Simulated Microgravity Increases β-Adrenergic Lipolysis in Human Adipose Tissue¹

The effect of a sustained decrease in sympathetic nervous activity, achieved through 5-day head-down bed rest (HDBR), on the β-adrenergic lipolytic activity of sc adipose tissue was studied in eight healthy men. The in situ β-adrenoceptor (AR) sensitivity was studied using the microdialysis method. Local perfusion of increasing concentrations of isoprenaline showed an increased β-AR sensitivity to lipolysis (assessed by extracellular glycerol concentration) and to vascular tone (assessed by the ethanol clearance). The adrenergic sensitivity of isolated adipocytes was studied in vitro. Basal lipolysis and the response to nonselective (isoprenaline) or selective (dobutamine, terbutaline, and CGP 12177) β-AR agonists were increased after HDBR as was the lipolytic effect of dibutyryl cAMP. When data were expressed as a percentage of the dibutyryl cAMP effect to rule out the postreceptor events, basal and lipolytic responses toβ -AR agonists where similar before and during HDBR. Theα ₂-AR-mediated antilipolytic effects of adrenaline were not modified. Lymphocyte β-AR number was unchanged during HDBR. Our results demonstrate that a sustained sympathoinhibition induces an increase in the lipolytic β-adrenergic response in adipose tissue and suggest that this hypersensitization is linked to an increase in the postreceptor steps of the lipolytic cascade in the adipocyte rather than to changes in β-adrenoceptors.

CATECHOLAMINES control membrane adenylyl cyclase activity in a large number of cells through stimulatoryβ -adrenoceptors (AR). Changes in sympathetic nervous system (SNS) activity or in plasma catecholamine levels have commonly been associated with altered AR functions in target cells. Sustained adrenergic stimuli or chronic administration of β-AR agonists led to a reduction of β-AR-dependent responses and down-regulation ofβ ₁- and β₂-AR (1–4). This phenomenon is called desensitization, tachyphylaxia, or refractoriness (5). On the other hand, chronic reduction of catecholamine levels led to supersensitization of inotropic and chronotropic β-AR-mediated effects in rat heart (6, 7). Vascular adrenergic supersensitivity, associated with low plasma noradrenaline levels, was also found in patients with orthostatic hypotension (8). These observations have only been made on animal models or in dysautonomic patients.

In healthy humans, the consequence of low SNS activity on AR-mediated functions is poorly documented. A low SNS activity state can be achieved during maintained −6° head-down bed rest (HDBR) (9–13). HDBR suppress the SNS stimulation linked to the baroreflex stimulation induced by orthostatism. In these experimental conditions, the hemodynamic consequences of the resting state on the SNS during HDBR have been established (14, 15). Adipose tissue is an interesting model to investigate the low SNS activity state because human fat cells possess various β-AR (β₁-, β₂-, andβ ₃-AR), the stimulation of which enhances adenylyl cyclase activity, controlling lipolysis (16). The β-adrenergic function of adipose tissue can be explored in vitro on isolated adipocytes obtained by needle biopsy and in situ using the microdialysis technique (17, 18). To our knowledge, the effect of HDBR on the adaptation of β-AR-mediated lipolysis in adipose tissue, has never been reported.

The present study was conducted to investigate the consequences of a sustained low SNS activity on the β-adrenergic lipolytic responses of adipose tissue. For this purpose, lipolytic responsiveness was studied 1) in vitro on isolated fat cells using various adrenergic agents, and 2) in vivo using the changes in the interstitial glycerol concentration (as the lipolysis index) induced by in situ infusion of isoprenaline, as assessed by microdialysis. All investigations were conducted before and during the last 5 days of HDBR.

Subjects and Methods

Subjects

Eight healthy young male subjects, 21–34 yr old (mean ± sem, 27.2 ± 1.5 yr), who had not been subjected to any nutritional or pharmacological protocol before the study, gave their informed consent to participate in this clinical protocol. All were drug free. Their mean body weight and height were 70.6 ± 2.4 kg (range, 59–79 kg) and 175.5 ± 2.6 cm (range, 168–189 cm), respectively. All had stable weight during the previous 3 months, and their mean body mass index was 22.9 ± 0.6 kg/m² (range, 21.7–25.5 kg/m²). All experimental procedures were approved by the ethical committee, and the investigations were carried out at the Center of Clinical Investigation of Purpan University Hospital.

Experimental protocol

The subjects underwent similar sets of investigations on days 1 and 6, i.e. before and during a 5-day period of −6° HDBR. For that, the volunteers were placed in beds propped up with blocks at the foot to achieve a −6° head down tilt. During this period, the subjects were supervised using a video camera to ensure that they remained in this position throughout the experiment. The mean daily caloric intake was 2341 ± 102 Cal. Dietary sodium and potassium intake were held constant at 90 and 80 mmol/day, respectively. Water intake was ad libitum (mean, 1236 ± 66 mL/day). The photoperiod was 16 h of light, 8 h of darkness, with lights off at 2300 h.

The day before the beginning of HDBR (day 1), the subjects entered the room at 2000 h and were maintained in a supine position. A venous blood sample of 20 mL for the preparation of leukocytes was drawn from an indwelling polyethylene catheter inserted in the cubital vein. A biopsy of abdominal sc adipose tissue was performed (after intradermal anesthesia with 1% lidocaine, Roger-Bellon, France) with a 2.3-mm diameter needle. By successive suctions, approximately 200–300 mg adipose tissue were drawn into a syringe.

At 0830 h, a microdialysis probe (20 × 0.5 mm; 20,000 dalton molecular mass cut-off; Carnegie Medicine, Stockholm, Sweden) was inserted percutaneously after light intradermal anesthesia (100 μL 1% lidocaine) into the abdominal sc adipose tissue 100 mm from the navel, on the opposite side of the biopsy. The probes was connected to a multichannel microinjection pump (Harvard Apparatus, South Natick, MA) and continuously perfused with sterile Ringer’s solution (154 mmol/L sodium, 4 mmol/L potassium, 2.5 mmol/L calcium, and 160 mmol/L chloride) supplemented with ethanol (40 mmol/L) at 0.8 μL/min.

No collection of the dialysate was performed during the first 30 min immediately after implantation of the probes. For each probe and during each experimental period, the in vivo recovery rate at 2.5μ L/min was assessed using the measure of dialysate glycerol concentrations at various perfusion rates as previously described (18, 19). Briefly, the probes were perfused at four successive rates (0.8, 1.5, 3.5, and 2.5 μL/min, respectively), and the dialysate glycerol concentrations were determined at the steady state for each perfusion rate. These concentrations were plotted (after log transformation) against the perfusion rates. Regression analysis was used to calculate the glycerol concentration at zero flow, corresponding to the extracellular glycerol concentration. The ratio (dialysate glycerol concentration at 2.5 μL/min/extracellular glycerol concentration) × 100 expressed the in vivo recovery rate of the probe at 2.5μ L/min. Previous studies have shown a good reproducibility of extracellular glycerol determinations over time when the same group of subjects is studied with 8- to 15-day intervals between experiments (18). After this calibration period, the perfusion flow rate was maintained at 2.5 μL/min, and 10-min fractions were collected. After three fractions were collected at baseline, the probe was perfused with increasing concentrations of isoprenaline, i.e. 0.01, 0.1, and 1 μmol/L. Three 10-min fractions were collected for each concentration. In each collected fraction, glycerol was assayed as the lipolysis index. The ethanol levels were measured in the perfusate and in the outgoing dialysate to assess the ethanol outflow/inflow ratio, indicating the changes in nutritive blood flow around the probes (20, 21). On day 6, the subjects performed an identical session of investigation, but in the head-down position.

Lipolysis measurements

After needle biopsy, adipocytes were isolated using the method described by Rodbell (22) in a Krebs-Ringer bicarbonate HEPES solution (pH 7.4) containing 2% BSA, 6 mmol/L glucose (KRBHA), and 0.5 mg/mL collagenase. Isolated adipocytes were washed three times, and the cells were used for lipolysis measurements in KRBHA buffer. Concentration-response curves were performed using isoprenaline (a nonselective β-AR agonist), adrenaline alone or in the presence of 10μ mol/L RX 821002 (a selective α₂-AR antagonist), and dobutamine or terbutaline (selective β₁- andβ ₂-AR-agonists, respectively). The lipolytic effect of CGP 12177 (a selective β₃-AR agonist) was studied at 100μ mol/L. All pharmacological compounds were added to a 5-μL volume at the start of the incubation performed with 2000–3000 isolated fat cells in a final volume of 100 μL KRBHA. The incubation was run for 90 min, and 30 μL infranatant were removed for the determination of glycerol (lipolytic index). Lipolytic activity was expressed as micromoles of glycerol released per 10⁶ adipocytes for 90 min. The average number of adipocytes added in 100 μL KRBHA was measured by dividing the total lipid in 100 μL (determined gravimetrically after lipid extraction) by the average cell volume (determined by counting under the microscope). The concentration of the agonist inducing a half-maximal lipolytic effect (EC₅₀) was determined using computer-fitting analysis of concentration-response curves obtained with the various β-AR agonists. The negative logarithm of the EC₅₀ value (pD₂) was defined as the AR sensitivity.

Preparation of leukocytes membranes and binding studies

Leukocytes were isolated from citrated blood samples. After dilution (1:1, vol/vol) with a saline buffer (140 mmol/L NaCl, 14.5 mmol/L Tris, 0.54 mmol/L KCl, 0.098 mmol/L MgCl₂, 0.005 mmol/L CaCl₂, and 0.55 mmol/L glucose), the samples were layered on 10 mL Ficoll-Paque (density = 1.077) and centrifuged for 30 min at 150 × g at 20 C. The upper layer containing the platelets was removed. The leukocytes remaining in a thin lower band were diluted in buffer and centrifuged at 80 × g for 10 min at 20 C, and the pellet was washed twice in buffer and stored at −60 C until use. Binding experiments were carried out using [¹²⁵I]cyanopindolol. Lymphocyte membranes were homogenized and washed in 30 mL 5 mmol/L Tris-HCl, 5 mmol/L ethylenediamine tetraacetate, and 120 mmol/L NaCl buffer (pH 7.5) before centrifugation (40,000 × g, 15 min, 4 C). Binding studies were performed in a final volume of 200 μL of the corresponding buffer. Incubations were carried out at 37 C for 40 min. Nonspecific binding was determined in the presence of 200 μmol/L adrenaline. The binding assay was stopped by the addition of 4 mL ice-cold incubation buffer followed by rapid filtration through Whatman GF/C glass fiber filters (Clifton, NJ). The filters were then washed twice with 10-mL portions of ice-cold buffer. The radioactivity retained on the filters was measured in a Packard γ-counter (Downers Grove, IL) at an efficiency of 80%. Specific binding was defined as total binding minus nonspecific binding.

Biochemical determinations

Plasma adrenaline and noradrenaline were assayed by high pressure liquid chromatography using electrochemical (amperometric) detection, as previously described (23). The detection limit was 20 pg/sample. Day to day variability was 4%, and within-run variability was 3%. Glycerol was determined in plasma, dialysate, or KRBHA samples using an ultrasensitive radiometric method (18); the intra- and interassay variabilities were 5.0% and 9.2%, respectively. The ethanol concentration in 5-μL fractions of dialysate and perfusate was determined with an enzymatic method (18); the intra- and interassay variabilities were 3.0% and 4.5%, respectively. Plasma glucose and nonesterified fatty acids were assayed with a glucose oxidase technique (Biotrol, Paris, France) and an enzymatic method (Unipath, Dardilly, France), respectively.

Drugs and analytical methods

Isoprenaline hydrochloride was obtained from Winthrop (Clichy, France). Dobutamine hydrochloride was obtained from Lilly (Saint-Cloud, France), and terbutaline sulfate was obtained from Astra (Nanterre, France). CGP 12,177 (4-[3-t-butylamino-2-hydroxypropoxy]benzimidazol-2-one) came from Ciba-Geigy (Basel, Switzerland). The α₂-AR antagonist RX 821002[ 2-(2-methoxy-1,4-benzodioxan-2yl)-2-imidazoline] was a gift from Reckitt and Colman Laboratories (Kingston-upon-Hull, UK).[ ¹²⁵I]Cyanopindolol and [γ-³²P]ATP were obtained from Amersham (Les Ullis, France) and ICN Biochemical (Orsay, France), respectively. Crude collagenase and other products were purchased from Boehringer Mannheim (Mannheim, Germany).

Statistical analysis

All values are given as the mean ± sem. ANOVA for repeated measures with Scheffe’s post-hoc test and Wilcoxon’s paired test were used for comparisons when appropriate. P < 0.05 was considered statistically significant. All statistical comparisons were performed using a statistical software package (StatView 4.5, Abacus Concepts, Berkeley, CA).

Results

Hormonal, metabolic, and cardiovascular parameters are given in Table 1. Fasting plasma noradrenaline and adrenaline levels and other biological parameters were not modified during the HDBR period. No change in resting heart rate or systolic blood pressure and a weak significant increase in diastolic blood pressure were observed. Body weight, fluid intake, and urine output were unchanged during HDBR period (not shown). The saturation curves of[ ¹²⁵I]cyanopindolol on leukocyte membranes before and during HDBR are depicted in Fig. 1. No significant difference was found in either total number of binding sites or affinity for [¹²⁵I]cyanopindolol, evaluated after Scatchard analysis (Table 1).

Table 1.

Effect of 5 days of HDBR on catecholamines, metabolic and cardiovascular parameters, and leukocyteβ -adrenoceptors

	Before HDBR	During HDBR
Plasma noradrenaline (pg/mL)	178 ± 44	169 ± 17
Plasma adrenaline (pg/mL)	30 ± 5	44 ± 10
Plasma NEFA (μmol/L)	569 ± 73	669 ± 59
Plasma glycerol (μmol/L)	78 ± 11	70 ± 7
Plasma glucose (mmol/L)	5.1 ± 0.2	5.1 ± 0.2
Heart rate (beats/min)	64 ± 3	62 ± 2
Systolic blood pressure (mm Hg)	122 ± 3	123 ± 2
Diastolic blood pressure (mm Hg)	62 ± 3	70 ± 2a
Leukocyte β-adrenergic receptors
Binding sites (fmol/mg protein)	115 ± 20	99 ± 10
Kd (pmol/L)	53.4 ± 5.5	50.7 ± 6.9

All values are the mean ± se.

^a

P < 0.05 compared to values measured before HDBR (by Wilcoxon’s paired test).

Figure 1.

Specific [¹²⁵I]cyanopindolol binding on lymphocyte membranes before (○) or during (•) 5-day HDBR. The inset shows Scatchard plots of specific binding derived from the saturation curves. Values are the mean ± se.

In situ lipolysis

The calculated interstitial glycerol concentrations were not different before and during HDBR (mean of the first three fractions, 171 ± 18 and 177 ± 25 μmol/L, respectively). The in vivo recovery rates of the probes at 2.5 μL/min were not different either before or during HDBR (33 ± 2% and 31 ± 4%, respectively). Before HDBR, the addition of 0.01 μmol/L isoprenaline (nonselective β-AR agonist) into the perfusate did not modify extracellular glycerol levels. Higher concentrations (0.1 and 1μ mol/L) significantly increased the extracellular glycerol concentrations to 153 ± 12% and 237 ± 24% over the baseline (100%), respectively (Fig. 2). During HDBR, isoprenaline increased the extracellular glycerol levels to 144 ± 13% at 0.01μ mol/L, to 261 ± 27% at 0.01 μmol/L, and to 374 ± 44% at 1 μmol/L (Fig. 2). The lipolytic response to isoprenaline was significantly higher during than before HDBR. In both periods, isoprenaline decreased the ethanol outflow/inflow ratio, indicating an isoprenaline-dependent increase in nutritive blood flow in adipose tissue (Fig. 2). The vasodilating effects of isoprenaline were more pronounced during than before HDBR.

Figure 2.

In situ effect of isoprenaline on the extracellular glycerol concentration (A) and the ethanol ratio (B) in sc adipose tissue before (○) and during (•) 5-day HDBR in microdialysis experiments. After 30 min (basal period), different concentrations of isoprenaline, indicated by the horizontal lines, were added to the perfusate. Values are the mean ± se. The effects of isoprenaline were first analyzed in each period using one-way ANOVA, with time as the factor of the analysis and Scheffe’s post-hoc test (*, P < 0.05). Then a statistical comparison of the curves was performed using two-way ANOVA for repeated measures, with HDBR period (before vs. during) and time as factors of the analysis and Bonferroni-Dunn post-hoc test. Upper panel, Before HDBR, isoprenaline increased the interstitial glycerol concentration (F = 37.91; P < 0.001); the increment was significant from the 80 min fraction. After HDBR, isoprenaline increased the interstitial glycerol concentration (F = 29.61; P < 0.001); the increment was significant from the 50 min fraction. The time courses of the two curves assessed by the interaction term of the two factors were different (F = 8.53; P < 0.001), the glycerol concentrations were different before and after HDBR from the 60 min fraction to the 120 min fraction. Lower panel, Before HDBR, isoprenaline decreased the ethanol ratio (F = 6.99; P < 0.001); the decrease was significant from the 100 min fraction. After HDBR, isoprenaline decreased the ethanol ratio (F = 17.34; P < 0.001); the decrease was significant from the 80-min fraction. The time courses of the two curves were different (F = 4.83; P < 0.001) from the 40 min fraction to the 70 min fraction.

In vitro lipolysis

The distribution of fat cell sizes (expressed as the number of cells for a given diameter) was not changed before or during HDBR (not shown). Thus, the results of the lipolysis experiments are expressed as micromoles of glycerol released per 10⁶ cells. The spontaneous release of glycerol (basal lipolysis) was significantly increased during HDBR (Table 2). Concentration-response curves obtained with isoprenaline, dobutamine (β₁-AR agonist), and terbutaline (β₂-AR agonist), before and during HDBR, are depicted in Figs. 3 and 4⁴.

Table 2.

Effect of 5 days of HDBR on spontaneous lipolysis (basal) and maximal lipolysis induced by various pharmacological agents on isolated adipocytes

	Glycerol (μmol/10f cells)		% of dbAMPC effect
	Before HDBR	During HDBR	Before HDBR	During HDBR
Basal	0.23 ± 0.07	0.36 ± 0.11a	15.8 ± 3.8	17.8 ± 5.2
Isoprenaline (10 μmol/L)	1.06 ± 0.19b	1.79 ± 0.27ab	85.0 ± 4.3	94.7 ± 16.8
Adrenaline (10 μmol/L)	0.63 ± 0.11b	1.22 ± 0.21ab	58.1 ± 6.9	57.8 ± 5.7
Adrenaline (10 μmol/L)+ RX 821002 (10 μmol/L)	1.09 ± 0.18b	1.92 ± 0.29ab	94.3 ± 3.9	93.8 ± 5.2
Dobutamine (10 μmol/L)	0.65 ± 0.13b	1.04 ± 0.18b	41.5 ± 8.4	43.1 ± 6.9
Terbutaline (10 μmol/L)	0.70 ± 0.14b	1.09 ± 0.16b	59.5 ± 6.4	59.9 ± 8.8
CGP 12,177 (100 μmol/L)	0.32 ± 0.10b	0.51 ± 0.12b	24.9 ± 5.8	29.8 ± 7.2
db-cAMP (10 μmol/L)	1.21 ± 0.20b	1.97 ± 0.26ab	100	100

db-cAMP, Dibutyryl- cAMP. Results are presented as glycerol released by 10^f fat cells (left part of the table). All values are the mean ± se. Data were analyzed using two-way ANOVA, with drug effect (drug vs. control) and HDBR (during vs. before) as factors of the analysis. In the right part of the table, the values are expressed as a percentage of the maximal lipolytic effect of dibutyryl cAMP before and during HDBR.

^a

P < 0.05 compared to corresponding values measured before HDBR.

^b

P < 0.05 compared to corresponding values (basal or adrenaline for adrenaline plus RX521002).

Figure 3.

In vitro effect of increasing concentrations of isoprenaline on glycerol release in sc isolated adipocytes before (○) and during (•) 5-day HDBR. A, Glycerol production is expressed as micromoles of glycerol released by 10⁶ isolated fat cells. Values are the mean ± se. A statistical comparison of the curves was performed using ANOVA for repeated measures, with HDBR period (before vs. during) and isoprenaline concentrations as factors in the analysis and Bonferroni-Dunn post-hoc test. The two curves were different (F = 2.49; P = 0.02) from the 10⁻⁷ mol/L concentration to the 10⁻⁵ mol/L concentration of isoprenaline. B, Glycerol production is expressed as a percentage of the maximum lipolytic effect of dbcAMP (100%).

Figure 4.

In vitro effect of increasing concentrations of dobutamine (A) and terbutaline (B) on glycerol release in sc isolated adipocytes before (○) and during (•) 5-day HDBR. Values are the mean ± sem. The comparison of the curves was performed using ANOVA for repeated measures, with HDBR period (before vs. during) and drug concentrations as factors in the analysis. The courses of the concentration-response curves, assessed by the interaction term of the two factors, were not different before and after HDBR with either dobutamine (F = 1.41; P = NS) or terbutaline (F = 1.11; P = NS).

The effect of adrenaline on lipolysis is presented in Fig. 5. Before and during HDBR, adrenaline was devoid of lipolytic effect until it reached a level of 10 μmol/L (Fig. 5). In both situations a clear increase in the lipolytic effect of adrenaline was observed when RX 821002 (selective α₂ antagonist) was added to the incubation medium. The additive effect of RX 821002 was not different before and during HDBR (86.3 ± 33.2% and 74.6 ± 20.2% increase in lipolysis induced with 10 μmol/L adrenaline, respectively). The lipolytic effect of adrenaline, measured in the presence of RX 821002, was higher during than before HDBR (F = 7.76; P = 0.01).

Figure 5.

In vitro effects of increasing concentrations of adrenaline, alone (○ and •) or in the presence of 10⁻⁵ mol/L RX 821002 (□ and ▪), on glycerol release in sc isolated adipocytes before (open symbols) and during (filled symbols) HDBR. Values are the mean ± sem. The effects of adrenaline, alone and with 10⁻⁵ mol/L RX 821002, were first analyzed in each period using one-way ANOVA, with adrenaline concentration as the factor of the analysis and Scheffe’s post-hoc test. Adrenaline alone increased glycerol release (F = 11.09; P < 0.001); the increment was only significant with the 10⁻⁵ mol/L adrenaline concentration. In the presence of RX 821002, adrenaline increased glycerol release (F = 26.89; P < 0.001); the increment was significant from 10⁻⁷-10⁻⁵ mol/L adrenaline. Then a statistical comparison of the curves was performed using ANOVA for repeated measures, with HDBR period (before vs. during), drugs (adrenaline vs. adrenaline and RX 821002), and adrenaline concentrations as factors of the analysis and Bonferroni-Dunn post-hoc test. HDBR increased the lipolytic effect of adrenaline alone or in the presence of RX 821002 (F = 12.29; P < 0.001). The addition of RX 821002 increased the lipolytic of adrenaline (F = 10.26; P < 0.001); this effect was not different during HDBR (interaction term; F = 0.59; P = NS).

The maximal lipolytic effects of the different pharmacological agent are presented in Table 2. The results were analyzed using two-way ANOVA, with drug effect (drug vs. basal) and HDBR (during vs. before) as factors of the analysis. In each case, there was an increase in drug-induced glycerol release before as well as during HDBR (Table 2), but no significant modification of the response was assessed by the interaction term of the two factors (F = 2.01–4.11; P = NS). The lipolytic sensitivities (pD₂) of isoprenaline, adrenaline, and the selective β-AR agonists used were unmodified during HDBR (Table 3).

Table 3.

Effect of 5-day HDBR on lipolytic sensitivity to various adrenergic agonists

	pD2 (−log mol/L)
	Before HDBR	After HDBR
Isoprenaline	7.74 ± 0.13	7.79 ± 0.10
Adrenaline+ RX 821002	6.63 ± 0.04	6.76 ± 0.06
Terbutaline	6.36 ± 0.10	6.52 ± 0.20
Dobutamine	6.09 ± 0.11	6.10 ± 0.02

All values are the mean ± se. pD₂ is the negative logarithm of the agonist concentration that produced the half-maximal lipolytic effect.

To study the effect of HDBR on the postreceptor pathway, the lipolytic effects of dibutyryl cAMP (dbcAMP; an analog of cAMP that directly activates cAMP-dependent protein kinase A) were also evaluated (Table 2). dbcAMP increased the glycerol release before and after HDBR. The lipolytic efficacy of dbcAMP was not changed during HDBR (interaction term: F = 4.11; P = 0.06).

To rule out the postreceptor events on β-AR-mediated lipolysis, the lipolytic activities of fat cells were expressed as a percentage of the dbcAMP effect at the maximal post-receptor stimulation. Using this expression, the concentration curves of isoprenaline before and after HDBR (Fig. 3B) were superimposable. Identical results were found with the selective β₁- and β₂-selective agonists or with adrenaline alone or in presence of RX 821002 (not shown). Table 2 reports the comparative maximal lipolytic effects of all of the compounds, using a similar expression of the results; the lipolytic effects of the selective β-AR agonists were comparable before and during HDBR.

Discussion

The metabolic adaptations of a sustained decrease in sympathetic nervous activity on human adipose tissue have never been studied. The present study demonstrates that a reduction of sympathetic activity, achieved through a 5-day HDBR, was associated with an increment in the lipolytic response promoted by β-AR agonists in sc adipose tissue.

Previous reports have shown that HDBR promotes a sustained reduction of sympathoneural release and lower noradrenaline synthesis and turnover (10, 24, 25). The inhibition of SNS activity was not associated with a reduction of plasma catecholamine levels, as concurrent hypovolemia occurs (24). This was sustained by a decrease in urinary catecholamine excretion in the first 24 h of HDBR (10, 26). The accuracy of plasma catecholamine levels has been questioned, as a decreased plasma volume may maintain (10, 26) plasma catecholamine concentrations when spillover is reduced. However, the total circulating norepinephrine level was dramatically lowered when corrected by the plasma volume (13). The lack of changes in plasma catecholamine levels in our study is in agreement with these previous reports. Leukocytes are currently used as cell models for estimating the changes occurring in β-AR expression in humans during HDBR. In our study, the number of lymphocyte β-AR was not modified. This is in agreement with the findings of other researchers who measured either the number of β-AR or the production of cAMP after β-AR stimulation in leukocytes during 9 days of HDBR (26). Conversely, a selective increase inβ ₁-/β₂-AR responsiveness has recently been reported in the cardiovascular system after 14 days of HDBR (13).

In microdialysis experiments, infusion of increasing concentrations of isoprenaline (nonselective β-AR agonist) into the probes promoted a dose-dependent increase in extracellular glycerol concentrations: this lipolytic response was enhanced during HDBR (Fig. 2A). As previously shown by our group (18), the in situ isoprenaline infusions through microdialysis probes reduced the ethanol outflow/inflow ratio, suggesting an increase in nutritive blood flow in adipose tissue (Fig. 2B). The vasodilating effect of isoprenaline was also increased during HDBR. It is noticeable that this vascular effect could partly blunt the positive effect on lipolysis, since an increment in glycerol drainage in the circulation was simultaneously promoted (19).

In agreement with in situ results, the in vitro studies of isolated fat cells showed that HDBR promoted an increase in isoprenaline-induced glycerol release (Fig. 3). This effect during HDBR was also found with selective agonists of the three β-AR subtypes (Fig. 4 and Table 2). The in vitro study revealed an increase in basal lipolysis (i.e. spontaneous glycerol release in the absence of any agonist). However, the basal lipolysis can be largely artifacted, as adipocytes are not submitted in vitro to their physiological regulation factors. It is noticeable that no change in basal interstitial glycerol concentrations was observed in microdialysis experiments. The results were expressed as a percentage of maximal dbcAMP to target the modification of lipolysis measurements (i.e. modification of the transduction system). This expression seems to be more accurate than the use of basal lipolysis. After this transformation, the stimulating effects of the various β-AR agonists were similar before and during HDBR (Fig. 3 and Table 2). This strongly suggests that a low sympathetic activity state led to a modification the lipolytic cascade, downstream from theβ -AR/adenylyl cyclase system rather than a modification of β-AR number, affinity, or the coupling system with adenylyl cyclase. The unchanged pD₂ of β-AR agonists on the lipolytic process (Table 3) argues for this hypothesis, as does the absence of modification of leukocyte β-AR (Table 1), insofar as they reflect those in adipocytes.

Thus, an increment in the activity of enzymes such as cAMP-dependent protein kinase and hormone-sensitive lipase or a reduced activity of the cAMP-dependent phosphodiesterase could explain the improved lipolytic effect of β-AR agonists described in situ and in vitro. Larger amounts of adipose tissue, not available with the needle biopsy used in the present experiment, would be required to measure these enzymatic activities. The observed increase in basal or dbcAMP lipolysis during HDBR could be caused by the increased hormone-sensitive lipase activity (or expression) because this enzyme catalyzes the rate-limiting step in fat cell lipolysis. When considering the present results obtained in HDBR with other situations known to reduce SNS activity (fasting or calorie restriction), they fit with the appearance of an increased β-AR sensitivity of adipose tissue: energy restriction increased the hormone-sensitive lipase expression, the in vitro sensitivity of the β-adrenergic component of fat cell lipolysis (27), and the sensitivity of adipose tissue to exogenously infused (19, 28–30) or exercise-released catecholamines (31).

Adipocyte α₂-AR are involved in the control of lipolysis of human adipocytes. Their stimulation by physiological catecholamines induces antilipolysis (32). They were responsible for the weak lipolytic effect of adrenaline observed (Fig. 5). The blockade by anα ₂-AR antagonist unmasked the β-adrenergic component of the adrenaline effect. It was observed that the potentiating effect of RX 821002 on adrenaline-induced lipolysis was of similar amplitude before and during HDBR, suggesting that the α₂-AR pathway was not modifies by sympathoadrenal inhibition.

In conclusion, the present results demonstrate that a low sympathetic activity state induced by 5-day HDBR increases the responsiveness of lipolytic processes in adipose tissue assessed by in situ microdialysis assay. In vitro studies on adipocytes suggest an increased efficiency of the posttransductional steps of the lipolytic cascade rather than an alteration of adrenergic receptors. It should be noticed that the whole body sensitivity of lipid mobilization to physiological and pharmacological β-AR agonists was not assessed in this study and needs further investigations.

Acknowledgements

The authors express their gratitude to Marie-Adeline Marques and Marie-Thérèse Canal for their contributions to the study. We are also indebted to Ghislaine Portolan and Marie-Antoinette Tran for laboratory support in catecholamine measurements, and to the staff of the Clinical Center of Investigation of Purpan University Hospital.

¹

This work was supported by the Centre National d’Etudes Spatiales.