The regulation of triglyceride mobilization by catecholamines was investigated in the teleost fishOreochromis mossambicus (tilapia) in vivo and in vitro. In vitro experiments were carried out with adipocytes that were isolated for the first time from fish adipose tissue. For the in vivo experiments, cannulated tilapia were exposed to stepwise decreasing oxygen levels (20, 10, and 5% air saturation; 3.9, 1.9, and 1.0 kPa Po 2, respectively), each level being maintained for 2 h. Blood samples were taken at timed intervals and analyzed for plasma lactate, glucose, free fatty acids, epinephrine, norepinephrine, and cortisol. Hypoxia exposure did not change plasma epinephrine levels. In contrast, the plasma norepinephrine concentration markedly increased at all hypoxia levels. Over the same period, plasma free fatty acid levels showed a significant continuous decrease, suggesting that norepinephrine is responsible for the reduced plasma free fatty acid concentration, presumably through inhibition of lipolysis in adipose tissue. To elucidate the mechanism, adipocytes were isolated from mesenteric adipose tissue of tilapia and incubated with 1) norepinephrine, 2) norepinephrine + phentolamine (α1,α2-antagonist),3) isoproterenol (nonselective β-agonist), 4) isoproterenol + timolol (β1,β2-antagonist), 5) norepinephrine + timolol, and 6) BRL-35135A (β3-agonist). The results demonstrate for the first time that norepinephrine and isoproterenol suppress lipolysis in isolated adipocytes of tilapia. The effect of norepinephrine is not mediated through α2-adrenoceptors but, like isoproterenol, via β-adrenoceptors. Furthermore, this study provides strong indications that β3-adrenoceptors are involved.
- teleost fish
- fat cells
- free fatty acids
in teleost fish, catecholamines play an important role in the regulation of energy metabolism during stressful conditions like hypoxia (6, 32, 46,51). In these vertebrates, lipids dominate metabolism, whereas carbohydrates play a minor role in energy production since low amounts of carbohydrates are ingested and stored (38, 50). Nevertheless, most information on the regulation of energy metabolism by catecholamines deals with carbohydrates (9, 19, 25, 30,53), whereas relatively little information is known about the regulation of lipid metabolism by these neurohormones (34).
Exposure of carp to deep hypoxia (46) resulted in a strong increase of circulating norepinephrine (NE) accompanied with a marked decrease of plasma free fatty acid (FFA) levels. In rainbow trout, on the other hand, both epinephrine (Epi) and NE increased modestly, whereas the plasma FFA concentration ([FFA]) showed only a minor reduction. From these results, it was suggested that the Epi-to-NE ratio is important for the effect on the plasma [FFA]. Infusion experiments with catecholamines in carp (45) demonstrated that infusion of Epi results in elevated plasma FFA levels, whereas NE infusion, on the contrary, causes a marked decrease of plasma FFA levels. So far, in mammals, both Epi and NE have been shown to stimulate lipolysis via β-adrenoceptors, resulting in increased plasma [FFA] (33, 54). Accumulation of these amphiphilic compounds may be deleterious to membrane structure and membrane-dependent functions (17,21). Therefore, the observation that an increase in FFA levels does not occur in fish, showing even a marked reduction instead, may be considered as an adaptive strategy to survive hypoxic conditions.
Van Raaij et al. (45) suggested that the observed decline of FFA levels by NE involved a direct stimulation of α2-adrenoceptors in adipocytes, resulting in inhibition of lipolysis, whereas Epi was proposed to stimulate lipolysis by activation of β-adrenoceptors. This concept is based on the observations of mammalian systems in which stimulation of α2- and β-adrenoceptors results in a decrease and an increase of cAMP, respectively, and consequently in a reduction and enhancement of triglyceride lipase activity (36). In a recent in vivo study, carp were infused with adrenergic agonists and antagonists (39). It was observed that yohimbine (selective α2-adrenoceptor antagonist) delayed the FFA-suppressing effect of NE. However, infusion of the α2-adrenoceptor agonist clonidine only partially mimicked the lipolysis-inhibiting effect of NE followed by a strong rebound immediately after the infusion. These results did not support a direct involvement of α2-adrenoceptors on lipolysis. Surprisingly, infusion of isoproterenol (Iso; a nonselective β-adrenoceptor agonist) caused an unexpected marked decline of plasma [FFA]. This effect could be enlarged by preincubation with ICI-118,551 (selective β2-antagonist), whereas preincubation with atenolol (selective β1-antagonist) resulted in increased FFA levels. From these results, it was concluded that β1-adrenoceptors inhibit and β2-adrenoceptors stimulate lipolysis. In mammalian fat cells β1-, β2-, and/or β3-adrenoceptors have been found only to stimulate lipolysis (55), though in rat adipocytes β3-adrenoceptors were described to interact not only with Gs but also with Gi proteins, thereby restraining Gs-mediated adenylyl cyclase activation (7). Obviously, to establish unequivocally the occurrence of inhibitory β-adrenoceptors in fish adipose tissue, in vitro experiments with adipocytes isolated from lipid depots have to be carried out.
In the present study, we demonstrated that tilapia depresses its plasma [FFA] during hypoxia. Furthermore, we were able to show that the release of FFA by isolated adipocytes is suppressed through activation of β-adrenoceptors.
MATERIALS AND METHODS
Tilapia (Oreochromis mossambicus), of both sexes, weighing 400–700 g, were obtained from the department of Animal Physiology (University of Nijmegen, Nijmegen, The Netherlands). Fish were kept in aquaria with well-aerated running local tap water, fed daily with pelleted trout food (10 g/kg body mass, Trouvit, Putten, The Netherlands), and acclimated to a 14:10-h light-dark cycle. Before the in vivo experiments were carried out, fish were acclimated for at least a month at 20°C. In vitro experiments were performed at the department of Molecular Pharmacology (University of Groningen, Groningen, The Netherlands). Before these experiments, fish were adapted to 25°C, transported and kept for at least a week in a 200-liter tank with well-aerated running water at 25°C, and fed daily with pelleted trout food (10 g/kg body mass).
In Vivo Experiments
The experiments were performed in a recirculation system as described by Vianen et al. (47). The water of the system was kept at a temperature of 20.0 ± 0.5°C, air saturated at a normoxic level of 80–90% (15.5–17.5 kPa Po 2), and pumped through experimental flow chambers at a rate of 0.8–1.0 l/min. The air saturation (AS) level of the in-flowing water was measured by an oxygen electrode, which was connected to an oxygen controller (Applikon) set to a desired AS value. When the AS level reached values below the set point (as a consequence of oxygen consumption by the fish or N2 bubbling), a magnetic valve, built in the air supply, was activated, resulting in an increase of water Po 2.
Before the start of each experiment, three fishes were captured and placed individually in the experimental flow chambers. During the experimental period, the animals were deprived of food. After 3 days of acclimation, each fish was anesthetized in the flow chamber with MS-222 (150 mg/l tricaine methanesulfonate), which was buffered by the carbonate in the water. After cessation of breathing movements, the animal was placed on an operating table, which permitted continuous irrigation of the gills with aerated water containing 100 mg/l MS-222. In all fish, a polyurethane catheter (∼10 cm), which was connected to 50 cm of polyethylene tubing (PE-50; Rubber, Hilversum, The Netherlands), was inserted occlusively in the third afferent gill artery to permit withdrawal of blood. The cannula was filled with heparinized (200 IU/ml) saline and was secured to the skin with sutures after implantation. After surgery, the animals were placed in the flow chambers again and allowed to recover for 2 days before the experiments were carried out. This 5-day preexperimental protocol has been shown to minimize the effects of handling, anesthesia, and surgery (46). Two times per day each cannula was flushed with heparinized saline (200 IU/ml) and filled with a viscose solution of polyvinylpyrrolidone (PVP; 1 g/ml saline) containing 500 IU/ml heparin. With the use of this procedure, the diffusion of heparin in the circulation is negligible.
Before the in vivo experiments were started, one control blood sample was taken (between 0900 and 0930) for measuring the initial values of plasma substrates and stress hormones. Directly after this blood sample, tilapia (n = 6) were exposed to a controlled linear Po 2 decline (over 1 h) from 80–90% AS to 20% AS (∼17–3.9 kPa Po 2). This decline was followed by a stepwise hypoxic load of successively 20, 10, and 5% AS (3.9, 1.9, and 1.0 kPa Po 2, respectively), each level being maintained for 2 h. Blood samples were taken before and during the stepwise hypoxia exposure for measuring the substrates at the following time points (in h): 0 (control); 1.5 and 3 (20% AS); 3.75 and 5.25 (10% AS); 6 and 7.5 (5% AS). Samples for measuring the stress hormones were taken at the following time points (in h): 0 (control), 3 (20% AS), 5.25 (10% AS), and 7.5 (5% AS). Control experiments with tilapia (n = 6) were carried out at constant AS levels of 80–90% AS (15.5–17.5 kPa Po 2), and blood samples were taken at the same time points.
Withdrawal of blood (300 or 450 μl) was carried out with ice-cooled, heparin-flushed (10,000 IU/ml) microliter syringes (500 μl; Hamilton) that were placed immediately on ice. Thereafter, the volume of blood was replaced by Ringer saline (52), and the cannula was refilled with PVP. The blood samples were centrifuged for 5 min at 10,000 g (Eppendorf model 5415), and plasma was separated directly. For analysis of lactate and glucose, 100 μl plasma were added to 400 μl of TCA solution (6% vol/vol) to precipitate plasma proteins. After being mixed and incubated (minimal 20 min), these samples were centrifuged, and two aliquots of 200 μl supernatant were stored at −20°C and analyzed within 1 wk. For catecholamine measurements, 100 μl plasma were directly mixed with 15 μl EDTA solution (22 mg/ml) and stored at −80°C. For measurement of FFA and cortisol, 40 and 20 μl, respectively, of untreated plasma were stored at −80°C.
In Vitro Experiments
Adipose tissue collection and incubation procedure.
Adipocytes were isolated and incubated according to the procedure in rats described by Hollenga et al. (18) with some modifications to isolate fat cells from fish adipose tissue. Tilapia were killed by a sharp blow on the head followed by spinal transection at the cervical level. Samples of mesenteric adipose tissue were removed rapidly, obtaining portions of 8–10 g/fish, which were placed immediately into a petri dish with Krebs-Henseleit buffer [at 25°C (in mmol/l): 117.5 NaCl, 5.6 KCl, 1.18 MgSO4, 2.52 CaCl2, 1.28 NaH2PO4, 25.0 NaHCO3, and 5.5 d-glucose] pregassed with 5% CO2 in O2, pH 7.4. The portions of adipose tissue were chopped with a McIIwain tissue chopper, obtaining slices of 1 mm2. Cells were isolated by incubation at 25°C in a shaking water bath in a Teflon vessel containing 20 ml of Krebs-Henseleit buffer per portion with the addition of 1% BSA (fraction V) and collagenase (type II, 130 U/ml) under an atmosphere of 5% CO2 in O2, pH 7.4. After 1.5 h, the suspension of adipocytes was filtered through a nylon cloth and washed three times with Krebs-Henseleit buffer (1% BSA). In a shaking water bath equilibrated at 5% CO2 in O2, portions of 100 μl of adipocyte suspension (∼3 × 105 cells) were incubated in Teflon vessels for 5 h at 25°C in a total volume of 3 ml Krebs-Henseleit buffer containing 2% BSA with or without (controls) the addition of adrenoceptor (ant)agonist. Control measurements were performed in triplicate and experimental measurements in duplicate. When both an agonist and antagonist were used, the adipocytes were preincubated with the antagonist 15 min before the agonist was added. The incubation period was terminated by adding the content of each vessel to 3 ml of extraction medium [1-propanol-n-heptane-1 N H2SO4(40:20:1)]. The tubes were mixed for 60 s and centrifuged for 5 min at 2,000 g in a Hettich Rotixa/KS centrifuge. From the upper layer, 400 μl were used for FFA determination.
Chemicals and analytical procedures.
MS-222 (tricaine methanesulfonate), heparin, BSA (fraction V), collagenase (type II), (−)-isoproterenol hydrochloride, and (−)-norepinephrine hydrochloride were obtained from Sigma (St. Louis, MO). Phentolamine was obtained from Novartis (Arnhem, The Netherlands) and polyvinylpyrrolidone from Merck (Darmstadt, Germany). Timolol was a gift from Merck Sharpe & Dohme (Haarlem, The Netherlands), and BRL-35135A was donated by SmithKline Beecham (Welwyn, UK).
The concentrations of FFA released by the adipocytes in the incubation vessels were measured using a Technicon autoanalyzer according to the method of Antonis as described previously (Hollenga et al., Ref.18).
Lactate measurements were performed according to the method of Hohorst (see Bergmeyer, Ref. 3) and glucose measurements with the enzymatic test kit of Instruchemie (Hilversum, The Netherlands). Before analysis of both lactate and glucose, the stored aliquots of supernatant were neutralized with potassium phosphate solution (1 M K3PO4). Plasma FFA was measured using a commercial test kit from WAKO (NEFA C method; Instruchemie, Hilversum, The Netherlands). Catecholamines were measured by an HPLC method with electrochemical detection using dihydroxybenzylamine as the internal standard (31). Cortisol was measured by RIA, using a commercial antiserum. The procedure of the RIA using this antiserum (validated previously for the measurement of cortisol in fish plasma; see Ref. 29) was slightly modified as described by Balm et al. (2).
Presentation and Data Analysis
The data of the in vivo experiments are presented in Table1 and in Fig.1, A-C, as means ± SE for both the normoxia (n = 6) and hypoxia (n = 6) groups. Statistical differences between each time point and the initial value at time (t) = 0 h were analyzed with SigmaStat with the nonparametric Friedman test (Friedman repeated-measures ANOVA on ranks); multiple comparisons were made by the Dunnett's test, and the level of significance was set to P < 0.05. Statistical differences between the normoxia group and the hypoxia group were determined with the Mann-Whitney rank sum test.
The results of the in vitro experiments were normalized to the mean value of the controls [without (ant)agonist] and are presented in Figs. 2-4 as means ± SE, representing 5–8 measurements for each concentration. Statistically significant differences (P < 0.05) between a treatment and the control value were determined using one-way ANOVA followed by the multiple-comparisons Dunnett's test. Differences between a treatment with antagonist and without antagonist were tested for statistical significance using the paired Student's two-tailed t-test (P < 0.05).
In Vivo Experiments
The initial plasma lactate and glucose concentrations (Table 1) in tilapia were 0.36 ± 0.08 and 3.62 ± 0.29 mmol/l, respectively, for the normoxia group and 0.24 ± 0.08 and 3.46 ± 0.44 mmol/l, respectively, for the hypoxia group. In the control group, both plasma glucose and lactate remained at these levels during the whole experimental period. Exposure to 20% AS (3.9 kPa Po 2) did not change plasma lactate levels. After transition and 0.5-h exposure to 10% AS (1.9 kPa Po 2), the lactate concentration ([lactate]) was significantly elevated at t = 3.75 h followed by a further increase up to 2.18 ± 0.75 mmol/l at the end of 10% AS. Also, exposure to 5% AS (1.0 kPa Po 2) caused a continuous increase of plasma lactate levels, reaching a value of 5.65 ± 1.05 mmol/l at the end of the experiment. The continuous accumulation of lactate was accompanied with a hyperglycemia during the same period. During exposure to 10% AS, plasma glucose levels increased to 6.36 ± 0.75 mmol/l at the end of this period. This was followed by a further increase during the 5% AS period to a concentration of 7.79 ± 0.70 mmol/l at t = 7.5 h. The initial level of plasma FFA was 0.47 ± 0.05 and 0.38 ± 0.05 mmol/l for the normoxia and hypoxia groups, respectively (Fig. 1 A). During subsequent normoxia exposure, plasma FFA levels fluctuated around the initial value. In the hypoxia group, plasma FFA levels showed a continuous depression that was significant at 10 and 5% AS, reaching a value of 0.22 ± 0.02 mmol/l at t = 7.5 h.
The initial plasma concentrations of cortisol (Table 1), NE (Fig.1 B), and Epi (Fig. 1 C) were 55.0 ± 9.2, 0.05 ± 0.01, and 0.15 ± 0.07 ng/ml, respectively, for the control group and 41.0 ± 12.2, 0.07 ± 0.01, and 0.21 ± 0.07 ng/ml, respectively, for the hypoxia group. In the control group, these values did not change during normoxic conditions. In the hypoxia group, plasma Epi concentration ([Epi]) fluctuated around basal levels during the whole experimental period. In contrast, both plasma NE and cortisol increased significantly at all hypoxia levels. At 20% AS, these hormones were elevated to 0.22 ± 0.05 ng/ml for plasma NE and 99.0 ± 12.9 ng/ml for plasma cortisol. Both during 10 and 5% AS, there was a further increase to 2.47 ± 0.51 ng/ml for NE and 317.2 ± 93.1 ng/ml for cortisol at t = 7.5 h.
In Vitro Experiments
Figure 2 A shows the effect of increasing concentrations of NE in the absence and presence of the α1,α2-adrenoceptor antagonist phentolamine (10 μM) on the FFA release from tilapia adipocytes. Exposure to increasing NE concentration ([NE]) gradually decreased the FFA release, with a maximal inhibition of ∼55% at 10 μM NE. Phentolamine (10 μM) had no significant effect on the concentration-dependent inhibition of NE. In Fig. 2 B, the effect of the β1,β2-adrenoceptor antagonist timolol (1 μM) on the dose-response curve of NE is shown. Timolol completely blocked the effects of NE, up to the concentration of 10 μM; the effect of 100 μM NE was hardly antagonized, however.
Exposure of the adipocytes to Iso also resulted in a concentration-dependent inhibition of FFA release, as shown in Fig.3. The FFA release already decreased significantly at 10 nM Iso, with maximal inhibition to 31.8 ± 5.6% FFA release being reached at 100 μM. Timolol (1 μM) completely antagonized the antilipolytic response induced by Iso concentrations up to 1 μM; the two higher Iso concentrations (10 and 100 μM) were hardly (nonsignificantly) antagonized.
Figure 4 shows the results of increasing concentrations of the selective β3-adrenoceptor agonist BRL-35135A. The FFA release was significantly decreased from 1 μM on up to an almost complete inhibition of the FFA release below 10% at 100 μM BRL-35135A.
In Vivo Experiments
The plasma [lactate] in cannulated tilapia was not affected during exposure to 20% AS (3.9 kPa Po 2), suggesting that the anaerobic metabolism was not stimulated at this oxygen level. This is supported by the observation that tilapia is able to suppress its metabolic rate below the standard metabolic rate, which is an important strategy to survive hypoxia conditions (41,42), revealing this fish species to be rather resistant to hypoxia. In contrast to the 20% AS period, there was a significant and continuous increase in plasma [lactate] during 10% AS (1.9 kPa) and 5% AS (1.0 kPa), which indicates that these oxygen levels are below the anaerobic threshold (47). In agreement with these findings, Van Ginneken et al. (43) have shown that the critical oxygen level with respect to the anaerobic threshold in tilapia is around 18% AS. The lactate accumulation was accompanied by hyperglycemia during the same period. Because in fish uptake rates of glucose by the peripheral tissues are low (4, 22), hyperglycemia is mainly the result of an increased hepatic glucose release to the blood via stimulation of glycogenolysis and/or gluconeogenesis (46). These processes can be stimulated by both catecholamines and cortisol (20, 27, 49, 53); catecholamines mainly stimulate glycogenolysis (27), whereas cortisol mainly stimulates gluconeogenesis (20, 40,49). Plasma [Epi] did not increase in tilapia during the stepwise decreasing oxygen levels. In contrast, there were marked elevations in plasma [NE] and cortisol concentration, especially at 10 and 5% AS; hence, both glycogenolysis and gluconeogenesis might account for the observed hyperglycemia. However, because cortisol is a steroid hormone acting specifically via DNA transcription in its target cells, it is generally slow acting (40), inducing its hyperglycemic effect after hours (5, 24) or even days (49). On the other hand, catecholamines are able to induce their metabolic effects within minutes (9, 53); therefore, it is likely that the increase in plasma glucose in tilapia was the result of adrenergic stimulation of hepatic glycogenolysis. In fish, like in mammals, catecholamines stimulate this metabolic process via activation of glycogen phosphorylase (35, 53) through stimulation of β2- and α1-adrenoceptors (11, 25, 30).
The continuous reduction of plasma FFA levels in tilapia during stepwise hypoxia is in agreement with a previous study (48) in which exposure of carp (and trout; Vianen G, van den Thillart G, van Kampen M, van Heel T, and Steffens A, unpublished observations) to prolonged hypoxia resulted in a marked continuous suppression of plasma FFA levels in both fish species. This suggests a general protection mechanism of fish against accumulation of these amphiphilic compounds, which may cause damage to the membrane structure and membrane-dependent functions in mammals after ischemic and hypoxic insults (17, 21). Van Raaij et al. (45) demonstrated in carp that depression of plasma [FFA] is induced by NE, most probably via inhibition of lipolysis, which is in contrast to the situation in mammals (33, 54). In accordance with this finding, Vianen et al. (48) observed that the permanently reduced FFA levels in carp (and trout; Vianen G, van den Thillart G, van Kampen M, van Heel T, and Steffens A, unpublished observations) during prolonged hypoxia were accompanied in both species with chronically elevated NE levels (new steady state around 4 ng/ml). In these fish species, the FFA reduction became significant when [NE] was increased to ∼2 ng/ml. In the present study, the significantly reduced FFA levels in tilapia were also accompanied with increased plasma NE levels. The FFA reduction became significant when the [NE] was ∼0.5 ng/ml, which is lower compared with carp and trout. However, because catecholamines are rapidly (minutes) cleared from the plasma (16, 28), the turnover rate of NE may have already increased during exposure to 20% AS. Therefore, the results of the present in vivo experiments may support the idea that NE suppresses plasma FFA levels in tilapia most probably via inhibition of lipolytic activity in adipose tissue.
In Vitro Experiments
The main objective of the present study was to investigate whether in fish NE decreases lipolysis via direct activation of adipocyte α- and/or β-adrenoceptors. For this purpose, adipocytes were isolated from mesenteric adipose tissue of tilapia and incubated with adrenoceptor agonists in the presence or absence of an adrenoceptor antagonist. Exposure to increasing concentrations of NE resulted in a marked reduction of FFA release from tilapia adipocytes, thus in line with our hypothesis that NE indeed inhibits lipolysis from adipose tissue. Migliorini et al. (26) observed that catecholamines did not affect the release of FFA by adipose tissue of the fish Hoplias malabaricus; in that study, slices of adipose tissue were used in contrast to isolated adipocytes in the present study, which may have resulted in an insufficient exposure of fat cells to the catecholamines in the first case. In an early study by Farkas (13), on the other hand, the FFA production by 1- to 2-mm-thick slices of bream (Abramis brama) adipose tissue did decrease in the presence of NE. In the present study, addition of NE in the concentrations of 1–100 nM slightly but nonsignificantly affected the FFA production. However, FFA release from the adipocytes decreased considerably in the presence of 1 μM NE, which remained sustained at higher [NE]. Likewise, Farkas (12) observed a marked reduction in FFA release by carp adipose tissue incubated with 6 μM NE, which did not change as [NE] was raised gradually to 120 μM. This suggests that the threshold of its antilipolytic action lies between 10−7 and 10−6 M.
The NE-induced reduction of FFA production in tilapia adipocytes was not affected by the addition of phentolamine, indicating that in tilapia the effect of NE is not mediated through α2 (or α1-)-adrenoceptors. These results are in contrast to the in vivo observations of Van den Thillart et al. (39) in carp, which showed that yohimbine (α2-adrenoceptor antagonist) antagonized the NE effect. However, a reduction in plasma FFA could be mimicked only partially by the α2-adrenoceptor agonist clonidine. It may be possible that, in carp, other lipid depot tissues like liver and muscle are affected by NE via α2-adrenoceptors. An additional explanation may be that catecholamines are potent vasomodulators in fish (1) like in mammals. During NE infusion in vivo, the FFA reduction in the plasma of carp may have been partly due to an NE-induced inhibition of visceral blood flow via vasoconstriction through α2-adrenoceptors (15). This may have resulted in a decreased release of FFA from visceral lipid stores, which could be antagonized by yohimbine.
Van den Thillart et al. (39) observed a marked reduction in plasma FFA after infusion of Iso in vivo in carp, indicating the involvement of β-adrenoceptors. Incubation of tilapia adipocytes with Iso in the present study also resulted in a decline of FFA release, which was already significant at 10 nM, indicating that Iso is more potent (by a factor of 10–100) than NE, as is also the case for the lipolytic effects in mammalian adipocytes (55). The effects of the lower concentrations of Iso and NE could be inhibited by timolol (a potent β-adrenoceptor antagonist) at 1 μM, a concentration that does not yet antagonize β3-adrenoceptors (Obels and Zaagsma, unpublished observations). Therefore, these results strongly suggest that the antilipolytic effect of NE and Iso under in vivo conditions is mediated to an important extent through β1- and/or β2-adrenoceptors.
Although β-adrenoceptors generally activate Gs proteins and thereby adenylyl cyclase activity, β1- and β3-adrenoceptors in rat adipocytes have been postulated to interact with inhibitory Gi proteins as well. Thus Chaudhry et al. (7) demonstrated that pertussis toxin, which locks Gi in the GDP (inactive) form, markedly potentiates the β3- and, to some extent, the β1-adrenoceptor-mediated cAMP accumulation by Iso. In the present study, it is observed that at higher concentrations of Iso (10 and 100 μM) timolol (1 μM) did not significantly antagonize the Iso effect. It is known that, in mammalian adipocytes, Iso preferentially stimulates β1- and β2-adrenoceptors at low concentrations, whereas higher Iso concentrations also activate β3-adrenoceptors (7, 8, 14). Therefore, the observed reduction in FFA release at the highest Iso concentrations in the presence of timolol (1 μM) suggests the involvement of β3-type adrenoceptors. In agreement, exposure of the tilapia adipocytes to increasing concentrations of BRL-35135A (a selective β3-adrenoceptor agonist) reduced the FFA production by 90% at the highest concentration. This provides evidence that β3-adrenoceptors are also present in fish adipose tissue and inhibit FFA mobilization in contrast to the mammalian situation (14, 37, 44, 55).
BRL-35135A is the methyl ester of BRL-37344, the prototype and most frequently studied β3-adrenoceptor agonist. For the present study, BRL-35135A was chosen because this compound has been shown to relax guinea pig taenia cecum solely through β3-adrenoceptors, with an even fivefold higher potency than BRL-37344 (23). The observation that significantly higher concentrations of BRL-35135A are required to inhibit lipolysis of tilapia adipocytes (present study) than to activate rodent β3-adrenoceptors (23) is reminiscent of observations made in human adipocytes in which the parent compound BRL-37344 in stimulating lipolysis also has a low potency (18). However, it should be mentioned that the structure and properties not only of β3- but also of β1- and β2-type adrenoceptors in tilapia (and in fish, in general) are yet completely unknown and may deviate from those of mammalian species. Hence, the selectivity of BRL-35135A for rodent β3-adrenoceptors does not necessarily imply selectivity for tilapia β3-adrenoceptors. Remarkably, however, in tilapia adipocytes the maximal efficacy of BRL-35135A exceeds that of Iso, which was not seen before in mammalian systems. This finding would indicate that inhibition of lipid mobilization in tilapia by activation of β3-adrenoceptors is more efficacious than by β1- and/or β2-adrenoceptors.
It could be argued that the suppression of lipid mobilization, as observed in the present study, may be the result of enhancement of reesterification rather than, or in addition to, inhibition of lipolysis. However, to our knowledge, direct activation of reesterification of FFA by stimulation of adipocyte β1-, β2-, and/or β3-adrenoceptors with catecholamines has never been observed. Furthermore, in our previous study in carp (39), a single injection of Iso reduced plasma FFA levels acutely (within 30 min) and simultaneously increased plasma glucose. Remarkably, the antilipolytic effect was significantly accelerated and increased by pretreatment with the selective β2-adrenoceptor antagonist ICI-118,551, whereas the glycogenolytic effect of Iso was decreased by ICI-118,551. These results strongly indicate that reesterification of FFA does not underly the antilipolytic effect of the catecholamines.
From the present results, it is not clear to what extent NE (released from sympathetic nerves) exerts its effect through β3-adrenoceptors. In mammalian systems, β3-adrenoceptors have been found to be in close proximity to sympathetic nerve terminals, leaving the possibility of high local concentrations of NE to be attained during enhanced nerve activity (10). Timolol at 1 μM almost completely antagonized the reduction of FFA release up to 10 μM of NE. At the highest [NE], however, timolol was not or at most partially effective, suggesting the (additional) involvement of β3-adrenoceptors. It is obvious that further research is required to characterize the β-adrenoceptor subtype in more detail, including their capacity to couple to inhibitory Gi proteins in tilapia adipocytes.
In summary, the results of the present study indicate that, in tilapia, endogenous (hypoxia-induced) and exogenously applied NE suppresses lipolysis in tilapia adipose tissue, resulting in a decline of FFA levels. Whereas in carp this effect is mediated partially through α2-adrenoceptors, in tilapia it appeared to be mediated solely through β-adrenoceptors. Furthermore, for the first time, strong indications are provided that, in fish, β3-adrenoceptors are involved in the reduction of lipolysis.
We thank the biology students (University of Leiden) for assistance with the in vivo experiments and Frans Brouwer and Tamara van Heel for skillful assistance with the analyses of the catecholamines and cortisol, respectively. We thank Dr. J. R. S. Arch (SmithKline Beecham, Epson, UK) for the gift of BRL-35135A, Dr. M. Axelsson for the generous gift regarding the polyurethane tube, and Dr. A. D. F. Addink for reading the manuscript critically.
Address for reprint requests and other correspondence: G. J. Vianen, Institute of Evolutionary and Ecological Sciences, Dept. of Integrative Zoology, Van der Klaauw Laboratories, PO Box 9516, 2300 RA Leiden, The Netherlands (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society