotypes were CC (Pro/Pro), CG (Pro/Ala) and GG (Ala/Ala). GG genotype was not detected in the obesity and the control 2, which only contained two genotypes, namely CC (Pro/Pro) and CG (Pro/Ala). In the C161-T polymorphism, all four groups showed three genotypes, namely genotypes CC, CT and TT. Statistical analysis of the two SNP loci showed that there were no signi cant differences in the Pro12Ala genotype, C161-T genotype and the two combined genotypes between the extremely overweight and the control 1, either between the obesity and the control 2 (P > 0.05). Further, the alleles of Pro12Ala and C161-T did not exhibit any signi cant association with extreme overweight or obesity among the complete sample population (P > 0.05). Notably, in the two SNP polymorphisms, allele C was the dominant allele in our study and in all previous studies. Conditional logistic regression analysis was conducted for the extreme overweight and control 1, the obesity group and control 2, respectively. Table 3 describes the odds ratio analysis before and after covariates adjustment. The results of unadjusted analysis for covariates showed that no statistically signi cant association was found between PPARγ2 Pro12Ala genotypes, C161-T genotypes, combined genotypes and extreme overweight or obesity among our sample population (P > 0.05). Interestingly, after adjusting for covariates, there was a statistically signi cant association between the two combined genotype of "CC + TT" and the extreme overweight (Calculated with the combined genotype of "CC + CC " as the reference category; P = 0.014). Compared with subjects who with combined genotype of "CC + CC", the OR for extreme overweight of those with "CC + TT" was 4.04 (95% CI = 1.33-12.33).  (14,28). In contrast to our study and the aforementioned studies, the G allele of PPARγ2 Pro12Ala has been suggested to be signi cantly associated with obesity in populations of diverse ethnicities, including Indians, Caucasians in the United States, Finns, Spanish, and Iranians (10,(29)(30)(31)(32). The potential reasons for differences between these studies and our results may be the use of diverse phenotype assessments, insu cient sample size, ethnic differences, effects of gender, different study designs, and a lack of adjustment for confounders. The distribution of Pro12Ala polymorphism also varies among different populations. According to the HAPMAP database, the Ala allele frequency ranged between 2.7% and 4.1% in Han population, 0 to 4.4% in Africans, 4.6-16% in European population, 5.3% in Koreans, and 13% in Pakistani population. The previous studies of our research group in Yunnan Province have showed that the G allele frequency of Pro12Ala was 3.17% in Blang people, 2.27% in Wa people, 1.44% in Hani people and 4.66% in Naxi people (data unpublished). In this study, the frequency of G allele in the extreme overweight and control 1 was 3.87% and 5.11%, and in the obese and control 2 was 5.37% and 5.58%, which was close to the Korean, Hui and Naxi ethnic groups, but much lower than that in Pakistan. It may be in uenced by ethnic and environmental factors, thus showing variations between different regions.
Several studies indicated that the distribution of C161-T polymorphism may varies greatly among different countries and ethnical populations. Based on our results, the T allele distribution in the extreme overweight and control 1 was 24.82% and 21.83%, and in the obese and control 2 was 24.11% and 24.95%. Our previous studies in Yunnan Province have found that the T allele frequency of C161-T was 24.50% in Blang nationality, 27.20% in Wa people, 17.35% in Hani people and 18.87% in Naxi people respectively(data unpublished). The carrying rate of T mutation gene in Blang nationality and other domestic studies was basically the same as that in this study (19,33), however, compared with foreign studies, our results were signi cantly higher than that of other ethnic groups (16.30% in Caucasians and 13.30% in Japanese women), which may be caused by ethnic differences (34,35 (37). There are also studies suggesting the opposite conclusion. The Chia PP et al. suggest that C161-T polymorphism is not associated with obesity(38). In our previous studies, we found that the association between C161-T polymorphism and obesity was not found in Yunnan Blang, Wa and Hani nationalities (data unpublished). In our study, no statistical signi cance was found between the genotype of C161-T polymorphism frequency distribution and the extreme overweight or obesity. We speculated that this might be partly due to the differences in the haplotypes of PPARγ among different ethnic groups, and Pro12Ala and C161-T were associated with other SNPs in PPARγ gene. Therefore, other variants around PPARγ Pro12Ala and C161-T may in uence the prevalence of overweight or obesity.
Our study showed that the combined genotypes "CC + TT" of PPARγ2 Pro12Ala and C161-T were associated with an increased risk of extreme overweight (OR = 4.04; 95%CI = 1.33-12.33; P = 0.014), using "CC + CC" combination genotype as reference category. As far as we know, the present study is the rst analysis of the Pro12Ala and C161-T polymorphisms in the PPARγ2 gene and obesity.
The study population in the present analysis were all from Yongsheng, Lijiang county of Yunnan province, where belongs to the typical plateau slants cold regions. Special geographical environment may produce certain effect to our study results. As a major regulator of adipocyte development, PPARγ regulates adipogenesis and adipogenesis pathways in white adipocytes (39), and is required for the differentiation and control of the thermogenesis program in brown or beige adipocytes (40). Exposure to low temperatures modulates the expression of heat shock protein (HSP) in humans and animals, and it has been reported that the HSP20-FBXO4-ubiquitin dependent pathway plays a role in mediating adipocytic function by inhibiting PPARγ activity of β-adrenergic signaling (41). On this basis, we speculated other possibilities for the results of this study, that is whether the high expression of HSP under cold exposure inhibits the activity of PPARγ and reduces its regulatory function on adipocytes, thus covering the possible association between PPARγ gene polymorphism and obesity, because the study objects are located in a cold region. Of course, this hypothesis needs to be explored in further experimental studies.
One limitation of the study is that the sample was too small to draw rm conclusions. Secondly, the results of our study may not be generalized to other populations, and it is necessary to conduct ethnomatching studies to understand whether such results exist in Han subjects from non-cold areas of the Yunnan Plateau (42). Due to the multifactorial inheritance of obesity, more research must be done to elucidate the role of different genes and SNPs. In future work, it is necessary to conduct large-scale trials to answer the correlation between this SNP and obesity and obesity-related metabolic characteristics.

Conclusion
In conclusion, our study is the rst to investigate whether the main effects of Pro12Ala and C161-T polymorphisms of PPARγ2 genes are related to the risk of obesity in Han nationality in the cold region of Yunnan Plateau. The results suggest that the combined genotypes "CC + TT" of PPARγ2 Pro12Ala and C161-T were associated with an increased risk of extreme overweight, however, Pro12Ala and C161-T polymorphisms may not be the main cause of obesity in Han nationality in the colder areas of the Yunnan Plateau. This may be due to the complexity of obesity, which needs to be further studied. individual differentiation provided a written informed consent before enrolment.

Consent for publication
Not applicable.

Availability of data and materials
The datasets supporting the conclusions of this study are included in this article and its Additional le 1.