The common rs9939609 gene variant of the fat mass- and obesity-associated gene FTO is related to fat cell lipolysis

Kerstin Wåhlén, Eva Sjölin, Johan Hoffstedt

Journal of Lipid Research · 2007 · 242 citations · 12 references

DOIFull text

Open access

Concepts

Abstract

We investigated the rs9939609 single nucleotide polymorphism of the FTO gene in relation to fat cell function and adipose tissue gene expression in 306 healthy women with a wide range in body mass index (18–53 kg/m2). Subcutaneous adipose tissue biopsies were taken for fat cell metabolism studies and in a subgroup (n = 90) for gene expression analyses. In homozygous carriers of the T-allele, the in vitro basal (spontaneous) adipocyte glycerol release was increased by 22% (P = 0.007) and the in vivo plasma glycerol level was increased by ∼30% (P = 0.037) compared with carriers of the A allele. In contrast, there were no genotype effects on catecholamine-stimulated lipolysis or basal or insulin-induced lipogenesis. We found no difference between genotypes for adipose tissue mRNA levels of FTO, hormone-sensitive lipase, adipose triglyceride lipase, perilipin, or CGI-58. Finally, the adipose tissue level of FTO mRNA was increased in obesity (P = 0.002), was similar in subcutaneous and omental adipose tissue, was higher in fat cells than in fat tissue (P = 0.0007), and was induced at an early stage in the differentiation process (P = 0.004). These data suggest a role of the FTO gene in fat cell lipolysis, which may be important in explaining why the gene is implicated in body weight regulation. We investigated the rs9939609 single nucleotide polymorphism of the FTO gene in relation to fat cell function and adipose tissue gene expression in 306 healthy women with a wide range in body mass index (18–53 kg/m2). Subcutaneous adipose tissue biopsies were taken for fat cell metabolism studies and in a subgroup (n = 90) for gene expression analyses. In homozygous carriers of the T-allele, the in vitro basal (spontaneous) adipocyte glycerol release was increased by 22% (P = 0.007) and the in vivo plasma glycerol level was increased by ∼30% (P = 0.037) compared with carriers of the A allele. In contrast, there were no genotype effects on catecholamine-stimulated lipolysis or basal or insulin-induced lipogenesis. We found no difference between genotypes for adipose tissue mRNA levels of FTO, hormone-sensitive lipase, adipose triglyceride lipase, perilipin, or CGI-58. Finally, the adipose tissue level of FTO mRNA was increased in obesity (P = 0.002), was similar in subcutaneous and omental adipose tissue, was higher in fat cells than in fat tissue (P = 0.0007), and was induced at an early stage in the differentiation process (P = 0.004). These data suggest a role of the FTO gene in fat cell lipolysis, which may be important in explaining why the gene is implicated in body weight regulation. Obesity is becoming a serious health problem worldwide as a result of its association with various disorders, including type 2 diabetes mellitus and cardiovascular disease. In searching for gene variation implicated in obesity development, the recent discoveries of strong associations with obesity of common single nucleotide polymorphisms (SNPs) in the fat mass- and obesity-associated gene FTO in large populations are of particular interest. In a genome-wide search, Frayling et al. (1.Frayling T.M. Timpson N.J. Weedon M.N. Zeggini E. Freathy R.M. Lindgren C.M. Perry J.R. Elliott K.S. Lango H. Rayner N.W. et al.A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science. 2007; 316: 889-894Crossref PubMed Scopus (3250) Google Scholar) found that several SNPs of the FTO gene, including rs9939609, were associated with increased body mass index (BMI), and after additional studies in almost 39,000 subjects, it was concluded that homozygous carriers of the risk allele weighed 3 kg more than those devoid of the allele, with an increased odds ratio for obesity of 1.67 and a population-attributable risk of 20%. Interestingly, these findings have been replicated in a case-control study by Dina et al. (2.Dina C. Meyre D. Gallina S. Durand E. Korner A. Jacobson P. Carlsson L.M. Kiess W. Vatin V. Lecoeur C. et al.Variation in FTO contributes to childhood obesity and severe adult obesity.Nat. Genet. 2007; 39: 724-726Crossref PubMed Scopus (1206) Google Scholar) identifying an at-risk haplotype of the FTO gene, which showed a population-attributable risk of 22% for common obesity, and in a genome-wide association scan associating the FTO gene with obesity-related traits (3.Scuteri A. Sanna S. Chen W-M. Uda M. Albai G. Strait J. Najjar S. Nagaraja R. Orrú M. Usala G. et al.Genome-wide association scan shows genetic variants in the FTO gene are associated with obesity-related traits.PLoS Genet. 2007; 3: e115Crossref PubMed Scopus (1268) Google Scholar). All of the at-risk FTO gene variants are located within a 47 kb region encompassing the second exon as well as parts of the two first introns of the FTO gene (1.Frayling T.M. Timpson N.J. Weedon M.N. Zeggini E. Freathy R.M. Lindgren C.M. Perry J.R. Elliott K.S. Lango H. Rayner N.W. et al.A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science. 2007; 316: 889-894Crossref PubMed Scopus (3250) Google Scholar, 2.Dina C. Meyre D. Gallina S. Durand E. Korner A. Jacobson P. Carlsson L.M. Kiess W. Vatin V. Lecoeur C. et al.Variation in FTO contributes to childhood obesity and severe adult obesity.Nat. Genet. 2007; 39: 724-726Crossref PubMed Scopus (1206) Google Scholar). Hitherto, there have been no clues to help define any of these SNPs as functional. The human FTO gene is widely expressed in both fetal and adult tissues, including adipose tissue, with the highest relative levels found in the brain (1.Frayling T.M. Timpson N.J. Weedon M.N. Zeggini E. Freathy R.M. Lindgren C.M. Perry J.R. Elliott K.S. Lango H. Rayner N.W. et al.A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science. 2007; 316: 889-894Crossref PubMed Scopus (3250) Google Scholar). The findings of higher waist circumference as well as higher subcutaneous fat mass in individuals carrying the rs9939609 risk allele (1.Frayling T.M. Timpson N.J. Weedon M.N. Zeggini E. Freathy R.M. Lindgren C.M. Perry J.R. Elliott K.S. Lango H. Rayner N.W. et al.A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science. 2007; 316: 889-894Crossref PubMed Scopus (3250) Google Scholar) suggest that the FTO gene's effects on adipocyte function may be of physiological importance. Therefore, this study was designed to investigate the association of the rs9939609 SNP of the FTO gene with the metabolic function of fat cells and to explore the gene expression levels in various adipose materials. This study comprised 306 women. They were healthy and free of medication and were recruited to study the influence of genetic variance on adipocyte metabolism. BMI ranged between 18 and 53 kg/m2, and age ranged between 20 and 72 years. All subjects were living in the Stockholm area and were at least second generation Scandinavian. None was completely sedentary or involved in athletic performances. All ate a standard Swedish diet. None had undergone a slimming effort or experienced a change in body weight (>1 kg) within 6 months before the study, according to self-report. At ∼7:30 AM after an overnight fast, a venous blood sample was obtained for DNA extraction and for analyses of plasma levels of glucose, insulin, triglycerides, cholesterol, HDL cholesterol, and glycerol, which were performed by the hospital's accredited chemistry laboratory. Insulin sensitivity was indirectly assessed (homeostasis model assessment) by a formula based on glucose and insulin values (4.Matthews D.R. Hosker J.P. Rudenski A.S. Naylor B.A. Treacher D.F. Turner R.C. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man.Diabetologia. 1985; 28: 412-419Crossref PubMed Scopus (24799) Google Scholar). Systolic and diastolic blood pressures were measured in the supine position after 15 min of rest. Measurement of total body fat was obtained by a formula based on age, BMI, and sex (5.Gallagher D. Visser M. Sepulveda D. Pierson R.N. Harris T. Heymsfield S.B. How useful is body mass index for comparison of body fatness across age, sex, and ethnic groups?.Am. J. Epidemiol. 1996; 143: 228-239Crossref PubMed Scopus (1070) Google Scholar). This formula is accurate compared with more direct estimates of body fat content (6.Dahlman I. Dicker A. Jiao H. Kere J. Blomqvist L. Harmelen V.van Hoffstedt J. Borch-Johnsen K. Jorgensen T. Hansen T. et al.A common haplotype in the G-protein-coupled receptor gene GPR74 is associated with leanness and increased lipolysis.Am. J. Hum. Genet. 2007; 80: 1115-1124Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar). Plasma glycerol divided by body fat was used as an indirect measure of lipolysis in vivo. An adipose sample (1–2 g) was obtained by needle biopsy from the abdominal subcutaneous area under local anesthesia. The study was approved by the ethics committee at Karolinska University Hospital, and informed consent was obtained from all participants. The adipose tissue was treated with collagenase, and isolated fat cells were collected and subjected to lipolysis and lipogenesis experiments as described (7.Lofgren P. Hoffstedt J. Naslund E. Wiren M. Arner P. Prospective and controlled studies of the actions of insulin and catecholamine in fat cells of obese women following weight reduction.Diabetologia. 2005; 48: 2334-2342Crossref PubMed Scopus (63) Google Scholar). In lipolysis studies, adipocytes were incubated in the absence (basal) or presence of increasing concentrations of noradrenaline. After 2 h of incubation, the medium was removed the for measurement of glycerol, which is a quantitative marker for lipolysis. Glycerol release was related to the number of incubated adipocytes. Noradrenaline caused a concentration-dependent stimulation of glycerol release that reached a plateau at the highest agonist concentration. Consequently, it was always possible to determine the concentration of agonist producing a half-maximum effect (sensitivity) as well as the maximal effect. In lipogenesis studies, adipocytes were incubated with radioactive glucose without (basal) or with increased concentrations of insulin. Incorporation into lipid was related to adipocyte number (nmol/2 h/107 cells), and the EC50 corresponding to insulin receptor sensitivity and maximum effect were calculated. Genotype determination of the rs9939609 T>A polymorphism of the FTO gene (http://www.ncbi.nlm.nih.gov/SNP) was accomplished using a TaqMan-based method (Applied Biosystems, Foster City, CA; C_30090620_10). Genotyping failed in seven subjects, who were excluded from further studies. In 90 subjects, 76 obese (BMI > 30) and 14 nonobese (BMI < 26), there was subcutaneous fat tissue available also for gene expression studies of mRNA. In 18 obese women, we also obtained omental adipose tissue in association with abdominal gastric banding surgery. These mRNA analyses were performed as described (8.Mairal A. Langin D. Arner P. Hoffstedt J. Human adipose triglyceride lipase (PNPLA2) is not regulated by obesity and exhibits low in vitro triglyceride hydrolase activity.Diabetologia. 2006; 49: 1629-1636Crossref PubMed Scopus (75) Google Scholar) using a SYBR Green-based real-time PCR (Bio-Rad Laboratories, Inc., Hercules, CA). The primers used for 18S (reference gene), hormone-sensitive lipase (HSL), adipose triglyceride lipase (ATGL), perilipin, and peroxisome proliferator-activated receptor γ transcript variant 2 have been described elsewhere (8.Mairal A. Langin D. Arner P. Hoffstedt J. Human adipose triglyceride lipase (PNPLA2) is not regulated by obesity and exhibits low in vitro triglyceride hydrolase activity.Diabetologia. 2006; 49: 1629-1636Crossref PubMed Scopus (75) Google Scholar, 9.Ryden M. Arvidsson E. Blomqvist L. Perbeck L. Dicker A. Arner P. Targets for TNF-alpha-induced lipolysis in human adipocytes.Biochem. Biophys. Res. Commun. 2004; 318: 168-175Crossref PubMed Scopus (152) Google Scholar, 10.van Harmelen V. Ryden M. Sjolin E. Hoffstedt J. A role of lipin in human obesity and insulin resistance: relation to adipocyte glucose transport and GLUT4 expression.J. Lipid Res. 2007; 48: 201-206Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). The primers used for FTO were 5′-GACTGCCGAGGAACGAGAG-3′ (sense) and 5′-GGGTCAGATAAGGGAGCCAAG-3′ (antisense), and those used for the lipase-regulating gene CGI-58 were 5′-AGACCCAGGTTTGACAGTGATG-3′ (sense) and 5′-AGTAAGCAGCAGCCAAGAATCC-3′ (antisense). In 14 subjects, there was subcutaneous tissue available also for the comparison of adipose cell versus tissue mRNA expression. From 10 subjects, some of the subcutaneous adipose tissue was used to study preadipocytes. These cells were isolated and differentiated into adipocytes, as described previously (9.Ryden M. Arvidsson E. Blomqvist L. Perbeck L. Dicker A. Arner P. Targets for TNF-alpha-induced lipolysis in human adipocytes.Biochem. Biophys. Res. Commun. 2004; 318: 168-175Crossref PubMed Scopus (152) Google Scholar). The cells reached full differentiation after 12–14 days. At days 4, 8, and 12, cells were lysed for the isolation of total RNA. The Hardy-Weinberg equilibrium test was applied to ensure independent segregation of alleles. Parameter distributions were normalized when necessary by 10logarithm transformation. Testing for differences among parameters between genotype groups was performed using ANOVA, Student's unpaired t-test, or analysis of covariance with BMI as covariate. The analyses were performed using StatView version 6.0 (Stata Corp., College Station, TX). Values are expressed as means ± SEM. The genotype distribution of the FTO rs9939609 T>A polymorphism was in Hardy-Weinberg equilibrium, and the frequency of the minor allele A was 43%, which is in agreement with the allele frequency of Centre d'Etude du Polymorphisme Humain Europeans (45%) reported by the International HapMap project (www.hapmap.org). This study was designed to investigate the association of the rs9939609 polymorphism on various aspects of fat cell metabolism. As shown in Table 1, no association of the FTO genotypes and various clinical data were found. With respect to fat cell function, there were no associations between the FTO genotypes and maximal noradrenaline-induced lipolysis, basal or maximal insulin-stimulated lipogenesis, or EC50 for noradrenaline or insulin (Table 2). However, there was a significant difference in basal adipocyte lipolysis between genotypes. Therefore, we pooled the AA homozygotes (n = 49) with the heterozygotes (AT; n = 158) and made comparisons with the TT homozygous women (n = 92). As seen in Fig. 1, a 22% greater level of in vitro basal unstimulated adipocyte glycerol release (μmol glycerol/107 cells) was found in TT subjects (10.1 ± 0.8) than in TA/AA subjects (8.3 ± 0.5; P = 0.007). This was accompanied by a 28% greater in vivo plasma glycerol level of TT homozygotes (3.2 ± 0.4 vs. 2.5 ± 0.1 glycerol/kg body fat; P = 0.037) (Fig. 1).TABLE 1.Clinical data in subjects genotyped for the rs9939609 SNP of the FTO geneGenotypeCharacteristicTTATAAPNumber9215849Age (years)38 ± 139 ± 140 ± 10.72BMI (kg/m2)33 ± 134 ± 135 ± 10.44Waist (cm)103 ± 2103 ± 2107 ± 30.43Body fat (%)43 ± 144 ± 146 ± 20.40P-Glucose (mmol/l)5.4 ± 0.15.3 ± 15.3 ± 10.69P-Insulin (mU/l)12 ± 112 ± 112 ± 10.82Homeostasis model assessment (index)3.1 ± 0.32.9 ± 0.23.0 ± 0.30.84P-Triglycerides (mmol/l)1.4 ± 0.11.4 ± 0.11.2 ± 0.10.57P-Cholesterol (mmol/l)4.8 ± 0.15.0 ± 0.15.1 ± 0.10.11P-HDL cholesterol1.3 ± 0.041.3 ± 0.031.4 ± 0.060.60P-Glycerol/kg body fat3.2 ± 0.42.5 ± 0.22.5 ± 0.20.11Systolic blood pressure (mmHg)125 ± 2124 ± 1123 ± 20.66Diastolic blood pressure (mmHg)76 ± 176 ± 177 ± 10.91BMI, body mass index; SNP, single nucleotide polymorphism. Values are means ± SEM and were compared using ANOVA. Open table in a new tab TABLE 2.Fat cell function in subjects genotyped for the rs9939609 SNP of the FTO geneGenotypeVariableTTATAAPNumber9215849Lipolysis (μmol glycerol/107 cells) Basal10.1 ± 0.88.2 ± 0.58.8 ± 0.90.03 Maximal (−basal)12.9 ± 0.612.9 ± 0.612.9 ± 0.90.98 EC50−8.3 ± 0.1−8.2 ± 0.1−8.1 ± 0.20.85Lipogenesis (nmol glucose/2 h/107 cells) Basal2.4 ± 0.32.5 ± 0.22.4 ± 0.40.89 Maximal (−basal)3.4 ± 0.43.4 ± 0.23.1 ± 0.40.95 EC50−12.9 ± 0.2−12.8 ± 0.1−13.2 ± 0.20.10Values are means ± SEM and were compared using analysis of covariance with BMI as covariate. Open table in a new tab BMI, body mass index; SNP, single nucleotide polymorphism. Values are means ± SEM and were compared using ANOVA. Values are means ± SEM and were compared using analysis of covariance with BMI as covariate. Table 3 shows the association of the FTO polymorphism on mRNA expression levels of FTO and various genes implicated in lipolysis regulation, including HSL, ATGL, perilipin, and CGI-58. However, no difference in mRNA expression levels between TT, AT, and AA subjects was found for any of the genes analyzed.TABLE 3.FTO rs9939609 SNP in relation to adipose tissue gene expression levelsGenotypemRNATTATAAPNumber274617FTO/18S (au)1.8 ± 0.11.7 ± 0.11.8 ± 0.10.21Hormone-sensitive lipase/18S (au)4.6 ± 0.44.8 ± 0.35.0 ± 0.50.52Adipose triglyceride lipase/18S (au)5.9 ± 0.45.8 ± 0.35.6 ± 0.50.72Perilipin/18S (au)3.7 ± 0.23.8 ± 0.23.7 ± 0.40.76CGI-58/18S (au)1.4 ± 0.11.1 ± 0.11.3 ± 0.20.55au, arbitrary units. Values are means ± SEM and were compared using analysis of covariance with BMI as covariate. Open table in a new tab au, arbitrary units. Values are means ± SEM and were compared using analysis of covariance with BMI as covariate. Finally, we investigated the relative expression levels of FTO mRNA in various adipose materials. As seen in Fig. 2A, whereas obese (BMI > 30) compared with nonobese (BMI < 26) subjects showed an increased level of subcutaneous adipose tissue FTO mRNA, no difference in FTO gene expression between omental and subcutaneous adipose tissue was found (Fig. the relative expression levels in adipocytes and adipose tissue, we found that there is an of FTO mRNA in fat cells (Fig. In we that FTO mRNA expression was induced at an early stage in the differentiation process (Fig. between FTO mRNA and lipolysis was for a with plasma glycerol (P = The the effect of FTO on body fat is L. From in to human obesity.Nat. Genet. 2007; 39: PubMed Scopus Google Scholar). This study some for the first on the possible by which this gene may fat We that healthy women, who are homozygous for the more common FTO allele, have ∼30% increased in vivo as glycerol for total body compared with independent of In the (basal) lipolysis in fat cells is increased by in homozygous women. The the increased in vivo the effect of a in on adipocyte lipolysis was not by the FTO polymorphism. findings were obtained with a number of subjects However, it is by the for genetic lipolysis studies, and the FTO polymorphism is common allele frequency We women and subcutaneous fat and there are in lipolysis between the and adipose tissue E. differences in fat PubMed Scopus Google Scholar, M. M. differences in adipocyte new Full Text Full Text PDF PubMed Scopus Google Scholar). Therefore, we the findings are also for or for fat as the The subcutaneous adipose tissue, on the is by the fat How FTO findings suggest that the gene is expressed in adipocytes and that the mRNA for FTO is in these cells compared with total adipose FTO expression is regulated in human adipose adipocyte differentiation and is increased in obesity, it is not by adipose However, there to be no between FTO gene expression and lipolysis in adipose We found no between FTO mRNA levels and lipolysis, the in vivo to (P = The polymorphism is not associated with the level of mRNA expression of We also investigated the association of the FTO genotype with expression levels of genes that have been shown to be important in basal adipocyte lipolysis, including the two and and the two lipid and CGI-58 P. Langin D. The role of in human adipose tissue 2007; PubMed Scopus Google no were found. However, a number of additional genes may be associated with the FTO is also possible that FTO some indirect effects on lipolysis, which are not at there were no differences in BMI or metabolic parameters between these effects are not to body fat that in vivo is an important for body at least at the level of gene We that an receptor GPR74 is involved in the of human fat cell lipolysis and that a common in the GPR74 gene was associated with from obesity, in and increased lipolysis in fat cells (6.Dahlman I. Dicker A. Jiao H. Kere J. Blomqvist L. Harmelen V.van Hoffstedt J. Borch-Johnsen K. Jorgensen T. Hansen T. et al.A common haplotype in the G-protein-coupled receptor gene GPR74 is associated with leanness and increased lipolysis.Am. J. Hum. Genet. 2007; 80: 1115-1124Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar). as well as (6.Dahlman I. Dicker A. Jiao H. Kere J. Blomqvist L. Harmelen V.van Hoffstedt J. Borch-Johnsen K. Jorgensen T. Hansen T. et al.A common haplotype in the G-protein-coupled receptor gene GPR74 is associated with leanness and increased lipolysis.Am. J. Hum. Genet. 2007; 80: 1115-1124Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar) of lipolysis in fat cells may be the of The data are in agreement with genetic of obesity, including which are by increased lipolysis, in resistance to obesity J. D. A. D. Chen L. of in leanness and obesity in Genet. PubMed Scopus Google Scholar, C. J. C. et in a with adipocyte lipolysis, and resistance to Scopus Google Scholar). In the more common allele in the FTO gene is in its homozygous associated with increased adipocyte both in vivo and in that FTO at least in body fat mass lipolysis, the of to be This study was by the Swedish and the Swedish

References

12