Am J Physiol Endocrinol Metab 285: E1230–E1236, 2003.
First published September 3, 2003; 10.1152/ajpendo.00197.2003.
White adipose tissue contributes
to UCP1-independent thermogenesis
J. G. Granneman, M. Burnazi, Z. Zhu, and L. A. Schwamb
Center for Integrative Metabolic and Endocrine Research, Departments of Pathology
and Psychiatry, Wayne State University School of Medicine, Detroit, Michigan 48201
Submitted 1 May 2003; accepted in final form 25 August 2003
Granneman, J. G., M. Burnazi, Z. Zhu, and L. A.
Schwamb. White adipose tissue contributes to UCP1-inde-
pendent thermogenesis. Am J Physiol Endocrinol Metab 285:
E1230–E1236, 2003. First published September 3, 2003;
10.1152/ajpendo.00197.2003.—␤3-Adrenergic receptors (AR)
are nearly exclusively expressed in brown and white adipose
tissues, and chronic activation of these receptors by selective
agonists has profound anti-diabetes and anti-obesity effects.
This study examined metabolic responses to acute and
chronic ␤3-AR activation in wild-type C57Bl/6 mice and con-
genic mice lacking functional uncoupling protein (UCP)1, the
molecular effector of brown adipose tissue (BAT) thermogen-
esis. Acute activation of ␤3-AR doubled metabolic rate in
wild-type mice and sharply elevated body temperature and
BAT blood flow, as determined by laser Doppler flowmetry.
In contrast, ␤3-AR activation did not increase BAT blood flow
in mice lacking UCP1 (UCP1 KO). Nonetheless, ␤3-AR acti-
vation significantly increased metabolic rate and body tem-
perature in UCP1 KO mice, demonstrating the presence of
UCP1-independent thermogenesis. Daily treatment with the
␤3-AR agonist CL-316243 (CL) for 6 days increased basal and
CL-induced thermogenesis compared with naive mice. This
expansion of basal and CL-induced metabolic rate did not
require UCP1 expression. Chronic CL treatment of UCP1 KO
mice increased basal and CL-stimulated metabolic rate of
epididymal white adipose tissue (EWAT) fourfold but did not
alter BAT thermogenesis. After chronic CL treatment, CL-
stimulated thermogenesis of EWAT equaled that of inter-
scapular BAT per tissue mass. The elevation of EWAT me-
tabolism was accompanied by mitochondrial biogenesis and
the induction of genes involved in lipid oxidation. These
observations indicate that chronic ␤3-AR activation induces
metabolic adaptation in WAT that contributes to ␤3-AR-
mediated thermogenesis. This adaptation involves lipid oxi-
dation in situ and does not require UCP1 expression.
metabolic plasticity; obesity; diabetes; tissue remodeling
␤3-ADRENERGIC RECEPTOR (AR) agonists have profound
therapeutic effects in rodent models of obesity and
diabetes (2, 9, 14). ␤3-AR in rodents are expressed
almost exclusively in brown (BAT) and white adipose
tissue (WAT), and the therapeutic effects of ␤3-AR
agonists have been attributed to dramatic stimulation
of adipose tissue triglyceride hydrolysis coupled to el-
evated fatty acid oxidation in BAT. However, the direct
and indirect contribution of adipose tissue types to the
Address for reprint requests and other correspondence: J.
Granneman, CIMER/WSU, 2309 Scott Hall, 540 E. Canfield, Detroit
MI 48201 (E-mail: jgranne@med.wayne.edu).
E1230 0193-1849/03 $5.00 Copyright © 2003 the American Physiological Society
therapeutic effects of these compounds is largely un-
known.
␤3-AR agonists elicit strong thermogenesis in vivo,
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017
and this effect has been largely attributed to activation
of BAT, a thermogenic organ. The mechanism of brown
fat thermogenesis involves fatty acid oxidation and
uncoupling of oxidative phosphorylation through un-
coupling protein (UCP)1. Thus ␤3-AR stimulation does
not produce thermogenesis in brown adipocytes of mice
lacking UCP1 (3, 7, 21, 22). Furthermore, ␤3-AR-medi-
ated whole animal thermogenesis is sharply reduced in
UCP1 knockout (KO) mice (8).
Although UCP1-dependent BAT thermogenesis
clearly plays a dominant role in the overall metabolic
response to acute effects of ␤3-AR agonists, various
lines of evidence indicate that WAT plays a role as well.
For example, restoring ␤3-AR expression in BAT of
␤3-AR KO mice does not fully restore thermogenesis to
acute ␤3-AR agonist treatment, whereas restoration in
both brown and white fat does (12, 17). Furthermore,
mice lacking UCP1, and hence brown adipocyte ther-
mogenesis, still exhibit 20–40% of wild-type thermo-
genesis in response to ␤3-AR stimulation (8). These
observations suggest that WAT directly or indirectly
participates in the overall thermogenic response to
acute ␤3-AR stimulation and that a component of ther-
mogenesis is UCP1 independent.
The above analysis concerns the effects of acute
␤3-AR activation. The anti-obesity and anti-diabetes
effects of ␤3-AR agonists, however, require chronic
drug exposure. Chronic exposure to ␤3-AR agonists
elevates UCP1 expression in BAT, and it is thought
that BAT mediates the augmented thermogenic capac-
ity seen after chronic treatment (6, 14). However, be-
cause BAT cannot account fully for ␤3-AR-mediated
thermogenesis to acute agonist stimulation, it is possi-
ble that other tissues contribute to the augmented
thermogenic capacity seen after chronic exposure to
␤3-AR agonists. In this regard, recent work by Himms-
Hagen et al. (15) is of interest. These investigators
demonstrated that chronic exposure to ␤3-AR agonists
induces mitochondrial biogenesis in WAT, suggesting
that this tissue might be capable of oxidizing lipid in
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.
http://www.ajpendo.org
 UCP1-INDEPENDENT THERMOGENESIS
situ and could contribute to the thermogenic effects of
␤3-AR activation (15).
The aim of this study was to examine the acute and
chronic effects of the ␤3-AR stimulation on thermogen-
esis and WAT and BAT function. The results indicate
that ␤3-AR activation does not elevate BAT thermogen-
esis in UCP1 KO mice, in vivo or in vitro, but nonethe-
less significantly elevates body temperature and oxy-
gen consumption. Chronic treatment of mice with the
␤3-AR agonist CL-316243 (CL) increases CL-induced
thermogenesis, but that augmentation occurs indepen-
dently of UCP1. Furthermore, chronic CL treatment
elevates oxygen consumption fourfold in WAT of UCP1
KO mice but has no effect on BAT respiration. The
elevation of WAT thermogenesis correlates with induc-
tion of mitochondrial biogenesis and expression of
genes involved in fatty acid oxidation. These results
indicate functional plasticity in WAT that has impor-
tant implications for development of obesity and dia-
betes therapeutics.
MATERIALS AND METHODS
Animals. Wild-type C57Bl/6 (WT) and congenic UCP1 KO
mice were obtained from L. Kozak of Pennington Research
Institute (Baton Rouge, LA), bred at Wayne State University,
and used when 2–6 mo old. For all experiments, animals
were matched for age, and only male mice were used. Mice
were maintained at 23°C in a 12:12-h light-dark cycle and
had continuous access to food and water. Testing was per-
formed during the light phase.
Experiment 1. WT and UCP1 KO mice were anesthetized
with pentobarbital sodium (50 mg/kg) and immediately
placed on a warm surface to stabilize body temperature.
Interscapular BAT (IBAT) was carefully exposed through a
1-cm incision, a 21-gauge laser Doppler probe (Transonic,
Ithaca, NY) was placed into the tissue, and a temperature
probe (Physitemp, Clifton, NJ) was placed into the peritoneal
cavity. After stabilization of temperature and tissue blood
perfusion, mice were injected intraperitoneally with norepi-
nephrine (100 nmol) or BRL-37344 (50 nmol; RBI, Natick,
MA). Temperature and blood flow data were recorded at
2-min intervals. Whole body oxygen consumption (V̇O2) was
determined using an open-circuit system (Accuscan Oxyplot,
Columbus, OH). After 10 min of stable baseline data had
been collected, animals were injected with adrenergic ago-
nists, as indicated in Figs. 1–4, and V̇O2 was measured over
the following 20–30 min.
Experiment 2. The second experiment examined the effects
of acute and chronic exposure to the ␤3-AR agonist CL-
316243 (CL; 30 nmol; Wyeth, Princeton, NJ) on whole body
metabolic rate, WAT and BAT respiration, WAT mitochon-
drial biogenesis, and WAT gene expression. Age-matched WT
and UCP1 KO mice were injected daily for 6 days with saline
or CL. Twenty-four hours after the last injection, animals
were anesthetized with Avertin, and V̇O2 was determined as
described in experiment 1.
WAT and BAT respiration in vitro. UCP1 KO mice were
treated with saline or CL for 6 days as described. Mice were
anesthetized with Avertin and acutely challenged with saline
or CL (20 nmol). Fifteen to twenty minutes after injection,
epididymal WAT (EWAT) and IBAT were removed and im-
mediately placed in HEPES-buffered Krebs-Ringer solution
containing 1% bovine serum albumin. The tissue was minced
into 5- to 10-mg fragments and placed in a closed chamber
AJP-Endocrinol Metab • VOL 285 • DECEMBER 2003 •
E1231
(700 ␮l volume) for measurement of V̇O2 with a Clark-style
electrode (Qubit Systems, Kingston, ON, Canada) at 35°C
under constant stirring. Respiration rates were determined
by linear regression of the O2 concentration traces (Vernier
Software, Beaverton, OR).
Labeling and confocal imaging of WAT mitochondria.
WAT mitochondria were labeled using a modification of the
method described by DeMartinis (1). Briefly, EWAT was
dissected and minced into 10- to 20-mg fragments and placed
into Delbecco’s modified Eagle’s medium containing the flu-
orescent mitochondrial dye MitoTracker Red (100 nM; Mo-
lecular Probes, Eugene, OR). Tissue minces were incubated
for 30 min, washed twice in media, and then fixed in phos-
phate-buffered saline containing 4% paraformaldehyde. For
imaging, tissue minces were placed between coverslips in a
Leiden chamber and optically sectioned using confocal mi-
croscopy (Olympus Fluoview).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017
Quantitative RT/PCR analysis. Total RNA from EWAT
was isolated, and cDNA was synthesized, as previously de-
scribed (11). Synthesized cDNA was subjected to quantitative
(q)RT-PCR (iCycler, Bio-Rad) in duplicate, with SYBR-green
as the reporter fluorophore. Cycle threshold (Ct) values were
calculated using the Bio-Rad software, and data were nor-
malized to values obtained for peptidylprolyl isomerase 1A
(PPIA). The following primers were used: PPIA (⫹) GTG-
GTCTTTGGGAAGGTGAA, (⫺) TTACAGGACATTGCGAG-
CAG; pyruvate dehydrogenase kinase 4 (PDK4) (⫹) GGC-
CATCCATGTAGGAGAGA, (⫺) GAGGGAGACCCACAGAA-
GAA; long-chain acyl-CoA dehydrogenase (Acad1) (⫹)
CTCATGCAAGAGCTTCCACA, (⫺) CCACAAAAGCTCTG-
GTGACA; cytochrome c oxidase subunit 8b (Cox8b) (⫹) TGC-
GAAGTTCACAGTGGTTC, (⫺) CTCAGGGATGTGCAAC-
TTCA.
Statistical analysis. Data are presented as means ⫾ SE.
Data were evaluated by ANOVA. Post hoc analysis was
performed with the Neuman-Keuls test for multiple compar-
isons among means.
RESULTS
Experiment 1. Injection of norepinephrine (NE; 100
nmol) approximately doubled the metabolic rate of WT
mice, as expected (Fig. 1). The effect of NE on metabolic
rate was severely reduced in UCP1 KO mice. Nonethe-
less, NE significantly elevated metabolic rate in UCP1
Fig. 1. Norepinephrine (NE)-induced thermogenesis in wild-type
(WT) and uncoupling protein (UCP)1 knockout (KO) mice. Pentobar-
bital sodium-anesthetized mice were injected ip with 100 nmol of NE
at time 0 (arrow). NE significantly elevated metabolic rate [oxygen
consumption (V̇O2)] in WT and UCP1 KO mice relative to baseline
(P ⬍ 0.01), and WT mice were significantly more responsive to NE
than were UCP1 KO mice (P ⬍ 0.01). Values are means ⫾ SE; n ⫽ 4.
www.ajpendo.org
E1232 UCP1-INDEPENDENT THERMOGENESIS

KO mice by 25%, indicating that a component of the

NE-dependent thermogenesis is independent of UCP1.
NE activates several AR subtypes in numerous tis-
sues. In contrast, ␤3-AR are expressed nearly exclu-
sively in BAT and WAT, and expression in these tis-

Fig. 3. Effects of chronic CL-316243 (CL) treatment on CL-induced

Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017

thermogenesis in WT and UCP1 KO mice. Mice were treated with
saline (Sal) or CL daily for 6 days (CL-6d). On the 7th day, mice were
anesthetized with Avertin, and metabolic rate (V̇O2) was determined
under basal conditions (B) and after stimulation (S) by ip injection of
CL (30 nmol). Basal values are averages of the 5-min period before
injection, and stimulated values are averages of the 5-min period
20–25 min after injection. Two-way ANOVA indicated significant
effects of drug treatment and genotype (P ⬍ 0.0001) but no signifi-
cant interaction. Values are means ⫾ SE; n ⫽ 7–9.

sues is necessary and sufficient to account for the full

thermogenic effects of ␤3-AR agonists (1, 9, 12). We
therefore examined the effects of the selective ␤3-AR
agonist BRL-37344 (BRL; 50 nmol) on thermogenesis,
body temperature, and BAT blood flow in WT and
UCP1 KO mice.
Acute injection of BRL sharply elevated metabolic
rate (Fig. 2A) and body temperature (Fig. 2B) in WT
mice. The elevation of metabolic rate was greatly
blunted, but not eliminated, in UCP1 KO mice. Inter-
estingly, the elevation in body temperature was largely
intact in UCP1 KO mice, suggesting that BRL might
affect body temperature independently of metabolic
rate. BAT blood flow is sharply elevated by increased
local tissue respiration and is a sensitive physiological
measure of BAT metabolism in situ (4, 24). Injection of
BRL produced an immediate and sustained elevation
of BAT blood flow in WT mice, as determined by laser

Fig. 2. Effects of the selective ␤3-adrenergic receptor (AR) agonist

BRL-37344 on metabolic rate (V̇O2), body temperature, and brown
adipose tissue (BAT) blood flow in WT, heterozygous, and UCP1 KO
mice. Pentobarbital sodium-anesthetized WT (⫹/⫹), heterozygous
(⫹/⫺), and UCP1 KO (⫺/⫺) mice were injected ip with BRL-37344
(BRL; arrow). Metabolic rate (A), change in core body temperature
(B), and BAT blood flow (C) were determined as detailed in MATERIALS Fig. 4. Time course of CL-induced thermogenesis (shown by increase
AND METHODS. TPU, tissue perfusion units. Values are means ⫾ SE; in V̇O2) in naive and chronically-treated UCP1 KO mice. Values are
n ⫽ 5–11. means ⫾ SE; n ⫽ 4–5.

AJP-Endocrinol Metab • VOL 285 • DECEMBER 2003 • www.ajpendo.org

 UCP1-INDEPENDENT THERMOGENESIS
Fig. 5. Effect of chronic CL treatment on in vitro respiration of white
adipose tissue (WAT, A) and BAT (B) from UCP1 KO mice. CL
treatment for 6 days (CL6) significantly elevated basal (⫺; P ⬍ 0.001)
and CL stimulated (⫹; P ⬍ 0.001) respiration of epididymal WAT
from UCP1 KO mice but had no significant effect on respiration of
BAT. Values are means ⫾ SE; n ⫽ 4–6.
AJP-Endocrinol Metab • VOL 285 • DECEMBER 2003 •
E1233
Doppler flowmetry (Fig. 2C). In contrast, BRL had no
effect on tissue perfusion in UCP1 KO mice, indicating
that the tissue is thermogenically unresponsive to
␤3-AR activation.
Experiment 2. The therapeutic effects of ␤3-AR ago-
nists on diabetes and obesity require chronic exposure
to the compounds. We therefore compared the acute
and chronic effects of ␤3-AR activation on metabolic
rate and WAT function, gene expression, and histology.
WT and UCP1 KO mice were injected daily for 6 days
with saline or the highly selective ␤3-AR agonist CL (30
nmol). On the 7th day, metabolic rate was determined
under basal conditions and after stimulation by a su-
pramaximal dose of CL (20 nmol; Fig. 3). As with NE
and BRL, CL challenge strongly elevated metabolic
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017
rate in naive (i.e., saline-treated) WT mice, and this
effect was significantly reduced, but not eliminated, in
UCP1 KO mice. A 6-day course of CL treatment ele-
vated both basal and stimulated V̇O2. Importantly,
ANOVA indicated significant effects of drug treatment
and genotype, but no interaction among these vari-
ables. In other words, the enhancement of thermogen-
esis by chronic CL treatment was independent of
UCP1.
Figure 4 illustrates the time course of thermogenesis
induced by CL in UCP1 KO mice previously treated
with saline or CL. CL appeared to increase metabolic
rate in two temporally distinct phases. The rapid phase
occurred during the first 6 min after injection, with the
sustained elevation occurring thereafter. Although
both phases appeared to be augmented by chronic CL
Fig. 6. Effect of chronic CL treatment
on mitochondrial biogenesis in adipose
tissue of WT and UCP1 KO mice. Ep-
ididymal WAT of naive mice and mice
treated for 6 days with CL was labeled
with MitoTracker Red and fixed, and
images were acquired by confocal mi-
croscopy. Chronic CL treatment frag-
mented single lipid droplets and in-
duced mitochondrial proliferation.
www.ajpendo.org
E1234 UCP1-INDEPENDENT THERMOGENESIS

treatment, the first phase was affected more dramati-

cally.
Conditions that produce sustained activation of
WAT lipolysis, such as cold exposure and ␤3-AR ago-
nist treatment, induce mitochondrial biogenesis in
WAT and the appearance of multilocular cells, some of
which express UCP1 (15). These observations suggest
that chronic ␤3-AR activation induces metabolic adap-
tation in WAT, which might contribute significantly to
the thermogenic response. Therefore, the above exper-
iment was repeated, and the thermogenic responses of
EWAT and IBAT were determined in vitro under basal
conditions and after systemic challenge by CL.
As shown in Fig. 5, metabolic rate was low in EWAT
of naive mice, and acute CL treatment produced a

Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017

slight, but nonsignificant, increase in tissue respira-
tion. Chronic treatment with CL dramatically in-
creased basal metabolic rate in EWAT (P ⬍ 0.001). CL
challenge significantly elevated thermogenesis in
EWAT of chronically-treated mice (P ⬍ 0.01), and this
elevation was fourfold greater (P ⬍ 0.001) than that
observed in naive mice.
In control UCP1 KO animals, IBAT tissue thermo-
genesis was 7–8 times greater than that of EWAT (Fig.
5) and ⬃30% lower than that of WT mice (not shown).
IBAT tissue thermogenesis was not significantly af-
fected by either acute or chronic treatment with CL.
Importantly, EWAT thermogenesis equaled that of
IBAT (relative to tissue weight) after chronic CL treat-
ment and CL challenge.
The elevated metabolic capacity of EWAT was re-
flected in the histological appearance and gene expres-
sion pattern of the tissue. As shown in Fig. 6, 6 days of
CL treatment resulted in the fragmentation of the
single large lipid droplet characteristic of white adipo-
Fig. 7. Effect of chronic CL treatment on gene expression in WAT of
cytes into multiple lipid droplets. These lipid droplets WT and UCP1 KO mice. RNA from saline- and CL-treated mice was
were completely surrounded by mitochondria in nu- isolated and evaluated for expression of pyruvate dehydrogenase
merous cells, as demonstrated by confocal imaging of kinase 4 (PDK4), long-chain acyl-CoA dehydrogenase (LCAD), and
MitoTracker staining. The magnitude of the Mito- subunit 8b of cytochrome c oxidase (Cox8b) by quantitative (q)RT-
PCR. Values are means ⫾ SE; n ⫽ 3–6. CL treatment significantly
Tracker staining was highly similar in WT and UCP1 (P ⬍ 0.01) increased expression of Pdk4, LCAD, and Cox8b in UCP1
KO mice. KO mice, and LCAD and Cox8b in WT mice.
WAT mRNA from naive mice and mice treated
chronically with CL was subjected to qRT-PCR analy-
sis, with focus on genes that are involved in substrate vated blood flow in the tissue in UCP1 KO mice. These
selection and fatty acid oxidation. Chronic CL treat- observations confirm results of in vitro studies and
ment induced PDK4 expression similarly in WT and extend them to the functioning of the intact tissue in
UCP1 KO mice (Fig. 7A). Similar results were obtained situ (3, 7, 21, 22).
for expression of LCAD (Fig. 7B), a key enzyme in The dependence of BAT thermogenesis on UCP1
mitochondrial ␤-oxidation. Finally, chronic CL treat- indicates that UCP1 KO mice represent a model with
ment strongly (⬎20-fold) induced expression of Cox8b which to explore mechanisms of thermogenesis that
(Fig. 7C). are largely independent of BAT. In naive mice, UCP1-
independent thermogenesis constitutes 20–40% of the
DISCUSSION total thermogenic response to various ␤-AR agonists.
As mentioned above, the therapeutic effects of ␤3-AR
Recent work with transgenic and KO mouse models activation require chronic treatment. The present re-
strongly indicates that thermogenic responses to ␤3-AR sults indicate that chronic CL treatment significantly
agonists requires activation of both BAT and WAT (5, elevates both basal and stimulated thermogenesis. Im-
12). BAT clearly contributes importantly to the acute portantly, the present data indicate that this effect
effects of ␤3-AR stimulation. BAT thermogenesis in- does not require UCP1.
duced by ␤3-AR activation appears to be entirely de- In addition to elevating the maximal level of thermo-
pendent on UCP1, as indicated by the absence of ele- genesis, chronic CL treatment also accelerated the rate
AJP-Endocrinol Metab • VOL 285 • DECEMBER 2003 • www.ajpendo.org
 UCP1-INDEPENDENT THERMOGENESIS
at which that maximum was achieved. The slow eleva-
tion of thermogenesis in naive UCP1 KO mice con-
trasts with the rapid induction of V̇O2 seen in WT mice
in which fatty acids are rapidly oxidized by BAT in
situ. The relatively slow activation of thermogenesis in
naive KO mice suggests that UCP1-independent ther-
mogenesis might involve oxidation of mobilized lipid by
nonadipose tissues like liver and muscle. The time
course of CL-induced thermogenesis in UCP1 KO mice
chronically treated with CL resembled that of WT mice
and suggests that this augmentation of thermogenesis
might involve the direct effects of CL on adipocytes.
In vitro analysis of EWAT from UCP1 KO mice
demonstrated that chronic CL treatment dramatically
increased basal and CL-stimulated thermogenesis. In
contrast, neither acute nor chronic CL treatment af-
fected thermogenesis in IBAT from UCP1 KO mice.
Importantly, the mass-specific metabolic rate of EWAT
equaled that of BAT after chronic CL with acute CL
challenge. Although the quantitative significance of
the elevated WAT metabolism to total in vivo thermo-
genesis is difficult to extrapolate from in vitro data, it
is likely that WAT contributes more than BAT in UCP1
KO mice given the much greater mass of WAT in these
animals.
The present study addressed the molecular profile
and metabolic activity of adipose tissue after chronic
treatment with CL. The results demonstrate that
chronic activation of WAT ␤3-AR induces genes in-
volved in substrate selection, fatty acid oxidation, and
mitochondrial biogenesis. PDK4 phosphorylates the
pyruvate dehydrogenase complex, thereby suppressing
glucose utilization in favor of lipid oxidation (23);
Acad1 (LCAD) is a key step in mitochondrial ␤-oxida-
tion and has been shown to be essential for nonshiver-
ing thermogenesis (13); and Cox8b is a subunit of
cytochrome c oxidase that is induced during mitochon-
drial biogenesis. The strong upregulation of these
genes, along with the pronounced induction of mito-
chondrial biogenesis, indicates that CL induces a novel
plasticity in WAT that shifts the metabolic profile from
lipid storage to lipid oxidation in situ.
Kozak’s laboratory [Hofmann et al. (16) and Liu et al.
(20)] has provided compelling evidence for the presence
of UCP1-independent thermoregulatory thermogene-
sis. For example, Liu et al. reported that mildly cold-
stressed UCP1 KO mice exhibited lower levels of
plasma free fatty acids, higher levels of plasma ke-
tones, and a lower cumulative respiratory quotient, all
indicative of greater lipid oxidation. Similar effects
occur during chronic treatment with ␤3-AR agonsts (2,
14, 18, 24). Furthermore, mild cold stress produces
morphological changes in inguinal fat of UCP1 KO
mice, similar to those produced by CL treatment in
epididymal fat. These observations raise the possibility
that CL and mild cold stress engage similar thermo-
genic mechanisms in UCP1 KO mice and that one
component of that thermogenesis is likely to involve
activation of white fat thermogenesis.
The quantitative significance of WAT to overall ther-
mogenesis is presently unclear and is likely to depend
AJP-Endocrinol Metab • VOL
E1235
on the nature and magnitude of the stimulus, as well
as the thermogenic mechanisms available to the ani-
mal. For example, chronic CL treatment targets adi-
pose tissue and produces more extensive WAT remod-
eling than does mild cold stress. In contrast, cold stress
might be expected to elevate sympathetic activity and
thereby activate thermogenesis in liver and muscle,
which are important extra-adipose sites of lipid oxida-
tion/thermogenesis. Finally, it is important to recog-
nize that, although chronic CL treatment increases the
thermogenic capacity of WT and UCP1 KO mice simi-
larly, the mechanisms involved could differ, with WT
mice relying mainly on BAT and UCP1 KO mice en-
gaging WAT, muscle, and liver. Clearly, an important
goal of future work will be to identify the tissues that
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017
contribute to UCP1-independent thermogenesis.
␤3-AR agonists are highly effective therapeutics in
rodent models of obesity and diabetes (9, 10, 19). None-
theless, the mechanisms and sites of ␤3-AR action
remain poorly understood. The dominant role of UCP1-
dependent thermogenesis in the acute effects of ␤3-AR
agonists has cast doubt as to whether adipokinetic
drugs will be useful as human therapeutics, given the
paucity of BAT in adult humans. The present results,
however, indicate that WAT is capable of pronounced
thermogenic activity in response to sustained lipolytic
stimulation. Clearly, the abundance of the tissue tar-
get is not limiting in human obesity and type 2 diabe-
tes. Understanding the mechanisms controlling WAT
metabolic plasticity could lead to the identification of
novel points of therapeutic intervention for treatment
of obesity and diabetes in humans.
We thank Dr. L. Kozak and C. Bearden for supplying UCP1 KO
mice and for advice on anesthesia. We also thank Dr. A. Chaudhry
for use of the confocal microscope, and Drs. R. MacKenzie and T. Leff
for helpful comments.
DISCLOSURES
This work was supported by National Institute of Diabetes and
Digestive and Kidney Diseases Grant DK-62292, Michigan Life Sci-
ences Corridor Grant 27, and the Fund for Medical Research and
Education.
REFERENCES
1. DeMartinis FD, Ashkin KT, and Lampe KT. Fluorescence
detection of mitochondrial clusters in mammalian white fat cells
in vivo. Am J Physiol Cell Physiol 253: C783–C791, 1987.
2. De Souza CJ, Hirshman MF, and Horton ES. CL-316,243, a
beta3-specific adrenoceptor agonist, enhances insulin-stimulated
glucose disposal in nonobese rats. Diabetes 46: 1257–1263, 1997.
3. Enerback S, Jacobsson A, Simpson EM, Guerra C, Ya-
mashita H, Harper ME, and Kozak LP. Mice lacking mito-
chondrial uncoupling protein are cold-sensitive but not obese.
Nature 387: 90–94, 1997.
4. Foster DO and Frydman ML. Tissue distribution of cold-
induced thermogenesis in conscious warm- or cold-acclimated
rats reevaluated from changes in tissue blood flow: the dominant
role of brown adipose tissue in the replacement of shivering by
nonshivering thermogenesis. Can J Physiol Pharmacol 57: 257–
270, 1979.
5. Gavrilova O, Marcus-Samuels B, and Reitman ML. Lack of
responses to a beta3-adrenergic agonist in lipoatrophic A-ZIP/
F-1 mice. Diabetes 49: 1910–1916, 2000.
6. Ghorbani M and Himms-Hagen J. Appearance of brown adi-
pocytes in white adipose tissue during CL 316,243-induced re-
285 • DECEMBER 2003 • www.ajpendo.org
E1236 UCP1-INDEPENDENT THERMOGENESIS
versal of obesity and diabetes in Zucker fa/fa rats. Int J Obes
Relat Metab Disord 21: 465–475, 1997.
7. Golozoubova V, Hohtola E, Matthias A, Jacobsson A, Can-
non B, and Nedergaard J. Only UCP1 can mediate adaptive
nonshivering thermogenesis in the cold. FASEB J 15: 2048–
2050, 2001.
8. Gong DW, Monemdjou S, Gavrilova O, Leon LR, Marcus-
Samuels B, Chou CJ, Everett C, Kozak LP, Li C, Deng C,
Harper ME, and Reitman ML. Lack of obesity and normal
response to fasting and thyroid hormone in mice lacking uncou-
pling protein-3. J Biol Chem 275: 16251–16257, 2000.
9. Granneman JG. Why do adipocytes make the beta 3 adrenergic
receptor? Cell Signal 7: 9–15, 1995.
10. Granneman JG. The putative ␤4-adrenergic receptor is a novel
state of the ␤1-adrenergic receptor. Am J Physiol Endocrinol
Metab 280: E199–E202, 2001.
11. Granneman J, Skoff R, and Yang X. Member of the peroxi-
some proliferator-activated receptor family of transcription fac-
tors is differentially expressed by oligodendrocytes. J Neurosci
Res 51: 563–573, 1998.
12. Grujic D, Susulic VS, Harper ME, Himms-Hagen J, Cun-
ningham BA, Corkey BE, and Lowell BB. Beta3-adrenergic
receptors on white and brown adipocytes mediate beta3-selective
agonist-induced effects on energy expenditure, insulin secretion,
and food intake. A study using transgenic and gene knockout
mice. J Biol Chem 272: 17686–17693, 1997.
13. Guerra C, Koza RA, Walsh K, Kurtz DM, Wood PA, and
Kozak LP. Abnormal nonshivering thermogenesis in mice with
inherited defects of fatty acid oxidation. J Clin Invest 102:
1724–1731, 1998.
14. Himms-Hagen J, Cui J, Danforth E Jr, Taatjes DJ, Lang
SS, Waters BL, and Claus TH. Effect of CL-316,243, a ther-
mogenic beta 3-agonist, on energy balance and brown and white
adipose tissues in rats. Am J Physiol Regul Integr Comp Physiol
266: R1371–R1382, 1994.
15. Himms-Hagen J, Melnyk A, Zingaretti MC, Ceresi E, Bar-
batelli G, and Cinti S. Multilocular fat cells in WAT of CL-
316243-treated rats derive directly from white adipocytes. Am J
Physiol Cell Physiol 279: C670–C681, 2000.
AJP-Endocrinol Metab • VOL 285 • DECEMBER 2003 •
16. Hofmann WE, Liu X, Bearden CM, Harper ME, and Kozak
LP. Effects of genetic background on thermoregulation and fatty
acid-induced uncoupling of mitochondria in UCP1-deficient
mice. J Biol Chem 276: 12460–12465, 2001.
17. Ito M, Grujic D, Abel ED, Vidal-Puig A, Susulic VS, Lawitts
J, Harper ME, Himms-Hagen J, Strosberg AD, and Lowell
BB. Mice expressing human but not murine beta3-adrenergic
receptors under the control of human gene regulatory elements.
Diabetes 47: 1464–1471, 1998.
18. Kato H, Ohue M, Kato K, Nomura A, Toyosawa K, Furu-
tani Y, Kimura S, and Kadowaki T. Mechanism of ameliora-
tion of insulin resistance by beta3-adrenoceptor agonist AJ-9677
in the KK-Ay/Ta diabetic obese mouse model. Diabetes 50: 113–
122, 2001.
19. Kraegen EW, Cooney GJ, Ye JM, Thompson AL, and
Furler SM. The role of lipids in the pathogenesis of muscle
insulin resistance and beta cell failure in type II diabetes and
obesity. Exp Clin Endocrinol Diabetes 109, Suppl 2: S189–S201,
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on September 8, 2017
2001.
20. Liu X, Rossmeisl M, McClaine J, Riachi M, Harper ME,
and Kozak LP. Paradoxical resistance to diet-induced obesity
in UCP1-deficient mice. J Clin Invest 111: 399–407, 2003.
21. Monemdjou S, Kozak LP, and Harper ME. Mitochondrial
proton leak in brown adipose tissue mitochondria of Ucp1-defi-
cient mice is GDP insensitive. Am J Physiol Endocrinol Metab
276: E1073–E1082, 1999.
22. Nedergaard J, Golozoubova V, Matthias A, Asadi A, Ja-
cobsson A, and Cannon B. UCP1: the only protein able to
mediate adaptive non-shivering thermogenesis and metabolic
inefficiency. Biochim Biophys Acta 1504: 82–106, 2001.
23. Sugden MC and Holness MJ. Recent advances in mechanisms
regulating glucose oxidation at the level of the pyruvate dehy-
drogenase complex by PDKs. Am J Physiol Endocrinol Metab
284: E855–E862, 2003.
24. Takahashi H, Nakamura S, Shirahase H, Yoshida T, Nish-
imura M, and Yoshimura M. Heterogenous activity of BRL
35135, a beta 3-adrenoceptor agonist, in thermogenesis and
increased blood flow in brown adipose tissue in anaesthetized
rats. Clin Exp Pharmacol Physiol 21: 539–543, 1994.
www.ajpendo.org