| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Departments of Medicine (A.M., S.O.), Clinical Biochemistry (A.M., D.H., S.O.), and Public Health and Primary Care (J.L., P.S.-F., S.H., J.M., N.J.W.), University of Cambridge, Cambridge CB2 2SR, United Kingdom
Address all correspondence and requests for reprints to: Dr. Nicholas J. Wareham, Institute of Public Health, University of Cambridge, Robinson Way, Cambridge CB2 2SR, United Kingdom.
Abstract
The lipolytic effects of catecholamines are mediated through members of the ß2-adrenergic receptor (BAR-2) family. Previous studies have suggested that genetic variants in the BAR-2 gene may be associated with obesity in some populations. To our knowledge, no studies have directly examined the effects of this polymorphism on circulating nonesterified fatty acid (NEFA) levels. To explore this issue further, a cohort of 604 Caucasian individuals (aged 40–65 yr) was genotyped for a common polymorphism in the BAR-2 gene (Gly16Arg), and the relationships between genotype, body mass index (BMI), NEFA, and lipid levels were examined. Women bearing the Arg16 allele had higher BMI values (P < 0.01) than Gly16Gly women. Women carriers of the Arg16Arg genotype had lower fasting plasma NEFAs (P < 0.01) and greater suppression of NEFAs (P < 0.01) after an oral glucose load than women bearing the Gly16 allele. In multivariate analysis after adjustment for age, sex, and smoking status, the interaction between the BAR-2 genotype and BMI in determining fasting NEFA concentrations was statistically significant (P < 0.05). The availability of objective measures of total energy expenditure in this population permitted the further examination of interactions, particularly that between genotype and physical activity. In the population as a whole, after adjustment for confounding by age, smoking, and BMI, the effect of the Arg16Arg genotype on the suppression of NEFA levels was modified by physical activity level (P for interaction <0.05). These data suggest the existence in this population of a gene-physical activity interaction on NEFA levels.
THE ß2-ADRENERGIC
RECEPTORS (BAR-2) bind the endogenous catecholamines epinephrine and
norepinephrine and transfer the signals to the interior of cells via
the stimulatory guanine nucleotide-binding protein, Gs
(1). The three ß-adrenergic receptors
(ß1-, ß2-, and
ß3-), all of which are expressed in adipocytes,
stimulate lipolysis, whereas the 
2 receptor
inhibits it. Several polymorphisms of the human BAR-2 gene have been
described, including two common ones in the extracellular domain of the
receptor, Gly16Arg and Gln27Glu. The BAR-2 gene contains a short open
reading frame located 102 bp upstream of the receptor coding block
called the 5' leader cistron, which encodes a 19-amino acid peptide
that inhibits the cellular expression of the BAR-2 by obstructing the
translation of its mRNA (2). A common polymorphism at the
last amino acid of this 5' leader cistron, Cys19Arg, has also been
reported.
Each of these three BAR-2 polymorphisms is known to be functionally relevant. When compared with the Arg16 receptor, the Gly16 receptor has been reported to undergo significantly greater down-regulation after prolonged agonist stimulation in vitro (3, 4). The receptor with the Glu27 allele does not appear to be expressed in the fully mature form on Western blots (3). The Cys19Arg polymorphism in the 5' leader cistron results in an approximately 2-fold lower expression of the BAR-2 receptor at a translational level (5).
Of note, given the role of ß-adrenergic receptors in the control of lipolysis, several studies have reported associations between polymorphisms in the BAR-2 gene and obesity (6, 7, 8, 9, 10, 11, 12, 13). However, this association has not been detected in all studies reported to date (14, 15, 16). Other studies have reported that polymorphisms in the BAR-2 gene might influence the effects of physical activity (11) or diet (17) in the determination of body fat mass.
Although a direct functional effect of these polymorphisms on adipocyte lipolysis appears plausible, none of the previous studies have directly examined the association between these polymorphisms and in vivo measures of circulating free fatty acids. We therefore examined this association in the Ely Study, a population-based cohort study in which measures of fasting and postglucose load nonesterified fatty acid (NEFA) concentrations were available (18, 19, 20, 21). This study is also unique because physical activity has been assessed using an objective and quantitative method rather than simple self-reported behavior (22, 23, 24). The ability to detect gene-environment interactions is highly dependent not only upon the strength of the interaction, but also on the precision with which the environmental exposure is measured (25). Therefore, the detection of such interactions in this study is aided by the use of heart rate monitoring with individual calibration, a method previously validated against the gold standard techniques of doubly labeled water and whole body calorimetry (26, 27). The technique relies on the fact that there is a definable linear relationship between heart rate and oxygen consumption, and therefore energy expenditure, above a critical level (the flex heart rate) below which energy expenditure can be assumed to be equal to resting. As an objective, cheap, and noninvasive method, it is, therefore, potentially suitable for small and medium-sized epidemiological studies in which prospective assessment of energy expenditure is required (28, 29).
Materials and Methods
The Ely Study
Selection of the subjects and metabolic tests. The volunteers were all participants in the Isle of Ely Study, a continuing population-based cohort study in Ely, Cambridgeshire, United Kingdom, the design of which has been described previously (19, 21). The original sample of 1122 individuals without known diabetes were recruited between 1990 and 1992 at random from a population-based sampling frame consisting of all people in Ely, aged 40–65 yr in 1990 (19). The initial response rate was 74%. Fifty-one individuals were found to have prevalent but undiagnosed diabetes (19). Between 1994 and 1997, a second examination was performed at a 4.5-yr interval in all those individuals who did not have diabetes by World Health Organization criteria at baseline (n = 1071). Twenty subjects had died in the interim, and 937 of the remaining volunteers attended the second examination (89% restudy rate) (21). These individuals constituted the sample for this particular study. Because this analysis was focused on the effects of the BAR-2 polymorphism, we excluded individuals from the analysis if they were being treated with ß-blockers at any stage in this study.
At each phase of the study, volunteers attended the clinic at 0830 h, having fasted since 2200 h the previous evening, and underwent a standard 75-g oral glucose tolerance test (OGTT). Blood samples were taken at fasting and 30 and 120 min after oral glucose. Plasma glucose was measured in the routine National Health Service laboratory at Addenbrooke’s Hospital using the hexokinase method and triglycerides measured using the RA 1000 (Bayer Corp., Basingstoke, UK), with a standard enzymatic method. Plasma specific insulin was determined by two-site immunometric assays with either 125I or alkaline phosphatase labels. Cross-reactivity was less than 0.2% with intact proinsulin at 400 pmol/liter and less than 1% with 32–33 split proinsulin at 400pmol/liter. Interassay coefficients of variation were 6.6% at 28.6 pmol/liter (n = 99), 4.8% at 153.1 pmol/liter (n = 102), and 6.0% at 436.7 pmol/liter (n = 99), respectively. The 30-min insulin incremental response is used in the analysis as a measure of insulin secretion, because validation studies comparing this derived measure with the 3-min insulin concentration in an iv glucose tolerance test have shown high correlations (0.55 in normoglycaemic individuals and 0.69 in those with impaired glucose tolerance (30). Plasma NEFA measurements were determined enzymatically on the basis of the activity of acyl-CoA synthetase (Roche Molecular Biochemicals, Lewes, Sussex). The resultant acyl-CoA is oxidized to yield hydrogen peroxide, which is measured colormetrically. NEFA concentrations were measured in time 0, 30- and 120-min samples during the OGTT in both phases of the study. NEFA area under the curve (AUC) was calculated as a measure of the overall NEFA response during the OGTT and was defined as the area under the trapezoid described by the NEFA measurements at time 0, 30 and 120 min (18, 20). The units of this measure are millimoles per liter per hour. Height and weight were measured in light clothing. Body circumferences were measured in duplicate using a metal tape. Ethical permission for the study was granted by the Cambridge Local Research Ethics Committee.
Assessment of resting and exercise oxygen consumption–heart rate
relationship. The protocol for undertaking the individual
calibration between heart rate and energy expenditure has been reported
previously (22, 28, 29). The oxygen consumption–heart
rate relationship was assessed at rest with the subject lying prone and
then seated, using an oxygen analyser calibrated daily using 100%
nitrogen and fresh air as standard gases. Subjects bicycled on a cycle
ergometer at several different workloads to provide the slope and the
intercept of the line relating energy expenditure to heart rate. Each
subject cycled at 50 rpm, and the workload was progressively increased
from 0 W, through 37.5, 75, and 125 W in stages each lasting 5 min. At
each workload, three separate readings were made of heart rate, minute
volume, and expired air oxygen concentration. The 125-W level was
undertaken only if the heart rate had not reached 120 beats per minute
by the end of the 5 min at 75 W. The oxygen concentration in the
expired air and minute volume data were used to calculate oxygen
consumption after correction for standard temperature and pressure.
Energy expenditure (kilojoules per minute) was calculated at each time
point as oxygen consumption (milliliters per minute) x 20.35
(31). Mean resting energy expenditure was taken as the
average of the lying and sitting values. The slope and intercept of the
least-squares regression line of the exercise points were calculated.
Flex heart rate was calculated as the mean of the highest resting pulse
rate and the lowest on exercise. This point was used in the analysis of
heart rate data to discriminate between rest and exercise. Below this
point, energy expenditure was assumed to be equivalent to rest. Above
it, it was predicted from the slope and intercept of the regression
line that was calculated during the exercise test.
O2max was measured from the linear
regression as predicted oxygen consumption at maximal heart rate (220
- age) and is expressed in the results per unit body weight.
The volunteers wore the heart rate monitor (Polar Electro,
Kempele, Finland) continuously during the waking hours over the
following 4 d. Heart rate readings were directly downloaded into a
computer via a serial interface, and the individual calibration data
were used to predict minute energy expenditure for each person.
Sleeping energy expenditure was calculated as 95% of basal metabolic
rate, where this was derived from published prediction equations
(32, 33). A physical activity level (PAL), which is the
ratio of total energy expenditure to basal metabolic rate, was computed
for each day and averaged over the 4-d period.
Genetic analyses
DNA was extracted from white blood cells. The Gly16Arg polymorphism was amplified by PCR using the following oligonucleotides: (sense) 5'-CTT CTT GCT GGC ACG CAA T-3' and (antisense) 5'-CCA GTG AAG TGA TGA AGT AGT TGG-3. The PCR was performed with 0.5 mM MgCl2 and 30 cycles of (94 C 1 min, 56 C 1 min, and 72 C 1 min), generating a 200-bp amplicon. This PCR product was digested with BsrDI at 65 C (the Gly16 allele is distinguished by two bands at 130 and 70 bp; the Arg16 allele by three bands at 22, 108, and 70 bp).
Statistical analyses
The means and SD of anthropometric and physical
activity data were calculated by genotype, and comparison between
genotypes was analyzed by comparison of means for recessive and
dominant models separately. Comparison of the proportion of current,
ex-smokers, and nonsmokers between genotype was performed by the
2 test. The dataset includes repeated
measures, and therefore the mean within an individual over the two
repeats is a closer representation of the usual level than either value
separately. We compared the adjusted means of the repeated data across
individuals stratified by genotype after adjustment for age and smoking
status. The adjustment was undertaken using ANOVA. Those variables that
were not normally distributed were normalized by logarithmic
transformation and are presented in the results as geometric means. We
undertook a similar analysis including body mass index (BMI) as a
covariate. These multivariate analyses were undertaken using the
repeated observations in the Mixed Model procedure MIXED in SAS (SAS
System for Windows, version 6.12, SAS Institute, Cary,
NC). A mixed model is used to describe the variance-covariance
structure of responses at successive time points and to estimate the
corresponding parameters along with fixed effects. Fixed effects
include factors like the genotype, and the model allows for
time-dependent covariates such as BMI. Finally, we examined the
possibility of an interaction between physical activity and genotype.
Because the physical activity data were only collected at the second
visit, this particular multivariate model included the data from the
second visit only.
Results
In this population, the Arg16 allele was common (35%) and
was found in Hardy-Weinberg equilibrium.
The characteristics of the participants stratified by sex and by
genotype for Gly16Arg polymorphism are shown in Table 1
. There
was no statistically significant difference in BMI or waist to hip
ratio (WHR) between genotypes in men. However, in women BMI was
significantly greater in carriers of the Arg allele. There were no
differences between genotypes in the proportion of smokers, nor in PAL
or cardiorespiratory fitness. The
association between genotype and various
anthropometric and biochemical parameters stratified by sex is
shown in Table 2 after adjustment for age and cigarette smoking. The
higher BMI in women carriers of the Arg16 allele persisted after
adjustment for age and cigarette smoking status (P <
0.01 in a dominant model). However, NEFA concentrations at fasting and
30 min and NEFA AUC were all significantly lower in women who were
homozygous for the Arg16 allele (P < 0.01), suggesting
an interaction between genotype and obesity in determining NEFA levels.
The effect on neither BMI nor NEFA concentrations was apparent in the
men. Other differences between groups were only significant at the 5%
level, and given the number of statistical comparisons undertaken in
Table 2, differences at this level should be interpreted with caution.
There was no interaction between overall and central obesity in
determining NEFA levels. In multivariate analysis, after adjustment for
age, sex, and smoking status, a significant interaction between the
BAR-2 genotype and BMI was demonstrated for fasting NEFA
(P < 0.05) and total NEFA AUC (P <
0.05).
|
|
). The previously demonstrated
differences in NEFA concentrations in women by genotype were unaffected
by the addition of BMI as a covariate, suggesting that the apparent
association had not arisen through confounding by obesity. Because of
the previously demonstrated interaction between this polymorphism and
physical activity, we examined the possibility of this gene-physical
activity interaction in separate multivariate models for the NEFA
concentration at each time point in the OGTT, including age, BMI, sex,
and smoking as covariates. In these models, the interaction term
between the genotype and BMI was not significant and was not included.
Table 4 shows the regression coefficients
for the different terms in these models, showing that there is a
statistically significant interaction between the genotype and the PAL
in determining both NEFA AUC and the 30 min NEFA concentration.
Although the interaction term was not significant for the fasting NEFA,
the direction of the interaction was similar, and its magnitude was
comparable. The interactions are displayed graphically in Fig. 1, which demonstrates how at high PAL,
individuals who are homozygous for the Arg16 allele have lower levels
of NEFA at fasting and 30 min and total NEFA AUC. The differences
between genotype are less apparent at low PAL. In an analysis
stratified by sex, the interaction term was stronger in men than in
women, but because men have higher values of physical activity, more of
the data in the male population lies in the range where the interaction
is apparent.
|
|
|
In this study, women carriers of the Arg16 variant of the BAR-2 gene were characterized by higher BMI, but those who were homozygous for Arg16 had lower fasting plasma NEFA concentrations and lower NEFA AUC during the OGTT. Given the overall positive relationship between obesity and NEFA concentrations (20), one would have expected those with the Arg 16 allele to have higher levels. Thus, these observations initially suggested an interaction between the BAR-2 genotype and obesity, an interaction that was statistically significant when tested for in multivariate analysis. However, when physical activity was included in the models, the interaction between BAR-2 and obesity was no longer significant, whereas that between the genotype and physical activity was. Thus the differences between groups in NEFA concentrations can be explained on the basis of an interaction between the habitual level of physical activity and the BAR-2 genotype, with those who are homozygous for the Arg16 variant having lower levels of NEFA when they are more physically active.
It is difficult to explain these phenotypic characteristics with the functional data previously published on the Gly16Arg polymorphism. First, it has been shown that the receptor undergoes a greater down-regulation after prolonged agonist stimulation with the Gly16 allele (less receptor number) than the Arg16 allele (4). In accordance with this result, Martinez et al. (34) reported that children with or without asthma and Arg16Arg homozygous showed more often (five times) a positive response to the bronchodilator albuterol than Gly16Gly carriers. On the other hand, Large et al. (6) showed that adipocytes from subjects carrying the Gly16 allele displayed a 5-fold better sensitivity to terbutaline than Arg16Arg subjects, but the maximum lipolytic action was similar in both groups. More recently, it has been shown that Arg16Arg asthmatic subjects who had regularly used albuterol had a decline in the peak expiratory flow (35). This decline was not observed in asthmatic subjects with other genotypes, and this suggests that the Arg16Arg patients exhibit a tachyphylactic effect to albuterol, maybe because their receptors have not yet been down-regulated. Our study shows that Arg16Arg women have lower plasma NEFAs and greater NEFA suppression after glucose load than other women, and it may be that the catecholamine-induced lipolysis is less efficient in Arg16Arg women. However, it is impossible on the basis of the data in this study alone to determine whether the lower NEFA levels are the product of diminished release or enhanced clearance.
The univariate associations between the BAR-2 polymorphism and body
weight, BMI, and plasma lipids were only seen in women. Many studies
reported sex- and fat regional-related differences in lipolysis
(36), and some have suggested sex differences in
insulin-mediated NEFA metabolism (37). An alternative
explanation for the difference between men and women is that the
interaction between the Gly16Arg polymorphism and physical activity is
stronger in men. We have shown in the context of the interaction
between the peroxisome proliferator-activated receptor 
-gene
and the ratio of polyunsaturated to saturated fat in the diet that
interaction can obscure a main genetic effect if it is not taken into
account (38). Because the interaction in this study is
stronger in men than in women, it follows that the likelihood of the
interaction obscuring the genetic effect would also be greater in
men.
The Ely population is unique in the fact that direct measurements of free living energy expenditure have been undertaken in a large population-based cohort. This provides unprecedented opportunities for the examination of interactions between genotype and physical activity. Although a previous study had suggested an interaction between the BAR-2 locus and questionnaire-based measures of physical activity in the determination of BMI (11), we were unable to show such an interaction using the direct measures of physical activity undertaken in this study. This result is not surprising because we could not find an association between the BAR-2 polymorphism and obesity per se in the present study. However, we did find that the effect of the Gly16Arg polymorphism on NEFA levels was modulated by physical activity, with the effects of the Arg16 variant being accentuated by high PAL. The effects of physical activity on NEFA levels are complex. Acute exercise promotes lipolysis, initiating the release of energy-rich fatty acids from adipocytes (39). However, the use of NEFAs is also increased, an effect that persists for up to several days after exercise (40, 41). The chronic effects of physical activity on NEFA concentrations have been studied in small-scale studies comparing trained and untrained individuals (42), but larger studies in population-based samples have not previously been undertaken, principally because of the difficulties of assessing usual physical activity.
In summary, we have demonstrated that variation at codon 16 in the BAR-2 gene is associated with alterations in fasting and postglucose load NEFA levels and a moderate effect on body weight in women. The finding of a significant interaction between genotype and direct measurement of physical activity demonstrates the importance of considering gene-environment interaction when examining the polygenic determinants of human metabolic variation.
Acknowledgments
We are grateful to the staff of the St. Mary’s Street Surgery (Ely, Cambridgeshire, UK) and to H. Shannasy, S. Curran, P. Murgatroyd, and Drs. M. Hennings and A. M. Prentice for their help with the fieldwork for this study. The staff of the Department of Clinical Biochemistry, Addenbrooke’s Hospital (Cambridge, UK), led by Prof. C. N. Hales, performed the biochemical analyses.
Footnotes
This work is supported by a program grant from the Medical Research Council (MRC) of the United Kingdom (to S.O.). The Ely Study was funded by the British Diabetic Association, the Anglia and Oxford Regional Health Authority, and the MRC. A.M. is funded by the Marie Curie Association. P.S.-F. is an MRC Ph.D. student. N.J.W. is an MRC Clinician Scientist Fellow.
Abbreviations: AUC, Area under the curve; BAR-2, ß2-adrenergic receptor; BMI, body mass index; NEFA, nonesterified fatty acid; OGTT, oral glucose tolerance test; PAL, physical activity level; WHR, waist to hip ratio.
Received January 24, 2001.
Accepted August 31, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  M. Tomaszewski, F. J. Charchar, B. Lacka, U. Pesonen, W. Y.S. Wang, E. Zukowska-Szczechowska, W. Grzeszczak, and A. F. Dominiczak Epistatic Interaction Between {beta}2-Adrenergic Receptor and Neuropeptide Y Genes Influences LDL-Cholesterol in Hypertension Hypertension, November 1, 2004; 44(5): 689 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dallongeville, N. Helbecque, D. Cottel, P. Amouyel, and A. Meirhaeghe The Gly16->Arg16 and Gln27->Glu27 Polymorphisms of {beta}2-Adrenergic Receptor Are Associated with Metabolic Syndrome in Men J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4862 - 4866. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Pereira, M. S. Floriano, G. F.A. Mota, R. S. Cunha, F. L. Herkenhoff, J. G. Mill, and J. E. Krieger {beta}2 Adrenoceptor Functional Gene Variants, Obesity, and Blood Pressure Level Interactions in the General Population Hypertension, October 1, 2003; 42(4): 685 - 692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Martinez, M. S. Corbalan, A. Sanchez-Villegas, L. Forga, A. Marti, and M. A. Martinez-Gonzalez Obesity Risk Is Associated with Carbohydrate Intake in Women Carrying the Gln27Glu {beta}2-Adrenoceptor Polymorphism J. Nutr., August 1, 2003; 133(8): 2549 - 2554. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
