| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
1- and ß1-Adrenoceptor Signaling Fully Compensates for ß3-Adrenoceptor Deficiency in Brown Adipocyte Norepinephrine-Stimulated Glucose Uptake
The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, SE-106 91 Stockholm, Sweden
Address all correspondence and requests for reprints to: Tore Bengtsson, The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, SE-106 91 Stockholm, Sweden. E-mail: tore.bengtsson{at}zoofys.su.se.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1A-, 1D-, and moderate ß1-adrenergic receptor mRNA levels with no apparent expression of other adrenergic receptors. The levels of 1A-, 1D-, and ß1-adrenergic receptor mRNA were not changed in the ß3-KO brown adipocytes, indicating that the ß3-adrenergic receptor ablation does not influence adrenergic gene expression in brown adipocytes in culture. As expected, the ß3-adrenergic receptor agonists BRL-37344 and CL-316 243 did not induce 2-deoxy-D-glucose uptake in ß3-KO brown adipocytes. Surprisingly, the endogenous adrenergic neurotransmitter norepinephrine induced the same concentration-dependent 2-deoxy-D-glucose uptake in wild-type and ß3-KO brown adipocytes. This study demonstrates that ß1-adrenergic receptors, and to a smaller degree 1-adrenergic receptors, functionally compensate for the lack of ß3-adrenergic receptors in glucose uptake. ß1-Adrenergic receptors activate glucose uptake through a cAMP/protein kinase A/phosphatidylinositol 3-kinase pathway, stimulating conventional and novel protein kinase Cs. The 1-adrenergic receptor component (that is not evident in wild-type cells) stimulates glucose uptake through a phosphatidylinositol 3-kinase and protein kinase C pathway in the ß3-KO cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-, 2-, and ß-adrenergic receptors) (2, 3, 4, 5) including the ß3-adrenergic receptor, which is a potential target for antiobesity and antidiabetic drug therapy (6). Glucose uptake in BAT is markedly increased during two opposite metabolic conditions: during active anabolic processes (stimulated by insulin) and activation of thermogenesis (activation of the sympathetic nervous system) (7, 8, 9). Cultured brown adipocytes have previously been characterized to express an intact adrenergic- and insulin-signaling system (10, 11, 12, 13, 14), and in vitro glucose uptake is stimulated in brown adipocytes by norepinephrine and adrenergic agents (7, 15, 16, 17, 18).
We have shown that norepinephrine increases glucose uptake in these cells through ß3-adrenergic receptors, with no apparent contribution of 1-, 2-, ß1-, or ß2-adrenergic receptors (18). Evidence for an important role of ß3-adrenergic receptors in lipolysis, thermogenesis, and glucose homeostasis comes from studies using the ß3-adrenergic receptor knockout (KO) model (19, 20, 21, 22, 23), but significant uncertainty exists regarding the relative role of ß3-adrenergic receptors vs. ß1-adrenergic receptors in mediating signal transduction in brown adipose tissue. The usage of ß3-adrenergic receptor KO mice has made it possible to address some of these questions (19, 20), but it has been concluded that ß3-adrenergic receptor ablation in these mice alters ß1-adrenergic receptor gene expression, probably due to physiological compensatory mechanisms (19, 20, 24).
We have here analyzed adrenergic activation of glucose uptake in primary cultures of brown adipocytes from control and ß3-adrenergic receptor KO mice that should not be affected by in vivo compensatory mechanisms. The aim of the present study was to assess the relative roles and potential contribution of other adrenergic receptor subtypes and their signaling systems in norepinephrine-mediated glucose uptake in brown adipocytes. This study demonstrates that ß1-adrenergic receptors and to a smaller degree 1-adrenergic receptors functionally compensate for the lack of ß3-adrenergic receptors in ß3-adrenergic receptor-deficient brown adipocytes, without compensatory receptor up-regulation. We show here that ß1-adrenergic receptors activate glucose uptake through the same cAMP/protein kinase A/phosphatidylinositol 3-kinase (PI3K) pathway stimulating conventional and novel protein kinase Cs (PKCs) as ß3-adrenergic receptors in intact brown adipocytes. We also observe a 1-adrenergic receptor component that is not evident in wild-type cells, which stimulates glucose uptake through a PI3K and PKC pathway in the ß3-adrenergic receptor KO cells. Thus, both ß1-adrenergic receptors and 1-adrenergic receptors compensate for the lack of ß3-adrenergic receptors in the KO cells. Our results therefore demonstrate an important role for ß3-adrenergic receptors but also a remarkable redundancy in the adrenergic receptor system in mediating glucose uptake in brown adipocytes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The experiments were conducted with ethical permission from the North Stockholm Animal Ethics Committee.
Primary cell culture of brown adipocytes
The cell culture medium consisted of DMEM supplemented with 10% newborn calf serum (Life Technologies, Paisley, Scotland, UK), 2.4 nM insulin, 10 mM HEPES, 50 IU/ml penicillin, 50 µg/ml streptomycin, and 25 µg/ml sodium ascorbate (25). Aliquots of 0.1 ml cell suspension were cultivated in 12-well culture dishes with 0.9 ml cell culture medium. Cultures were incubated at 37 C in a water-saturated atmosphere of 8% CO2 in air (Heraeus CO2-auto-zero B5061 incubator, Hanau, Germany). On d 1, 3, and 5, the medium was discarded, cells washed with prewarmed DMEM, and fresh medium added. After 5 d in culture, the brown adipocyte precursor cells spontaneously convert from displaying fibroblast-like morphology to acquiring typical mature brown adipocyte features; this conversion occurs at the time of cellular confluence without any addition of differential mix (26, 27, 28). In these cells, spontaneous induction of ß3-adrenergic receptor mRNA and other mature adipocyte markers reaches a steady-state level at d 5 that coordinates with the ability of norepinephrine to induce the expression of the most specific brown adipocyte differentiation marker: the uncoupling protein 1 (UCP1) (29). UCP1 is currently the only specific marker for brown fat cells (for longer discussion about mature brown adipocytes, see review in Ref. 30).
Analysis of adrenergic mRNA levels in primary brown adipocytes (RT-PCR)
Primary brown adipocytes were grown as described above and RNA prepared after 7 d of differentiation. Tissues (brain, ventricle, liver, white adipose tissue, BAT) were obtained from one FVB or one ß3-adrenergic receptor KO mouse (male, 4 wk old). Mice were anesthetized with 80% CO2-20% O2 and decapitated and tissues rapidly removed, frozen in liquid nitrogen, and stored at –80 C until use. For both cells and tissues, total RNA was extracted by homogenization in Ultraspec (Biotecx, Houston, TX) according to the manufacturer’s instructions. The yield and quality of RNA was assessed by measuring absorbance at 260 and 280 nm and electrophoresis on 1.3% agarose gels. There was no degradation of any RNA samples.
cDNAs were synthesized by reverse transcription of 1 µg of each total RNA using oligo (dT)12–18 as a primer according to the SuperScript first-strand synthesis system for RT-PCR kit (Invitrogen, Paisley, Scotland, UK). PCR amplifications were carried out in a Hybaid Omni Gene thermocycler. cDNA equivalent to 100 ng of starting RNA was amplified using primers specific for 1A-, 1B-, 1D-, ß1-, ß2-, or ß3-adrenergic (Invitrogen, Cybergene, Huddinge, Sweden; see Table 1
). For 1A-, ß2-, or ß3-adrenergic PCR, PCR mixes contained cDNA, 2 U Taq DNA polymerase (Invitrogen), 1x PCR buffer, 200 µM deoxynucleotide triphosphates (dNTPs), 1.5 mM MgCl2, forward primer (1 ng/µl), and reverse primer (1 ng/µl) in a total volume of 50 µl. For 1B- or ß1-adrenergic PCR, PCR mixes contained cDNA, 1.25 U Platinum Pfx DNA polymerase (Invitrogen), 1x Pfx amplification buffer, 1 mM MgSO4, 300 µM dNTPs, forward primer (1.5 ng/µl), and reverse primer (1.5 ng/µl) in a total volume of 50 µl. For 1D-adrenergic PCR, PCR mixes contained cDNA, 2.5 U platinum Taq DNA polymerase (Invitrogen), 1x PCR amplification buffer, 1.5 mM MgSO4, 200 µM dNTPs, forward primer (1 ng/µl), and reverse primer (1 ng/µl) in a total volume of 50 µl.
|
1D-adrenergic PCR). Samples were extended for a further 2 min at 72 C before reactions where cooled to 20 C. The annealing temperature for the PCR was 64 C (except 1D- or ß1-adrenergic PCR, which was 60 C). For each set of primers, the log (PCR product) vs. cycle number was plotted and a cycle number chosen within the linear portion of the graph. The number of cycles performed were 28 for 1A-, 30 for 1B-, 28 for 1D-, 26 for ß1-, 26 for ß2-, and 30 for ß3-adrenergic receptor. After amplification, PCR products (10 µl) were electrophoresed on 1.3% agarose gels and visualized.
Analysis of UCP1 and adrenergic mRNA in primary brown adipocytes (Northern blot)
Total RNA (10 µg) was separated by electrophoresis in an ethidium bromide-containing agarose-formaldehyde gel. The intensity of the 18S and 28S rRNA band under UV light was checked to verify that the samples were equally loaded and that no RNA degradation had occurred. The mRNA levels were analyzed by Northern blotting as described earlier (29).
The UCP1 cDNA was used as described previously (3). The probes were labeled with [-32P]dCTP using Ready To Go DNA labeling beads (Amersham, Aylesbury, UK) according to the manufacturer’s instructions. Adrenergic receptor mRNA was also analyzed with Northern blot technique. The rat ß1-adrenergic receptor cDNA probe used has been previously characterized (31). It was cloned in the EcoR1 site of the PVZ1 plasmid (
2.7 kb) and the 1.5 kb fragment obtained by Nar1 digestion was used for hybridization. The ß2-adrenergic receptor probe was an 896-bp EcoR V-BstEII fragment obtained from the human ß2-cDNA in pUC 18 (32). The ß3-adrenergic receptor probe was a 306-bp Nhe I-Xho I fragment obtained from the ß3-adrenergic receptor cDNA clone IMAGE: 4189430 corresponding to ß3-adrenergic receptor residues 120–222. The phosphoglycerate kinase (PGK)-NEO (neomycin)-poly(A) vector replaces this 306-bp in ß3-adrenergic receptor KO mice (19). The 1A-adrenergic receptor probe was made from 1A-adrenergic receptor PCR product from mouse BAT, which was randomly labeled with [-32P]dCTP. The 1D-adrenergic receptor probe was made from reamplified 1D-adrenergic receptor PCR product from mouse brain with the standard PCR protocol, except cold 0.1 mM dCTP and 2.5 µl [-32P]dCTP was used.
Analysis of ß-adrenergic receptor protein levels in primary brown adipocytes (Western blot)
Mice (FVB or ß3-adrenergic receptor KO mice, 3 wk old) were killed with 80% CO2-20% O2, and decapitated and tissues rapidly excised, removed of visible white fat and connective tissues, and quickly frozen in liquid nitrogen before being stored at –80 C. Tissues were homogenized in ice-cold homogenization buffer [20 mM Tris (pH 7.4) at room temperature, containing 1 mM EDTA and Miniprotease inhibitor (one tablet per 10 ml buffer; Roche, Stockholm, Sweden)], and centrifuged at low speed (800 x g, 10 min, 4 C) to remove cell debris and unhomogenized tissue. The supernatant was retained and centrifuged (20,000 x g, 60 min, 4 C) and the resulting pellet suspended in homogenization buffer and stored at –20 C until further use. Primary brown adipocytes (differentiated for 7 d) were washed in ice-cold PBS, scraped, homogenized in homogenization buffer with a Dounce homogenizer (approximately 20 strokes), and centrifuged at low speed (800 x g, 10 min, 4 C) to remove cell debris. The supernatant was retained and the pellet rehomogenized and centrifuged again. Supernatants were pooled and centrifuged (20,000 x g, 60 min, 4 C). The pellets were resuspended in homogenization buffer and stored at –20 C until further use. Proteins were measured (33) using BSA as a standard sample.
Samples (50 µg protein except for ß1-adrenergic receptor in BAT that was loaded with 25 µg protein) were mixed with an equal volume of sodium dodecyl sulfate sample buffer [62.5 mM Tris (pH 6.8), 2% sodium dodecyl sulfate, 10% glycerol, 50 mM dithiothreitol, and 0.1% bromophenol blue] and boiled for 3 min. Samples were separated on 10% polyacrylamide gels and transferred to Hybond-C Extra nitrocellulose (Amersham Biosciences, Arlington Heights, IL) membranes (pore size 0.45 µm). After transfer, membranes were washed in Tris-buffered saline [TBS; 20 mM Tris, 140 mM NaCl, (pH 7.6)] for 5 min followed by quenching of nonspecific binding in blocking buffer (5% nonfat dry milk, 0.1% Tween 20 in TBS) for 1 h at room temperature. Membranes were incubated at 4 C overnight with gentle shaking with primary antibody ß1-adrenergic receptor (V-19), ß2-adrenergic receptor (H-73), and ß3-adrenergic receptor (M-20) (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at 1:500 dilution factor in buffer (TBS containing 0.1% Tween 20, 5% wt/vol fraction V BSA). Primary antibody was detected using a secondary antibody (horseradish peroxidase-linked antirabbit for ß1-adrenergic receptor, ß2-adrenergic receptor, horseradish peroxidase-linked antigoat for ß3-adrenergic receptor) at 1:2000 dilution factor for 1 h at room temperature in blocking buffer and detected with enhanced chemiluminescence (Amersham Pharmacia Biotech, Little Chalfont, UK) according to the instructions of the manufacturer.
2-Deoxy-D-[1-3H]-glucose uptake in primary brown adipocytes
Glucose uptake studies were performed as previously described (18, 34). All experiments were performed on d 7 of cell culture. From d 6, the cells were serum starved overnight in DMEM/nutrient mix F12 (1:1) with 4 mM L-glutamine, 0.5% BSA, 2.4 nM insulin, 10 mM HEPES, 50 IU/ml penicillin, 50 µg/ml streptomycin, and 25 µg/ml sodium ascorbate. To reduce basal glucose uptake, this medium was changed to DMEM without insulin (containing 0.5% BSA, 0.125 mM of sodium ascorbate) for 30 min [longer periods did not further reduce basal uptake (data not shown)] before the cells were challenged with drugs for a total of 2 h [time courses showed that 2 h stimulation gave the maximum response to insulin or to adrenergic agonists (data not shown)]. Detailed protocols for the inhibitors used are found in the description of each experiment. After 110 min of incubation with drugs, the medium was discarded, cells washed with prewarmed PBS buffer [10 mM phosphate buffer, 2.7 mM KCl, 137 mM NaCl (pH 7.4)] before glucose-free DMEM (containing 0.5% BSA, 0.125 mM sodium ascorbate) was added and drugs readded with trace amounts of 2-deoxy-D-[1-3H]-glucose (50 nM) (Amersham; specific activity 9.5–12 Ci/mmol) for 10 min. Reactions were terminated with washing in ice-cold PBS, cells lysed (500 µl of 0.2 M NaOH, 1 h at 55 C) and the incorporated radioactivity determined by liquid scintillation counting.
cAMP determinations in primary brown adipocytes
All experiments were performed on d 7 of cell culture. The cells were treated according to the first part of the protocol for the glucose uptake studies. After indicated times with drugs, the culture medium was aspirated, 0.8 ml of 95% ethanol added to each well, and the cells scraped off. The samples were routinely washed with 70% ethanol (0.4 ml) and the combined suspensions dried in a Speedvac centrifuge. The dried samples were dissolved in 150–500 µl of the buffer 1 provided with the cAMP (3H) assay system from Amersham (TRK 432), sonicated briefly, and centrifuged at 14,000 x g for 10 min. One 50-µl aliquot of the supernatant was analyzed for every sample according to the description in the assay system; for every concentration of any agonist in each experiment, duplicate wells were analyzed.
Analysis of results
For analysis of concentration-response curves, the curve-fitting option of the KaleidaGraph 3.0 program (Synergy Software, Reading, PA) was used. Monophasic concentration-response data were analyzed with the Michaelis-Menten equation VA = basal + Vmax/(1 + (EC50/[A]), where A is the concentration of adrenergic agent added, VA the response observed at that concentration, and Vmax the estimated maximal increase. In some calculations, basal was set as a constant to avoid a singular matrix that would make the fitting unsolvable.
Results are presented as the mean values ± SE. Student’s paired t test was used to test for significance between the different treatments and/or controls. In studies in which inhibitors affected basal levels, differences between stimulated and inhibited effects were calculated as delta values above each corresponding control (i.e. control and inhibitor alone). Statistical significance in text means at least P 
0.05 (statistical significance in figures: *, P 0.05 and **, P 0.01). In experiments in which antagonists were used, log (dose ratio-1)-log[antagon] (pKB) values were calculated according to the method of Furchgott (35) and values given as mean ± SE.
Chemicals
L-Norepinephrine bitartrate (Arterenol), (±)-isoproterenol, DL-propranolol, collagenase (type II), BRL-37344, cirazoline, prazosin, 8-bromoadenosine-cAMP (8-Br-cAMP), 12-O-tetradecanoylphorbol-13-acetate (TPA), LY294002, CL-316243, and forskolin were obtained from Sigma-Aldrich (St. Louis, MO) and Ro-81–3220, Gö6976, and Gö6983 from Calbiochem (La Jolla, CA). ICI-89406 and ICI-118551 were from Zeneca (Wayne, PA). 2',5'-Dideoxyadenosine (DDA) was from Biomedicals Inc., Irvine, CA. Insulin (Actrapid) was from Novo Nordisk (Bagsvaerd, Denmark) and cAMP kit (TRK 432), 2-deoxy-D-[1-3H]-glucose (specific activity 9.5–12 Ci/mmol) from AmershamBiosciences (Little Chalfont, UK). All cell culture media and supplements were from Life Technologies (Carlsbad, CA). The concentrations of the blockers in this study have been carefully selected for best selectivity and to limit putative side effects (36, 37, 38, 39, 40).
All adrenergic agents, 8-Br-cAMP, and Ro 81–3220 were dissolved in water. Norepinephrine was dissolved in water with 0.125 mM Na ascorbate. TPA, LY 294002, Gö 6983, Gö 6976, DDA, and ICI-89406 were dissolved in dimethyl sulfoxide (final concentration was maximally 0.1%).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1A-Adrenergic receptors.1A-adrenergic receptor subtype gene was highly expressed in mouse BAT (as compared semiquantitatively with brain that is known to express high levels of the 1A-adrenergic receptor subtype) (Fig. 1A). This is in accordance with results from rat BAT in which the 1A-adrenergic receptor subtype gene is highly expressed, compared with other tissues (12, 41, 42). In contrast, primary brown adipocytes in culture showed very low level of 1A-adrenergic receptor subtype expression, which did not differ between wild-type and the ß3-KO brown adipocytes with RT-PCR (Fig. 1A) or Northern blotting (Fig. 1B).
|
1B-Adrenergic receptors.1B-adrenergic receptor gene expression in BAT and primary brown adipocytes, in accordance with results from rat BAT (12) (Fig. 1A
).
1D-Adrenergic receptors.
The levels of 1D-adrenergic receptors gene expression was high in mouse BAT, in accordance with results from rat BAT (12, 41, 42). There were similar levels of 1D-adrenergic receptor gene expression in the primary brown adipocytes, with no difference between wild-type and ß3-KO brown adipocytes as observed with RT-PCR (Fig. 1A) or Northern blotting (Fig. 1B).
ß1-Adrenergic receptors.
ß1-Adrenergic receptor gene and protein expression in BAT was confirmed by RT-PCR (Fig. 1A), Northern blotting (Fig. 1B) as previously described (11), and Western blotting (Fig. 1C). The ß1-adrenergic receptor mRNA levels were up-regulated in ß3-KO BAT in both the RT-PCR experiment and Northern blots (Table 2) in agreement with earlier findings (19). This correlated with elevated ß1-adrenergic receptor protein amount. However, in ß3-KO brown adipocytes in culture, the ß1-adrenergic receptor mRNA levels were not significantly different from wild-type levels in both the RT-PCR and Northern blots (Fig. 1, A and B, and Table 2).
|
ß2-Adrenergic receptors.
We detected high levels of ß2-adrenergic receptor mRNA with RT-PCR [and Northern blot (not shown)], compared with ventricle (Fig. 1A), and high levels of ß2-adrenergic receptor protein in BAT, compared with soleus muscle (Fig. 1C). However, there were very low levels of ß2-adrenergic receptor gene expression in primary brown adipocytes, compared with BAT [in accordance with our previous studies (we earlier suggested that the ß2-adrenergic receptor is expressed in blood vessels) (11, 43)], and we could not detect any ß2-adrenergic receptor protein in primary brown adipocytes (Fig. 1C).
ß3-Adrenergic receptor.
The mouse ß3-adrenergic receptor gene contains two exons, which undergo alternative splicing and produce expressed splice variants of the ß3-adrenergic receptor (44). The two PCR products obtained from the ß3-adrenergic PCR reflect the two different isoforms of the ß3-adrenergic receptor; the 234-bp fragment represents the ß3A-adrenergic, whereas the 337-bp fragment represents the ß3B-adrenergic (Fig. 1A), which is downstream the PGK-NEO-poly(A) cassette used for the ß3-adrenergic receptor KO (19). The ß3A-adrenergic receptor variant was the most abundant variant in BAT, as reported previously (44). The ratio between the different transcripts was not changed in primary brown adipocytes, and there was no difference in wild-type and ß3-KO cultures [also in the ß3-KO lane, there are ß3-adrenergic bands because this is fully downstream of the PGK-NEO-poly(A) cassette used for the ß3-adrenergic receptor KO mice (19)]. As expected, in Northern blots with BAT and primary cultures from ß3-KO mice (Fig. 1B and Table 2), we did not detect any ß3-adrenergic bands because the probe we used was homolog with the disrupted region of the ß3-adrenergic receptor gene (19). In agreement, no ß3-adrenergic receptor protein in ß3-KO mice was observed (Fig. 1C).
The result of the expression level of the adrenergic receptor genes show that the PGK-NEO-poly(A) cassette upstream the ß3-adrenergic gene does not influence the transcription levels per se and that the ß3-ablation does not lead to a compensatory mechanism on the gene transcription in the cultured primary brown adipocytes.
UCP1 induction in ß3-KO brown adipocytes in culture
In mature cells (d 7), unstimulated brown adipocytes expressed the UCP1 gene at low levels with a tendency to a lower basal expression in the ß3-KO cells (n = 2 in duplicate). The addition of the ß-adrenergic agonist isoprenaline led to a large increase in UCP1 mRNA both in control cells (Fig. 2) [in agreement with our earlier observations (18)] and ß3-KO cells. Thus, brown adipocyte differentiation was confirmed by induction of UCP1 gene expression.
|
. There was no significant difference between the –log(p)EC50 or maximal insulin-stimulated glucose uptake for the wild-type cells (pEC50 9.1 ± 0.2; Vmax 217 ± 7%) and the ß3-KO cells (pEC50 9.4 ± 0.6; Vmax 204 ± 9%).
|
, A and B).
|
1- and ß1-adrenergic receptors), implying that adrenergic receptors other than the ß3-adrenergic receptor can induce glucose uptake.
|
). We also used the selective ß1-adrenergic receptor antagonist ICI-89406. This compound antagonized the norepinephrine concentration-response curve with a pKB of 7.6 ± 0.1 (Fig. 6B). Both ß1- and ß2-adrenergic receptors are characterized by high affinity for classical ß-adrenergic receptor antagonists such as propranolol with pA2 values of about 9 on ß1-/ß2-adrenergic receptors (49). It is therefore evident that ß1-adrenergic receptors play a major role in the norepinephrine-induced glucose uptake in ß3-KO brown adipocytes (we found no evidence for ß2-adrenergic receptors in these cells) but that the stimulatory effect of norepinephrine cannot be explained fully by ß1-adrenergic receptors (pKB values). This is in contrast to control cells, in which we found no evidence that ß1-receptors or other receptors than ß3-receptors are involved in norepinephrine-induced increase in glucose transport in brown adipocytes (18).
|
). However, the response to norepinephrine (Vmax 348 ± 12%; pEC50 6.8 ± 0.1) was significantly greater than that to isoprenaline in the ß3-KO cells (Fig. 6C), implying an involvement of -adrenergic receptors. To investigate whether norepinephrine may act at 1- or 2-adrenergic receptors to increase glucose uptake in cultures from ß3-KO cells, we performed glucose uptake experiments in the presence of a 1-adrenergic receptor antagonist (prazosin) or an 2-adrenergic antagonist (yohimbine). Norepinephrine-induced glucose uptake was not affected by yohimbine (Fig. 7A) but was significantly reduced by prazosin (Fig. 7B), indicating that norepinephrine-induced glucose uptake in ß3-KO cells is partially mediated by 1-adrenergic receptors. This was confirmed in studies showing that the 1-adrenergic receptor agonist cirazoline produced a concentration-dependent increase in glucose uptake in these cells (Fig. 7C). The ß-adrenergic agonist isoprenaline in combination with the 1-adrenergic agonist cirazoline gave the same maximal response as norepinephrine alone (Fig. 7D).
|
To further investigate the role of cAMP formation in activation of glucose uptake, we blocked cAMP formation with DDA, which blocks adenylyl cyclase (50). DDA significantly inhibited norepinephrine- and isoprenaline-activated cAMP formation (84 and 87% inhibition) (Fig. 8A). Insulin did not significantly increase cAMP levels (data not shown). Parallel cultures were analyzed for 2-deoxy-D-glucose uptake. DDA significantly inhibited norepinephrine- and isoprenaline-mediated 2-deoxy-D-glucose uptake (62 and 64% inhibition) but not insulin stimulated 2-deoxy-D-glucose uptake (Fig. 8B). The stimulation was made with a concentration of 1 µM for the agonists (at higher norepinephrine concentrations, cAMP levels decrease, compared with 1 µM in brown adipocytes; for a longer discussion, see Ref. 10), and we concordantly found no statistical difference between norepinephrine and isoprenaline in 2-deoxy-D-glucose uptake as we did at a higher concentration (Fig. 6C). This indicates that a large component, if not all, of the ß1-adrenergic receptor-mediated increase in glucose uptake in brown adipocytes is mediated through cAMP.
|
1- and ß1-adrenergic receptor-induced glucose uptake1- and ß1-adrenergic receptor signaling increases glucose uptake through PI3K in ß3-KO cells, we used two specific PI3K inhibitors (wortmannin and LY294002). Isoprenaline-induced glucose uptake was inhibited to a large degree by wortmannin (61% inhibition) and LY 294002 (66% inhibition), and norepinephrine-mediated glucose uptake was also inhibited by LY294002 (52% inhibition) (Figs. 9 and 10A). Insulin-stimulated glucose uptake was inhibited by the PI3K inhibitors to a very high degree (wortmannin, 90% inhibition and LY 294002, 84% inhibition). The 1-adrenergic receptor (cirazoline) increase was completely inhibited (101% inhibition) by LY294002 (Fig. 10B). To investigate the 1-adrenergic receptor signaling pathway, we examined whether the phorbol ester TPA, which stimulates conventional and novel PKCs, stimulated glucose uptake. TPA activated glucose uptake in ß3-KO brown adipocytes to a large extent (Fig. 10B), which was also reduced by LY294002 (56% inhibition) (P = 0.06) but not to the same extent as cirazoline-induced glucose uptake. The Ca2+ ionophore A23187 did not increase glucose uptake, indicating that intracellular Ca2+ levels do not influence glucose uptake in brown adipocytes.
|
|
1- and ß1-adrenergic receptors activate glucose uptake through PKCs1- and ß1-adrenergic receptor-activated glucose uptake in brown adipocytes. We used Gö 6976, which inhibits conventional (, ß1) PKC isoforms (36), Ro-31–8220, which inhibits conventional (, ß1, ß2, 
) and novel (
) but not atypical PKC isoforms (52), and Gö 6983, which inhibits conventional (, ß, ), novel (
), and atypical PKC (
) isoforms (38).
Interestingly, the 1- and ß1-adrenergic receptor (norepinephrine)-induced increase in glucose uptake was largely inhibited by blocking conventional PKCs [Gö 6976, 92% inhibition (Fig. 11)], conventional and novel PKCs [Ro-31–8220, 49% inhibition (Fig. 12A)], and conventional, novel, and atypical PKCs [Gö 6983, 64% inhibition, data not shown (n = 3 in duplicate)]. Isoprenaline-induced glucose uptake was also largely inhibited by Gö 6976 (87% inhibition), Ro-31–8220 (79% inhibition), and Gö 6983 [80% inhibition, data not shown (n = 3 in duplicate)]. 1-Adrenergic receptor (cirazoline) activation was blocked by Ro-31–8220 (50% inhibition) (Fig. 12B). TPA-activated glucose uptake in ß3-KO was largely inhibited by Ro-31–8220 (88% inhibition) (Fig. 12B). The insulin-mediated increase in glucose uptake was inhibited by Gö 6976 (39% inhibition) (Fig. 11), Ro-31–8220 (60% inhibition) (Fig. 12B), and Gö 6983 [41% inhibition (n = 3 in duplicate)].
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been stated by ourselves and others that ß3-adrenergic receptors mediate most or all of the norepinephrine-induced glucose uptake in brown primary adipocytes (14, 18, 48), even though the presence of ß1-adrenergic receptors have been verified by radioligand binding and molecular studies (ß2-adrenergic receptors are hard to detect and probably play a very minor/nonexistent role) (11, 62). Furthermore, brown adipocytes also express coupled 1- and 2-adrenergic receptors (see review in Ref. 30).
Significant uncertainty exists regarding the physiological importance and the relative role of ß3- vs. ß1-/ß2-adrenergic receptors in mediating signal transduction in BAT. To investigate the importance of ß3-adrenergic receptors and the possibility that ß3-adrenergic receptors could have a distinct coupling to glucose uptake in brown adipocytes, we performed experiments in primary cultures of brown adipocytes from ß3-KO mice. These mice demonstrate a modest increase in body fat, which is not associated with an increase in food intake but rather a decrease in energy expenditure (19, 20). Several explanations for the relative lack of obesity can be discussed, such as a relative marginal role of ß3-adrenergic receptors mediating norepinephrine-induced responses or that mechanisms develop that compensate for the lack of ß3-adrenergic receptors (such as up-regulation of other adrenergic receptors). The two existing models of ß3-KO mice show opposite results on ß1-adrenergic receptor gene expression, one showing an up-regulation of ß1-adrenergic receptor mRNA in BAT (19) and the other a decrease (20). We here used an in vitro model system, primary cultures of brown adipocytes, which have previously been established as a good model system for mature brown adipocytes (11, 26, 27, 28) in which whole-body regulatory and compensatory mechanisms, i.e. sympathetic feedback loop in the intact mouse, would not be present.
It is beneficial to explore which adrenergic receptors influence glucose uptake in our system because BAT includes more cell types besides mature brown adipocytes, including brown preadipocytes, fibroblasts, endothelial cells, macrophages, and nerves, most of which express adrenergic receptors at various degrees. Indeed, both the brown adipocytes and the blood vessels of the brown adipose tissue are intensely innervated from the sympathetic nervous system. Vasoconstriction in the tissue is induced via 1-adrenergic receptors and vasodilation via ß-adrenergic receptors (63). It is established in several blood vessel systems that ß2-adrenergic receptor couple to vasodilation (64), and this is probably also the case in BAT. Our results confirm that ß1-adrenergic receptor mRNA levels are up-regulated in BAT from ß3-KO mice as shown before (19). This up-regulation did not occur in primary ß3-KO brown adipocytes, indicating that the ß3-adrenergic receptor gene disruption does not affect ß1-adrenergic receptor gene expression (or protein levels) in itself and that such a compensatory mechanism is present in vivo but not in vitro. Furthermore, there was no up-regulation of 1-adrenergic receptor gene expression, reinforcing that ß3-adrenergic receptor gene disruption does not effect basal adrenergic gene expression in vitro.
Induction of UCP1 gene expression
ß-Adrenergic agonists induce UCP1 gene expression in mature brown adipocytes (3). UCP1 is heavily induced in wild-type and, rather surprisingly, ß3-KO cells using a nonselective ß-adrenergic receptor agonist, indicating that ß3-adrenergic receptor ablation does not affect the cells ability to differentiate into mature brown adipocytes. Because ß2-adrenergic receptors are not present in these cells, this indicates that ß1-adrenergic receptors are inherently as effective as ß3-adrenergic receptors in increasing UCP1 gene expression. This is interesting because our earlier results show that even though ß1-adrenergic receptors are expressed, they are not coupled to any significant extent in wild-type mature brown adipocytes (29).
BRL-37344 and CL-316 243 do not increase glucose uptake in ß3-KO brown adipocytes
Norepinephrine and ß3-adrenergic receptor agonists increase glucose uptake in wild-type brown adipocytes (14, 18, 48). BRL-37344, a ß3-adrenergic receptor agonist (49) that stimulates glucose uptake through ß2-adrenergic receptors in L6 muscle cells (65), had no stimulatory effect on glucose uptake in ß3-KO brown adipocytes, indicating that BRL-37344 stimulates glucose uptake through ß3- and not ß1-/ß2-adrenergic receptors. Similarly, there was no induction of glucose uptake in ß3-KO brown adipocytes with another ß3-adrenergic receptor agonist (CL-316 243; Refs. 66 and 67).
No difference between norepinephrine-stimulated glucose uptake in wild-type and ß3-KO brown adipocytes
The effect of norepinephrine was similar in both wild-type and ß3-KO cells, with no significant difference in pEC50 or Vmax. This result was surprising because we have previously shown that norepinephrine induces glucose uptake in wild-type brown adipocytes through ß3-adrenergic receptors, with no evident involvement of other adrenergic receptors (18). Because there is no compensatory up-regulation of ß1-adrenergic receptors at the level of mRNA or protein, the results indicate efficient coupling of ß1-adrenergic receptors or/and an involvement of -adrenergic receptors in activation of glucose uptake in ß3-KO cells.
Both 1- and ß1-adrenergic receptors activate glucose uptake in ß3-KO brown adipocytes
Using several ß-adrenergic receptor antagonists we show that a large portion of the norepinephrine-induced glucose uptake in ß3-KO cells is mediated through ß1-adrenergic receptors. This was unexpected because our earlier studies have shown that ß1-adrenergic receptors, even though they are expressed, are coupled only to cAMP formation in brown preadipocytes and not in mature wild-type brown adipocytes in culture (29). At this point it could therefore be postulated that either ß1-adrenergic receptors couple to cAMP and increase glucose uptake in mature ß3-KO cells or that ß1-adrenergic receptors couple to another signaling pathway to increase glucose uptake in ß3-KO cells.
The ß-adrenergic receptor agonist (isoprenaline) could not evoke the same Vmax as norepinephrine, indicating a possible -adrenergic receptor component. Because the 1-adrenergic receptor agonist cirazoline increased glucose uptake and 1A- and 1D-adrenergic receptor mRNA is present, we conclude that 1-adrenergic receptors are involved in increasing glucose uptake in ß3-KO brown adipocytes. The relative role of the 1-subtypes in contributing to increased glucose uptake has to be examined in a more detailed investigation.
In our previous study, there were indications for a small 1- but not 2-adrenergic receptor component in norepinephrine-induced glucose uptake in wild-type cells (18). We similarly found no evidence for a 2-adrenergic receptor component increasing glucose uptake in ß3-KO brown adipocytes. The 1-adrenergic receptor component was much larger in ß3-KO cells than the very small effect in control cells. These ß1- and 1-adrenergic receptor pathways were additional (not synergistic), making it plausible that all of the norepinephrine effect on glucose uptake is through 1- and ß1-adrenergic receptors in ß3-KO cells. This indicates a redundancy in the adrenergic receptor system in stimulating glucose uptake. ß1-Adrenergic receptors have the possibility to carry out some or most of the ß3-adrenergic receptor function, but to reach maximum effect, both 1- and ß1-adrenergic receptors have to be activated in ß3-KO cells.
ß-Adrenergic receptor-induced glucose uptake is mediated by cAMP
ß3-Adrenergic receptors induce glucose uptake through cAMP in mature brown adipocytes (18). It is not fully established which adenylyl cyclase isoform(s) mediate the ß-adrenergic receptor signal in BAT, but it is possible to inhibit all known mammalian isoforms with DDA (50). By blocking adenylyl cyclases with DDA, we conclude that cAMP formation is very important and necessary in ß1-adrenergic receptor-activated glucose uptake in mature ß3-KO brown adipocytes.
PI3K is necessary for 1- and ß1-adrenergic receptor glucose uptake
Because PI3K is crucial in ß3-adrenergic receptor-induced glucose uptake in mature wild-type brown adipocytes (18), we examined whether PI3K was involved in norepinephrine-induced glucose transport in ß3-KO brown adipocytes, using the established PI3K inhibitors wortmannin and LY294002, which are chemically distinct from each other. ß1-Adrenergic receptor-mediated increases in glucose uptake were almost fully inhibited by these specific PI3K inhibitors (glucose uptake mediated by 1-adrenergic receptors was also blocked with PI3K inhibition). TPA-induced glucose uptake was partly blocked with PI3K inhibition. This would indicate that some of the TPA activated PKCs would be upstream of PI3K or that phorbol esters are capable of activating PI3K. Although phorbol esters are capable of activating PI3K in fat and muscle cells (68, 69, 70), glucose uptake induced by TPA is still mediated primarily through conventional and novel PKCs (as opposed to PI3K) in white adipocytes (69), making it plausible that this would also be the case in brown adipocytes.
Both 1- and ß1-adrenergic receptors activate glucose uptake through conventional and novel PKCs
We hypothesized that it is likely that after activation of PI3K, the adrenergic receptor pathway uses the same kinases as the insulin pathway with respect to glucose uptake. Both insulin and TPA stimulate conventional and novel PKCs (71). Atypical PKC isoforms are involved in insulin-stimulated glucose uptake in adipocytes but are not stimulated by TPA (69, 72, 73, 74), whereas conventional and novel PKCs do not appear to be required for insulin-stimulated glucose transport in white adipocytes (72, 74). To delineate whether PKCs are involved in ß1-adrenergic receptor activation of glucose uptake we used three different PKC inhibitors. All three PKC inhibitors inhibited isoprenaline-induced glucose uptake to a large degree. Interestingly the blocking of conventional PKCs lead to a large inhibition. This inhibition was not larger when blocking conventional and novel or conventional, novel, and atypical PKCs, making it likely that conventional PKCs are very important in ß1-adrenergic receptor activation of glucose uptake. In brown adipocytes 1-adrenergic receptor stimulation is likely mediated via an increase in intracellular Ca2+ (10, 75, 76) and/or via PKC (77). Because the Ca2+ ionophore A23187 did not increase glucose uptake, we examined whether PKC stimulation by TPA would mimic 1-adrenergic receptor stimulation. TPA was a potent activator of glucose uptake in brown adipocytes, which could be blocked with Ro-31–8220, a potent inhibitor of conventional and novel PKCs [inhibits PKC- only at relatively high concentrations (72)]. These results confirm that TPA activates glucose uptake mainly through conventional and novel PKCs in brown adipocytes. This makes it plausible that 1-adrenergic receptor stimulation activates the same pathway. On the other hand, 1-adrenergic receptor stimulation with cirazoline was inhibited to a lesser degree by Ro-31–8220 and not statistically significant, making it hard to reject that atypical PKCs also are involved in 1-adrenergic receptor activation of glucose uptake in brown adipocytes as they are in muscle cells (70).
The insulin response was partially inhibited by blocking conventional PKCs with Gö 6976 and conventional and novel PKCs with Ro-31–8220, indicating that conventional/novel PKCs could to some extent be involved in insulin stimulation of glucose uptake in brown adipocytes. However, the Ro-31–8220 concentration used (5 µM) slightly inhibits insulin-induced glucose uptake in rat white adipocytes through inhibition of PKC-, indicating that this could be the case also in brown adipocytes (72).
Further studies would be needed to delineate whether these conventional and novel PKCs can be activated directly by ß-adrenergic receptors. However, it is generally accepted that PKCs are downstream effectors of PI3K (68, 72, 73, 74). It is therefore likely that these conventional and novel PKCs are downstream of PI3K and not directly activated by elevated cAMP levels.
A hypothesis for the mechanism of the receptor switch
The ß3-adrenergic receptor is the main adrenergic receptor involved in norepinephrine-stimulated glucose uptake in intact brown adipocytes. In intact mice, the ablation of the ß3-adrenergic receptor leads to compensatory mechanisms and modulation of the adrenergic receptor signaling system. We show here that brown adipocytes in primary culture are a good model system to isolate the inherent effects of ß3-adrenergic receptor ablation because these cells are devoid of such compensatory mechanisms. Surprisingly, ß3-adrenergic receptor ablation did not influence norepinephrine-stimulated glucose uptake. ß1-Adrenergic receptor signaling can partly compensate the signal to glucose uptake through the same signaling pathway as ß3-adrenergic receptors via the second messenger cAMP and PI3K (18). This activation leads to stimulation of PKCs and subsequently activation of glucose transport. In the ß3-KO cells, 1-adrenergic receptors became able to increase glucose uptake through a PI3K/PKC pathw
) or wild-type cells (
). Vehicle-treated wells were set to 100% for each cell culture series and the other values given relative to this. The results were fitted with the general curve-fitting procedure for Michaelis-Menten kinetic.
; wild-type cells, 
; stippled line, data from cell cultures examined in direct parallel (in Ref. 
; in wild-type cells (n = 6 series in duplicates, 
). Vehicle-treated wells were set to 100% for each cell culture series and the other values given relative to this. The results were fitted with the general curve-fitting procedure for Michaelis-Menten kinetics.
) in ß3-KO cells. B, Norepinephrine (n = 2 series in quadruplicate, 
) in ß3-KO cells. Vehicle-treated wells were set to 100% for each cell culture series and the other values given relative to this. The results were fitted with the general curve-fitting procedure for Michaelis-Menten kinetics.
) in ß3-KO cells. C, Cirazoline (n = 3 series in duplicate, 
) in ß3-KO cells. D, Isoprenaline (ISO, 10 µM), cirazoline (CIR, 10 µM), and norepinephrine (NE, 10 µM) (n = 3 series in quadruplicate). *, P 
