See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/20250549

α2‐Antagonist compounds and lipid mobilization: evidence for a lipid

mobilizing effect of oral yohimbine in healthy male volunteers

Article  in  European Journal of Clinical Investigation · December 1988

DOI: 10.1111/j.1365-2362.1988.tb01272.x · Source: PubMed

CITATIONS READS

60 221

6 authors, including:

Jean Galitzky Mohammed Taouis

French Institute of Health and Medical Research Université Paris-Sud 11
168 PUBLICATIONS   7,452 CITATIONS    152 PUBLICATIONS   3,733 CITATIONS   

SEE PROFILE SEE PROFILE

Max Lafontan
French Institute of Health and Medical Research
323 PUBLICATIONS   15,161 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

RESISTIN AND NEUROINFLAMMATION View project

All content following this page was uploaded by Jean Galitzky on 18 March 2019.

The user has requested enhancement of the downloaded file.

European Journal of Clinical Investigation (1 988) 18, 587-594
a2-Antagonist compounds and lipid mobilization:
evidence for a lipid mobilizing effect of oral
yohimbine in healthy male volunteers
J. GALITZKY, M. TAOUIS, M. BERLAN, D. RIVIERE*, M. GARRIGUES* & M. LAFONTAN,
INSERM U-317, Laboratoire de Pharmacologie Clinique et Medicale & Institut de Physiologie, Universitir
Paul Sabatier and *Laboratoire de Physiologie, H6pital Purpan, Toulouse, France
Received 19 April 1988 and in revised form 4 July 1988
Abstract. Investigations were carried out to analyse the
interactions of a2-antagonists (yohimbine, idazoxan,
SK & F-86,466) with human fat cell a2-adrenoceptors.
All the a2-antagonists enhanced the lipolytic potencies
of epinephrine with an order of potency: yohimbi-
ne > idazoxan > SK & F-86,466; the same order was
also found in 3H-yohimbine competition studies on
human fat cell membranes. The most potent agent,
yohimbine, was administered orally in humans to
define the conditions of appearance and the time-
course of a putative lipid-mobilizing action. Oral
yohimbine administration (0.2 mg kg- ') elevated
plasma glycerol and non-esterified fatty acids in fasting
healthy subjects without significant action on heart
rate or blood pressure during the time-course of the
experiment. The lipid-mobilizing action of yohimbine
was reinforced during physical exercise, completely
suppressed after a meal and partially blocked by
administration of propranolol (0.5 mg kg-'; 60 min
before yohimbine). Plasma norepinephrine concentra-
tions were increased (40-50%) after oral yohimbine
administration. The rise in plasma catecholamine
concentration elicited by yohimbine was not modified
by propranolol treatment. The lipid-mobilizing effect
of yohimbine could be attributable to: (i) the increase
in synaptic norepinephrine with a resultant increment
in lipolysis by fi-adrenergic agonism; (ii) a decrease in
a2-adrenoceptor stimulation of human fat cell a2-
adrenoceptors; (iii) a blockade of presynaptic a2-
adrenoceptors. The use of highly selective a2-antago-
nists will allow investigations into a2-adrenoceptors,
which may represent a novel locus for pharmacologi-
cal intervention in lipid-mobilization strategies. It is
questioned whether az-antagonists would be useful in
treatment or catecholamine-refractoriness and obes-
ity.
Abbreviations: KRBA: Krebs Ringer bicarbonate albumin
buffer; FFA: non-esterified fatty acids.
Correspondence: Dr Michel Berlan, INSERM U-3 17, Labora-
toire de Pharmacologie Medicale et Clinique, 37 Allkes J. Guesde,
31073 Toulouse Cedex. France.
Keywords. Human adipocytes, lipolysis, a2-receptors,
al-antagonists, yohimbine, idazoxan, free fatty acids,
lipid mobilization, insulin release, physical exercise, p-
blocking agents, propranolol.
Introduction
Human fat cells possess both a2- and P-adrenergic
receptors mediating opposing effects on adenylate
cyclase activity, CAMPproduction and lipolysis [I -41.
The balance between these adrenoceptors appears to
be essential in the control of the final lipolytic response
initiated by catecholamines in human adipocytes. The
number of a*-adrenoceptors is higher and more sus-
ceptible to variation than that of P-adrenoceptors in
human fat cells. Alpha2-adrenoceptors are predomin-
ant in the adipocytes of subcutaneous fat deposits; this
is related to decreased catecholamine-induced fat
mobilization, which can be responsible for an
enhanced tendency to accumulate fat in these regions
[3,41.
For the moment, the physiological relevance of the
in-uitro studies performed on isolated adipocytes from
various deposits is not clearly understood. Theoreti-
cally, since blockade of a2-sitespromotes an increased
lipolytic activity in fat cells, the post-junctional a2-
adrenoceptors of the fat cell may represent a novel
locus for pharmacological intervention for the en-
hancement of lipid mobilization during fasting or
caloric restriction.
In the recent years there has been an interest in the
promotion of lipid mobilization and of energy expen-
diture for the treatment of obesity; various investiga-
tions have been carried out in rodent species while
clinical approaches have been very limited [5,6]. The
pharmacological strategy was mainly based upon the
utilization of fi-agonist compounds considered to act
rather specifically on the white and brown adipocytes
of the rodents [7-lo]. Alpha2-antagonist administra-
tion could also represent an attractive lipid-mobilizing
alternative, since it is expected that the compounds will
activate sympathetic tone through central and local
587
588 J. GALTTZKY et al.
effects on noradrenergic neurotransmission and pro-
mote a blockade of the post-junctional az-adrenocep-
tors, identified as being important in adipocytes of
subcutaneous fat deposits [3,4]. This aspect requires a
more complete exploration, the first step being pro-
posed in the present paper.
Two recent attempts to include yohimbine (az-
antagonists) in a weight-reducing regimen have been
made. For one group [I 11, yohimbine administration
(18 mg day-': delivered 3 times each day) associated
with a mild hypocaloric diet (1000 kcal day-') for 8
weeks was without any impact on weight loss; more-
over, blood pressure, heart rate and standard lipid
parameters were unchanged. The other group [I21
demonstrated that yohimbine application (1 5 mg
day-I for 7 weeks) associated with a diet of 400 kcal
day-' during the slimming treatment increased weight
loss and had an impact on exercise-induced energy
expenditure without other side-effects. These investi-
gations led to conflicting results, moreover, the endo-
crine-metabolic impact of yohimbine has not been
investigated in details.
The purpose of the present studies was to: (i) select
an a*-antagonist from among older and recently
discovered compounds on the basis of their impact on
the human adipocyte (yohimbine was seen to be the
most efficient); (ii) define the early metabolic and
endocrinological impact of yohimbine administration
on normal-weight subjects with special attention for
their nutritional status (fasting or after a standardized
meal); (iii) clarify the contribution of yohimbine
administration to lipid mobilization under physiologi-
cal activation of orthosympathetic activity promoted
by a calibrated exercising period; (iv) dissociate the
contribution of the noradrenergic activation (and its p-
adrenergic-driven effect on lipolysis) from a presynap-
tic effect on nerve terminals or postsynaptic action on
fat cell az-adrenoceptors.
Patients and methods
Isolation of human adipocytes
Abdominal subcutaneous adipose tissue samples were
obtained from premenauposal women (25-40 years of
age) undergoing elective surgical lipectomy. Although
having a mean body mass index (ratio:weight(kg)/
height(m2)) of 25.8 f 1.5 (mean k SEM, n = 6), the
patients were otherwise healthy and none had any
identified metabolic or endocrinological disorders.
Subjects were fasted overnight before tissue removal;
general anaesthesia was induced by an association of
penthotal, droleptan and fentanyl. The patients did
not receive drugs active on the autonomic nervous
system or modifying catecholamine levels. After surgi-
cal excision, the adipose tissue was quickly transported
to the laboratory in cold sterile physiological saline
(NaCI 0.9%; Hepes 5 mmol I-'; 6 mmol 1-' glucose,
pH 7.4) and used within 15-20 min of removal.
Isolated adipocytes were obtained, according to the
method of Rodbell [ 131, by collagenase digestion of
adipose fragments in Krebs-Ringer bicarbonate
buffer containing albumin (3.5 g 100; ml -') (KRBA)
and glucose (6 mmol 1-I) at pH 7.4 and 37°C under a
95% 0 2 / 5 % COZgas phase. At the end of incubation,
the fat cells were filtered through a silk screen and
washed three times with KRBA buffer to eliminate
collagenase.
Lipolysis measurements
Isolated fat cells obtained after collagenase treatment
were incubated in 1 ml KRBA (pH 7.4) containing
glucose (6 mmol I-'), at 37°C under gas phase (95%
0 2 / 5 % COZ)with gentle shaking (60 cycles min-I) in a
water-bath, as previously described [2,3]. Pharmacolo-
gical agents at suitable dilutions were added to the cell
suspension just before the beginning of the assay in 10-
pl portions to obtain the desired final concentration.
After 90 min of incubation, the polyethylene tubes
were placed in an ice bath and 200-p1 aliquots of
infranatant were removed for enzymatic determina-
tion of glycerol released in the incubation medium,
which was taken as the index of fat-cell lipolyses. Total
lipid was evaluated gravimetrically after extraction.
Ascorbic acid (0.1 mmol I-') was included in the
incubation medium in order to prevent catecholamine
degradation.
Clinical pro tocoIs
Informed consent was obtained from 14 normal
healthy male volunteers and the experimental protocol
was approved by the Ethical Committee of the Hospi-
tal. All the volunteers (age 30 k 6, weight 72 k 3 kg;
body mass index 23 f0.5 kg m2; means fSEM) had a
normal medical status, with normal glucose tolerance
and were taking no medications. They had no family
history of diabetes or any other endocrinological
disorder. All the subjects were consuming their usual
normal diet (2200-2400 kcal day-!) before the experi-
ments and had a stable weight for at least 1 month
before the beginning of the study.
The experimental procedures used throughout the
experiments are summarized in Fig. 1. All the experi-
ments began at 08-30 h after an overnight fast. An
indwelling teflon catheter was placed in the antecubital
vein and a blood sample was always taken at zero time.
The first experimental protocol was fixed to define
the impact of yohimbine intake on plasma concentra-
tions of non-esterified fatty acids (FFA) and glycerol in
subjects after an overnight fast or 1 h after ingestion of
a standard balanced French breakfast (400 kcal, 60%
of kcal originating from carbohydrates). The patients
were their own controls, each experiment was separ-
ated by at least a week. The yohimbine dose (0.2 mg
kg-') used throughout the present studies was
selected, after preliminary assays with lower (0.05 mg
kg-I) or higher (0.5 mg kg-I) doses, for its substantial
metabolic effects and the absence of noticeable side-
 YOHIMBINE-INDUCED LIPID MOBILIZATION 589

bine administration. In the second group of experi-

PROTOCOL I
ments, the patients were submitted to the standardized
Yo h imbine period of exercise (without yohimbine) with collection
0.2 mg kg-' of blood as in protocol 1 (Fig. 1). The last group of
I experiments (protocol 2) corresponds to the associ-
ation of both previously described protocols, i.e.
1 t ' ' ' Time ( m i d yohimbine intake and exercise (Fig. 1).
-60 0 45 60 75 90 105 In the final protocol, 40 mg of propranolol (B-
I Blood sampling antagonist) were given per 0s 60 rnin before yohimbine
ingestion. Blood samples were collected as previously
PROTOCOL 2
described in the other experimental situations (proto-
Yohimbine col 3, Fig. I).
0.2 m g kg-'
Blood glucose and insulin were determined with a
glucose oxidase technique and a radioimmunoassay
1 1 (commercial radioimmunoassay kit, Institut Pasteur,
t
0 45 60 75 90 105
' Time(min) Paris), respectively. Plasma non-esterified fatty acids
were determined by a titrimetric method after extrac-
Blood sampling tion using the method of Dole & Meinertz [ 141. Plasma
PROTOCOL 3
glycerol was evaluated by an enzymatic procedure [ 151.
For norepinephrine determination, blood was col-
Yohlmbine lected on lithium heparine with 10 mM sodium metabi-
0.2 m g kg-' sulphite and centrifuged for 10 rnin at 10 000 g at 0°C;
Pro pronolol
40 mg

t
1 I
t ' ' Time(mrn1
-60 0 45 60 7 5 90 105
Blood sampling
Figure 1. Diagram of the experimental protocols. All the protocols
began at 08.30 h. The exercise was performed on an ergometric
bicycle at 60% of maximal aerobic capacity (VOz max) defined
according to the following formula taking in account the resting
heart rate (HR). "A) VOz max is calculated as function of the heart
rate reserve; 60% VOz max is defined according to the calculation:
[HR]+[(220-age of subject)-HR] x60/100 [33].
effects on heart rate and blood pressure. Yohimbine
pills were given orally with 50 ml water, immediately
after blood collection. A placebo was given instead of
yohimbine in control experiments. Another blood
sample was taken 45 rnin after yohimbine ingestion
and every 15 rnin thereafter, as shown in Fig. 1. In fed
patients, time 0 was taken 60 min after ingestion of the
meal, then yohimbine was orally administered as
described for fasting patients (Fig 1). Heart rate and
blood pressure were determined just before every
blood collection.
The second experimental protocol was established
to define the impact of yohimbine ingestion at rest and
during a calibrated period of physical exercise (30 rnin
at 60% of maximal aerobic capacity on an ergometric
bicycle). Plasma concentrations of FFA, glycerol,
glucose, norepinephrine and insulin were assayed.
Eight fasting subjects were tested in three different
situations, each experiment was separated by at least a
week; the order in which each patient participated in
the different experimental groups was randomly
defined. In the first group of experiments, correspond-
ing to protocol 1, fasting subjects rested after yohim-
the plasma was stored at - 80°C. Norepinephrine was
selectively isolated from the plasma sample by adsorp-
tion on activated alumina, then eluted with 0.1 ml I-'
perchloric acid. Dihydrobenzylamine was used as
internal standard to monitor recovery from the extrac-
tion step. Norepinephrine was assayed by high pres-
sure liquid chromatography using electrochemical
(amperometric) detection (Waters HPLC system) as
previously described [16]. All samples for each indi-
vidual were analysed at the same time.
Chemicals
Bovine serum albumin, fraction V, yohimbine hydro-
chloride, epinephrine bitartrate and ascorbic acid were
purchased from Sigma Chemical Company (St Louis,
MO, U.S.A.). The following drugs, were kindly
donated: yohimbine chlorhydrate@ pills by Labora-
toires Houde (Paris, France), propranolol tablets
(Avlocardyl@ by ICI Pharma (Enghien les Bains,
France), idazoxan HCI by Drs J. Doxey and M.
Stillings, Reckitt and Colman (Kingston upon Hull,
U.K.), SK & F-86,466 by Dr J. P. Hieble from Smith
Kline and French Company (Swedeland, PA, U.S.A.).
Enzymes for glycerol assay came from Boehringer
Mannheim (Mannheim, FRG). All other chemicals
and organic solvents were of reagent grade.
Results
Comparative study of az-antagonist potencies on human
fat cells
The first purpose of our study was to define the
potencies of yohimbine and of two recently developed
a2-antagonists on human fat cell cr2-adrenoceptors.
Several chemical families of compounds having selec-
tive activity at a2-adrenoceptors have been described.
590 J. GALITZKY et al.

on fat cell membranes after labelling of the a*-siteswith

3H-yohimbine (not shown).
Yohimbine was found to be the most potent com-
pound at human fat cell az-adrenoceptors. Yohimbine
was selected for the in-uivo investigations for this
action and since it has been also demonstrated to
increase the release of norepinephrine from both
peripheral and central noradrenergic neurons in var-
ious conditions [17].

Metabolic efects of oral yohimbine in fasted and fed

suhjec ts
During fasting, definition of the increment of free fatty
acids and glycerol concentrations in the circulation can
give a reasonable evaluation of lipid mobilization. In
the present experiments our attention was mainly
focused on the early metabolic events following
yohimbine ingestion with the purpose of defining the
time-course of their appearance.
As expected, after an overnight period of fasting,
plasma FFA concentrations were significantly
(P<O.OI) higher than those found in fed subjects
(423f43 vs. 240k20 pmol I-', n=6). In the fasting
state, oral yohimbine (0.2 mg kg-I) was followed, 45
r 9 8 7 6 5 I min later, by a significant ( P c 0 . 0 5 ) increment of
r- plasma FFA concentrations; the effect was sustained
I
0 over the period of time explored (Fig. 3 ) . Plasma
- 0 glycerol concentrations followed a parallel increment
Ei Log (antagonist) (rnol L-') z (not shown). In the experiments with placebo there was
no noticeable action of placebo ingestion; plasma FFA
Figure 2. Enhancement of epinephrine-induced lipolysis by various
a*-adrenergic antagonist compounds in isolated human fat cells. Fat concentrations did not significantly differ from those
cells were isolated as described in the Patients and methods section, measured at 0 time. In the fed state, when yohimbine
and were incubated in KRBA buffer with M epinephrine and was given orally 60 min after a standardized meal,
increasing concentrations of yohimbine ( 0 ) . idazoxan (H)and there was no effect of yohimbine on plasma FFA or
SK&F-86,466 (A).Values are meanskSEM (bars) of five different
experiments.
glycerol levels. In this situation, plasma insulin levels

They include rauwolfia alkaloids such as yohimbine Yohimbine

[I 71, imidazolines such as idazoxan (RX-78 I , 094) [ 181 (0-2 mg k g - I ) ** **
**
and 3-benzazepines such as SK & F-86,466 [19].
The definition of their relative potencies was based
on the determination of their ability to enhance the
lipolytic effect of epinephrine on isolated human
subcutaneous fat cells. All three compounds used were
able to promote an increment of epinephrine-induced
lipolysis. At the higher concentrations (10 pmol I-')
they were equipotent and the iipolytic efficiency of 1
pmol I-' epinephrine (associated with the a2-antago-
nists) approached that obtained with 0.1 pmol I-'
isoproterenol (Fig. 2). The dose-response curves
revealed the following order of potency: yohim- 45 60 75 90 105
bine > idazoxan > SK & F-86,466; their computerized Time (min)
analysis using DOSEFIT program [20] gave the following Figure 3. Effect of orally administered yohimbine on plasma non-
ECso values of 1 5 1 +_ 55,460 +_ 90 and 1260+_ 400 nmol esterified FFA concentrations in six resting subjects after a n
1 F ' (mean fSEM; n = 5), respectively; the ECso value ) or 1 h after a calibrated meal (Q), as described in
for yohimbine was significantly (P-=0.02) lower than the Patients and methods section and protocol (Fig. I). Data are
means k SEM of six experiments. * P < 0.05, ** P < 0.02; signifi-
those found for the two other compounds. The same cantly different from control values (time O),according to Student's
order of potency was also found in competition studies paired t-test.
 YOHIMBINE-INDUCED LIPID MOBILIZATION 591

were significantly ( P < 0.01) increased (72 f 8 pU ml-’
vs. 1 8 k 4 pU ml-’ after the overnight fast). The
nutritional status (and the resulting plasma insulin
concentrations) is of major importance for the incre-
ment of plasma FFA and glycerol levels after oral
yohimbine. Thus, all the following experiments were
carried out after an overnight fast in order to explore
the importance and mechanisms of the effects of
yohimbine.
Metabolic and endocrinological responses to oral
yohimbine-eflect of exercise
Another set of experiments was carried out on eight
subjects in order to test the impact of yohimbine
administration alone, exercise alone and yohimbine
administration associated with exercise on plasma
concentrations of FFA, glycerol, glucose, norepineph-
rine and insulin.
As shown in Fig. 4a, yohimbine administration to
subjects at rest promoted an increment of plasma
concentrations of FFA and glycerol. Plasma glucose
levels as well as plasma insulin concentrations were
unaffected in these conditions. The heart rate and
sitting systolic blood pressure were unchanged (Table
1). Plasma norepinephrine concentrations were mea-
sured 75 min after beginning the experiment; they were
higher than in control patients (Table 2). The yohim-
bine dose administered in the present investigation did
not result in noticeable autonomic symptoms.
When the same patients were subjected to a period
of physical exercise over 30 min, as described in
protocol 2, there was a significant ( P < 0.02) increase in

2ool
I50 (a)
plasma glycerol concentrations, while plasma FFA
levels were unchanged. Plasma concentrations of
glucose remained stable during the experiment while
** 8
1
*
** plasma insulin levels were significantly (P< 0.01) de-
creased at the end of the exercise period (Fig. 4b).
Norepinephrine levels were increased as expected; they
reached the levels obtained with yohimbine alone
J 45 60 75 90 105 (Table 2).
-50
2001 (b)

Table 1. Influence of yohimbine administration (0.2 mg kg-

per 0s) on systolic blood (SBP), diastolic blood pressure
(DBP) and heart rate (HR) in seven volunteer patients after
-8 100.
an overnight fast and at rest

!
50-

o-+ 45
*
h
I05
0
Time after yohimbine (min)

45 75 I I0
8
SBP (mmHg) 122f3.9 125k5.5 121k5.5 125+6

r- \,
FI
EXERCISE

*
*
DBP (mmHg) 73.6f2.9 73.1 k 3 . 6 71.4k3.3 78.8k5.2
H R (permin) 71.7+3.7 68.2k3.5 70.5k5.6 63.5k2.7

Results are means fSEM.

I . I

Table 2. Variation in plasma norepinephrine concentrations in eight

subjects in the different experimental conditions defined in Fig I .

Norepinephrine
Experimental conditions ’
(pg ml- plasma k SEM)

At rest (time 0 ) 302k15

* Exercise 421 +20*
Time(min1 Yohimbine alone (0.2 mg kg-’) 471 +87*
Exercisefyohimbine (0.2 mg kg-I) 559*67**
Figure 4. Variation of plasma concentrations of FFA (El), glycerol At rest-propranolol alone (0.5 mg kg-l) 283 k 30
(m), glucose (0)and immunoreactive insulin (IRI: (8)in eight fasting Yohimbine+propranolol (0.5 mg kg-’) 675+83**
male volunteers in three different experimental conditions: (a) at
rest, after oral administration of yohimbine (0.2 mg ml-I) a t time 0
~~~ ~

(protocol 1); (b) before, during and after a period of exercise
(protocol 2, without yohimbine); (c) when yohimbine administration
and exercise are associated. Values are meanfSEM of eight
different experiments. Percentage variation was calculated from the
parameters defined at time 0. * P < 0.05, ** P < 0.02; significantly
different from corresponding control values taken at 0 time accord-
ing to Student’s paired r-test.
Plasma norepinephrine concentrations were assayed by HPLC
(see Patients and methods section) after an overnight fast at rest in
sitting position (time 0) and 75 min after the beginning of the
experiments. Propranolol was administered 60 min before blood
sampling (Table 1).
* P<0.05, ** P i 0.02; significantly different from control values at
0 time according to Student’s paired t-test.
592 J. GALITZKY et 01.
The last set of experiments was based on the
association of oral yohimbine administration with the
30-min period of physical exercise. Under these condi-
tions there was a significant increase of exercise-
induced plasma glycerol. Plasma FFA concentrations
were increased ( P < 0.0 1) mainly after cessation of
exercise; the time-course FFA concentrations revealed
the additive effect of yohimbine on exercise-induced
metabolic events. The decline in plasma insulin con-
centrations was not modified by oral yohimbine
administration while the highest plasma norepineph-
rine concentration was observed on association of
both treatments (Table 2). Plasma glucose concentra-
tions were unchanged under our working conditions.
EfSect of oral yohimbine under p-blockade
A dose of propranolol (0.5 mg kg-'), known to
promote a clear P-adrenoceptor-blockade in humans,
was administered 60 min before yohimbine ingestion
in order to suppress the P-adrenergic effects promoted
by the increment of plasma norepinephrine levels
observed after yohimbine. Beta-blockade was assessed
by evaluation of the heart rate (73 k 4 vs. 60 f4; 30 min
after propranolol administration); it remained stable
over the experimental period. Moreover, propranalol
alone had no effect on plasma norepinephrine levels
(Table 2).
One hour after the administration of propranolol,
plasma FFA concentrations were 301 f21 vs. 379 f39
pmol I-' before its administration ( P < 0.05). The
changes of plasma FFA concentrations promoted by
yohimbine administration were considerably reduced
** **
I ** I
* taken as indices of lipid mobilization. Since total
45 60 75 90 105
Time ( m i n )
Figure 5. Effect of oral yohimbine administration on plasma FFA
concentrations under P-blockade. Variation (expressed on a percent-
age basis) of plasma FFA concentrations after orally administered
yohimbine (0.2 mg kgg') without (B) or after p-adrenoceptor
blockade (a) by 0.5 mg ml-' propranolol given 60 min before
yohimbine. Values are means k SEM of eight different experiments.
* P i 0.05, ** P i 0.02; significantlydifferent from control values (at
rest) according to Student's paired r-test.
after propranolol treatment, there was still a signifi-
cant (P<O.O5) residual increment of plasma FFA
levels (20%))but 80% of the initial increment was lost
(Fig. 5) after yohimbine administration. The yohim-
bine-dependent increment of plasma norepinephrine
concentrations was still observed after propranolol
pretreatment (Table 2). However, there was no modifi-
cation of heart rate (61 f 5; n = 6), a result that assesses
the efficiency of p-adrenoceptor blockade promoted by
propranolol.
Discussion
The purpose of the present studies was to determine
whether @*-antagonist administration in normal
humans results in enhanced lipid mobilization, and to
determine the conditions of expression of this effect as
well as the mechanisms involved in its origin.
There are few studies investigating the application of
yohimbine as an az-adrenergic antagonist in man,
although the drug has been used for a long time, before
its az-adrenergic potencies were known, for the phar-
macological treatment of impotence [ 17,211. Adminis-
tration of higher doses of yohimbine i.v. (0.5 mg kg-')
were seen to raise blood pressure and heart rate in man
and various species, while lower oral doses (5 mg
patient- I ) were used in patients with autonomic failure
[17,22]. The paucity of pharmacokinetic data at the
time we set up our experimental schedule directed us
towards the oral route for the administration of the
drug with the idea of looking for an oral dose of
yohimbine having a minor impact on heart rate and
blood pressure over the experimental period. In the
patients selected for this study we found that the oral
dose of 0.2 mg kg-' had a very limited impact on both
parameters (Table 1; no significant changes in mean
values of heart rate and blood pressure). This result is
consistent with previous results obtained in human
subjects with equivalent doses of yohimbine [23].
Increments in systolic blood pressure greater than
those observed in the present study have been reported
after intravenous administration of higher doses [24].
The plasma FFA and glycerol concentrations were
energy intake and composition of daily food play an
important role in the regulation of metabolite avail-
ability, the first set of experiments was devoted to the
evaluation of the impact of oral yohimbine administra-
tion on the two lipolytic indices in patients fasted
overnight and also 1 h after ingestion of a calibrated
meal after the overnight fast.
Yohimbine administration promoted lipid mobili-
zation in fasting subjects, as shown in Fig 3, while the
effect was completely blunted 1 h after the meal. Thus,
under a situation of substrate utilization, the lipid
mobilization is favoured by yohimbine intake. On the
contrary, in the postprandial situation, the lipid-
mobilizing effect is completely blocked; insulin is
probably responsible for the lack of postprandial
effects of yohimbine. If we refer to the recent results
 YOHIMBINE-INDUCED LIPID MOBILIZATION
obtained with 1x2-antagonists,a*-adrenergic blockade
increases glucose-potentiated insulin secretion in man
[25-281, a situation that strongly reinforces inhibition
of lipid mobilization. However, differences in the
yohimbine absorption and distribution during fasting
and fed states cannot be rejected; this aspect requires
further studies that are considered to be out off the
scope of the present paper. Our data focus attention on
the major importance of the period of administration
of yohimbine for the facilitation of its lipid-mobilizing
efficiency. The failure of its effect previously reported
in a slimming diet [ l l ] can be the result of an
inappropriate schedule of administration, while the
positive results obtained by another group can be
explained by the fact that it was associated with a very
low caloric intake (400 kcal day-’) probably leading to
a rather poor insulin-secreting efficiency [12]. This
aspect needs further experimental support and opens
some clinical perspectives.
The second set of experiments, summarized in Fig.
4, clearly demonstrates that yohimbine had no impact
on insulin release in the fasting state (Fig. 4a) while
keeping its lipid-mobilizing action. Moreover, yohim-
bine was seen to enhance the lipid-mobilizing effect of a
standard period of physical activity (Fig. 4c). Since it
has been demonstrated that the fractional extraction of
glycerol does not increase with elevated arterial con-
centrations, and also since it does not serve as a
significant glyconeogenic substrate, the use of plasma
glycerol response during exercise, as an index of
lipolytic activity, should be preferred to the FFA
response [29]. During this period plasma concentra-
tions of glycerol increased, while the absence of an
increase in plasma FFA levels (Fig. 4b) can be
interpreted as being linked to a concomitant utilization
of mobilized FFA by the exercising muscles. The
results depicted in Fig. 4c support this idea since
plasma FFA levels increased significantly after the end
of physical activity. Moreover, the data also demon-
strate that the effects of yohimbine are additive or
synergistic with those of exercise (Fig.4~).It is notice-
able that the well-known hypoinsulinaemia, observed
during physical exercise, was not modified by yohim-
bine administration.
Concerning the interpretation of the lipid mobiliza-
tion induced by yohimbine, the increase of plasma
norepinephrine concentrations strongly suggests that
lipid mobilization is consecutive to an increase in
synaptic cleft norepinephrine concentrations after
yohimbine [30], as also seen during physical exercise
[31] (Table 2). The associated of both experimental
situations leads to a higher activation of the sympathe-
tic nervous system and to a concomitant increment of
lipid mobilization (Fig. 4c).
A better understanding of the lipid-mobilizing
mechanism requires a dissociation of orthosympathe-
tic activation from the effects of a putative blockade of
post-junctional fat-cell a*-adrenoceptors. An experi-
ment was made to dissociate the involvement of the b-
adrenergic receptor-mediated lipolytic effects linked to
593
activation of the sympathetic nervous system from the
fat-cell 1x2-adrenoceptor antagonizing effect of the
drug. For this purpose, the last set of experiments was
carried out under p-blockade with propranolol. Under
j3-blockade, assessed by the stable lowered heart rate
before or after yohimbine administration, there was
still a residual lipid-mobilizing effect of yohimbine
administration (20% of the effect observed with
yohimbine alone) (Fig. 5) with equivalent increments
of norepinephrine levels (Table 2). Although the major
part of the action of yohimbine could be attributed to a
sympathetic activation, the experiments under b-
blockade suggest that all the effects of yohimbine could
not completely be explained by the effects on the level
of noradrenaline. The residual lipid mobilization
observed under b-blockade may be attributable to the
blockade of fat-cell a*-adrenoceptors.
In spite of the potential value of yohimbine as a
direct and indirect lipid-mobilizing compound, further
investigations are needed to assess the pharmacokine-
tic and pharmacodynamic behaviour of yohimbine.
During completion of the present investigations, a
paper has reported the kinetic disposition of yohim-
bine in young male subjects following a single oral dose
of 10 mg of yohimbine in conditions similar to ours
[32]. It is noticeable that the compound is absorbed
very rapidly and eliminated from the plasma with an
elimination half-life of 0.60 0.26 h [32]. The rapid fall
of plasma levels of yohimbine is not due to urinary
excretion of the drug or to its sequestration by red
blood cells; a pharmacodynamic study based on the
evaluation of metabolic impact would probably be
valuable in further approaches to extend pharmacoki-
netic data.
The lipid-mobilizing strategies based on cc2-antago-
nist utilization would be compared with the thera-
peutic lipolytic strategies based on the use of b-agonist
compounds [7-lo], which exhibit some limitations and
still require extended explorations in humans. If we
refer to a recent paper that appeared during comple-
tion of the present work [6], it is clear that the acute
administration of a new p-adrenoceptor agonist
(Ro16-8714), which is rather poorly selective, in
healthy male volunteers promotes lipid mobilization
and an increase in the rate of energy expenditure.
However, among its major limitations, Ro16-8714
administration produced substantial tachycardia.
In summary, we have shown that oral yohimbine
administration promoted lipid mobilization in fasting
subjects without noticeable action on heart rate and
blood pressure. The lipid-mobilizing action is rein-
forced during physical exercise and completely sup-
pressed after a meal. Concerning the mechanisms of
action, the lipid mobilizing effect of yohimbine is
attributable to: (i) an increase in synaptic norepineph-
rine with a resultant increment of lipolysis by p-
adrenergic agonism within the fat deposits; (ii) a
decrease in chronic excessive 1x2-adrenoceptorstimula-
tion of human fat-cell a*-adrenoceptors; (iii) a block-
ade of presynaptic Ixz-adrenoceptors.
 594 J. GALITZKY et of.
Acknowledgments
A part of the present work has been published under an
abstract form at the Annual Meeting of the British
Pharmacological Society held in London, U.K., 6-8
January 1988 and at the 1st European Congress on
Obesity held in Stockholm, Sweden, 5-6 June 1988.
The investigations reported here were supported in
part by an INSERM grant (C.R.E. 867 007).
References
~~
I Fain JN, Garcia-Sainz JA. Adrenergic regulation of adipocyte
metabolism. J Lipid Res 1983;24:945-66.
2 Berlan M, Lafontan M. Evidence that epinephrine acts preferen-
tially as an antilipolytic agent in abdominal human subcutaneous
fat cells: assessment by analysis of beta and alphal-adrenoceptor
properties. Eur J Clin Invest 1985;15:341 -8.
3 Mauriege P, Galitzky J, Berlan M, Lafontan M. Heterogenous
distribution of beta- and alphaz-adrenoceptor binding sites in
human fat cells from various fat deposits: functional conse-
quences. Eur J Clin Invest 1987;17:156-65.
4 Richelsen B. Increased alpha2- but similar beta-adrenergic
receptor activities in subcutaneous gluteal adipocytes from
females compared with males. Eur J Clin Invest 1986;16:302~9.
5 Munro JF, Chapman BJ, Robb GH, Zed C. Clinical studies with
thermogenic drugs. In: Berry EM, Blondheim SH, Eliahou HE,
Shafrir E, eds. Recent Advances in Obesity Research: V.
London: John Libbey, 1987:155-9.
6 Henry C, Schutz Y, Buckert A, Meylan M, Jecquier E, Felber JP.
Thermogenic effect of the new p-adrenoceptor agonist Ro 16-
8714 in healthy male volunteers. Int J Obesity 1987;11:473-83.
7 Levin BE. Thermogenic agents as a treatment of obesity. In
Hirsch J, Van Itallie TB, eds. Recent Advances in Obesity
Research: IV. London: John Libbey, 1984:217-33.
8 Arch JRS, Ainsworth AT, Cawthorne MA. er ul. Atypical b-
adrenoceptor on brown adipocytes as targets for anti-obesity
drugs. Nature 1984;309:163-5.
9 Arch JRS, Ainsworth AT. Thermogenic and antiobesity activity
of a novel p-adrenoceptor agonist (BRL 26830 A) in mice and
rats. Am J Clin Nutr 1983;38:549-58.
10 Arch JRS, Piercy V, Thurlby PL, Wilson C, Wilson S. Thermo-
genic and lipolytic drugs for the treatment of obesity: old ideas
and new possibilities. In: Berry EM, Blondheim SH, Eliahou HE,
Shafrir E, eds. Recent Advances in Obesity Research: V.
London: John Libbey, 1987;300-1 I .
11 Berlin I, Stalla-Bourdillon A, Thuillier F, Turpin G, Puech AJ.
Absence d’efficacite de la yohimbine dans le traitement de
I’obtsitt. J Pharmacol (Paris) 1986;17:343-7.
12 Zahorska-Markiewicz B, Kucio C, Piskorska D. Adrenergic
control of lipolysis and metabolic responses in obesity. Horm
Metab Res 1986;18:693-7.
13 Rodbell M. Metabolism of isolated fat cells. I . Effects of
hormones on glucose metabolism and lipolysis. J Biol Chem
1964;239:375-380.
14 Dole VP, Meinertz H. Microdetermination of long chain fatty
acids in plasma and tissues. J Biol Chem 1960;235:2595-9.
View publication stats
15 Wieland 0. Eine enzymatische methode zur Bestimmung von
glycerin. Biochem Z 1957;239:313-9.
16 Damase-Michel C, Valet P, Montastruc JL. Nicardipine causes
sympathetic activation that does not involve baroreceptor reflex
tachycardia in conscious sino-aortic denervated dogs. Eur J
Pharmacol 1987;142:145-9.
17 Goldberg MR, Robertson D. Yohimbine: a pharmacological
probe for study of the alphaz-adrenoceptor. Pharmdcol Rev
I983;35: 143-80.
18 Doxey JC, Roach AG, Smith CFC. Studies on RX 781094: a
selective, potent and specificantagonist of alphd2-adrenoceptors.
Br J Pharmacol 1983;78:489-505.
19 Roesler JM, McCafferty JP, DeMarinis RM, Matthews WD,
Hieble JP. Characterization of the antihypertensive activity of
SK&F-86,466, a selective alpha2-antagonist, in the rat. J Phar-
macol Exp Ther 1986;236:1-7.
20 Barlow RB. Biodata Handling with Microcomputers. Amster-
dam: Elsevier Science Publishers, 1983.
21 Morales A, Surridge DHC, Marshall PG, Fenemore J. Nonhor-
monal pharmacological treatment of organic impotence. J Urol
1982;128:45-7.
22 Onrot MD, Goldberg MR, Biaggioni I, Willey RG, Hollister AS,
Robertson D. Oral yohimbine in human autonomic failure.
Neurology 1987;37:215-20.
23 Charney DS, Heninger GR, Sternberg DE. Assessment of
alpha2-adrenergic autoreceptor function in humans: effects of
oral yohimbine. Life Sci 1982;30:2033-41.
24 Goldberg MR, Hollister AS, Robertson D. Influence of yohim-
bine on blood pressure, autonomic reflexes, and plasma catecho-
lamines in humans. Hypertension l983;5:772-8.
25 Kashiwagi A, Harano Y, Susuki M.et al. New alphaz-adrenergic
blocker (DG-5128) improves insulin secretion and in vivo
glucose disposal in NIDDM, patients. Diabetes 1986;35:1085-9.
26 Robertson RP, Halter JB, Porte Jr D. A role for alpha-
adrenergic receptors in abnormal insulin secretion in diabetes
mellitus. J Clin Invest 1976;57:791-5.
27 Broadstone VL, Pfeiffer MA, Bajaj V, Stagner JI, Samols E.
Alpha-adrenergicblockade improves glucose-potentiatedinsulin
secretion in non-insulin-dependent diabetes mellitus. Diabetes
1987;36:932-7.
28 Morgan NG, Montague W. Studies of the mechanism of
inhibition of glucose-stimulated insulin secretion by noradrena-
line in rat islets of Langerhans. Biochem J 1985;226:571-6.
29 Nygaard E. Energy supply and utilization during exercise in
women and men-a comparison. In: Vague J, Bjorntorp P, Guy-
Grand B, Rebuffe-Scrive M, Vague P, eds. Metabolic Complica-
tions of Human Obesities. Amsterdam: Excerpta Medica,
1985:249-58.
30 Holmberg GS, Gershon LH, Beck L. Yohimbine as an autono-
mic test drug. Nature 1962;193:13134.
31 Christensen NJ, Galbo H. Sympathetic nervous activity during
exercise. Ann Rev Physiol 1983;45:139-53.
32 Owen JA, Nakatsu SL, Fenemore J, Condra M, Surridge DHC,
Morales A. The pharmacokinetics of yohimbine in man. Eur J
CIin Pharmacol 1987;32:577-82.
33 Karvonen M, Kentala E, Mustala 0. The effects of training on
heart rate. A longitudinal study. Ann Med Exp Biol Fenn
1957;35:307-15.