Clinical Science (1995) 89, 421-429 (Printed in Great Britain)
421
Effects of weight reduction on the regulation of lipolysis
in adipocytes of women with upper-body obesity
Signy REYNISDOTIIR, Dominique LANGIN*, Kjell CARLSTROMt, Cecilia HOLMt,
Stephan ROSSNER§ and Peter ARNER
Department of Medicine and Research Centre, tDepartment of Obstetrics and Gynecology,
Huddinge University Hospital, Huddinge, Sweden, §Department of Medicine, Karolinska
Hospital, Karolinska Institute, Stockholm, Sweden, *INSERM Unite 317, Institut Louis Bugnard,
Faculte de Medicine, CHU Rangueil, Toulouse, France, and tDepartment of Medical and
Physiological Chemistry, Lund University, Lund, Sweden
(Received 24Januaryj22 May 1995; accepted 29June 1995)
1. The regulation of lipolysis was studied in 14
upper-body obese women aged 26-55 years. Isolated
subcutaneous adipocytes from the abdominal region
were examined before and after an 8 to 12-week-long
weight reduction programme, during which the mean
body mass index (kg/m Z ) of the group was reduced
from 36.4 ±0.9 to 30.3 ± 1.0.
2. Fat cell volume was reduced from 891 ± 39 to
655 ± 45 pl, The sensitivity to noradrenaline stimulation of lipolysis in vitro increased fivefold after weight
reduction. A corresponding increase in sensitivity was
found with the liz-selective adrenoceptor agonist
terbutaline. However, the number of IIz-adrenoceptor
binding sites as assessed by radioligand binding with
12sI_labelied cyanopindolol was not changed. No
changes were observed when dobutamine (a 111selective adrenoceptor agonist) and clonidine (an
Otz-adrenoceptor agonist) were used.
3. The basal lipolysis rates decreased by about 50%
after weight reduction and the maximum enzyme
activity of hormone-sensitive lipase was also reduced
by almost 50%.
4. Plasma concentrations of insulin, noradrenaline
and total testosterone decreased and sex hormonebinding globulin increased after weight reduction.
Calculated apparent free testosterone levels decreased
by more than 40% after weight reduction.
5. In conclusion, weight reduction leads to increased
efficiency of adipocyte lipolysis with decreased resting
lipolysis rate but increased sensitivity to stimulation
by catecholamines, which may be attributed to a
decreased activity of hormone-sensitive lipase and an
increased sensitivity of IIz-adrenoceptors. Changes in
circulating levels of catecholamines, insulin and
testosterone may play a role in these modifications of
adipocyte function. The increased lipolytic efficiency
may be of importance for amelioration of the meta-
bolic complications of obesity that are observed after
weight reduction.
INTRODUCTION
Dietary treatment of obesity results in a reduction
in body weight, mainly through a decrease in fat
mass [1]. The decrease in fat mass is due to a
decrease in fat cell size, which suggests that changes
in adipocyte metabolism may be of importance for
the overall metabolic effects of weight reduction.
In fat cells, the hydrolysis of intracellular triglyceride stores to glycerol and non-esterified fatty
acids by lipolysis is one of the most important
metabolic processes. Adipocyte lipolysis is under
intense hormonal control, catecholamines being the
most important lipolytic hormones in humans. They
act via dual G-protein-coupled adrenergic receptors,
stimulating lipolysis via fJ 1-' fJ2~ and 133adrenoceptors and inhibiting lipolysis via 0(2adrenoceptors [2]. The final and rate-limiting step is
the activation of hormone-sensitive lipase through
cyclic AMP-mediated phosphorylation [3]. The
effect of catecholamines can be modulated by a
number of hormones. The most important of these
are insulin and steroid hormones [2].
Decreased sensitivity to catecholine-stimulated
lipolysis has previously been demonstrated in obese,
in particular upper-body obese, subjects in vitro
[4, 5] as well as in vivo [6-8]. An increased sensitivity to catecholamine action has been found in
vivo in previously obese subjects [9]. These data
indicate that weight reduction may modify the
lipolytic activity of adipocytes. Although few studies
have examined the effect of weight reduction on
lipolysis regulation at the cellular level, a decreased
overall lipolysis rate is a common observation
Key words: adrenoceptors, catecholamines, cyclic AMP, hormone-sensitive lipase.
Abbreviations: A4, 4,-androstene-3,17-dione; DHEA, dehydroepiandrosterone; DHEA(S). dehydroepiandrosterone sulphate; SHBG, sex hormone-binding globulin; VLCO, very
low;:alorie diet.
Correspondence: DrSigny Reynisdottir, Department of Medicine. Huddinge University Hospital, 5-141 86 Huddinge, Sweden.
422
S. Reynisdottir et al.
[10, 11]. The mechanism behind this change is not
well understood.
In the present study we have examined the
mechanisms of lipolysis activation at the cellular
level in 14 upper-body obese women before and
after diet-induced weight reduction when the subjects were in a new steady state of body weight. The
effect of weight loss on adrenoceptors was studied in
lipolysis experiments and radioligand binding
studies. Hormone-sensitive lipase activity was determined and the levels of circulating hormones that
may be involved in lipolysis regulation were
assessed.
MATERIALS AND METHODS
40
~ 35
b:o
~
x
]
<.
t
'. . . .
<.
.....
~
E
-?;0
a>
30
·········+·······+······t
t
o-t----,----,----r----.,
o
10
15
w
Time (weeks)
Subjects
Fourteen obese (body mass index 30.8-41.5 kg/
rn"), but otherwise healthy, women participated in
the study. They were selected for upper-body distribution of body fat by means of the waist-hip ratio.
The waist and hip circumferences were measured at
the level of umbilicus and at the widest part of the
hip region respectively. The lower limit for the
definition of upper-body obesity is usually set at
0.85 [12]. All subjects included in the present study
had a waist-hip ratio of > 0.87 and thus were
markedly upper-body obese. All were sedentary and
had had stable body weight for at least I year
preceding the study. Their mean age was 42.2 ± 2.1
(26-55) years. Two of the women were postmenopausal. This was not expected to affect the
results since androgen satus is essentially unaffected
by menopause [13J, the reduced levels of testosterone and steroid hormone-binding globulin
(SHBG) commonly observed in post-menopausal
women being instead the result of a gradual
decrease in these levels with age [14]. All subjects
had given their informed consent before entering the
study. The study was approved of by the ethics
committee of the Karolinska Institute.
The subjects were first examined 2-4 weeks before
entering a weight reduction programme. They were
then given a balanced very low-calorie liquid
formula diet (VLCD), administered five times daily,
a total of 400 kcal/day for 8-12 weeks. The diet kit
(Nutrilette, Nycomed, Oslo, Norway) consisted of
63()" protein, 6"" fat and 31';" carbohydrate and was
supplemented with essential vitamins and minerals.
The subjects were then re-examined 4-6 weeks after
having returned to an ordinary isocaloric diet
directly after the period on the liquid formula diet
(Fig. I). At the second time of investigation they
were in a new steady state of body weight.
On both occasions the subjects were examined at
08.00 hours after an overnight fast in the follicular
phase of the menstrual cycle. First, venous blood
samples for analysis of hormones and metabolites
were collected. Blood pressure was measured in the
supine position. Second, a subcutaneous fat biopsy,
Fig. I. Mean body mass beforeand after &-12 weeks of treatment of
14 upper-body obesewomen with a very low-alorie diet. Thearrows
indicate when the fat biopsies were obtained.
of about 2-4 g, was surgically removed from the
paraumbilical region under local anaesthesia as previously described [15]. The tissue was immediately
transported to the laboratory in saline. Small pieces
of the tissue, about 0.3-0.5 g, were immediately
frozen in liquid nitrogen for later analysis of
hormone-sensitive lipase activity. The remaining
tissue was used for lipolysis and radioligand binding
experiments.
For separate methodological experiments, intraabdominal (omental) adipose tissue was obtained
from three otherwise healthy subjects who underwent general surgery for treatment of obesity. None
of these subjects was on any medication. The tissue
was treated exactly as described for the speciments
obtained through biopsy.
Drugs and chemicals
Bovine serum albumin (fraction V, lot 63F-0748),
Clostridium histiolyticum collagenase type I, propranolol, forskolin, dibutyryl cyclic AM P and glycerol
kinase from Escherichia coli (G4509) were obtained
from Sigma (St Louis, MO, U.S.A.). Isoprenaline
came from Hassle (Molndal, Sweden), terbutaline
from Draco (Lund, Sweden), dobutamine from
Lilly (Indianapolis, IN, U.S.A.) and ICI 118,551
from Cambridge Research Biochemicals Limited
(Cheshire, U.K.). ATP monitoring reagent containing firefly luciferase was from LKB Wallac (Turku,
Finland). 125-labelled cyanopindolol came from
New England Nuclear (Boston, MA, U.S.A.). Antipain and leupeptin were obtained from Sigma
(L'isle d'Abeau, France). 1(3)-monoeHJoleoyl-2oleylglycerol was manufactured at the Department
of Medical and Physiological Chemistry, Lund
University (Lund, Sweden). All other chemicals were
of the highest grade of purity commercially available. Collagenase and other ingredients in the incu-
Lipolysis and weight reduction
bation buffers were from the same batches throughout the study.
Isolation of fat cells and lipolysis experiments
Isolated fat cells were prepared by incubation in
collagenase according to Rodbell [16]. Fat cell
diameter, weight and volume as well as the number
of fat cells were determined as previously described
[4].
The lipolysis assay has previously been described
in detail [17]. Briefly, a diluted suspension of fat
cells (about 10000 cells/ml) was incubated for 2 h at
37°C in duplicate samples with or without increasing concentrations of different lipolytic agents. We
used the non-selective natural catecholamine noradrenaline, which acts on all adrenoceptors present
in human fat cells, i.e. the C(2-' Pl-, P2- and
P3-adrenoceptors [18]. We also used isoprenaline,
an adrenergic agonist that acts on all three Padrenoceptor subtypes, the pz-adrenoceptor selective
agonist terbutaline and the PI -adrenoceptor selective agonist dobutamine. Furthermore, we used the
C(2-adrenoceptor agonist clonidine; forskolin, which
stimulates adenylate cyclase; and dibutyryl cyclic
AMP, which activates the cyclic AMP-dependent
protein kinase, which in turn phosphorylates and
activates hormone-sensitive lipase. Glycerol release
to the incubation medium was determined using an
automated bioluminescence assay [19] and was
used as a measure of the lipolysis rate.
All the agonists caused a dose-dependent increase
or inhibition of glycerol release, reaching a plateau
at the highest agonist concentrations. The sensitivity
to agonist action was defined as the pD 2 value, i.e.
the negative logarithm of the ED 50 value, defined as
the concentration of each agonist giving halfmaximum effect. The ED 50 value was determined by
linear regression analysis after log-logit transformation of each individual dose-response curve. The
responsiveness to each agonist was defined as the
lipolysis rate at the maximum effective agonist
concentration. The lipolysis rates in the presence or
absence of agonist were related to the number of
incubated cells and expressed per cell number.
p-Adrenoceptor binding assay
The radioligand binding studies have previously
been described in detail previously [17]. Briefly,
isolated adipocytes (20000 cells/rnl) were incubated
for 60min at 37°C with 1251-labelled cyanopindolol
and cell-bound radioactivity was measured after
vacuum filtration through a Whatman GF/C filter.
Saturation experiments were performed in duplicate with increasing concentrations of the radioligand (1G--750pmol/I). Non-specific binding was
measured in the presence of propranolol (10- 5mol/l)
and was about 30% at low and 45% at high
radioligand concentrations. The total number of Padrenoceptor binding sites and the K d values for the
423
radioligand were determined through linear regression analysis of Scatchard plots [20].
Displacement experiments were performed in
which the radioligand (l00pmol/l) was displaced by
12 increasing concentrations of the P2-selective
antagonist ICI 118,551 (0, 10- 11 to 10- 4 mol/l).
Non-specific binding, defined as the binding
at 10- 4 mol/l, was about 20%. The experiments
gave shallow biphasic displacement curves. A nonlinear least-squares regression analysis was performed using a computerized evaluation program
(LIGAND) to determine the proportions of high(P2) and low- (PI) affinity binding sites.
When the results of the saturation and displacement experiments were taken together, the number
of Pl- and pz-adrenoceptor binding sites could be
estimated.
Previous studies indicate that the concentrations
of the radioligand used are too low to identify
p3-adrenoceptors, for which the radioligand has
been reported to have low affinity [21]. To confirm
that this was also true in the present setting, a
separate set of methodological experiments were
performed using omental adipocytes obtained from
three subjects who underwent surgical treatment of
obesity. Saturation experiments were performed as
described above, but with the addition of several
higher concentrations of 125 1-labelled cyanopindolol
(up to 1500pmol/I).
Assay of hormone-sensitive lipase activity
This assay was performed essentially as described
previously [22, 23]. Pieces of adipose tissue (about
0.3 g) were homogenized in 3 ml of a buffer containing 0.25 mol/l sucrose,
1 mmol/l EDTA,
1 mmol/l dithioerythritol and the protease inhibitors
leupeptin and antipain, both at 20 Jlg/ml. The samples were then centrifuged at 100000 g for 45 min at
4°C in a Beckman ultracentrifuge L8-60M. The
fat-free infranatant was recovered for analysis of
enzymatic activity, which was determined using
1(3)-mono[3H]0Ieoyl-2-0Ieylglycerol as substrate.
All samples were incubated in triplicate for 30"min
at 37°C. The use of a diacylglycerol analogue as
substrate enhances the sensitivity of the assay, since
hormone-sensitive lipase has 10-fold higher activity
towards diacylglycerol than triacylglycerol [22].
Moreover, since this substrate only has one hydrolysable ester bond at the 1(3)-position, neither the
substrate itself nor its hydrolysis products can be
hydrolysed by monoacylglycerol lipase, which is
abundant in adipose tissue. Furthermore, under the
conditions for the assay, i.e. pH 7.0 and no apo CII
present, very low lipoprotein lipase activity is
measured [22]. One unit of enzyme activity is
defined as 1 Jlmol of fatty acid released per minute
at 37°C. Lipase activity was related to the protein
concentration of the infranatant, which was
measured according to Bradford [24] using bovine
serum albumin as a standard.
424
S. Reynisdottir et al.
Assay of hormones and binding proteins
Plasma concentrations of insulin, cortisol, transcortin (corticosteroid-binding globulin), testosterone
and SHBG were determined by radioimmunological
or (SHBG) immunoradiometric methods using commercial kits from Pharmacia, Uppsala, Sweden
(insulin), Orion Diagnostics, Turku, Finland (cortisol, SHBG), Diagnostic Products, Los Angeles, CA,
U.S.A., and Medgenix Diagnostics, Fleurns, Belgium
(transcortin). Free cortisol concentrations were calculated from total cortisol and transcortin values
according to instructions from the manufacturer of
the transcortin kit. Free testosterone was calculated
from values for total testosterone, SHBG and a
fixed albumin concentration of 40 gil using an
equation derived from the law of mass action as
described by Sodergard et al. [25]. Plasma concentrations of 4-androstene-3,17-dione (A4), dehydroepiandrosterone (DHEA) and its sulphate
[DHEA(S)] were determined by radioimmunoassay
after extraction with diethylether [26]. In the
DHEA(S) assay the conjugate was cleaved by
thermal hydrolysis before ether extraction. Plasma
catecholamine levels were determined as previously
described [17]. Blood glucose was analysed at the
hospital's routine chemistry laboratory.
Statistical analysis
Statistical analysis was performed using Wilcoxon's test for paired observations. All values are
expressed as means ± SEM. In some cases the Spearman rank correlation test was applied. The values
for K d from the radioligand binding experiments
were transformed into their logarithmic form before
statistical evaluation.
RESULTS
The mean changes in body mass before and after
the period with VLCD are shown in Fig. 1. All
subjects achieved a substantial (7-24%, 10-25 kg)
weight loss after, on average, 10 weeks of diet. The
subjects had a stable body weight for on average 5
weeks when re-examined.
Clinical data are presented in Table 1. The mean
body mass index in the whole group was reduced by
almost 17% with a slight concomitant decrease in
the waist-hip ratio (3%). Fat cell size decreased by
about 25%. Blood pressure and heart rate decreased
significantly. This was also true for plasma glycerol
levels. Blood glucose and insulin levels decreased as
well, indicating increased insulin sensitivity. The
circulating levels of noradrenaline decreased significantly, but plasma adrenaline levels remained
unchanged. Plasma concentrations of cortisol, transcortin and free cortisol and of the purely adrenocortical androgens DHEA and DHEA(S) were not
affected by weight reduction. Total testosterone
concentrations were markedly reduced and a signifi-
Table I. Clinical data, hormones and metabolites before and after
weight reduction. BMI. body mass index; SHBG, sex hormone-binding
globulin; A4, 4-androstene-3, 17~ione; DHEA, dehydroepiandrostenedione;
DHEA(S), dehydroepiandrostenedione sulphate. The values are means ± SE.
All values were compared using Wilcoxon's test. NS, not significant.
Obese
~6.4 ± 0.9
BMI (kg/m')
0.9S±0.01
Waist-hip ratio
Heart rate (beats/min)
7H3
Systolic blood pressure (mmHg)
129±3
Diastolic blood pressure (mmHg)
78±2
5.3±0.2
Blood glucose (mmol/l)
10.9±1.2
Plasma insulin (mUll)
1.82±0.13
Plasma noradrenaline (nmol/I)
0.10±0.01
Plasma adrenaline (nmol/l)
Plasma glycerol (mmol/l)
148± 18
Serum cortisol (nmol/l)
350±34
23,9±4.3
Free serum cortisol (nmol/I)
1.43±0.22
Testosterone (nmol/l)
22.9±3,5
SHBG (nmolll)
Free testosterone (nmol/l)
0.037 ± 0.006
4.17±O.74
A4 (nmolll)
DHEA (nmol{l)
15.O±H
2831 ±646
DHEA(S) (nmol/l)
Cell volume (pi)
891 ±39
Weight reduced
obese
30.3± 1.0
0.92±0.01
68±2
117±2
70±2
4.8±0.2
7.0±0.8
1.28±0.14
0.IHO.03
III ± 12
342± 18
22.9±2.3
0.98±0.18
33,7±3,7
0.022 ± 0.004
3,83 ± 0.76
12.7±3,1
2586 ± 665
655 ±45
<0.001
<0.05
<0.01
<0.01
<0.005
<0.005
<0.05
<0.005
NS
<0.05
NS
NS
<0.01
<0.01
<0.005
<0.05
NS
NS
<0.0005
cant decrease was also observed in the A4 levels.
SHBG concentrations were significantly increased
by weight reduction. Together with the decrease in
total testosterone, this resulted in a pronounced fall
in circulating concentrations of free testosterone to
a mean level of 59% of the pretreatment value. The
two post-menopausal women did not differ from the
remaining 12 subjects as regards the effect of weight
reduction on SHBG and steroid hormone levels. A
significant negative correlation (r= -0.67, P<0.05)
was found between SHBG and insulin after weight
reduction but not before (r= -0.45, P=0.12).
Weight reduction caused an overall decrease in
the response of adipocytes to lipolysis activation by
noradrenaline (Fig. 2). However, the dose-response
curve for noradrenaline was shifted to the left after
weight loss, indicating an increased sensitivity to
hormone action. This is illustrated in Fig. 3, in
which the mean results of lipolysis stimulation are
expressed as a percentage of the maximum lipolysis
rate. The dose-response curves for each of the
adrenoceptor
subtype-specific
agonists
were
examined in the same manner to evaluate the
relative importance of each receptor subtype for the
shift in noradrenaline sensitivity (Fig. 3). There was
a shift to the left of the dose-response curve for the
fJz-selective agonist terbutaline, while the curves for
the fJ l-adrenoreceptor-selective agonist dobutamine
were practically superimposed, indicating a selective
change in fJz-adrenoceptor status. There was no
significant change in the sensitivity to the inhibitory
effect of the !Xz-agonist clonidine. Statistical comparison of the pD z values for each agonist are found in
Table 2. There was a fivefold increase in nor-
Lipolysis and weight reduction
40
~ 30
<5
E
..m
..:;, 20
e
~
.. 10
!;:.
i3
O+---,----r--r---r-----r----,-.-,
o
-10
-9
-8
-7
-6
Noradrenaline (log mol/I)
-5
-4
Fig. 2. Mean dose-response curves for lipolysis before (0) and after
(e) weight reduction in 14 obese subjects. Isolated fat cells were
incubated with increasing concentrations of noradrenaline. The lipolytic
effect is expressed as glycerol release per cell number.
adrenaline sensitivity and an eightfold increase in
terbutaline sensitivity but no significant change in
the pD z values of do but amine or clonidine after
weight reduction. The increase in terbutaline sensitivity was found to correlate significantly with the
change in body mass (r=0.55 and P<0.05).
The increase in sensitivity to f3z-adrenoceptormediated lipolysis was, however, not accompanied
by any significant change in the density of either 131or f3z-adrenoceptor binding sites (Table 3). This was
regardless of the denominator used: cell number or
cell surface area respectively. The K d values for the
radioligand (calculated from the Scatchard plots)
and the K d values for the high- and low-affinity
binding sites for the displacing drug, fCl 118,551
(calculated from the displacement experiments), were
not affected by weight reduction (data not shown).
The results of methodological experiments performed to evaluate the possible contribution of
f3radrenoceptors to the results of the binding
studies are shown in Fig. 4. Saturation experiments
with high concentrations of the radioligand
(1500pmol/l) gave linear curves, with a Hill coefficient near unity, indicating that all identified binding sites have the same affinity for the ligand. This
suggests that no f33-adrenoceptor binding sites are
identified by the radioligand since previous reports
indicate that the radioligand has 10 to 100-fold
lower affinity for the human 133- than for 131" or
f3z-adrenoceptors [21]. The x, value (-logmol/l)
for the radioligand, calculated from the Scatchard
plots, was 10.1 ±0.1. Non-specific binding determined with propranolol (10- 5 mol/I) was about 60%
at the highest radioligand concentrations.
The overall decrease in the lipocytic effect of
noradrenaline (Fig. 2) suggests additional effects of
weight reduction on fat cell lipolysis, which are not
related to adrenoceptor function. A marked decrease
in the basal lipolysis rate in the absence of the
lipolytic agent was observed after weight reduction.
425
There was also a marked decrease in the maximum
lipolytic effect of agents acting at different levels in
the lipolytic cascade: the non-selective f3-agonist
isoprenaline, forskolin, which activates the adenyl ate
cyclase, and dibutyl cyclic AMP, which activates the
protein kinase-hormone-sensitive lipase complex
(Fig. 5a). However, when these results are expressed
with the basal rates subtracted (Fig. 5b), there is no
significant change in the maximum lipolysis rate.
Maximum terbutaline- and dobutamine-stimulated
lipolysis was also decreased after weight reduction;
again this difference was abolished when basal
lipolysis rates were subtracted (data not shown).
In order to investigate a possible influence of
changes in adipocyte volume on changes in lipolysis, a correlation analysis was performed. The
changes (before minus after) in basal, maximum
noradrenaline- and maximum isoprenaline-induced
glycerol release as well as adipocyte volume were
calculated. There was no significant relationship
between the change in cell volume on one hand and
the changes in the lipolysis parameters on the other
(r=0.18-0.42).
The required amount of adipose tissue for the
assay of hormone-sensitive lipase activity could be
obtained at both occasions from 11 of the 14
subjects. There was a decrease in lipase activity of
about 50% after weight reduction. The values were
45 ± 7 m-unit/mg protein before and 23 ± 4 m-units/mg
protein after weight reduction (P < 0.05).
DISCUSSION
Marked-effects of body weight reduction on lipolysis regulation in obese subjects were observed in
the present study. Two major changes in
catecholamine-stimulated lipolysis were induced.
First, an approximately fivefold increase in the
sensitivity to noradrenaline was observed, shifting
the mean dose-response curve to the left. This effect
of weight loss was adrenoceptor subtype specific.
There was an increase in the sensitivity to the
f3z-adrenoceptor subtype-specific agonist terbutaline,
but no change in the sensitivity to dobutamine
(f31-adrenoceptor selective) or to clonidine (!Xzadrenoceptor selective), indicating that the change in
noradrenaline sensitivity was due to a selective
increase in f3z-adrenoceptor sensitivity. This was,
however, not explained by an increased number of
f3z-adrenoceptor binding sites as determined by
radioligand binding. One possibility is that the
interaction of f3z-adrenoceptors with the Gs protein
is facilitated. Alternatively, some other modification
of f3z-adrenoceptor function may be induced,
improving receptor coupling to lipolysis. However,
the amount of adipose tissue obtained from each
subject did not allow investigations of these
mechanisms.
The possible influence of f33-adrenergic receptors
on lipolysis activation was not evaluated in this
study. Although the results with the selective agents
S. Reynisdottir et al.
426
(a)
100
<U
!2
~
(c)
80
]>-
""60
E
E
.~
E
40
(;
c
<U
20
<U
a..
-10
(b)
<U
!2
~
I
-9 -8 -7 -6 -5
Noradrenaline (log mol/l)
20
s:
0
-4
(d)
100
0
-11-10-9 -8 -7 -6 -5 -4
Dobutamine (log mol/I)
100
r.I
<U
80
80
~
e
~
""
~
"" 60
E
E
E
'x 40
60
40
-'"
(;
(;
c
~
e
U
c
20
20
<U
U
<U
U
<U
a..
<U
o+-~--'-----,---,--r---'F===-'
c;
-12-11-10-9-8-7-6-5-4-3
Terbutaline (log mol/l)
0
-12 -II -10-9 -8
Clonidine (log mol/l]
-7-6
Fig. 3. Lipolysis before (0) and after (e) weight reduction in 14 obese subjects. Mean dose-response curves for (a)
noradrenaline (non-specific adrenergic agonist). (b) terbutaline (selective fi,-receptor agonist) and (c) dobutamine (selective
/I,-agonist). To illustrate sensitivity to agonist action, the data are expressed as a percentage of the glycerol release at the maximum
effective agonist concentration with basal lipolysis rates subtracted. (d) The antilipolytic effect of the !X,-agonist clonidine isshown asa
per cent inhibition of the basal lipolysis rate.
Table 2. Mean pDz values before and after weight reduction. pD, is
defined as the negative logarithm of ED", the concentration of each agonist
giving half-maximum effect. NA, noradrenaline (non-specific adrenergic
agonist); TER, terbutaline; a selective fl,-agonist; DOBU, dobutamine. a
selective {I,-agonist; CLO, c1onidine, an !X,-agonist. Values are means t SE.
They were compared using Wilcoxon's test. NS, not significant,
0.10
008
<U
<U
0.06
•
•
~
§
ED" (log rnol'l)
Obese
Weight reduced
obese
0
<0
NA
TER
DOBU
CLO
-7.63
to.29
-8,17
tOl6
<O,OS
-7.75
to,17
-8.65
to,28
<0,05
-743
to,08
-7.54
to.13
NS
-9.75
to.30
-9.30
to.26
NS
Table 3. Radiological binding data in obese subjects before and after
weight reduction. The number of adrenoceptor binding sites was determined by saturation experiments with '''I-labelled cyanopindolol and displacement experiments using ICI 118,551 as displacing drug, Data are both
expressed per unit cell surface area (amol/pm') and per cell number
(pmol/lO' cells). Values are means ± SE. They were compared using
Wilcoxon's test. NS, not significant.
{I-Adrenoceptor binding sites
amolipm'
Obese
Weight reduced
obese
pmolllO' cells
{i,
{I,
/I,
{i,
147tOl0
1.64±0.18
0.63 t 0.03
0.7ltO,08
1.39±0.09
NS
181 ±0.19
NS
0.50 t 0.05
NS
0,6310,06
NS
0.04
•
0.02
0.00
0.000
0001
•• •
0.002
0.003
Bound
Fig. 4. Evaluation of the contribution of p...adrenoceptors to radioligand binding with Ilsi-labelled cyanopindolol. Saturation experiments
using 10-1500 pmolll of the radioligand were performed. A Scatchard plot
showing one representative experiment out of three is depicted,
terbutaline and dobutamine indicate that the
changes in noradrenaline sensitivity are exclusively mediated by changes in p2-adrenoceptor
function, we cannot, on the basis of the present
data, completely exclude the possibility that
P3-adrenoceptors may also have been involved. This
receptor is less well expressed on a functional basis
in subcutaneous adipocytes (presently investigated)
than on visceral fat cells [18]. Intra-abdominal
adipose tissue is, however, for technical and ethical
reasons, not available for this type of study. Further-
lipolysis and weight reduction
(0) SO
:ii!40
Q;
v
~
] 30
.=,
.,
:Q
~ 20
~
~
G 10
0
Basal
NA
ISO
Forskolin
dcAMP
(b) SO
~40
e
0
] 30
.=,
.,
:Q
~ 20
e
tt
G 10
NA
ISO
Forskolin
dcAMP
Fig. 5. Basal and maximum lipolysis in 14 obesesubjects before(D)
and after (.) weight reduction. Glycerol release is shown in the
absence (basal) or presence of maximum effective concentrations of noradrenaline (non-selective catecholamine), isoprenaline (iso, non-selective
p-agonist), forskolin (which activates adenylate cyclase) and dibutyryl cyclic
AMP (dc-AMP, activating the cAMP-dependent protein kinase-hormone
sensitive lipase complex). Results are expressed per cell number with the
values for basal Iypolysis included (0) or excluded (b). The values were
compared using Wilcoxon's test (*P<0.05, **P<O.OI).
more, methodological data revealed that the 133adrenoceptor did not contribute in a significant way
to the results of radioligand binding, which is in
accordance with previous findings [21].
We have previously observed decreased catecholamine sensitivity due to decreased f3z-adrenoceptor
function in upper-body obese women when compared with healthy control subjects [4]. The present
data indicate that catecholamine resistance in
upper-body obesity is reversed by weight reduction,
at least where women are concerned.
The circulating levels' of some of the most important hormones that can modulate catecholamine
effects in peripheral tissues [2, 27] were determined
before and after weight reduction. A decrease in
insulin levels was observed. This may contribute to
the observed increase in f3-adrenoceptor-mediated
catecholamine sensitivity, since one effect of insulin,
the most important antilipolytic hormone, is acute
down-regulation of f3-adrenoceptors [28]. Whether
this also applies for chronic hyperinsulinaemia
remains to be established. It should be noted that
427
the present assay cannot separate insulin from proinsulin. It is thus possible that true changes in
circulating insulin after weight reduction might be
slightly different from the presently observed
changes.
Plasma noradrenaline levels decreased after body
weight reduction as well. It can be speculated that
this occurs in response to the observed increase in
catecholamine sensitivity in adipose tissue, although
some caution should be exercised in extrapolating
results with venous noradrenaline to adipose tissue
noradrenaline. Unfortunately, it is at present difficult to determine accurately the noradrenaline
concentration at the fat cell level. However, the
selective f32-adrenoceptor-mediated resistance to
catecholamines found in obese women [4] may also
be secondary to an increased sympathetic drive in
the obese state. This is in accordance with animal
studies, which indicate that f32-adrenoceptors are
more susceptible than f31-adrenoceptors to downregulation in response to elevated catecholamine
levels [29].
The levels of total and free testosterone as well as
A4 decreased markedly after weight reduction. The
ovary is responsible for about 40% of circulating
testosterone and 25% of A4 in adult women [30].
Insulin is known to act synergistically with luteinizing hormone in the stimulation of ovarian androgen
synthesis [31]. The observed decrease in androgen
levels after weight reduction may therefore be
explained by decreased insulin levels. The concomitant increase in SHBG levels may be the result of
decreased insulin levels as well, since insulin is
known to be a powerful inhibitor of SHBG synthesis [31,32]. There was a negative correlation
between insulin levels and SHBG in the present
study. In contrast to the steroids of partial ovarian
origin, the purely adrenocortical steroids cortisol,
DHEA and DHEA(S), were not affected, indicating
an inertness of the pituitary-adrenocortical axis to
the metabolic changes provoked by weight
reduction.
In the group examined here, the androgen levels
decreased markedly, while f3-adrenoceptor sensitivity
increased after weight reduction. This is an apparent
contrast to previous findings regarding testosterone,
which enhances lipolysis via increased 13adrenoceptor expression [27]. However, these findings have been divergent, reflecting the complexity
of androgen effects on lipolysis regulation, which
involve gender as well as species differences [2].
The second change in adipocyte metabolism
observed after weight loss was opposite to that
described above, causing an overall decrease in the
lipolysis rate. This was found with all lipolytic
agents used, acting via adrenoceptors as well as at
post-receptor levels. However, only the basal lipolysis rate was reduced, the stimulatory agonist effect
being essentially the same before and after weight
loss when the basal lipolysis rates were subtracted.
As discussed previously [33], it is well known that
428
S. Reynisdottir et al.
the rate of spontaneous lipolysis during incubation
is closely correlated with fat cell size. We presently
observed a decrease in cell size as well as in basal
lipolysis after weight reduction. Whether the
decrease in lipolysis is merely an artefact of the
incubation procedure or also occurs in vivo is not
clear, since the corresponding 'basal' conditions are
not easily defined in vivo. However, the present
results are probably not only secondary to a
decrease in fat cell size. Although adipocyte volume
decreased in parallel with the changes in lipolysis
after body weight reduction, there was no direct
relationship between the alterations in cell size and
lipolysis rates. In a recent in vivo study with the
microdialysis technique, an increase in resting lipolysis was found in obesity, regardless of body fat
distribution, but not when the enlarged fat mass was
taken into account [34]. When the resting lipolysis
rate was measured as the turnover of labelled
palmitate, an increased lipolysis rate was found in
the subcutaneous abdominal fat depot in upperbody, but not lower-body, obese women [8, 35].
It is possible that the observed decrease in
hormone-sensitive lipase function may be responsible for the decrease in the basal lipolysis rate after
weight reduction. The present method of measuring
hormone-sensitive lipase does not differentiate
between phosphorylated and dephosphorylated
enzyme; instead, the total amount of activatable
lipase is measured [2, 23]. Thus, weight reduction
may also have caused changes in the state of
phosphorylation of the enzyme or affected the components involved in the interaction of the lipase
with the lipid droplet [3, 36]. However, at present,
it is not possible to quantify phosphorylated and
unphosphorylated forms of hormone-sensitive lipase
in adipose tissue biopsies. Basal lipolysis rates and
hormone-sensitive lipase activity increase during
starvation in rats [37]. This is not in conflict with
the present findings since the subjects were
examined after several weeks on an isocaloric diet,
when there should be no remaining influence of the
preceding hypocaloric diet.
Women with a predominantly upper-body fat
distribution were selected for this study. This type of
obesity is more strongly associated with the adverse
metabolic effects of obesity than lower-body obesity
[38]. Lipolysis regulation may also differ between
individuals with the same body mass but different
body fat distribution. Furthermore, there may be
gender differences in lipolysis regulation and the
impact of body weight reduction. Thus, the present
study does not address the question of weight
reduction on lipolysis regulation in men or in
women with lower-body obesity. The visceral fat
depot is also of interest, delivering lipolytic products
directly to, and thereby influencing the metabolism
of, the liver. Adipocytes in the visceral fat depot are,
however, not accessible for this type of study. On
the other hand, results of recent studies have indicated that catecholamine-induced lipolysis rates in
subcutaneous and visceral fat cells are interrelated
[17].
Taken together, these results show decreased
basal lipolysis but increased sensitivity to catecholamine stimulation of lipolysis as a result of weight
reduction in obese subjects, suggesting a more
efficient regulation of lipolysis with less free fatty
acids released at rest and lower catecholamine levels
required for lipolysis activation. In view of the
central role proposed for elevated free fatty acids in
the complications of obesity [38-40], these data
may, at least partly, explain the positive effects of
weight reduction on the metabolic derangements in
upper-body obesity. The net effect of the observed
changes in fat cell metabolism on the metabolic rate
can only be speculated upon since this was unfortunately not examined in the present study. As
regards the delivery of lipolytic products to the
circulation over a 24-h period, it is possible that the
decrease in basal lipolysis rate is counterbalanced
by the increase in catecholamine sensitivity. These
questions will require further study.
In conclusion, reduction in body weight in upperbody obese women results in increased lipolytic
noradrenaline sensitivity and a decreased basal lipolysis rate, indicating an increased efficiency in the
regulation of lipolysis. These changes are presumably mediated by an increase in fJradrenoceptor
sensitivity and decreased activity of hormonesensitive lipase. Decreased circulating levels of insulin, catecholamines and androgens may play a role
in these events.
ACKNOWLEDGMENTS
The laboratories involved in this study participate
in the EUROLIP network, supported by the
European Union. The study was supported by
grants from the Swedish Medical Research Council
(19X-01034 and 3362), the Institut National de la
Sante et de la Recherche Medicale, the Swedish
Diabetes Association, the Karolinska Institute and
the foundations of Osterman, Nordic Insulin and
Golje. The skilful technical assistance of Catharina
Sjoberg, Eva Sjolin, Kerstin Wahlen and Michele
Dauzats is appreciated.
REFERENCES
I. jequier E. Energy metabolism in obese patients before and after weight loss.
and in patients who have relapsed. IntJObes 1990; 14 (Suppl. I): 59~7.
2. Lafontan M. Berlan M. Fat cell adrenergit receptors and the control of white
and brown fat tell function. J Lipid Res 1993; 34: 1057-91.
3. Stralfors P. Olsson H. Belfrage P. Hormone sensitive lipase. In: Boyer PD.
Krebs EG. eds. The enzymes. Vol. 18. New York: Academic Press. 1987.
147-77.
4. Reynisdottir S. Wahrenberg H. Carlstrom K. Rossner S. Arner P.
Catecholamine resistance in fat tells of women with upper-body obesity due
to increased expression of beta-adrenoceptors. Diabetologia 1994; 37: 428-35.
5. Reynisdottir S. Ellerfeldt K. Wahrenberg H. Lithell H, Arner P. Multiple
lipolysis defects in the insulin resistance (metabolic) syndrome. J Clin Invest
1994; 93: 2590-9.
6. Connacher AA, Bennet WM. Jung RT. et al. Effect of adrenaline infusion on
fatty acid and glucose turnover in lean and obese human subjects in
post-absorptive and fed states. Clin Sci 1991; 81: 635-44.
Lipolysis and weight reduction
7. Wolfe RR. Peters EJ. Klein S. Holland OB. Rosenblatt J. Gary H. Effect of
short term fasting on lipolytic responsiveness in normal and obese human
subjects, Am J Physiol 1987; 2523: EI89-96.
8. Jensen MD. Haymond MW. Rizza RA. Cryer PE. Miles JM. Influence of body
fat distribution on free fatty acid metabolism in obesity. J Clin Invest 1'989; 83:
1163-73.
9. Leibel RL. Berry EM. Hirsch J. Metabolic and hemodynamic responses to
endogenous and exogenous catecholamines in formerly obese subjects.
Am J Physiol 1991; 260: R78).-91.
10. Sorbris R. Petersson ~. Nilsson-Ehle P. Effecn of weight reduction on
plasma lipoproteins and adipose tissue metabolism in obese subjects.
Eur J Clin Invest 1981; II: 491-8.
II. Smith U. Hammersten J. Bjorntorp p. Kral JG. Regional differences and effect
of weight reduction on human fat cell metabolism. Eur J Clin Invest 1979; 9:
327-32.
12. Kaplan NM. The deadly quartet. Upper-body obesity. glucose intolerance.
hypertriglyceridemia and hypertension. Arch Intern Med 1989; 149: 1514--20.
13. Dahlgren E. Johansson S. Lindstedt G. et al. Women with polycystic ovary
syndrome wedge resected in 1956 to 1965: a long term follow-up focusing on
natural history and circulating hormones. Fertil Steril 1992; 57: 50),-13.
14. Gershagen S. Doebri A. Jeppson S. Rannevik G. Decreasing serum levels of
sex hormone binding globulin around the menopause and temporary relation
to changing levels of ovarian steroids. as demonstrated in a longitudinal study.
Fertil Steril 1989; 51: 616-21.
15. Arner P. Arner O. Ostman J. The effect of local anaesthetic agents on lipolysis
by human adipose tissue. Life Sci 1973; 13: 161-9.
16. Rodbell M. Metabolism of isolated fat cells. I.Effects of hormones on glucose
metabolism and lipolysis. J Bioi Chem 1964; 239: 375-80.
17. Lonnqvist F. Wahrenberg H. Hellstrom L. Reynisdottir S.Arner P. Lipolytic
catecholamine resistance due to decreased beta,-adrenoceptor expression in fat
cells. J Clin Invest 1992; 90: 217)-86.
18. lonnqvlst F. Krief S. Strosberg D. Nyberg B. Emorine L. Arner P. Evidence
for a functional p,-adrenoceptor in man. BrJ Pharmacol 1993; 110: 929-36.
19. Hellmer J. Arner p. Lundin A. Automatic luminometric kinetic assay of
glycerolfor lipolysis studies. Anal Biochem 1989; 177: 132-7.
20. Scatchard G. The attractions of proteins for small molecules and ions.
Ann NY Acad Sci 1949; 51: 66ll-672.
21. Tate KM. Briend-Sutren MM. Emorine LJ. Delavier-Klutchko C. Marullo S.
Strosberg AD. Expression ofthree human p-adrenergic receptor subtypes in
transfected chinese hamster ovary cells. Eur J Biochem 1991; 196: 357-61.
22. Fredrikson G. Stralfors p. Nilsson NO. Belfrage P. Hormone-sensitive lipase
from adijiose tissue of rat. Methods Enzymol 1981; 71: 636-46.
23. Frayn KN. Langin D. Holm C. Belfrage P. Hormone sensitive lipase:
quantitation of enzyme activity and mRNA level in small biopsies of human
adipose tissue. Clin Chim Acta 1993; 216: 183-9.
429
24. Bradford MM. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding.
Anal Biochem 1976; n: 246-54.
25. SOdergard R. Backstrom 1. Shanbag V. Carstensen H. Calculation offree and
bound fractions of testosterone and estradiol-17 p to human plasma proteins at
body temperature. J Steroid Biochem 1982; 18: 801-10.
26. Eriksson A. Carlstrom K. Prognostic value of serum hormone concentrations
in prostatic cancer. Prostate 1988; 13: 249-56.
27. Bjorntorp P. Adipose tissue distribution and function. Int JObes 1991; 15:
67-81.
28. Engfeldt P. Hellmer J, Wahrenberg H, Arner P. Effects of insulin on
adrenoceptor binding and the rate of catecholamine induced lipolysis in
isolated human fat cells. J Bioi Chem 1988; 263: I55S3-60.
29. Valet P, Montastruc J-L. Berlan M. Tran M-A. Lafontan M, Montastruc P.
Differential regulation of fat cell beta-2 and beta-l adrenoceptors by
endogenous catecholamines in dog. J Pharmacol Exp Ther 1989; 249: 271-77.
30. Crilly RG. Francis RM, Nordin BEC. Steroid hormones, ageing and bone.
Clin Endocrinol Metab 1993; 10: 11),-39.
31. Poretsky L. On the paradox of insulin-induced hyperandrogenism in
insulin-resistant states. Endocr Rev 1991; 12: 3-13.
32. Plymate SR, Matej LA, Jones RE. Friedl KE. Inhibition of sex hormone-binding
globulin production in the human hepatoma (Hep G2) cell line by insulin and
prolactin. J Clin Endocrinol Metab 1988; 67: 460-4.
33. Arner P. Control of lipolysis and its relevance to the development of obesity
in man. Diabetes Metab Rev 1988; 4: S07-15.
H. Jansson P-A. Larsson A, Smith U. Lonnroth P. Glycerol production in
subcutaneous adipose tissue in lean and obese humans. J Clin Invest 1992; 89:
161(l...17.
35. Martin ML. Jensen MD. Effecn of body fat distribution on regional lipolysis in
obesity. J Clin Invest 1991; 88: 609-13.
36. Egan Greenberg AS, Chang M-K, Wek SA. Moos Jr MC, Londos C.
Mechanism of hormone stimulated lipolysis in adipocytes: translocation of
hormone sensitive lipase to the lipid storage droplet. Proc Nat! Acad Sci USA
1992; 89: 8537-41.
37. Sztalryd C, Kraemer FB. Regulation of hormone sensitive lipase during fasting.
Am J Physiol 1994; 266: E179-85.
38. Groop LC, Saloranta C. Shank M, Bonnadonna RC. Ferranini E, DeFronzo RA.
The role offree fatty acid metabolism in the pathogenesis of insulin resistance
in obesity and noninsuiin-dependeut diabetes mellitus. J Clin Endocrinol Metab
1991; n: 96-107.
39. Reaven GM. The fourth Musketeer - from Alexandre Dumas to Claude
Bernard. Diabetologia 1995; 38: 3-13.
40. Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty acid
cycle: its role in insulin sensitivity and the metabolic disturbances of diabetes
mellitus. Lancet 1963; ii: 78),-9.
n.