| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, S-106 91 Stockholm, Sweden
Address all correspondence and requests for reprints to: Jan Nedergaard, The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, SE-106 91 Stockholm, Sweden. E-mail: jan{at}metabol.su.se
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The development of brown adipocytes may be studied in primary culture initiated with precursors freshly isolated from brown adipose tissue (2, 3, 4, 5). In the fibroblast-like brown preadipocytes that have only been 3–4 days in culture, norepinephrine stimulates cell proliferation (6, 7). In these cells, an increase in cellular cAMP level is also able to promote proliferation (6). Concordantly, the adrenergic effect is mediated via ß-receptors, which are of the ß1-subtype (6).
After about 5 days 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 (2, 3, 4). In these cells, norepinephrine very competently induces the expression of the most specific differentiation marker: the tissue-specific uncoupling protein (UCP1) (5). The response to norepinephrine in this case displays characteristic semibell-shaped kinetics (5, 8). In these cells, an increase in cellular cAMP level is also able to induce this gene expression (5, 8). In agreement with this, the adrenergic effect is mediated via ß-receptors, but of the ß3-subtype.
Thus, during their development, authentic brown adipocytes demonstrate a transition from one stage where ß1-stimulation induces proliferation to a stage where ß3-stimulation enhances cellular differentiation (functionally defined as an induction of UCP1 expression).
This originally clear picture of distinct ß-subtypes controlling distinct responses in authentic brown adipocytes in primary culture has, however, been challenged experimentally by later observations in immortalized brown adipocyte-like cell lines. In such cell lines, UCP1 gene expression was reported to be promiscuously controlled: both ß3- and ß1-stimulation (and even ß2) could induce UCP1 expression (9, 10, 11, 12, 13, 14). Further, the physiological relevance of ß3-mediated responses for the induction of UCP1 expression has been questioned, because it has been surmised that ß3-mediated processes display an inherently lower affinity for norepinephrine than do ß1-mediated processes (15). Thus, even if both ß1- and ß3-receptors should be capable of inducing UCP1 expression, in the physiological situation the expression would be stimulated predominantly through the ß1-receptor (14, 15, 16).
For this reason alone, a clear delineation of the receptor-signal coupling process in brown adipocytes during development may be considered a prerequisite for further development in this area. Such a delineation may also be of significance for the development of our ability to influence brown adipose tissue function and capacity in the intact organism (including possibly man).
Further, as two very distinct cellular processes, proliferation and enhanced differentiation, in the primary cell cultures have been demonstrated to be mediated through two different ß-receptor subtypes, a basic question concerning signal mediation may be formulated. Principally, two possibilities may be considered. One is that in addition to their effects on cAMP levels, the two ß-receptor subtypes may convey additional independent and distinct intracellular information to the cell interior that enables the cell to respond differentially—by proliferation or differentiation—to signals emanating from either of the two subtypes, even though both of them would be expected to increase cAMP levels. Such pathways are presently discussed (17, 18). Alternatively, it would be solely through their effect on cAMP levels that the distinctive effect of the two ß-subtypes is mediated. In that case, during cellular development, a switch in the ß-receptor subtype that (dominantly) controls cAMP levels, from ß1 to ß3, must occur in parallel with a switch in the innate response of the cell to an increase in cAMP levels, from proliferation stimulation to differentiation stimulation.
To distinguish between these two possibilities and to establish the nature of the receptor controlling UCP1 expression in authentic brown adipocytes, the subtype of ß-adrenergic receptor controlling cAMP levels was investigated during the development of these cells in culture. We found that during development a switch occurred, such that cAMP levels were ß1-controlled in brown preadipocytes but ß3-controlled in mature brown adipocytes; thus, it was through their regulation of intracellular cAMP levels that the ß-receptor subtypes conveyed their subtype specificity to the cellular responses. The underlying switch in coupled ß-receptor subtype was due to a spontaneously induced expression of ß3-receptors during cell differentiation. In the mature authentic brown adipocytes, there was no detectable contribution to the functional response from the ß1-receptors still existing on the cells, but the ß3-receptors had a functional norepinephrine affinity higher than that of the ß1-receptors that were functionally active in the brown preadipocytes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture
Cells were cultivated in 6-well or 12-well plates (growth area
9.4 cm2 and 3.83 cm2/well, respectively). An
inoculation density of 80–90 x 103
cells/cm2 was used in both types of wells. The cells were
cultivated in 2 ml (6-well) or 1 ml (12-well) of a culture medium
consisting of DMEM [Flow 1x liquid without glutamine; 4
mM glutamine (Flow) added], supplemented with 10%
newborn-calf serum (Flow), 4 nM insulin (Actrapid Human,
Novo), 10 mM HEPES (Flow), and with 50 IU penicillin, 50
µg streptomycin and 25 µg sodium ascorbate (Sigma Chemical Co.) (19) per ml, at 37 C in a water-saturated atmosphere of 8%
CO2 in air in a Heraeus
CO2-auto-zero B5061 incubator. On the next day (day 1), the
cultures were first washed with 2 ml prewarmed DMEM and fresh prewarmed
medium was then added. If the cultures were analyzed on day 3, no
further medium changes were made (Savant Instruments, Inc., Hicksville,
NY); otherwise the medium was fully exchanged with fresh prewarmed
medium on day 3 (without wash), and similarly on days 6 and 9, where
relevant.
cAMP determinations
On the indicated day of culture, 5–10 µl of agonist or
antagonist stock solution in DMEM were added into each well and the
cultures replaced in the incubator. After the indicated times, the
culture medium was aspirated, 0.8 ml 95% ethanol was added to each
well, and the cells were scraped off. The wells were then washed with
0.5 ml 70% ethanol, and the combined suspensions were dried in a
Speedvac centrifuge. The dried samples were dissolved in
150–500 µl of the Buffer 1 provided with the Cyclic AMP
[3H] Assay System from Amersham Pharmacia Biotech (Solna, Sweden) and centrifuged at 14000 rpm for 10 min.
Two 50 µl aliquots of the supernatants for every sample were analyzed
according to the description in the assay system, and for every
concentration of any agonist in each experiment, duplicate wells were
used; thus, each value initially obtained was the average of four
measurements of cAMP.
In each series of experiments, carried out with the same culture on the same day, the cAMP value was either stated as the mean ± SE based on the number of series, or the level observed with 1 µM (day 3 and 4) or 0.1 µM (day 6 and 8) norepinephrine was set in each series to 100% and mean values ± SE for these normalized values were then calculated.
Determination of messenger RNA (mRNA) levels for ß1-
and ß3-adrenoceptors and for UCP1
On the indicated day of culture, the medium was discarded (for
UCP1 mRNA determination 4 h after the addition of norepinephrine),
the cells were dissolved in 1 ml of an Ultraspec (Biotecx, Houston, TX)
solution, and the manufacturer’s procedure for RNA isolation was
followed. The final pellet was suspended in 75 µl 10 mM
EDTA and the RNA extracted at 70 C for 5 min and thereafter vortexed.
The RNA concentration was measured and absence of protein contamination
was checked on a Beckman Coulter, Inc. DU 50
spectrophotometer with readings at 260 nm and 280 nm. The ratio of
260/280 nm was routinely higher than 1.7.
The RNA solution was lyophilized in a SpeedVac. The RNA was then dissolved in 18 µl of RNA cocktail consisting of 50% (vol/vol) formamide, 5 mM MOPS and 9% (vol/vol) formaldehyde, and 2 µl of loading buffer consisting of 50% (wt/vol) glycerol and 0.1 mg/ml bromophenol blue. The solution was incubated for 8 min at 70 C and then chilled on ice. The samples were loaded on a gel (1.25% agarose, 10 mM MOPS, 6.2% (vol/vol) formaldehyde and 20 µl 1 mg/ml ethidium bromide). The gel was run in 20 mM MOPS-buffer for 2–3 h at 4–5 V/cm. After electrophoresis, it was verified under UV-light from the intensity of the 18S-28S rRNA bands that all samples were equally loaded and that no degradation was observable.
The RNA was blotted overnight from the gel to a Hybond-N membrane
(Amersham Pharmacia Biotech) in 20x SSC. Three sheets of
Whatman 3MM (Kebo Lab., Sp
nga, Sweden)
soaked in 20 x SSC were placed on top of the Hybond-N membrane.
The gel and the Hybond-N membrane were examined under UV light. The RNA
was cross-linked to the Hybond-N membrane [UV Stratalinker 1800
(Stratagene, La Jolla, CA)] with the auto cross-link
program). The Hybond-N membrane was prehybridized with 10 ml of a
solution containing 5 x SSC, 5 x Denhardt’s solution,
0.5% SDS, 50 mM sodium phosphate, 50% formamide and 100
µg/ml of degraded DNA from herring sperm (Sigma Chemical Co.) in a hybridization oven (Hybaid, Middlesex, UK) at 45 C for
2 h. After this prehybridization, the Hybond-N membrane was
transferred to a similar solution containing the denatured probe (see
below) at a final concentration of 1–3 · 4 106 cpm per
ml. The hybridization was carried out for at least 16 h at 45 C.
The Hybond-N membrane was then washed twice in 2 x SSC, 0.2% SDS
at 30 C for 20–30 min each and then twice in 0.1 x SSC, 0.2%
SDS at 50 C for 45 min. The membrane was sealed in a plastic envelope
and exposed to a PhosphorImager screen. The screens were analyzed on a
Molecular Dynamics, Inc. PhosphorImager with the
ImageQuant program. When the same membrane was analyzed for several
mRNA species, the previous probe was removed by boiling it in 0.1% SDS
solution.
The rat ß1-complementary DNA was that previously characterized by Revelli et al. (20). It was cloned in the EcoRI site of the PVZ1 plasmid (size about 2.7 kb). The 1.5-kb fragment obtained by Har 1 digestion was used for the hybridizations. The mouse ß3-probe originated from the A43 probe earlier characterized (21). A fragment of the mouse ß3-adrenoceptor gene was subcloned in pUC18 at the XbaI site. This genomic DNA fragment has a length of 300 bp and corresponds to the 5'-coding region of the ß3-adrenoceptor from ATG to the second transmembranal loop (TM2). To generate the ß3-adrenoceptor probe used here, the plasmid was cut with the restriction enzymes BamHI and SalI to a length of 0.5 kb. The UCP1 probe was that earlier used (5). The probes were labeled with a DNA labeling kit (Roche Molecular Biochemicals) to an activity of 7–10 cpm/µg DNA (10.000–60.000 cpm/µl stock solution).
Cell membrane preparations and equilibrium binding studies
On culture day 7, cells were first rinsed with 1 ml of
incubation buffer (50 mM Tris-HCl, 10 mM
MgCl2, 0.5 mM ascorbic acid, 1 mM
EDTA, 0.1 mM PMSF, pH 7.4) also containing 0.25
M sucrose and then scraped into incubation buffer (without
sucrose). After centrifugation (2 min, 14000 rpm,
Eppendorf centrifuge), cells were homogenized in a
Potter-Elvehjem homogenizer with a Teflon pestle in the incubation
buffer without sucrose. The homogenate was filtered through one layer
of silk cloth and centrifuged for 30 min at 100,000 x
g in a Beckman Coulter, Inc. high-speed
ultracentrifuge at +4 C. The pellet was rehomogenized in
the same incubation buffer at a protein concentration of about 1 mg/ml
and filtered again through silk cloth. Protein was determined by the
method of Bradford with fatty-acid-free BSA (fraction V, Roche Molecular Biochemicals) as standard. Samples, containing
approximately 150 µg protein in a total volume of 0.32 ml assay
buffer (50 mM Tris-HCl, 10 mM
MgCl2, 0.5 mM ascorbic acid, 1 mM
EDTA, 0.1 mM PMSF, pH 7.4) were incubated for 60 min at 30
C with seven concentrations of (-)-[3H] CGP-12177
(0.04–2.4 nM; 52 Ci/mmol, Amersham Pharmacia Biotech). The incubation was stopped with 2 ml ice-cold
incubation buffer diluted 10-fold and the samples were filtered through
a Whatman GF/C filter on a semiautomatic Skatron cell
harvester 7019 (Skatron, Lier, Norway) and washed with 8 ml/well of the
incubation buffer (diluted 1/10). The radioactivity was determined in 5
ml scintillation mixture [Emulsifier Scintillator Plus (Packard Instrument Co.)] in a Beckman Coulter, Inc.
scintillation counter. The total binding (Btot) was
measured as above. The nonspecific binding (Bns) was
determined by parallel incubations with 1 µM
(-)-alprenolol (Ciba-Geigy, Basel, Switzerland). The
specific binding (Bs) was estimated as the difference
between Btot and Bns.
Analysis of dose-response curves
For analysis of dose-response curves, the curve-fitting option
of the KaleidaGraph 3.0 application was used. Monophasic dose-response
data were analyzed with the rearranged Michaelis-Menten equation
VA = basal +
Vmax/(1+(EC50/[A])h), where h is
the Hill coefficient. If h was estimated to be close to 1 in the
initial analysis, the data were recalculated with h = 1.
For the analysis of biphasic ("semibell-shaped") dose-response data (observed for norepinephrine, isoprenaline and dobutamine in mature brown adipocytes), a model (22) for the interaction of a ligand with two different receptors, one stimulatory (S) and one inhibitory (I), was used: VA = basal + Vmax(S)/(1+(EC50/[A])h) + Vmax(I)/(1+(IC50/[A])h). IC50 here denotes the EC50 of the inhibitory component. In some calculations, basal was set as a constant to avoid a singular matrix that would make the fitting unsolvable.
Chemicals
Stock solutions (10 mM) of agents used (obtained
from Sigma Chemical Co. if not otherwise indicated) were
made in 0.05% ascorbic acid (NE, isoprenaline) or in DMEM and stored
at -80 C. The following agents were used: norepinephrine
((-)-Arterenol bitartrate), BRL-37344 (gift from SmithKline Beecham Pharmaceuticals, Brentford, UK), isoprenaline
((-)-isoproterenol (+)-bitartrate), CGP-12177 (CGP-12177A, from
Ciba-Geigy), D,L-propranolol, yohimbine,
dobutamine (Dobutrex from Lilly France S.A., Fegersheim,
France).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, dramatically
different profiles for adrenergic regulation of cAMP levels were
observed during brown adipocyte development. The responses in the brown
preadipocytes were simple to interpret (Fig. 1A). The dose-response
curve for norepinephrine was monophasic and fulfilled simple
Michaelis-Menten kinetics; the EC50 was 450 nM
(Table 1). There was hardly any response
to the ß3-selective agonist BRL-37344. Similarly, the
ß3-specific agonist CGP-12177—which is an agonist only
on ß3-receptors (23) and does not activate
ß1-receptors [it is virtually a full antagonist on this
receptor (24, 25)]—was fully devoid of effect. This response pattern
is in itself indicative of a dominating ß1-response in
the brown preadipocyte stage (but see also below).
|
|
). Norepinephrine, which in the proliferating
brown preadipocytes was the best of the tested agonists, in the mature
brown adipocytes gave the lowest response of the three. Further, the
response to norepinephrine did not demonstrate simple Michaelis-Menten
kinetics; instead it was now biphasic, with a clear maximal response at
0.1 µM; higher concentrations yielded markedly lower cAMP
levels. Notably, the functional affinity had increased: the
EC50 had decreased from 450 nM to only 35
nM. In contrast to what was the case in the brown
preadipocytes, the two ß3-agonists tested were now both
highly potent, maximally inducing nearly double the cAMP level induced
by optimal norepinephrine concentrations, and each demonstrating a
monophasic response. There are at least three implications of these dramatically altered characteristics of the response of the brown adipocytes. Firstly, the contrasting responses between the brown preadipocytes and the mature brown adipocytes make it clear that, as a consequence of the spontaneous differentiation of the brown adipocytes, a sudden switch had occurred in the complement of coupled ß-receptors. In that there was a response in the mature brown adipocytes to the ß3-specific agonist CGP-12177 (and to BRL-37344), it is evident that in these cells ß3-receptors existed that were coupled to cAMP production. (Whether it is solely (or at all) through these ß3-receptors that norepinephrine functions in these cells is not in itself demonstrated by this response pattern but will be investigated below.) The existence of a switch in response is principally in agreement with implications from earlier, less detailed single-dose investigations (6, 26).
Secondly, the fact that the maximal cAMP level achieved by the ß3-agonists in the mature brown adipocytes had become higher than that reached by norepinephrine, implies that in the mature brown adipocytes, norepinephrine stimulated inhibitory receptors not stimulated by BRL-37344 or CGP-12177 (see below).
Thirdly, the biphasic nature of the response to norepinephrine raises questions concerning the mechanisms and possible cellular consequences of this unusual receptor response. An analysis of this phenomenon will be presented elsewhere (Bronnikov, G., S.-J. Zhang, B. Cannon, and J. Nedergaard, submitted).
It could be suggested that the apparent switch in ß-receptor subtype
could in some way be due to different cAMP accumulation kinetics in the
different cell stages. We therefore followed the time course of the
response to norepinephrine and CGP-12177 (Fig. 2). In the brown preadipocytes, no
detectable response to CGP-12177 was observable at any time point after
stimulation (Fig. 2A), whereas the response to norepinephrine peaked
early and then slowly declined; it remained, however, well above
control levels even after several hours. In the mature brown
adipocytes, the response to CGP-12177 at every time point was higher
than (or similar to) the response to norepinephrine (Fig. 2B). Thus,
altered kinetics could not explain the apparent switch in functional
ß-receptor endowment. However, a differentiation-related alteration
in the kinetics was observable, in that during continued norepinephrine
stimulation, the cAMP level in the brown preadipocytes peaked earlier
than in the mature brown adipocytes (Fig. 2C).
|
We therefore analyzed the mRNA levels of the ß1- and the
ß3-receptors during the spontaneous differentiation
process in these cells (ß2-mRNA was not detectable in
unstimulated cells and was only transiently expressed 15–30 min after
norepinephrine stimulation [Bengtsson, T., B. Cannon, and J.
Nedergaard, submitted)]. As seen in Fig. 3A, a single band hybridizing with the
probe corresponding to ß1-mRNA was detectable in samples
from all stages investigated. The size of the
ß1-transcript was 2.6 kb, in agreement with what has been
observed in a variety of tissues, including brown adipose tissue (14, 27, 28, 29). In Fig. 3B, results from five cell culture series analyzed as
in Fig. 3A are presented. As seen, the level of ß1-mRNA
was stable through the switch period around day 4 and only decreased
after day 7, in the period during which the cells tend in general to
dedifferentiate [if they are not chronically norepinephrine-treated
(30)].
|
). The size
of 2.4 kb for the major transcript is in good agreement with previous
reports for ß3-receptors (21, 31, 32). Only in a few
blots could we detect small amounts of ß3-adrenoceptor
transcripts with sizes larger than the band of 2.4 kb (not shown), but
the occurrence of such bands seemed random. Also in other reports on
ß3-expression in mouse, the occurrence of larger
transcripts was inconsistently reported, with some groups reporting
only one transcript size in mouse tissue (21, 33) and other groups
several transcripts (14, 29), also in mouse-derived 3T3-F442A cells
(21, 34). It is therefore unclear whether these larger transcripts are
authentic products of the transcription process in mouse brown adipose
tissue. In Fig. 3C, we have compiled data from three cell cultures
analyzed as that in Fig. 3A; only the band of 2.4 kb was used for this
compilation. The ß3-receptor mRNA was hardly detectable
in brown preadipocytes (day 3), but the level then abruptly rose,
reaching a steady-state value at day 5. In Fig. 3C, observations (5, 35) on the ability of norepinephrine to induce expression of UCP1 in
this culture system are also included. As seen, these two events were
very well coordinated in time.
However, although the switch in receptor subtype mediating cAMP
accumulation (Fig. 1) was thus parallelled by a sudden induction in the
expression of the ß3-receptor, it was not associated with
an abrupt decrease in ß1-receptor gene expression;
rather, the level of ß1-mRNA decreased only slowly in the
mature brown adipocytes. This means that the switch in the ß-receptor
mediating the functional response to adrenergic stimulation is not
explainable simply at the level of gene expression. Instead, the switch
could be due to a selective inhibition of the translation of the mRNA
coding for the ß1-receptor, i.e. a possibility
would be that even though the ß1-mRNA remained abundant,
ß1-receptors were no longer synthesized and the existing
ß1-receptors were rapidly turned over and would therefore
disappear. To examine this possibility, we investigated with
ligand-binding techniques whether ß1-receptors were still
found in the mature brown adipocytes. We used the ligand
[3H]CGP-12177 which, within the concentration range
studied (0.02–2 nM), is expected to bind only to
ß1/ß2-receptors [it binds to and activates
ß3-receptors at much higher concentrations (36)].
Saturable binding was observed (Fig. 4),
and as seen on the Scatchard plot in the insert, only a single binding
site could be detected. The affinity of [3H]CGP-12177 for
this binding site (
0.5 nM) was very similar to that
observed in membrane preparations from brown adipose tissue (37, 38),
where it has been demonstrated to represent ß1-receptor
sites (38). There is therefore no doubt that also the mature brown
adipocytes investigated here possessed a substantial amount of
ß1-receptors, but whether these receptors are capable of
mediating the norepinephrine signal (or any other adrenergic signal) is
not known (but see below).
|
) demonstrates that the
ß3-receptor as such is coupled in the mature brown
adipocytes, but it does not necessarily indicate that it is through
this receptor that the physiologically relevant response to
norepinephrine is mediated. We therefore analyzed through which
receptors the norepinephrine effect was mediated. To do this, we used
the established pharmacological criterion for
ß3-receptors: that the inhibitory potency of propranolol
is low (pA2 of 6–7) compared with its potency at
ß1/ß2-receptors (pA2 of 8–9)
(39, 40).
Again, in the brown preadipocytes, the response to norepinephrine
demonstrated simple kinetics (Fig. 5A),
and the presence of as little as 0.1 µM propranolol
shifted the dose-response curve more than a decade to the right. The
picture was clearly different in the mature brown adipocytes that
exhibited the biphasic response described above. In these cells, the
same dose of propranolol had almost no effect on the dose-response
curve (not shown), and even a 10-fold higher propranolol concentration
was only able to shift the curve less than a decade (Fig. 5B).
|
),
the points corresponding to the brown preadipocytes and to the mature
brown adipocytes distributed as two clearly distinct lines. From these
data, it was calculated that in the brown preadipocytes, propranolol
was a potent inhibitor with a pA2 of 8.3, but in the mature
brown adipocytes, the pA2 was only 6.9. Thus, the response
to norepinephrine was mediated via two different receptors in the two
cell types—ß1 in the brown preadipocytes and
ß3 in the mature brown adipocytes—and had switched as a
result of the spontaneous differentiation process. In both cases the
slope of the line was very close to 1, indicating that in each cell
differentiation stage, propranolol was competitive with only one single
class of receptors.
The switch in receptor subtype from ß1 to
ß3 was also associated with an increased functional
sensitivity to norepinephrine [the EC50 was decreased from
450 nM to 35 nM (Table 1)]. This observation
is in contrast to the idea that ß3-pathways should be
much less sensitive to norepinephrine than are
ß1-receptors, and that ß1-receptors should
thus be responsible for stimulation under conditions of mild to modest
physiological stimulation (15, 16, 41). Clearly, the norepinephrine
signal is not transmitted to any appreciable degree through
ß1-receptors in these mature brown adipocytes.
Are coupled ß1-receptors detectable in
mature brown adipocytes?
The apparent discrepancy between the continued expression of the
ß1-receptor gene even after the functional switch during
the spontaneous differentiation process—vs. the apparent
sudden disappearance of the functional ß1-response at the
time of the switch—made it important to attempt to observe indications
of coupled ß1-receptors in the mature brown adipocytes.
For this, we stimulated the cells with a generally accepted
ß1-selective agonist dobutamine to attempt to induce cAMP
formation through the ß1-receptors.
In the brown preadipocytes (Fig. 6A),
dobutamine increased the cAMP level—although not with a higher
affinity than norepinephrine and actually only to half the level. In
the mature brown adipocytes (Fig. 6B), dobutamine had a lower affinity
than norepinephrine, albeit higher than that of dobutamine in the
preadipocytes (Table 1), and the total response was higher than that to
norepinephrine. If the generally accepted view that dobutamine is a
selective ß1-agonist were uncritically applied here, the
enhanced response to dobutamine in the mature brown adipocytes would
seem to indicate that ß1-receptors were well-coupled in
these cells. However, a Schild plot analysis of the effect of
propranolol on these responses (Fig. 6C) pointed to another
interpretation. Again, the points from the brown preadipocytes and from
the mature brown adipocytes did not overlap. In the brown
preadipocytes, the pA2 for propranolol was 8.4 and
dobutamine therefore (as expected) induced cAMP accumulation via
ß1-receptors [although it was apparently only a partial
agonist in this system, as has been seen in other
ß1-systems (42)]. However, in the mature brown
adipocytes, the pA2 for propranolol was only 7.1 and
dobutamine therefore induced cAMP accumulation through
ß3-receptors (on which it was a full or nearly full
agonist). The pA2 value of 7.1 is in good agreement both
with that observed above for propranolol inhibition of
norepinephrine-induced cAMP accumulation (Fig. 4C) and with that
estimated for propranolol inhibition of adrenergically stimulated
thermogenesis in freshly isolated hamster brown fat cells (43), a
cellular response earlier demonstrated to be ß3-mediated
(44).
|
We conclude that we were unable to identify any cAMP response mediated
through the ß1-receptors that, according to Fig. 4, are
present in the mature brown adipocytes. Thus, although we cannot
eliminate the possibility that a fraction of the response [below
reasonable detection (<10%)] is mediated via these receptors, the
functional significance of the ß1-receptors in the mature
brown adipocytes must be considered negligible. Therefore, it is
through the ß3-receptors that the physiologically
significant response is mediated.
cAMP controlled processes
The above experiments demonstrate that during the differentiation
process, a switch in the subtype of ß-receptors coupled to the cAMP
increase occurs. Although, as mentioned, an increase in cAMP can also
induce the same processes (5, 6, 8), this does not, however,
necessarily mean that it is the ensuing increase in cAMP level that
governs them. A further criterion to be fulfilled is that stimulation
of the measured functional parameter (proliferation or differentiation)
must be a monophasic function of the corresponding cAMP levels.
In Fig. 7A we have therefore plotted the
norepinephrine-induced increase in DNA amount per cell culture flask
(i.e. cell proliferation) (6) as a function of the
corresponding norepinephrine-induced cAMP levels. As seen, the data
were compatible with cAMP being the mediating agent, with stimulation
of cell proliferation being dependent on cAMP concentration.
|
) is qualitatively similar to that earlier
reported for norepinephrine-induced UCP1 mRNA and protein levels. In
Fig. 7B, we have plotted the reported norepinephrine-induced UCP1 mRNA
levels as a function of the corresponding norepinephrine-induced cAMP
levels. Indeed, a remarkable correlation is found: despite the unusual
kinetics of the underlying observations, all points in the correlation
fall on a single line. Thus, the unusual kinetics for
norepinephrine-induced UCP1 gene expression can readily be explained by
the special kinetics for cAMP accumulation reported here.
There was, however, a principal difference between the cAMP response
curve for cell proliferation (Fig. 7A) and that for UCP gene expression
(Fig. 7B). Whereas cell proliferation as a function of cAMP level
demonstrated saturation, this was clearly not the case for UCP1 gene
expression. This should imply that the cells possess a capacity to
enhance UCP1 expression further than that induced by norepinephrine,
provided that the cAMP level is further increased. It is, however,
evident that it is not possible to reach a higher level of cAMP by
simply altering the norepinephrine concentration (cf. Fig. 1B).
Effect of inhibition of 
2-receptors
As pointed out above, the inherent ability—observed solely in
mature brown adipocytes—of ß3-agonists to stimulate cAMP
accumulation to a higher level than does norepinephrine (Fig. 1)
implies that an inhibitory component is activated by norepinephrine but
not by the ß3-agonists. A natural candidate for the
inhibitory receptor would be the 2-receptor. We
therefore investigated whether the presence of the
2-antagonist yohimbine would augment the cAMP response
to norepinephrine stimulation in these cells.
In the brown preadipocytes, the presence of yohimbine had only a
marginal effect on norepinephrine-induced cAMP accumulation (Fig. 8A). However, in the mature brown
adipocytes (Fig. 8B), the presence of yohimbine had a dramatic effect
and led to a doubling of the maximal cAMP level attained. The maximal
cAMP levels now reached were similar to those generated by the
ß3-agonists BRL-37344 or CGP-12177 (Fig. 1B), which are
not expected to interact with 2-receptors. Thus, the
lower maximal effect of norepinephrine than of the
ß3-agonists (Fig. 1) in mature brown adipocytes is
apparently due to the presence on these of coupled
2-receptors (such receptors are thus apparently not
functional on the brown preadipocytes). It is, however, equally clear
that it was only the maximal level that was increased by the presence
of yohimbine; the semibell-shape persisted and could therefore not be
ascribed to the action of 2-receptors (further analysis
of these unusual kinetics will be presented elsewhere (Bronnikov, G.,
S.-J. Zhang, B. Cannon, and J. Nedergaard, submitted).
|
, this was indeed the case. Furthermore, when the
level of UCP1 mRNA was now plotted as a function of the cAMP level, a
saturating curve shape was obtained (Fig. 8D). It may further be
noticed that the elimination of the 2-component also led
to a potentiation of the ability of cAMP to induce UCP1 gene expression
(Fig. 8D). Thus, norepinephrine-induced cAMP levels fulfill the criteria for being the intracellular factors governing both cell proliferation and UCP1 expression. When cells are stimulated by norepinephrine, the maximal cAMP levels attained are limiting for the level of UCP1 expression reached.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
.
|
In this respect, the results presented here with authentic brown
adipocytes, derived from precursors freshly isolated from brown adipose
tissue, deviate from those obtained with immortalized brown
adipocyte-like cell lines. In such cell lines, expression of
ß3-receptors, even in the apparently mature state, is low
or absent (and also 2-adrenergic coupling is absent); to
a significant or dominating degree, it is ß1-receptors
that mediate the response to adrenergic stimulation (9, 11, 12, 13, 14).
Therefore, although such transformed cell lines are undoubtedly very
helpful in exploring important features in cell biology, it is clear
from the data presented here that such cell lines do not constitute
adequate reflections of certain regulatory features of nontransformed
brown adipocytes. However, in one white-adipocyte-like cell line,
3T3-F442A, spontaneous expression of ß3-receptors during
the differentiation process has also been observed (46, 47).
In the brown adipocytes studied here, the induction of ß3-expression is an innate part of the differentiation program and does not require any additional exogenous stimulation. Thus, to allow for differentiation to proceed, it is not necessary to change from a proliferation medium to a differentiation medium, as routinely done in certain brown adipocyte-like cell lines (HIB 1B and HIB 1B/8 cells) (10, 11, 12), nor is transient treatment with differentiation inducers [as done in HIB cells (10)] necessary. In the brown adipocyte-like B-clones, a full transition through the differentiation program does not occur; as pointed out by the original authors, the clones seem in certain respects to have been caught in a premature state and cannot proceed through to the fully mature state (9, 13, 14) and they therefore do not demonstrate the dominating coupling of ß3-receptors to the adenylyl cyclase observed in the authentic brown adipocytes.
High functional affinity between ß3-receptors and
norepinephrine
As the classical ß1/ß2-receptors and
the ß3-receptors are often coexpressed, at least in the
same tissue if not on the same cells, a discussion has been ongoing
concerning the physiological significance of the
ß3-receptors. The ß3-receptors have been
reported to have a low affinity (measured as ligand binding
competition) for norepinephrine, and it has therefore repeatedly been
suggested that in a situation where both receptors are expressed and
functional on the same cell, low (meaning physiologically relevant)
norepinephrine concentrations would preferentially stimulate
ß1-receptors, whereas the ß3-receptors
would only be activated at high, perhaps supraphysiological,
norepinephrine levels (14, 15, 16). In the present experiments, we find no
support for this tenet; in contrast, the switch in coupled receptor
subtype from ß1 to ß3 was parallelled by a
more than 10-fold increase in functional affinity, from
approximately 450 to approximately 30 nM for
norepinephrine. This increase in functional affinity (EC50)
is probably explained by alterations in the density of receptors, G
proteins, adenylyl cyclases, etc. However, it may also be noted that in
a direct comparison, the apparent functional affinity (for stimulation
of adenylyl cyclase through ß-receptors ectopically expressed in CHO
cells) has been reported to be 0.8 nM for norepinephrine
stimulation of ß1-receptors, 6.3 nM for
stimulation of ß3-receptors, and 36 nM for
ß2-receptors (48). Thus, there is no particularly low
functional affinity of the ß3-receptor for
norepinephrine, even in the ectopic system.
Coupling in mature brown adipocytes is solely through
ß3-receptors
The observation here that in the mature brown adipocytes, cAMP
accumulation is under ß3-adrenergic control, with no
detectable ß1 effect, is in very good agreement both with
observations on cAMP accumulation in isolated, in
situ-differentiated, brown fat cells (43) and with observations on
the acute thermogenic response in such brown fat cells from hamsters,
mice, and rats (44, 49, 50). ß1-Receptors are still
present on the mature brown adipocytes studied here [as they also are
on mature brown fat cells directly isolated from brown adipose tissue
(51)], and they are apparently still in quite high abundance. From the
present experiments, we cannot entirely exclude that they are still to
some extent coupled to cAMP production, although their contribution is
undetectable in our experiments.
Selective ß3-agonists can only stimulate
mature brown adipocytes
The present results indicate that during its spontaneous
development, the brown adipocyte switches from a state of
"ß1-receptors coupled to cell proliferation" to a
state of "ß3-receptors coupled to enhanced
differentiation." An extrapolation of this scheme would be that it
should not be possible to induce brown adipose tissue hyperplasia by
ß3-selective stimulation of intact animals. Indeed,
treatment of rats with the selective ß3-agonist
CL-316,243 increased the total thermogenic capacity of brown adipose
tissue but failed to increase the DNA content (52), an observation
understandable based on the results of the present study. In those
experiments, an increase in the thermogenic capacity of the tissue
could be achieved by enhancing the capacity of existing adipocytes,
without enhanced cell proliferation. However, under conditions (such as
probably in adult man) in which few brown adipocytes are initially
present, the ability of a selective ß3-treatment to
significantly augment thermogenic capacity may be much more limited. If
the same switch that is described here in mouse brown adipocytes occurs
during the development of human brown adipocytes, chronic
ß3-stimulation in man may therefore not lead to the
expected recruitment of brown adipose tissue.
A pivotal switch in the innate differentiation program of brown
adipocytes
From the observations presented here, in combination with earlier
analyses of the brown adipocyte differentiation process (5, 6, 7, 8), it is
clear that a pivotal switch occurs during the development of the brown
adipocyte: before the switch, only ß1-receptors are
coupled, and stimulation of the cAMP pathway results in accelerated
cell proliferation; after the switch, ß3- (and
2-) receptors are coupled and stimulation of the cAMP
pathway now results in promoted cell differentiation (notably UCP1 gene
expression). Observations on the expression of G proteins (Bourova, L.,
J. Novotny, Z. Pesanova, T. Bengtsson, B. Cannon, J. Nedergaard, and P.
Svoboda, submitted) and of the transcription factor C/EBP (53) also
indicate a qualitative switch occurring at this differentiation point.
Thus, the appearance of the mature phenotype is not a gradual
phenomenon, with the successive appearance of an array of
differentiation markers. Rather, the data are indicative of a
coordinated process in which a master controller within a short time
alters the entire character of the cell, including the nuclear
interpretation of the cAMP signal. The nature of this (brown) adipocyte
determinative factor remains unknown, but the
ß3-gene is clearly one of its targets.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipients of grants from the Royal Swedish Academy of Sciences.
3 On leave from The Institute of Cell Biophysics, Russian Academy of
Sciences, 142 292 Pushchino, Russia.
Received February 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and ß-Adrenergic induction of the
expression of the uncoupling protein thermogenin in brown adipocytes
differentiated in culture. J Biol Chem 265:16464–164711- and
ß-adrenergic receptors in brown adipose tissue of lean (Fa/?) and
obese (fa/fa) Zucker rats. Effects of cold-acclimation, sucrose feeding
and adrenalectomy. Biochem J 249:831–838[Medline]
1-Adrenergic stimulation potentiates the thermogenic
action of ß3-adrenoceptor-generated cAMP in brown fat
cells. J Biol Chem 272:32847–32856 and C/EBPß
in brown adipose tissue. FEBS Lett 318:235–241[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  R. Kou and T. Michel Epinephrine Regulation of the Endothelial Nitric-oxide Synthase: ROLES OF RAC1 AND beta3-ADRENERGIC RECEPTORS IN ENDOTHELIAL NO SIGNALING J. Biol. Chem., November 9, 2007; 282(45): 32719 - 32729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. S. Dallner, E. Chernogubova, K. A. Brolinson, and T. Bengtsson {beta}3-Adrenergic Receptors Stimulate Glucose Uptake in Brown Adipocytes by Two Mechanisms Independently of Glucose Transporter 4 Translocation Endocrinology, December 1, 2006; 147(12): 5730 - 5739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Chernogubova, D. S. Hutchinson, J. Nedergaard, and T. Bengtsson {alpha}1- and {beta}1-Adrenoceptor Signaling Fully Compensates for {beta}3-Adrenoceptor Deficiency in Brown Adipocyte Norepinephrine-Stimulated Glucose Uptake Endocrinology, May 1, 2005; 146(5): 2271 - 2284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Martinez-deMena and M.-J. Obregon Insulin increases the adrenergic stimulation of 5' deiodinase activity and mRNA expression in rat brown adipocytes; role of MAPK and PI3K J. Mol. Endocrinol., February 1, 2005; 34(1): 139 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, Y. Cui, X. Tong, H. Ye, and S. Li Effects of the Trp64Arg Polymorphism in the {beta}3-Adrenergic Receptor Gene on Insulin Sensitivity in Small for Gestational Age Neonates J. Clin. Endocrinol. Metab., October 1, 2004; 89(10): 4981 - 4985. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Golozoubova, H. Gullberg, A. Matthias, B. Cannon, B. Vennstrom, and J. Nedergaard Depressed Thermogenesis but Competent Brown Adipose Tissue Recruitment in Mice Devoid of All Hormone-Binding Thyroid Hormone Receptors Mol. Endocrinol., February 1, 2004; 18(2): 384 - 401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. CANNON and J. NEDERGAARD Brown Adipose Tissue: Function and Physiological Significance Physiol Rev, January 1, 2004; 84(1): 277 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Chernogubova, B. Cannon, and T. Bengtsson Norepinephrine Increases Glucose Transport in Brown Adipocytes via {beta}3-Adrenoceptors through a cAMP, PKA, and PI3-Kinase-Dependent Pathway Stimulating Conventional and Novel PKCs Endocrinology, January 1, 2004; 145(1): 269 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Bachman, H. Dhillon, C.-Y. Zhang, S. Cinti, A. C. Bianco, B. K. Kobilka, and B. B. Lowell beta AR Signaling Required for Diet-Induced Thermogenesis and Obesity Resistance Science, August 2, 2002; 297(5582): 843 - 845. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Himms-Hagen, A. Melnyk, M. C. Zingaretti, E. Ceresi, G. Barbatelli, and S. Cinti Multilocular fat cells in WAT of CL-316243-treated rats derive directly from white adipocytes Am J Physiol Cell Physiol, September 1, 2000; 279(3): C670 - C681. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Fredriksson, J. M. Lindquist, G. E. Bronnikov, and J. Nedergaard Norepinephrine Induces Vascular Endothelial Growth Factor Gene Expression in Brown Adipocytes through a beta -Adrenoreceptor/cAMP/Protein Kinase A Pathway Involving Src but Independently of Erk1/2 J. Biol. Chem., April 28, 2000; 275(18): 13802 - 13811. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Bronnikov, S.-J. Zhang, B. Cannon, and J. Nedergaard A Dual Component Analysis Explains the Distinctive Kinetics of cAMP Accumulation in Brown Adipocytes J. Biol. Chem., December 31, 1999; 274(53): 37770 - 37780. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Lindquist, J. M. Fredriksson, S. Rehnmark, B. Cannon, and J. Nedergaard beta 3- and alpha 1-Adrenergic Erk1/2 Activation Is Src- but Not Gi-mediated in Brown Adipocytes J. Biol. Chem., July 21, 2000; 275(30): 22670 - 22677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. V. Leaver and P. A. Pappone beta -Adrenergic potentiation of endoplasmic reticulum Ca2+ release in brown fat cells Am J Physiol Cell Physiol, May 1, 2002; 282(5): C1016 - C1024. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, Total
binding of [3H]CGP-12177; [
, nonspecific binding in
presence of 1 µM alprenolol; 
, specific binding. Lines
were drawn for best fit to the equations basal
+M ·[L] + Bmax ·[L]/([L] +
KD) for total binding, to basal
+M ·[L] for nonspecific binding, and to
Bmax ·[L]/([L] + KD) for specific
binding. The insert (B) is a Scatchard plot of the specific
[3H]CGP-12177 binding; the line is drawn for best linear
fit. Data from one series of experiments are shown here; mean values
from two experimental series: Bmax = 16 ± 1
fmol/mg protein, KD = 0.53 ± 0.28
nM.
