0021-972X/97/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1997 by The Endocrine Society
Vol. 82, No. 8 Printed in U.S.A.
Variants in the Human Intestinal Fatty Acid Binding Protein 2 Gene in Obese Subjects ¨ INEN, MATTI UUSITUPA, SAMI HEIKKINEN, AILA RISSANEN, RAISA SIPILA MARKKU LAAKSO
AND
Department of Clinical Nutrition (R.S., M.U.) and Department of Medicine (S.H., M.L.), University of Kuopio, 70211 Kuopio; Eating Disorder Unit (A.R.), University Hospital of Helsinki, 00270 Helsinki, Finland ABSTRACT Fatty acid binding protein 2 gene (FABP2) has been proposed to be an important candidate gene for insulin resistance; therefore, it also could be a promising candidate gene for obesity. We screened the whole coding region of the FABP2 gene in 40 obese nondiabetic Finnish subjects. Furthermore, we investigated the effects of the codon 54 polymorphism of this gene (Ala 3 Thr) on insulin levels and basal metabolic rate in 170 obese subjects. The frequencies of the variants found in exon 4 (GTA 3 GTG) and 39-noncoding region (GCGCA 3 GCACA), as well as the allele frequencies for the variable lengths of the ATT repeat sequence in intron 2 did not differ between the obese subjects and nonobese controls. The frequency of threonine-encoding allele in codon 54 of the FABP2 gene did not differ between obese and
I
NSULIN resistance has been shown to cluster within families, suggesting a genetic background for this condition (1–3). Because insulin resistance is also a characteristic finding in obese subjects, it is possible that factors that cause insulin resistance could lead to obesity as well. The fatty acid binding protein 2 gene (FABP2) locus has been associated with fasting insulin concentration (4) and with 2-h insulin levels (5). Furthermore, the codon 54 polymorphism (GCT 3 ACT) of the FABP2 gene has been related to insulin resistance in nondiabetic Pima Indians (6). However, in Caucasian populations no association was found between the FABP2 gene locus and insulin levels (7). The FABP2 gene encodes the protein that binds intestinal fatty acids, and it contains a high affinity binding site for both saturated and unsaturated longchain fatty acids, indicating that it might have a role in the absorption and intracellular transport of dietary long-chain fatty acids (8). In Pima Indians the codon 54 polymorphism of the FABP2 gene has been associated with increased lipid oxidation rate and also insulin resistance (6). To investigate the hypothesis that defects in the FABP2 gene could be associated with obesity, we screened the whole coding region of this gene in obese nondiabetic Finns. We also investigated the frequency of the codon 54 polymorphism of the FABP2 gene and the effects of this polymorphism on insulin levels and basal metabolic rate in a large group of obese Finns.
Received January 23, 1997. Revision received April 23, 1997. Accepted May 13, 1997. Address correspondence and requests for reprints to: Raisa Sipila¨inen, M.Sc., Department of Clinical Nutrition, University of Kuopio, P.O. Box 1627, 70211 Kuopio, Finland. E-mail:
[email protected].
control subjects (28 vs. 29%, respectively). In the obese group there were no differences in gender distribution, age, weight, body mass index, lean body mass, percentage of body fat, waist circumference, and waist-to-hip ratio among the individuals homozygous for Ala54, heterozygous for Thr54, and homozygous for Thr54-encoding alleles. Similarly, fasting serum insulin, glucose, lipids and lipoprotein concentrations, basal metabolic rate (adjusted for lean body mass and age), respiratory quotient, and rates of glucose and lipid oxidation did not differ among the groups. We conclude that obesity is not associated with specific variants in the FABP2 gene. Furthermore, the codon 54 Ala to Thr polymorphism of this gene does not influence insulin levels or basal metabolic rate in obese Finns. (J Clin Endocrinol Metab 82: 2629 –2632, 1997)
Subjects and Methods All the subjects participating in this study were Finnish. The Finnish population is genetically quite homogenous, descending mainly from a small number of founders of Baltic Finnish and German origin (9). Altogether 170 unrelated obese subjects participating in a weight reduction study were included in the present study (10). Clinical characteristics of the study subjects are presented in Table 1.
Initial screening Altogether 40 obese women, randomly drawn from the weight reduction study, were screened for the variants of the FABP2 gene by single strand conformation polymorphism (SSCP) analysis. Their mean age was 43 6 8 yr, body mass index (BMI) 33.9 6 3.4 kg/m2, and fasting blood glucose 5.2 6 0.5 mmol/L. None of them had impaired glucose tolerance or diabetes according to the criteria of the World Health Organization (11). All subjects had normal liver, kidney, and thyroid functions, and none had a history of excessive alcohol intake. The allele frequencies of the variants of the FABP2 gene were compared with 40 healthy normoglycemic subjects who were participants in our previous study investigating the relationship between insulin resistance and dyslipidemia (12). None of the control subjects had any chronic disease, any drug treatment, glucose intolerance, or hypertension.
Additional screening Screening for the codon 54 amino acid substitution of the FABP2 gene (6) was performed in an additional sample of 130 obese subjects participating in the same weight reduction study as those 40 subjects initially screened. They fulfilled the same inclusion criteria as did the subjects participating in the initial screening, and none of them was receiving drugs known to affect basal metabolic rate (BMR) or glucose metabolism. The frequency of the codon 54 polymorphism of the FABP2 gene in obese subjects was compared with that of 82 healthy normoglycemic subjects (an additional sample of 42 subjects was studied for this polymorphism) (12).
2629
¨ INEN ET AL. SIPILA
2630 Study protocol
Every subject in the initial screening program underwent an oral glucose tolerance test (75 g glucose) to exclude noninsulin-dependent diabetes mellitus and impaired glucose tolerance. The protocol was approved by the Ethics Committees of the University of Kuopio and the University Helsinki, and all subjects gave their informed consent.
Analytical methods All measurements were done in the morning after an overnight fast. Weight was measured by electronic scales. BMI was calculated from the following formula: BMI 5 weight (kg)/height2 (m). Waist and hip circumferences were measured as described previously (10). Body composition was determined by bioelectrical impedance (RJL Systems, Detroit, MI). Basal metabolic rate (BMR) was measured by indirect calorimetry (Deltatrac, TM Datex, Helsinki, Finland) after a 12-h fast, as previously reported in detail (13). Energy production rate (cal/min) was calculated according to Ferrannini (14) and expressed as kcal/day. Urinary nitrogen was measured in 119 subjects. The adjusted BMR was calculated according to Ravussin (15). Serum insulin was analyzed by TABLE 1. Clinical characteristics of the study group of obese Finns Men/women Age (yr) Weight (kg) BMI (kg/m2) Serum glucose (mmol/L) Serum insulin (pmol/L) Mean 6
29/141 43 6 8 95.4 6 12.9 34.7 6 3.8 5.5 6 0.8 96.2 6 46.4
Obese (n 5 40)
47/82 (29%)a
95/170 (28%)a
27 (34%)
28 (35%)
4 (5%)
2 (3%)
42 (53%) 12 (15%) 0 18 (23%) 7 (9%) 1 (1%)
38 (48%) 8 (10%) 3 (4%) 25 (31%) 6 (8%) 0
All the differences between the groups were nonsignificant. All subjects belonging in the control group (n 5 82) and in the obese group (n 5 170) were screened for the exon 2 polymorphism. a
Statistical analysis
Results
TABLE 2. Variants of the FABP2 gene between the control and obese subjects (rare allele frequency in parentheses) Controls (n 5 40)
radioimmunoassay with the double antibody-PEG technique (CIS Bio International, Gif-sur-Yvette, France) and serum glucose by kinetic photometry with glucose-dehydrogenase (16). Serum lipids and lipoproteins were analyzed after ultracentrifugation (92500 3g, 18 h, 5 C) and precipitation by enzymatic methods (17–19), the CHOD-PAP method (HiCo cholesterol reagents, Boehringer Mannheim, Mannheim, Germany) for cholesterol and high-density-lipoprotein (HDL) cholesterol, and the GPO-PAP method (Boehringer Mannheim) for triglycerides. DNA was prepared from peripheral blood leucocytes by the proteinase K-phenol-chloroform extraction method. All four exons and the intron-exon junctions of the FABP2-gene were amplified with the polymerase chain reaction (PCR) using the primers reported in a previous study (6). SSCP analysis was performed essentially according to the method of Orita et al. (20). Genomic DNA from individuals with variant single strand conformers was directly sequenced using Sequenase (US Biochemicals, Cleveland, OH) as previously described (21). Determination of the frequency of the GCT to ACT nucleotide substitution in codon 54 of the FABP2 gene was detected by PCR-RFLP (restriction fragment length polymorphism) assays as previously decribed (6).
All calculations were performed using the SPSS/WIN programs (Version 6.0, SPSS Inc., Chicago, IL). Data are presented as means 6 sd. Statistical significance among the three groups was evaluated using the x-squared test or ANOVA, when appropriate.
SD.
Exon 2 GCT 54 3 ACT 54 Ala Thr Exon 4 GTA 118 3 GTG 118 39 noncoding region GCGCA 3 GCACA ATT repeat sequence length in intron 2 10 11 12 13 14 15
JCE & M • 1997 Vol 82 • No 8
We found two variants in the coding region of the FABP2 gene. The frequency of the Thr54-encoding allele did not differ between the obese and control subjects (Table 2). Similarly, a silent substitution of GTA118GTG in exon 4 was as frequent in obese subjects as in normal weight controls. In addition, we found a substitution of GCGCA to GCACA in the 39-noncoding region. The frequencies of this variant as well as the allele frequencies for the variable lengths of the ATT repeat sequence in intron 2 did not differ between the obese and control subjects (Table 2). We also analyzed clinical characteristics, parameters of glucose, and lipid metabolism and the results of energy metabolism in three groups formed on the basis of the codon 54 polymorphism of the FABP2 gene: Ala54 homozygotes, (n 5 84, 49%), Thr54 heterozygotes (n 5 77, 45%), or Thr54 homozygotes (n 5 9, 5%). There were no differences in gender distribution, age, weight, BMI, lean body mass, percentage of body fat, waist circumference, and waist-to-hip ratio among the three groups (Table 3). Similarly, fasting serum insulin and glucose concentrations, serum lipids and lipoproteins, basal metabolic rate (adjusted for lean body mass and age), respiratory quotient, and rates of glucose and lipid oxidation did not differ among the three groups (Table 4).
TABLE 3. Clinical characteristics of the obese subjects according to the polymorphism in codon 54 of the FABP2 gene
Men/women Age (yr) Weight (kg) BMI (kg/m2) LBM (kg) Body fat (%) Waist (cm) Waist/hip ratio
Ala54 homozygotes
Thr54 heterozygotes
Thr54 homozygotes
ANOVA P-value
15/69 43.3 6 8.3 94.6 6 12.5 34.8 6 3.8 59.1 6 10.1 37.5 6 5.9 106.2 6 11.2 0.94 6 0.08
12/65 43.0 6 7.7 96.1 6 13.5 34.7 6 3.9 60.0 6 9.8 37.5 6 6.5 104.4 6 10.6 0.91 6 0.07
2/7 42.1 6 6.3 96.9 6 10.7 33.9 6 1.6 61.1 6 11.6 37.2 6 7.1 102.7 6 7.9 0.91 6 0.07
0.850 0.909 0.712 0.749 0.781 0.988 0.429 0.067
Mean 6 SD. BMI, body mass index; LBM, lean body mass.
FABP2 GENE IN OBESE FINNS
2631
TABLE 4. Metabolic characteristics of the obese subjects according to the polymorphism in codon 54 of the FABP2 gene
Men/women Serum glucose (mmol/L) Serum insulin (pmol/L) Cholesterol (mmol/L) Total HDL LDL VLDL Triglycerides (mmol/L) Free fatty acid (mmol/L)a BMR (kcal/day) Respiratory quotient Fasting glucose oxidationb (mmol/min/kg LBM) Fasting lipid oxidationb (mg/min/kg LBM)
Ala54 homozygotes
Thr54 heterozygotes
Thr54 homozygotes
ANOVA P-value
15/69 5.6 6 0.86 94.4 6 42.7
12/65 5.5 6 0.61 96.7 6 47.4
2/7 5.4 6 0.73 108.0 6 71.3
0.850 0.533 0.705
5.52 6 0.99 1.22 6 0.32 3.66 6 0.86 0.67 6 0.38 1.62 6 0.79 0.66 6 0.24 1648 6 138 0.82 6 0.05 7.8 6 3.9
5.56 6 1.03 1.22 6 0.27 3.63 6 0.91 0.75 6 0.57 1.82 6 1.95 0.61 6 0.23 1605 6 142 0.82 6 0.03 8.6 6 3.3
5.61 6 0.87 1.24 6 0.47 3.49 6 1.01 0.93 6 0.69 2.05 6 1.64 0.80 6 0.33 1625 6 131 0.80 6 0.02 8.1 6 2.2
0.938 0.973 0.846 0.268 0.561 0.091 0.512 0.461 0.517
1.12 6 0.31
0.98 6 0.26
1.12 6 0.21
0.038
Mean 6 SD. BMR, basal metabolic rate adjusted for lean body mass and age; LBM, lean body mass. a Number of subjects homozygous for Ala54-encoding allele was 47; heterozygous for Thr54-encoding allele, 51. b Number of subjects homozygous for Ala54-encoding allele was 52; heterozygous for Thr54-encoding allele, 52; and homozygous for Thr54encoding allele, 9.
Discussion
Our study suggests that obesity is not associated with specific variants of the FABP2 gene. Furthermore, the codon 54 polymorphism of the FABP2 gene did not modify basal metabolic rate, fasting insulin level, or other indicators of insulin resistance. In a previous study on nondiabetic, obese Pima Indians (6) fasting insulin levels or rates of lipid oxidation were associated with the codon 54 polymorphism of the FABP2 gene. This could not be found in the present study on obese Finns, however, nor was this polymorphism associated with basal metabolic rate, suggesting that the codon 54 polymorphism of the FABP2 gene is not related to energy metabolism in obese Finnish subjects. In the Pima Indian population higher mean fasting plasma insulin concentrations and lower insulin sensitivity were observed only after pooling of subjects who were homozygous and heterozygous for the Thr54-encoding allele of the FABP2 gene (6). Moreover, the reported differences were of borderline statistical significance. Our analysis was not based on direct measurement of insulin sensitivity, but on fasting insulin levels only. Although fasting insulin levels correlate quite closely with insulin sensitivity determined by the euglycemic hyperinsulinemic clamp (22), our findings cannot entirely rule out the possibility that insulin resistance measured by a direct method could be associated with the presence of the Thr54 allele of the FABP2 gene. Obesity, especially abdominal obesity and lipid abnormalities, are common features of insulin resistance syndrome. In our study, the three genotypes (Ala54 homozygotes, Thr54 heterozygotes, Thr54 homozygotes) of the FABP2 gene did not differ with respect to any clinical characteristics of insulin resistance syndrome. This finding implies that the codon 54 polymorphism of the FABP2 gene is not a significant determinant of insulin resistance syndrome in obese Finns. However, there was a tendency toward elevated concentrations of serum free fatty acids in Thr54 homozygotes; these subjects also had highly elevated very-low-density-lipid (VLDL) cho-
lesterol and total triglyceride levels. Therefore, modest effects of the Thr54 allele on lipid levels cannot be ruled out because our study included only a small number of Thr54 homozygotes. It should also be emphasized that our findings do not exclude the possibility that the effects of the Thr54 allele of the FABP2 gene could be population specific, and that other genes close to the FABP2 gene locus may have an effect on insulin action and thereby insulin resistance, as has been suggested earlier (5–7). References 1. Martin B, Warram J, Krolewski A, Bergman R, Soeldner J, Khan C. 1992 Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a 25-year follow-up study. Lancet. 340:925–929. 2. Beck-Nielsen H, Groop L. 1994 Metabolic and genetic characterization of prediabetic states. J Clin Invest. 94:1714 –1721. 3. Vaag A, Henriksen J, Madsbad S, Holm N, Beck-Nielsen H. 1995 Insulin secretion, insulin action, and hepatic glucose production in identical twins discordant for non-insulin-dependent diabetes mellitus. J Clin Invest. 95:690 – 698. 4. Prochazka M, Lillioja S, Tait J, et al. 1993 Linkage of chromosomal markers on 4q with a putative gene determining maximal insulin action in Pima Indians. Diabetes. 42:514 –519. 5. Mitchell B, Kammerer C, O’Connell P., et al. 1995 Evidence for linkage of postchallenge insulin levels with intestinal fatty acid-binding protein (FABP2) in Mexican-Americans. Diabetes. 44:1046 –1053. 6. Baier L, Sacchettini J, Knowler W., et al. 1995 An amino acid substitution in the human intestinal fatty acid binding protein is associated with increased fatty acid binding, increased fat oxidation, and insulin resistance. J Clin Invest. 95:1281–1287. 7. Humpreys P, McCarthy M, Tuomilehto J., et al. 1994 Chromosome 4q locus associated with insulin resistance in Pima Indians. Diabetes. 43:800 – 804. 8. Lowe J, Sacchettini J, Laposata M, McQuillan J, Gordon J. 1987 Expression of rat intestinal fatty-acid-binding protein in Escherichia coli. J Biol Chem. 262:5931–5937. 9. De la Chapelle A. 1993 Disease gene mapping in isolated human population; the example of Finland. J Med Genet. 30:857– 865. 10. Uusitupa M, Karhunen L, Rissanen A, Franssila-Kallunki A, Niskanen L, Kervinen K, Kesa¨niemi YA. 1996 Apolipoprotein E phenotype modifies metabolic and hemodynamic abnormalities related to central obesity in women. Am J Clin Nutr. 64:131–136. 11. World Health Organization: Diabetes Mellitus: Report of a WHO Study Group. 1985 Geneva, World Health Org. (Tech. Rep. Ser., no. 727). 12. Haffner SM, Karhapa¨a¨ P, Mykka¨nen L, Laakso M. 1994 Insulin resistance, body fat distribution and sex hormones in men. Diabetes. 43:212–219. 13. Laakso M, Uusitupa M, Takala J, Majander H, Reijonen T, Penttila¨ I. 1988 Effects of hypocaloric diet and insulin therapy on metabolic control and mech-
¨ INEN ET AL. SIPILA
2632
14. 15. 16. 17.
18.
anisms of hyperglycemia in obese non-insulin-dependent diabetic subjects. Metabolism. 37:1092–1100. Ferrannini E. 1988 The theoretical bases of indirect calorimetry: A review. Metabolism. 37:287–301. Ravussin E, Lillioja S, Knowler W, et al. 1988 Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med. 318:467– 472. Rindfrey H, Helger R, Lang H. 1977 Kinetic determination of glucose concentrations with glucose hydrogenase. J Clin Chem Clin Biochem. 15:217–220. Eggstein M. 1966 A new determination of the neutral fats in blood serum and tissue. II. Reliability of the method, other neutral fat determinations, normal values for triglycerides and glycerin in human blood. Klin Wochenschr. 44:267–273. Lopes-Virella M, Stone P, Ellis S, Colwell J. 1977 Cholesterol determination
19.
20.
21. 22.
JCE & M • 1997 Vol 82 • No 8
in high-density lipoproteins separated by three different methods. Clin Chem. 23:882–284. Siedel J, Hagele E, Ziegenhorn J, Wahlefeld A. 1983 Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin Chem. 29:1075–1080. Orita M, Suzuki T, Sekiya T, Hayashi K. 1989 Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics. 5:874 – 879. Kretz K, Carson G, O’Brien J. 1989 Direct sequencing from low-melt agarose with Sequenase. Nucleic Acids Res. 17:5864. Laakso M. 1993 How good a marker is insulin level for insulin resistance? Am J Epid. 137:959 –965.
The Cleveland Clinic Foundation Department of Continuing Education presents Endocrinology & Metabolism: Board Review 1997 Cleveland, Ohio September 7–9, 1997 About the symposium: The goal of this symposium is to provide the participant, an intensive preparation for the endocrinology and metabolism board examination. As a result of attending this symposium, the participant will be able to discuss and review key area of: ● Endocrinology, including diabetes control ● ● ● ● ●
and complications Thyroid disorders Hypo and hyperadrenalism Osteoporosis Calcium disorders Pituitary disorders
● Nuclear imaging ● Endocrine pathology ● Endocrine hypertension ● Renolithiasis ● Transplantation ● Important pediatric endocrine disorders and pregnancy
The faculty are known for their clinical and teaching experience. The lecture format will be heavily weighted to board simulation. An interactive Audience Response System will be employed to facilitate audience participation. The faculty will also be available (via e-mail) for board queries for two weeks following the symposium. A comprehensive syllabus will be provided. Accreditation: The Cleveland Clinic Educational Foundation is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to sponsor continuing medical education for physicians. The Cleveland Clinic Educational Foundation designates this educational activity for a maximum of 26 hours in Category 1 credit towards the AMA Physicians Recognition Award. This symposium may be submitted for A.O.A.-C.M.E. credit in Category 2-A. Information: For further information about the symposium, call the following numbers: Local: (216) 4445696; Toll Free: 800-762-8173; Fax Number: (216) 445-9406.
