Hereditary factors may be involved in the pathogenesis of type 2 diabetes. A polymorphism in the hormone-sensitive lipase (HSL) gene (HSLi6) is associated with obesity and diabetes, although it is unknown whether the polymorphism is functional and thereby influences lipolysis. We genotyped 355 apparently healthy nonobese male and female subjects for the HSLi6 polymorphism. Allele 5 was found to be the most common allele (allele frequency 0.57). In 117 of the subjects, we measured abdominal subcutaneous fat cell lipolysis induced by drugs acting at various steps in the lipolytic cascade. The lipolysis rate induced by norepinephrine isoprenaline (acting on β-adrenoceptors), forskolin (acting on adenylyl cyclase), and dibutyryl cyclic AMP (acting on HSL) were all decreased by ∼50% in allele 5 homozygotes, as compared with noncarriers. Heterozygotes showed an intermediate lipolytic rate. The difference in lipolysis rate between genotypes was more pronounced in men than in women. We conclude that allele 5 of the HSLi6 polymorphism is associated with a marked decrease in the lipolytic rate of abdominal fat cells. This may in turn contribute to the development of obesity.

There has been a rapid and global increase in the prevalence of obesity and type 2 diabetes, two interrelated disorders (1). Although environmental factors contribute, it is obvious that hereditary factors are of pathophysiological importance for these insulin-resistant conditions (2). Several candidate genes have been identified, including those regulating lipolysis. In this respect, the hormone-sensitive lipase (HSL) gene may be of particular interest. HSL catalyzes the rate-limiting step for lipolysis in fat cells, specifically the breakdown of triglycerides (3). Catecholamines are the main hormones stimulating this process in humans. A decreased ability of catecholamines to induce lipolysis in subcutaneous adipose tissue, partly explained by decreased expression/enzyme activity of HSL, has been found in obese subjects (4) as well as in normal-weight individuals with obesity among first-degree relatives (5). Recently, a polymorphism of a dinucleotide repeat including 16 alleles localized within intron 6 of the HSL gene was identified (HSLi6), and an increased frequency of allele 5 in both obese and diabetic subjects was demonstrated (6). However, it is not known whether this polymorphism is associated with altered lipolytic activity. We hypothesized that if the polymorphism is functional, then it should be associated with altered fat cell lipolysis in normal-weight subjects whose adipocytes are not influenced by factors related to obesity. The present study was designed to investigate lipolysis in fat cells of nonobese healthy subjects in relation to the occurrence of allele 5 of the HSLi6 polymorphism.

To investigate the frequencies of the different HSLi6 alleles in a Swedish population, we genotyped a large cohort (n = 355) of nonobese subjects. The following allele frequencies were found, using nomenclature previously described (6): A5 0.57, A4 0.16, A3 0.07, A7 0.05, and all other alleles <0.05. No difference in allele frequency between male and female subjects was found. The frequency of allele 5 in this population is higher than in a previously described French population (0.57 vs. 0.41), which may have several explanations, including two different populations (Swedish versus French), a different number of subjects (355 vs. 150), and different upper limits for BMI (30 vs. 25 kg/m2). In 117 of the subjects in our study, adipose tissue was obtained for lipolysis studies, and detailed data from these subjects are recorded in Table 1. In this group, 33% of the subjects were homozygous and 48% were heterozygous for allele 5. The frequency of allele 5 was higher in male subjects than in female subjects (0.68 vs. 0.52, P < 0.05 by χ2 analysis). This is probably a chance finding because there was no sex difference in the whole cohort. The distribution of alleles was in Hardy-Weinberg equilibrium. Subjects homozygous for allele 5 were termed 55, subjects heterozygous for allele five were termed 5X, and those subjects devoid of allele five were termed XX. In the male subgroup, values for BMI and waist-to-hip ratio (WHR) were higher in 55 and 5X subjects compared with XX subjects. In contrast, no difference between genotypes in either BMI or WHR was found in the female subgroup. The female 5X subjects were slightly younger than the 55 and XX subjects.

The influence of the HSLi6 polymorphism was investigated on lipolysis in abdominal subcutaneous adipocytes. Table 1 shows the results for maximal lipolysis rate (maximum minus basal release of glycerol) after stimulation with norepinephrine, isoprenaline, forskolin, and dibutyryl cyclic AMP (dcAMP). Because catecholamine-induced lipolysis may be influenced by age and BMI, all of the statistical calculations were corrected for these parameters using analysis of covariance (ANCOVA). In male subjects, the maximal lipolysis rate for all drugs was ∼50% lower in subjects homozygous for allele 5 compared with subjects with the XX genotype. Subjects with the 5X genotype showed an intermediate lipolytic response. The mean concentration-response curves in male subjects for norepinephrine and dcAMP are shown for illustrative purposes in Fig. 1. The female subgroup showed no significant differences for maximal lipolysis between the 55, 5X, and XX genotypes, although female XX subjects had on average 40% higher lipolytic activity compared with 55 and 5X subjects. Analysis of the combined data for men and women using ANCOVA, with age, BMI, and sex as covariates, revealed that the difference in lipolytic activity between genotypes was significant (P values from 0.003 to 0.02 for all of the drugs). Neither the basal lipolysis rate (no stimulating agent present) nor the lipolytic sensitivity (EC50) for norepinephrine was different among the three genotypes. We also divided the male and female subgroups with respect to high and low lipolytic capacity for forskolin and dcAMP (Table 2). In men, the frequency of allele 5 was increased in subjects with a low lipolytic response to both forskolin (P = 0.002) and dcAMP (P = 0.02). In women, subjects with a high lipolytic response to forskolin (P = 0.03), but not to dcAMP (P = 0.09), showed an increased frequency of allele X. No significant difference between subjects with high or low lipolytic capacity was observed for the frequency of the second most common allele, A4.

Thus, allele 5 of the HSLi6 polymorphism is associated with decreased catecholamine-stimulated lipolysis in abdominal subcutaneous adipose tissue. This finding is more apparent in men than in women. The reason for this is unknown. However, given the well-known difference between men and women regarding body fat distribution, we do not find it surprising that sex-related factors (e.g., sex hormones) may also interact with the genotype and cause different phenotypic effects in men and women. Because agents acting at different steps of the lipolytic cascade could mimic this defect, it is very likely that the observed impairment of lipolysis is located at the level of HSL. Because of the lack of sufficient amounts of adipose tissue, it was not possible to obtain direct measures of HSL expression or enzyme activity. However, dcAMP, which was used in the present study, acts at the protein kinase–HSL complex. Recent studies have also demonstrated correlations between the lipolysis rate and both HSL expression and enzyme activity (7,8). It should be noted, though, that the function of the polymorphism is not fully understood because it is found within an intron (6), which suggests that allele 5 may be in linkage disequilibrium with another as-yet unidentified gene polymorphism influencing HSL expression or activity.

An enlarged fat mass is one possible consequence of a decreased ability to mobilize lipids from adipose tissue. Indeed, both WHR, used as a measure of upper-body fat distribution, and BMI were found to be increased in male subjects with allele 5. However, this study was not designed to investigate the influence of the HSLi6 polymorphism on obesity. That would require a much larger study population including subjects with a high BMI. Furthermore, in a recent study, HSL-knockout mice did not develop obesity (9), indicating that lipolytic enzymes other than HSL are also involved in the regulation of body fat accumulation.

The 5X subjects were slightly younger than the other subjects. However, this has no important bearing on the results with lipolysis. First, the statistical analysis was corrected for age. Second, 55 and XX subjects had almost identical mean values for age, but they differed markedly with respect to lipolytic response.

In conclusion, the present data suggest that the HSLi6 polymorphism is important for human lipolysis. Subjects with allele 5 have a decreased catecholamine activation of HSL. This may in turn promote obesity, given the recent demonstration of an association between the HSLi6 polymorphism and obesity (6).

The cohort included 355 nonobese (BMI <30 kg/m2) male (n = 146) and female (n = 209) subjects of Scandinavian origin. These subjects were genotyped for the HSLi6 polymorphism to investigate the allele frequencies in a Swedish population. Some of these subjects (n = 117) also underwent a fat biopsy for lipolysis studies. All further data relate to this group, which consisted of male (n = 38) and female (n = 79) healthy subjects. All were nonobese (BMI range 17.8–29.4 kg/m2) according to the World Health Organization criteria, and all were between 19 and 72 years of age. None was completely sedentary or involved in athletic performance, and none had undertaken any weight-reduction effort during the previous 6-month period. None was on regular medication. The subjects were investigated in the morning after an overnight fast. Height, weight, and WHR (supine position) were determined. An antecubital venous blood sample was obtained for DNA preparation. Finally, a subcutaneous fat biopsy (1–2 g) was obtained under local anesthesia from the umbilical region. The study was approved by the Ethics Committee of Huddinge University Hospital.

Genotyping was performed by polymerase chain reaction, as previously described (6), using upstream primer 5′-AGGCAGAGGAATCTAAG-3′ and downstream primer 5′-CTGAGCGACATGACAC-3′. A detailed description of the fat cell experiments has been reported in detail elsewhere (4,5). Briefly, isolated fat cells were prepared by collagenase treatment, and their average size, volume, and weight were determined. Dilute fat cell suspensions were incubated in duplicate for 2 h, and at the end of the incubation, an aliquot of the medium was removed for the analysis of glycerol (lipolysis index). The following agents were added to the incubation medium in increasing concentrations (10−12 to 10−3 mol/l, depending on the type of agent) at the start of the experiment: norepinephrine (nonselective α2- and β-adrenoceptor agonist), isoprenaline (nonselective β-adrenoceptor agonist), forskolin (selective stimulator of adenylyl cyclase), and dcAMP (phosphodiesterase-resistant activator of the protein kinase A–HSL complex). All agonists caused a concentration-dependent stimulation of glycerol release that reached a plateau at the maximum effective drug concentration. Consequently, the maximum release of glycerol could be determined after subtraction of basal lipolysis values. This is in turn an indirect measure of postreceptor-related events, including HSL function (4,5). All of these values were related to fat cell number. The concentration response curves for norepinephrine were also analyzed for half-maximum effective concentrations on glycerol release (EC50), which reflects events at the adrenoceptor level, including binding and coupling to effector (10). For technical reasons, it was not possible to obtain enough tissue to perform all lipolysis measurements in each subject.

FIG. 1.

Mean concentration-response curves for lipolysis induced by norepinephrine and dcAMP in males.

FIG. 1.

Mean concentration-response curves for lipolysis induced by norepinephrine and dcAMP in males.

Close modal
TABLE 1

Clinical and lipolytic characteristics of allele 5 of the HSLi6 polymorphism

Men
Women
Men + Women
555XXXP555XXXP555XXXP
n 21 10  18 46 15  39 56 22  
Age (years) 40 ± 3 37 ± 4 43 ± 6 0.74 45 ± 2* 35 ± 2 42 ± 3 0.003 42 ± 2 35 ± 1 42 ± 3 0.01 
BMI (kg/m224.8 ± 0.4§ 25.3 ± 0.6 22.8 ± 0.7 0.02 24.1 ± 0.6 23.3 ± 0.3 24.1 ± 0.7 0.95 24.4 ± 0.4 23.7 ± 0.3 23.7 ± 0.5 0.53 
WHR 0.96 ± 0.01§ 0.96 ± 0.01 0.88 ± 0.02* 0.003 0.87 ± 0.01 0.85 ± 0.01 0.81 ± 0.02 0.08 0.92 ± 0.01 0.87 ± 0.01 0.84 ± 0.02 0.001 
Basal lipolysis 5.5 ± 0.8 6.0 ± 1.3 3.9 ± 0.9 0.97 5.5 ± 0.9 4.2 ± 0.6 4.4 ± 0.7 0.43 5.5 ± 0.6 4.5 ± 0.5 4.2 ± 0.6 0.49 
Maximal lipolysis             
Noradrenaline 6.1 ± 1.1§ 9.2 ± 1.6 15.6 ± 3.7 0.004 10.0 ± 1.5 11.2 ± 1.0 15.2 ± 3.8 0.10 8.0 ± 1.0§ 10.9 ± 0.8 15.3 ± 2.7 0.003 
Isoprenaline 16.7 ± 1.8 29.1 ± 3.8 34.7 ± 8.4§ 0.006 22.2 ± 2.6 22.8 ± 1.5 29.7 ± 6.1 0.16 19.2 ± 1.6§ 23.9 ± 1.4 31.3 ± 4.8 0.008 
Forskolin 14.4 ± 1.6 26.2 ± 3.4 30.7 ± 7.7§ 0.005 19.5 ± 2.2 20.6 ± 1.6 27.3 ± 5.4 0.12 16.8 ± 1.4§ 21.7 ± 1.5 28.5 ±4.3 0.006 
dcAMP 13.2 ± 1.3 21.0 ± 2.5 26.5 ± 6.9§ 0.01 19.2 ± 2.3 19.2 ± 1.6 27.1 ± 6.4 0.12 16.0 ± 1.4§ 19.5 ± 1.4 26.9 ± 4.7 0.02 
EC50             
Noradrenaline 8.6 ± 0.3 8.2 ± 0.3 8.2 ± 0.1 0.41 7.8 ± 0.4 8.3 ± 0.2 7.8 ± 0.4 0.36 8.2 ± 0.3 8.3 ± 0.2 7.9 ± 0.2 0.44 
Men
Women
Men + Women
555XXXP555XXXP555XXXP
n 21 10  18 46 15  39 56 22  
Age (years) 40 ± 3 37 ± 4 43 ± 6 0.74 45 ± 2* 35 ± 2 42 ± 3 0.003 42 ± 2 35 ± 1 42 ± 3 0.01 
BMI (kg/m224.8 ± 0.4§ 25.3 ± 0.6 22.8 ± 0.7 0.02 24.1 ± 0.6 23.3 ± 0.3 24.1 ± 0.7 0.95 24.4 ± 0.4 23.7 ± 0.3 23.7 ± 0.5 0.53 
WHR 0.96 ± 0.01§ 0.96 ± 0.01 0.88 ± 0.02* 0.003 0.87 ± 0.01 0.85 ± 0.01 0.81 ± 0.02 0.08 0.92 ± 0.01 0.87 ± 0.01 0.84 ± 0.02 0.001 
Basal lipolysis 5.5 ± 0.8 6.0 ± 1.3 3.9 ± 0.9 0.97 5.5 ± 0.9 4.2 ± 0.6 4.4 ± 0.7 0.43 5.5 ± 0.6 4.5 ± 0.5 4.2 ± 0.6 0.49 
Maximal lipolysis             
Noradrenaline 6.1 ± 1.1§ 9.2 ± 1.6 15.6 ± 3.7 0.004 10.0 ± 1.5 11.2 ± 1.0 15.2 ± 3.8 0.10 8.0 ± 1.0§ 10.9 ± 0.8 15.3 ± 2.7 0.003 
Isoprenaline 16.7 ± 1.8 29.1 ± 3.8 34.7 ± 8.4§ 0.006 22.2 ± 2.6 22.8 ± 1.5 29.7 ± 6.1 0.16 19.2 ± 1.6§ 23.9 ± 1.4 31.3 ± 4.8 0.008 
Forskolin 14.4 ± 1.6 26.2 ± 3.4 30.7 ± 7.7§ 0.005 19.5 ± 2.2 20.6 ± 1.6 27.3 ± 5.4 0.12 16.8 ± 1.4§ 21.7 ± 1.5 28.5 ±4.3 0.006 
dcAMP 13.2 ± 1.3 21.0 ± 2.5 26.5 ± 6.9§ 0.01 19.2 ± 2.3 19.2 ± 1.6 27.1 ± 6.4 0.12 16.0 ± 1.4§ 19.5 ± 1.4 26.9 ± 4.7 0.02 
EC50             
Noradrenaline 8.6 ± 0.3 8.2 ± 0.3 8.2 ± 0.1 0.41 7.8 ± 0.4 8.3 ± 0.2 7.8 ± 0.4 0.36 8.2 ± 0.3 8.3 ± 0.2 7.9 ± 0.2 0.44 

Values are means ± SEM. Age was analyzed using analysis of variance (ANOVA). BMI and WHR were analyzed using ANCOVA with age (men and women separately) or age and sex (men + women) as covariates. Values for lipolysis and EC50 were analyzed using ANCOVA with age and BMI (men and women separately) or age, BMI, and sex (men + women) as covariates. In post hoc analyses using Fisher’s PLSD test, 55 or XX subjects were compared with 5X subjects in men + women and in men separately:

*

P < 0.005,

P < 0.05,

P < 0.01,

§

P > 0.05,

P < 0.001. Values for basal maximal lipolysis are expressed as μmol glycerol · 10−7 cells · 2 h−1. EC50 for noradrenaline is expressed as pD2 = −log mol/l.

TABLE 2

Subjects with allele 5 and/or X in men and women with a high or low lipolytic response to forskolin and dcAMP

Lipolytic responseAlleles
555XXXP
Forskolin     
 Men    0.002 
  High  
  Low 15  
 Women    0.03 
  High 16 11  
  Low 26  
dcAMP     
 Men    0.02 
  High  
  Low 13  
 Women    0.09 
  High 10 18 10  
  Low 27  
Lipolytic responseAlleles
555XXXP
Forskolin     
 Men    0.002 
  High  
  Low 15  
 Women    0.03 
  High 16 11  
  Low 26  
dcAMP     
 Men    0.02 
  High  
  Low 13  
 Women    0.09 
  High 10 18 10  
  Low 27  

Values were analyzed using χ2 analysis.

This study was supported by grants from Pharmacia, Upjohn, the Swedish Medical Research Council, the Swedish Diabetes Association, the Swedish Heart and Lung Foundation, the Swedish Society of Medicine, and the Foundations of Novo Nordic, Söderberg, and Thuring.

The excellent technical assistance of Eva Sjölin, Kerstin Wåhlén, Britt-Marie Leijonhufvud, and Katarina Hertel is highly appreciated.

1.
Zimmet P: Globalization, coca-colonization and the chronic disease epidemia: can the doomsday scenario be averted?
J Intern Med
247
:
301
–310,
2000
2.
Pedersen O: Genetics of insulin resistance.
Exp Clin Endocrinol Diab
107
:
113
–118,
1999
3.
Langin D, Holm C, Lafontan M: Adipocyte hormone-sensitive lipase: an important regulator of lipid metabolism.
Proc Nutr Soc
55
:
93
–109,
1996
4.
Large V, Reynisdottir S, Langin D, Fredby K, Klannemark M, Holm C, Arner P: Decreased expression and function of adipocyte hormone-sensitive lipase in subcutaneous fat cells of obese subjects.
J Lipid Res
40
:
2059
–2065,
1999
5.
Hellström L, Langin D, Reynisdottir S, Dauzats M, Arner P: Adipocyte lipolysis in normal weight subjects with obesity among first-degree relatives.
Diabetologia
39
:
921
–928,
1996
6.
Magré J, Laurell H, Fizames C, Antoine PJ, Dib C, Vigouroux C, Bourut C, Capeau J, Weissenbach J, Langin D: Human hormone-sensitive lipase: genetic mapping, identification of a new dinucleotide repeat, and association with obesity and NIDDM.
Diabetes
47
:
284
–286,
1998
7.
Reynisdottir S, Dauzats M, Thörne A, Langin D: Comparison of hormone-sensitive lipase activity in visceral and subcutaneous human adipose tissue.
J Clin Endocrinol Metab
82
:
4162
–4166,
1997
8.
Large V, Arner P, Reynisdottir S, Grober J, Van Harmelen V, Holm C, Langin D: Hormone-sensitive lipase expression and activity in relation to lipolysis in human fat cells.
J Lipid Res
39
:
1688
–1695,
1998
9.
Osuga J, Ishibashi S, Oka T, Yagyu H, Tozawa R, Fujimoto A, Shionoiri F, Yahagi N, Kraemer FB, Tsutsimi O, Yamada N: Targeted disruption of hormone-sensitive lipase results in male sterility and adipose hypertrophy, but not in obesity.
Proc Natl Acad Sci U S A
97
:
787
–792,
2000
10.
Kenakin T: Drugs and receptors: an overview of the current state of knowledge.
Drugs
40
:
666
–687,
1990

Address correspondence and reprint requests to Johan Hoffstedt, CME, M61, Karolinska Institutet, Huddinge University Hospital, 141 86 Stockholm, Sweden. E-mail: [email protected].

Received for publication 18 December 2000 and accepted in revised form 21 May 2001.

ANCOVA, analysis of covariance; dcAMP, dibutyryl cyclic AMP; HSL, hormone-sensitive lipase; WHR, waist-to-hip ratio.