Associations of Fibroblast Growth Factor 21 Gene 3′ Untranslated Region Single-Nucleotide Polymorphisms with Metabolic Syndrome,Obesity,and Diabetes in a Han Chinese Population

Abstract

Fibroblast growth factor 21 (FGF-21) is a novel regulator for metabolic syndrome (MetS), diabetes, and obesity. However, no study has been performed on the association of these diseases with FGF-21 gene polymorphism. The aim of the study was to investigate the association of 3′ untranslated region (UTR) single-nucleotide polymorphisms (SNPs) in the FGF-21 gene with MetS, obesity, and diabetes in the Han Chinese population. A total of 291 subjects were recruited from the Han Chinese population in Sichuan province. The genotypes of FGF-21 were determined by polymerase chain reaction–restriction fragment length polymorphism. The genotypes were confirmed by sequencing. No polymorphisms were found in rs11665841 (1 of 291) and rs3745706 (2 of 291). We did not find an association between genotype frequencies of SNP rs11665896 and lipid concentration, glucose concentration, or blood pressure. The TG/GG genotype relative to the TT genotype had an age- and sex-adjusted odds ratio of 1.41 for MetS (p=0.149), 1.84 (p=0.016) for obesity (body mass index ≥25 kg/m²), and 1.19 (p=0.492) for diabetes. Genetic variation of the 3′ UTR of the FGF-21 gene was associated with obesity, however, not with MetS or diabetes.

Introduction

T he combination of excess caloric intake and a sedentary lifestyle has made metabolic syndrome (MetS), diabetes, and obesity a global epidemic in not only Western countries (Ford et al., 2002) but also in China (Gu et al., 2005; Wong and Wang, 2006). These diseases result from the interplay between adverse health behavior and genetic risk factors. Many glucose metabolism– and lipid metabolism–related gene polymorphisms were associated with obesity, MetS, and diabetes, such as adiponectin (Yang et al., 2007), peroxisome proliferator–activated receptor γ (PPARγ) (Dongxia et al., 2008), ghrelin (Choi et al., 2006), and retinol binding protein 4 (Rbp4) (Liu et al., 2008). However, despite major efforts to elucidate the underlying mechanisms whereby this variant increases the risk of this major health problem, its functional impact and its specific role in disease pathogenesis are still unclear.

Fibroblast growth factor 21 (FGF-21) is a novel metabolic regulator with multiple beneficial effects on glucose and lipid homeostasis in animal models and humans. It was suggested that FGF-21 could elevate glucose uptake in 3T3-L1 adipocytes and reduce blood glucose and triglyceride concentrations. Moreover, its overexpression in transgenic mice could prevent obesity (Kharitonenkov et al., 2005). FGF-21 was considered potentially therapeutic for obesity and fatty liver disease (Xu et al., 2009). Zhang et al. (2008) found that FGF-21 mRNA expression in both liver and adipose tissues was significantly higher in obese ob/ob mice than in normal-weight ob/ob mice, and FGF-21 expression levels in subcutaneous fat were in positive correlation with circulating FGF-21 concentrations.

MicroRNAs (miRNAs) constitute a growing class of noncoding RNAs that suppress target gene protein translation by combining with 3′ untranslated region (UTR) completely or incompletely (Bartel, 2004). miRNA binding may lead to phenotypic changes including disease susceptibility, cancer risk and drug-resistance of tumor cells and so on (Sethupahty et al., 2008). It is reported that ERα 3′UTR, which is the target binding site of miR-206 in vitro test, contains a C-T variant (rs 93410170) that enhanced binding with miR-206, and then reduced the level of ERa protein (Adams et al., 2007). In addition, SNPs near the miRNA target area also resulted in different phenotypes (Mishra et al., 2007). Therefore, the aim of our study is to investigate the association of 3′ UTR SNPs in FGF-21 gene with MetS, diabetes, and obesity in Chinese Han population.

Materials and Methods

Subjects

Subjects were recruited from Department of Physical Examination Center, West China Hospital, Sichuan University, China. Their age ranged from 50 to 90 years old. Initially, 550 subjects were recruited for this investigation. All subjects participated in a complete physical examination, including family history, history of past illness, and standard anthropometrical measurements. After exclusion of subjects with tumor, anorexia nervosa, serious liver disease, and serious kidney disease, 365 subjects were invited for further investigation, including 75 g oral glucose tolerance test (OGTT) and conventional biochemical tests. Finally, 291 subjects agreed to participate, including 130 MetS subjects, 103 diabetic subjects, 100 obesity subjects, and 141 healthy subjects. The diagnosis of MetS was according to International Diabetes Federation definition, which included central obesity (waist circumference ≥90 cm for men, ≥80 cm for women) plus any two of the following items: (1) triacylglycerol >1.7 mM; (2) high-density lipoprotein (HDL) cholesterol <1.03 mM for men and <1.29 mM for women; (3) systolic blood pressure ≥130 mmHg or diastolic blood pressure ≥85 mmHg or hypertension history; and (4) fasting plasma glucose ≥5.6 mM or diabetes history. The diagnosis of type 2 diabetes was based on 1999 World Health Organization criteria (fasting plasma glucose level ≥7.0 mM or 2-h OGTT plasma glucose level ≥11.1 mM). The World Health Organization's definition of obesity in Asians is body mass index (BMI; in kg/m²) ≥25.

All studies were approved by Ethics Committee at West China Hospital of Sichuan University. Written informed consent was obtained from all subjects before participation.

Measurement of biomarkers

Height, weight, waist circumferences, and blood pressure were measured in the fasting state by the same physician. The fasting and 2-h OGTT plasma glucose, serum triglycerides, total cholesterol, HDL and low-density lipoprotein were determined by Roche P800. Fasting plasma insulin was measured by electrochemiluminescence immunoassay (ECLIA) (Roche E170). Plasma FGF-21 levels were determined by radioimmunoassay (RIA). (Phoenix Pharmaceuticals, Inc.) using ¹²⁵I-labeled FGF-21 as tracer. Blood samples were collected with aprotinin (0.6 TIU/mL of blood) and detection was centralized.

The FGF-21 antibody was rabbit antihuman FGF-21 IgG, which is specific for human FGF-21; there are no cross-reaction between FGF-21 antibody and human FGF-6, FGF-10, FGF-18, FGF-19, FGF-20, adiponectin, visfatin, leptin, and retinol-binding protein-4 resistin. The linear range is 0.5–8.5 ng/mL, intraassay coefficient of variation (CV) <5%, and inter-CV <14%.

Selection of SNPs and genotyping

There are a total of 17 SNPs in the genomic region from ∼200 bp upstream to ∼200 bp downstream FGF-21. Thirty subjects (containing 10 healthy subjects, 8 MetS subjects, 6 diabetic subjects, and 6 obesity subjects) were randomly selected for screening the 17 SNPs by high-resolution melting. We found that only 3 SNPs had a high frequency (>0.1) of the minor allele in the 30 subjects.

Genomic DNA was extracted from peripheral blood leukocytes using the QIAamp DNA mini kit (Qiagen GmbH) following the manufacturer's instructions. Genotyping of the SNPs was performed by using polymerase chain reaction (PCR)–restriction fragment length polymorphism method. Locus-specific PCR primers were designed using the primer 5.0 software; primer pairs and specific restriction endonuclease are shown in Table 1.

Table 1.

Clinical and Biochemical Characteristics of Study Subjects

Variables	Value
Age (years)	66.6±10.9
Sex (male/female)	247/44
Waist circumference (cm)	87.9±9.4
BMI (kg/m²)	23.9±3.0
SBP (mmHg)	128.2±16.2
DBP (mmHg)	77.5±9.9
Fasting plasma glucose (mM)	6.0±1.9
2-h OGTT plasma glucose (mM)	8.7±4.4
Fasting plasma insulin (μU/mL)	5.2±3.5
Triglycerides (mM)	1.5±1.0
Total cholesterol (mM)	4.7±1.0
LDL cholesterol (mM)	2.8±0.7
HDL cholesterol (mM)	1.5±0.4
Obesity (%)	34.4
Diabetes (%)	35.4
Metabolic syndrome (%)	44.8

Data are mean±SD.

BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; OGTT, oral glucose tolerance test; LDL, low-density lipoprotein; HDL, high-density lipoprotein; SD, standard deviation.

The rs11665896 PCR primers were 5′ TGTGTGAGTGTC TGAGGGAAG 3′ (forward) and 5′ GAAGTCAAGAGAT GGAGAGCA 3′ (reverse). The rs3745706 PCR primers were 5′ GGGGGTTGTCCGGGCCAAGG 3′ (forward) and 5′ GGC CGAAGCCCCAGCTACGCTT 3′ (reverse). For rs11665896 and rs3745706, the PCR conditions were as follows: initial denaturation at 95°C for 3 min, amplification for 35 cycles of 95°C for 40 s, 64°C for 40 s, 72°C for 40 s. A final extension step at 72°C for 5 min followed the last PCR cycle. The PCR was performed in a final volume of 25 μL that contained 100 ng genomic DNA, 25 pM of each primer, 200 μM dNTP, 0.5 units Amplitaq Gold DNA polymerase (Applied Biosystems), and 2 mM MgCl₂. To analyze the rs11665896 and rs3745706 polymorphisms, PCR products were individually digested with 1 unit of BlnI and MboII restriction endonucleases at 37°C for 2 h and fragments obtained were visualized on a 2.5% agarose gel.

The rs11665841 PCR primers were 5′ CAGAGGAGGATA AGAAGAATGAGCA 3′ (forward) and 5′ TGGGCACAGT GACTCAAGC 3′ (reverse). For rs11665841, PCR conditions were as follows: initial denaturation at 95°C for 3 min, amplification for 35 cycles of 95°C for 40 s, 62°C for 40 s, and 72°C for 1 min. A final extension step at 72°C for 5 min followed the last PCR cycle. The rs11665896 PCR products were digested using 1 unit BsrI restriction endonuclease at 65°C for 4 h and fragments obtained were identified on a 2.5% agarose gel.

Statistical analysis

The normal distribution of data was expressed as mean±SD. Hardy–Weinberg equilibrium was computed for the expected genotype distribution. Genotypic and allelic distributions were compared using the Chi-square test. Odds ratios (ORs) were derived from multiple logistic regression analysis, which was carried out to evaluate the effect of the FGF-21 gene polymorphism. All statistical analyses were performed applying SPSS 11.5. A p<0.05 was considered statistically significant.

Results

Clinical and biochemical characteristics and 3′ UTR SNPs of FGF-21 in study subjects

The clinical and biochemical characteristics of all subjects are shown in Table 1. Frequencies of MetS, obesity, and diabetes were 44.67%, 34.36%, and 35.40%, respectively. No polymorphisms were found in rs11665841 (1 of 291) and rs3745706 (2 of 291). For rs11665896, the frequencies of genotypes TT, TG, and GG in our study subjects were 8.25%, 39.86%, and 51.89%, respectively. The genotypes of SNP rs11665896 were confirmed by sequencing. We did not find an association between rs11665896 and biochemical characteristics, as shown in Table 2.

Table 2.

Composition of Biochemical Characteristics Between TG/GG and TT Genotype

Variables	TT	TG/GG	p-Value
Number (N)	24	167
BMI (kg/m²)	23.1±3.4	24.0±3.0	0.178
Waist circumference (cm)	86.5±9.3	88.1±9.5	0.422
SBP (mmHg)	126.4±13.9	128.4±16.4	0.557
DBP (mmHg)	76.2±8.5	77.9±10.0	0.488
Fasting plasma glucose (mM)	5.8±1.5	6.0±2.0	0.602
2-h OGTT plasma glucose (mM)	9.0±5.6	8.7±4.3	0.747
Fasting plasma insulin (U/mL)	5.3±3.6	5.2±3.5	0.986
Triglycerides (mM)	1.6±0.3	1.5±0.3	0.290
Total cholesterol (mM)	2.8±0.7	2.8±0.7	0.894
LDL cholesterol (mM)	1.6±1.0	1.5±1.0	0.667
HDL cholesterol (mM)	4.8±0.8	4.7±1.0	0.558
FGF-21 (ng/mL)	1.8±0.3	1.8±0.5	0.947

Data are mean±SD.

Association of FGF-21 rs11665896 with MetS

The genotypes of rs11665896 were in Hardy–Weinberg equilibrium in both case and noncase groups (p>0.05 for all analyses). The genotype and allele frequencies in MetS, non-MetS group, and healthy groups are shown in Table 3. The genotype and allele frequencies were not significantly different between the MetS group and non-MetS group. No significant difference in genotype and allele frequencies was found between MetS group and healthy subjects. We also did not find an association between the genotype and allele frequencies of rs11665896 with MetS components (Table 4). As shown in Table 5, when the TG and GG genotypes were combined, the age- and sex-adjusted OR for MetS (vs. the TT genotype) was 1.41 (95% confidence interval [CI]: 0.88, 2.26; p=0.149). We estimated that the sample had a power of 83.40%.

Table 3.

The Frequency of TT, TG, and GG in Different Groups

	Genotype					Allele
	TT	TG	GG	χ²	p	T	G	χ²	p
MetS	13 (10.00)	44 (33.85)	73 (56.15)	3.83^a 0.21^b	0.147^a 0.902^b	70 (26.92)	190 (73.08)	0.37^a 0.15^b	0.545^a 0.695^b
Non-MetS	11 (6.83)	72 (44.72)	78 (48.45)			94 (29.19)	228 (70.81)
Diabetes	8 (7.77)	46 (44.66)	49 (47.57)	1.54^a 4.14^b	0.464^a 0.126^b	62 (30.10)	144 (69.90)	0.58^a 1.29^b	0.446^a 0.255^b
Non-diabetes	16 (8.51)	70 (37.23)	102 (54.26)			102 (27.13)	274 (72.87)
Obese	6 (6.00)	40 (40.00)	54 (54.00)	1.06^a 2.10^b	0.587^a 0.350^b	52 (26.00)	148 (74.00)	0.72^a 0.02^b	0.398^a 0.974^b
Non-obese	18 (9.42)	76 (39.79)	97 (50.79)			112 (29.32)	270 (70.68)
Healthy subjects^c	13 (9.22)	45 (31.91)	83 (58.87)			72 (25.44)	211 (74.56)

Compared with noncase subjects.

Compared with healthy subjects.

Healthy subjects: non-MetS, non-diabetes, and non-obese.

MetS, metabolic syndrome.

Table 4.

Comparison of Genotype and Allele Frequencies Between Metabolic Syndrome Components and Heathy Subjects

	MetS components
Polymorphisms	Central obesity (n=161)	Hypertension (n=155)	Hypertriglyceridemia (n=77)	Low-HDL (n=45)	Hyperglycemia or diabetes history (n=122)
TT (n=13)	14	16	5	2	10
TG (n=45)	67	66	34	17	53
GG (n=83)	80	73	38	26	59
χ ²	3.10	4.27	4.125	3.32	1.200
p	0.212	0.118	0.127	0.225	0.190
T (n=72)	95	98	44	21	73
G (n=211)	227	212	110	69	171
χ ²	1.24	2.76	0.50	0.16	1.32
p	0.265	0.097	0.479	0.687	0.251

Table 5.

Logistic Regression for Genetic Polymorphisms and Obesity, Diabetes and Metabolic Syndrome Components

	OR	95% CI	p
Obesity	1.84	1.12–3.01	0.016
Diabetes	1.19	0.73–1.94	0.492
MetS	1.41	0.88–2.26	0.149
MetS components
Central obesity	1.15	0.72–1.84	0.567
Hypertension	1.46	0.91–2.32	0.114
Hypertriglyceridemia	1.21	0.71–2.06	0.485
Low-HDL	0.65	0.34–1.25	0.195
Hyperglycemia	1.16	0.72–1.87	0.536

GG or TG versus TT.

Control for age and sex.

OR, odds ratio; CI, confidence interval.

Association of FGF-21 rs11665896 with diabetes

There were 103 diabetic subjects. The genotypic and allelic frequencies in the diabetes, nondiabetes, and healthy groups are shown in Table 3. The genotype and allele frequencies were not significantly different between the diabetes and nondiabetes groups. No significant difference of genotype and allele frequencies was found between the diabetes group and healthy subjects. When the TG and GG genotypes were combined, the OR of diabetes versus the TT genotype was 1.19 (95% CI: 0.73, 1.94; p=0.492) after adjustment for age and sex, as shown in Table 5. We estimated that the sample had a power of 96.08%.

Association of FGF-21 rs11665896 with obesity

The genotype and allele frequencies in the obese group, nonobese group, and healthy subjects are shown in Table 3. The genotype and allele frequencies were not significantly different between the obese and nonobese groups. No significant difference of genotype and allele frequencies was found between the obese group and healthy subjects. When the TG and GG genotypes were combined, the OR of obesity versus the TT genotype was 1.84 (95% CI: 1.12, 3.01; p=0.016) after adjustment for age and sex, as shown in Table 5. We estimated that the sample had a power of 84.61%.

Discussion

FGF-21, which is secreted by hepatic cells, has been proposed as a potent metabolic regulator in glucose and lipid metabolism. Many studies have shown a strong association between FGF-21 and obesity, hyperlipidemia, and diabetes (Kharitonenkov et al., 2005, 2007; Chen et al., 2008). When ketogenic diet was given to adenovirus-mediated FGF-21 knockdown mice, the mice had a fatty liver and hyperglycemia (Badman et al., 2007). In our study, plasma FGF-21 was positively correlated with BMI (data not shown), which is consistent with a previous study by Mashili et al. (2011). A list of obesity candidate genes, including FGF-21, was selected to draft “The Human Obesity Gene Map” (Rankinen et al., 2006). Here, we have sought to address whether 3′ UTR SNPs in the FGF-21 gene are associated with MetS, obesity, and diabetes in the Han Chinese population.

Our efforts focused on the 3′ UTR SNPs of FGF-21 because other 3′ UTR SNPs, such as PYY, have been associated with MetS traits (Shih et al., 2009). 3′ UTR SNPs impact on the regulation of miRNA by generating new targets for miRNA regulation, which could lower protein expression (Chen and Rajewsky, 2006; Landi et al., 2008); or altering miRNA targets, which could elevate protein expression (Clop et al., 2006; Adams et al., 2007). FGF-21 gene SNPs that had a high frequency (>0.1) of the minor allele in Asians were mainly concentrated in the 3′ region of the gene. In addition, it is reported that many novel loci located in the 3′ region, such as rs10811661 of CDKN2B, were associated with type 2 diabetes by genome-wide association analysis (Saxena et al., 2007). We identified 3 SNPs by screening the 3′ UTR of the FGF-21 gene, but only rs11665896 showed polymorphism in our study subjects.

FGF-21–transgenic mice were viable and resistant to diet-induced obesity (Kharitonenkov et al., 2005). Wang et al. (2008) identified an amino acid (F372) within helix 7 of the ligand-binding domain that is required for the response of PPARγ to endogenous ligands and also required for the expression of FGF-21, which shows that the FGF-21 gene was a direct target of PPARγ. Notably, rs11665896 was associated with higher risk of obesity in our study (OR: 1.84; 95% CI: 1.12, 3.01; p=0.016). Kaess et al. (2010) examined the associations of genetic variation in the FGF-21 signaling pathway with metabolic risk in subjects of white European ancestry. They found that FGF receptor 2 polymorphism (rs2071616) showed association with low-density lipoprotein cholesterol in both the CoLaus cohort (p=0.009) and men from the German Myocardial Infarction Family Study (p=0.017), but did not find association between FGF-21 gene polymorphism and metabolic risk (Kaess et al., 2010). We did not find association of rs11665896 with concentrations of plasma FGF-21, lipid concentration, and blood pressure. No study reported the relationship of rs11665896 with FGF-21 mRNA levels in liver or fat tissue. rs11665896 is about 100 bp from the 3′ end of the FGF-21 gene. rs11665896 is not predicted to affect mRNA stability and protein expression by bioinformaticsmethods (www.microrna.org; www.bioinformatics.ca). The mechanisms of FGF-21 gene rs11665896 in obesity require more study. Although obesity is a risk factor of MetS and diabetes, our study indicated that variation in the FGF-21 gene was not associated with the risk of MetS and diabetes; however, we cannot rule out that it takes part in the pathogenesis of MetS and diabetes. Further studies in other ethnic populations are needed to confirm our results.

Moreover, our study population was >50 years old. For the younger control subjects, the probability that they will develop a metabolic disease in the future is relatively high; therefore, the use of elderly subjects may be better. The sample size and the number of patients in each group were moderate in our study. In addition, the interaction between rs11665896 and the stability of FGF-21 mRNAs and protein expression levels remains to be evaluated.

In summary, FGF-21 rs11665896 was associated with a higher risk of obesity. We lack the power to assess the relationship between FGF-21 variants and MetS and diabetes in the Han Chinese population. However, further study on genetic targets may potentially improve understanding of metabolic disorders and lay a foundation for the treatment of MetS, obesity, and diabetes.

Footnotes

Acknowledgments

This study was supported by grants from the National Natural Science Foundation of China (#30900658), Sichuan University Young Scientist Funds (#2008095), and Sichuan Science and Technology Support Project (#2008SG0013).

Authors Contributions

Mei Zhang, Zhen-Mei An, and BinWu Ying participated in the research design. Mei Zhang, You-Juan Wang, and Li Zeng contributed to the performance of laboratory measurements. Mei Zhang and BinWu Ying were involved in data collection and analysis. Mei Zhang, Binwu Ying, and Zhen-Mei An wrote this article.

Disclosure Statement

The authors declare that they have no competing financial interests.

References

Adams

B.D.

, Furneaux

, White

B.A.

2007. The micro-ribonucleic acid (miRNA) miR-206 targets the human estrogen receptor alpha (ERalpha) and represses ERalpha messenger RNA and protein expression in breast cancer cell lines. Mol Endocrinol, 21:1132–1147.

Badman

M.K.

, Pissios

, Kennedy

A.R.

, Koukos

, Flier

J.S.

, Maratos-Flier

2007. Hepatic fibroblast growth factor-21 is regulated by PPARalpha and is a key mediator of hepatic lipid metabolism in ketotic states. Cell Metab, 5:426–437.

Bartel

D.P.

2004. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell, 116:281–297.

Chen

, Rajewsky

2006. Natural selection on human microRNA binding sites inferred from SNP data. Nat Genet, 38:1452–1456.

Chen

W.W.

, Li

, Yang

G.Y.

, Li

, Qi

X.Y.

, Zhu

, Tang

, Liu

, Boden

2008. Circulating FGF-21 levels in normal subjects and in newly diagnose patients with Type 2 diabetes mellitus. Exp Clin Endocrinol Diabetes, 116:65–68.

Choi

H.J.

, Cho

Y.M.

, Moon

M.K.

, Choi

H.H.

, Shin

H.D.

, Jang

H.C.

, Kim

S.Y.

, Lee

H.K.

, Park

K.S.

2006. Polymorphisms in the Ghrelin gene are associated with serum high-density lipoprotein cholesterol level and not with type 2 diabetes mellitus in Koreans. J Clin Endocrinol Metab, 91:4657–4663.

Clop

, Marcq

, Takeda

, Pirottin

, Tordoir

, Bibé

, Bouix

, Caiment

, Elsen

J.M.

, Eychenne

, Larzul

, Laville

, Meish

, Milenkovic

, Tobin

, Charlier

, Georges

2006. A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nat Genet, 38:813–818.

Dongxia

, Qi

, Lisong

, Jincheng

2008. Association of peroxisome proliferator-activated receptor γ gene Pro12Ala and C161T polymorphisms with metabolic syndrome. Circ J, 72:551–557.

Ford

E.S.

, Giles

W.H.

, Dietz

W.H.

2002. Prevalence of the metabolic syndrome among U.S. adults: findings from the third National Health and Nutrition Examination Survey. JAMA, 287:356–359.

10.

, Reynolds

, Wu

, Chen

, Duan

, Reynolds

R.F.

, Whelton

P.K.

, He

2005. Prevalence of the metabolic syndrome and overweight among adults in China. Lancet, 365:1398–1405.

11.

Kaess

B.M.

, Barnes

T.A.

, Stark

, Charchar

F.J.

, Waterworth

, Song

, Wang

W.Y.

, Vollenweider

2010. FGF21 signalling pathway and metabolic traits—genetic association analysis. Eur J Hum Genet, 18:1344–1348.

12.

Kharitonenkov

, Shiyanova

T.L.

, Koester

, Ford

A.M.

, Micanovic

, Galbreath

E.J.

, Sandusky

G.E.

, Hammond

L.J.

, Moyers

J.S.

, Owens

R.A.

, Gromada

, Brozinick

J.T.

, Hawkins

E.D.

, Wroblewski

V.J.

, Li

D.S.

, Mehrbod

, Jaskunas

S.R.

, Shanafelt

A.B.

2005. FGF-21 as a novel metabolic regulator. J Clin Invest, 115:1627–1635.

13.

Kharitonenkov

, Wroblewski

V.J.

, Koeste

R A.

, Chen

Y.F.

, Clutinger

C.K.

, Tigno

X.T.

, Hansen

B.C.

, Shanafelt

A.B.

, Etgen

G.J.

2007. The metabolic state of diabetic monkeys is regulated by FGF-21. Endocrinology, 148:774–781.

14.

Landi

, Gemignani

, Barale

, Landi

2008. A catalog of polymorphisms falling in microRNA-binding regions of cancer genes. DNA Cell Biol, 27:35–43.

15.

Liu

, Gao

J.Y.

, Zhang

J.P.

, Li

P.Q.

, Liu

J.X.

, Liu

, Xie

X.D.

2008. Evaluation of the association between retinal binding protein 4 polymorphisms and type 2 diabetes in Chinese by DHPLC. Endocrine, 34:23–28.

16.

Mashili

F.L.

, Austin

R.L.

, Deshmukh

A.S.

, Fritz

, Caidahl

, Bergdahl

, Zierath

J.R.

, Chibalin

A.V.

, Moller

D.E.

, Kharitonenkov

, Krook

2011. Direct effects of FGF21 on glucose uptake in human skeletal muscle: implications for type 2 diabetes and obesity. Diabetes Metab Res Rev, 27:286–297.

17.

Mishra

P.J.

, Humeniuk

, Mishra

P.J.

, Longo-Sorbello

G.S.

, Banerjee

, Bertino

J.R.

2007. A miR-24 microRNA binding-site polymorphism in dihydrofolate reductase gene leads to methotrexate resistance. Proc Natl Acad Sci U S A, 104:13513–13518.

18.

Rankinen

, Zuberi

, Chagnon

Y.C.

, Weisnagel

S.J.

, Argyropoulos

, Walts

, Pérusse

, Bouchard

2006. The human obesity gene map: the 2005 update. Obesity, 14:529–644.

19.

Saxena

, Voight

B.F.

, Lyssenko

, Burtt

N.P.

, de Bakker

P.I.

, Chen

, Roix

J.J.

, Kathiresan

, Hirschhorn

J.N.

, Daly

M.J.

, Hughes

T.E.

, Groop

, Altshuler

, Almgren

, Florez

J.C.

, Meyer

, Ardlie

, Bengtsson Boström

, Isomaa

, Lettre

, Lindblad

, Lyon

H.N.

, Melander

, Newton-Cheh

, Nilsson

, Orho-Melander

, Råstam

, Speliotes

E.K.

, Taskinen

M.R.

, Tuomi

, Guiducci

, Berglund

, Carlson

, Gianniny

, Hackett

, Hall

, Holmkvist

, Laurila

, Sjögren

, Sterner

, Surti

, Svensson

, Tewhey

, Blumenstiel

, Parkin

, Defelice

, Barry

, Brodeur

, Camarata

, Chia

, Fava

, Gibbons

, Handsaker

, Healy

, Nguyen

, Gates

, Sougnez

, Gage

, Nizzari

, Gabriel

SB.

, Chirn

GW.

, Ma

, Parikh

, Richardson

, Ricke

, Purcell

2007. Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science, 316:1331–1336.

20.

Sethupathy

, Collins

F.S.

2008. MicroRNA target site polymorphisms and human disease. Trends Genet, 24:489–497.

21.

Shih

P.A.

, Wang

, Chiron

, Wen

, Nievergelt

, Mahata

, Khandrika

, Rao

, Fung

M.M.

, Mahata

S.K.

, Hamilton

B.A.

, O'Connor

D.T.

2009. Peptide YY (PYY) gene polymorphisms in the 3′-untranslated and proximal promoter regions regulate cellular gene expression and PYY secretion and metabolic syndrome traits in vivo. J Clin Endocrinol Metab, 94:4557–4566.

22.

Wang

, Qiang

, Farmer

S.R.

2008. Identification of a domain within peroxisome proliferator-activated receptor gamma regulating expression of a group of genes containing fibroblast growth factor 21 that are selectively repressed by SIRT1 in adipocytes. Mol Cell Biol, 28:188–200.

23.

Wong

K.C.

, Wang

2006. Prevalence of type 2 diabetes mellitus of Chinese populations in Mainland China, Hong Kong, and Taiwan. Diabetes Res Clin Pract, 73:126–134.

24.

, Lloyd

D.J.

, Hale

, Stanislaus

, Chen

, Sivits

, Vonderfecht

, Hecht

, Li

Y.S.

, Lindberg

R.A.

, Chen

J.L.

, Jung

D.Y.

, Zhang

, Ko

H.J.

, Kim

J.K.

, Véniant

M.M.

2009. Fibroblast growth factor-21 reverses hepatic steatosis, increases energy expenditure, and improves insulin sensitivity in diet-induced obese mice. Diabetes, 58:250–259.

25.

Yang

W.S.

, Yang

Y.C.

, Chen

C.L.

, Wu

I.L.

, Lu

J.Y.

, Lu

F.H.

, Tai

T.Y.

, Chang

C.J.

2007. Adiponectin SNP276 is associated with obesity, the metabolic syndrome, and diabetes in the elderly. Am J Clin Nutr, 86:509–513.

26.

Zhang

, Yeung

D.C.

, Karpisek

, Stejskal

, Zhou

Z.G.

, Liu

, Wong

R.L.

, Chow

W.S.

, Tso

A.W.

, Lam

K.S.

, Xu

2008. Serum FGF-21 levels are increased in obesity and are independently associated with the metabolic syndrome in humans. Diabetes, 57:1246–1253.