| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
in the adult rodent heart
Division of Cardiology,
* Department of Integrative Biology, and the
Division of Organ Transplantation, University of Texas Houston Medical Center, Houston, Texas 77030, USA
1Correspondence: Division of Cardiology,, University of Texas Houston Medical School, 6431 Fannin, MSB 1.246, Houston, TX 77030, USA. E-mail: ht{at}heart.med.uth.tmc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
) administration affect UCP-2 and UCP-3 expression,
and 2) whether peroxisome proliferator-activated
receptor (PPAR) regulates the expression of either UCP-2 or
UCP-3. Direct comparisons were made between cardiac and skeletal
muscle. UCP-2, UCP-3, and PPAR expression were reduced when cardiac
workload was either increased (pressure overload by aortic
constriction) or decreased (mechanical unloading by heterotopic
transplantation). Similar results were observed during cytokine
administration. Reduced dietary fatty acid availability resulted in
decreased expression of both cardiac UCP-2 and UCP-3. However, when
fatty acid (the natural ligand for PPAR) supply was increased
(high-fat feeding, fasting, and STZ-induced diabetes), cardiac UCP-3
but not UCP-2 expression increased. Comparable results were observed in
rats treated with the specific PPAR agonist WY-14,643. The level of
cardiac UCP-3 but not UCP-2 expression was severely reduced (20-fold)
in PPAR-/- mice compared to wild-type mice. These
results suggest that in the adult rodent heart, UCP-3 expression is
regulated by PPAR. In contrast, cardiac UCP-2 expression is
regulated in part by a fatty acid-dependent, PPAR-independent
mechanism.—Young, M. E., Patil, S., Ying, J., Depre, C., Ahuja,
H. S., Shipley, G. L., Stepkowski, S. M., Davies,
P. J. A., Taegtmeyer H. Uncoupling protein 3 transcription is
regulated by peroxisome proliferator-activated receptor in the
adult rodent heart.
Key Words: diabetes • fasting • fatty acids • hypertrophy • unloading
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. UCP-1 is a
transmembrane protein found in the inner mitochondrial membrane, where
it uncouples ATP synthesis and mitochondrial electron transport,
dissipating the energy as heat (2)
. More recently, two
genes have been described with high sequence homology to UCP-1—UCP-2
and UCP-3 (3
4
5)
—both of which have been shown to possess
uncoupling activity in cells (6)
. UCP-2 is ubiquitously
expressed whereas UCP-3 is highly expressed in skeletal muscle, adipose
tissue, and, to a lesser extent, the heart. Although much research has
focussed on UCPs in skeletal muscle, the mechanism(s) by which cardiac
UCP expression alter(s) require elucidation.
There are four primary hypotheses concerning the physiological roles of
UCP-2 and UCP-3 (7)
: thermogenesis, regulation of fatty
acid oxidation, regulation of ATP synthesis, and reduction of
mitochondrial reactive oxygen species (ROS) generation. Whether UCPs in
the heart play a role in one or several of the above mentioned
functions is unknown. The roles of UCPs in tissues, such as skeletal
muscle, as well as white and brown adipocytes have been investigated by
determining the changes in gene expression of the UCPs in response to
various stimuli. For example, increases in plasma free fatty acid
levels result in increased skeletal muscle UCP expression, suggesting a
role in fatty acid utilization and/or prevention of lipotoxicity and
insulin resistance (7
8
9)
. Thus, determining UCP
expression in the heart in response to altered workload (in which ATP
demand, substrate utilization, and ROS generation are altered) and/or
substrate availability might aid an understanding of the role of
cardiac UCPs. In addition, tumor necrosis factor (TNF-) has been
shown to increase skeletal muscle UCP expression, suggesting a role in
cytokine-induced thermogenesis (10)
. TNF- levels are
known to increase in the failing heart(11), which has also been
described as ‘energy starved’ (12)
. Whether increased
TNF- levels result in increased UCP expression and whether the
latter affects the efficiency of the failing heart are unknown.
The purpose of the present study was twofold. First, we wanted to
investigate both physiological and pathophysiologic situations under
which cardiac UCP-2 and UCP-3 expression might be altered. These
included increased and decreased workload and altered substrate
availability (e.g., fasting/refeeding, altered dietary fatty acid
composition, streptozotocin-induced diabetes), as well as cytokine
(TNF-) challenge. Second, we were curious about the potential role
of peroxisome proliferator-activated receptor (PPAR, a nuclear
transcription factor that has been suggested to play a role in skeletal
muscle and adipocyte UCP expression) in the changes observed in UCP
expression. We 1) measured PPAR expression under
conditions in which UCP expression altered; 2) determined
the effects of specific activation of PPAR in vivo, with
WY-14,643, on UCP expression; and 3) measured the level of
UCP expression in PPAR-/- mice. Where
possible, we made direct comparisons between cardiac and skeletal
muscle to determine the similarities and differences of UCP control in
these two tissues. The results show that during increased availability
of fatty acids, the heart increases the expression of UCP-3 only, with
no effect on cardiac UCP-2 expression. In contrast, in skeletal muscle,
the expression of both UCP-2 and UCP-3 is increased with PPAR
activation. UCP-3, but not UCP-2, expression was severely reduced in
the heart of PPAR-/- mice compared to
wild-type mice. Together, these findings strongly implicate PPAR as
the major regulator of UCP-3, but not UCP-2, expression in the adult
rodent heart. During mechanical unloading and pressure overloading of
the heart, UCP-2, UCP-3, and PPAR expression all decrease. The
results provide new hypotheses for the roles of UCPs in the adult
heart.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Changes in workload
In the first series of experiments, cardiac unloading was
induced by heterotopic transplantation of a rat heart into the abdomen
of a recipient rat, as described earlier (13
, 14)
. After 2
wk, the animals were anesthetized and both donor (unloaded) and
recipient (control) hearts were removed, freeze clamped, and stored at
-80°C prior to RNA extraction.
In the second series of experiments, cardiac pressure overload was
induced by banding the ascending aorta, with a 20-gauge needle, as
described previously (14
, 15)
. In control animals, sham
operations were performed without banding of the aorta. Either 7 or 9
days after aortic constriction (to be specified), the animals were
anesthetized, hearts were removed, freeze-clamped, and stored at
-80°C.
Changes in substrate availability
In the first set of experiments, rats were fed either a high-
carbohydrate/low-fat (HC/LF) diet or a low-carbohydrate/high-fat
(LC/HF) diet (Purina Mills). These isocaloric diets varied only in the
proportion of energy obtained from carbohydrate and fat. The
contribution of carbohydrate, fat and protein to total energy available
were 71%, 6% and 23% for the HC/LF diet, and 24%, 53% and 23% for
the LC/HF diet, respectively. The source of carbohydrate was a
combination of sucrose and dextrin, while the source of fat was a
combination of corn oil and lard. Nonnutritive fiber was also increased
in the LC/HF diet. The length of time in which the rats were fed the
special diets is specified for individual experiments. In selected
experiments, soleus (skeletal) muscle was removed in addition to the
heart.
The effects of fasting and refeeding on cardiac gene expression were investigated. Rats were fasted for either 1 or 2 days, after which the heart and soleus muscles were isolated. A subset of fasted rats were refed with the HC/LF diet for an additional 4 days.
Diabetes was induced through a single injection of streptozotocin [STZ; 55 mg/kg intravenous (i.v.)]. Control animals were administered with buffer (Hank’s buffer; Life Technologies, Inc., Grand Island, N.Y.) only. Five, 7, 14, or 182 days (6 months) after the initial injection, the animals were anesthetized, hearts were removed, freeze-clamped, and stored at -80°C. Animals were considered diabetic if their blood glucose level was greater than 300 mg per deciliter.
Pharmacological interventions
To test the effects of specific PPAR activation, WY-14,643
was added to standard powdered Purina rodent chow at a concentration of
0.01% (w/w). Rats were fed a WY-14,643-containing diet for 4 days.
Control animals received powered rodent chow only.
In a separate set of experiments, rats received a single i.v. (tail
vein) injection with TNF- (human recombinant, 30 µg/kg body weight
in 1 ml 0.9% NaCl); controls were injected with NaCl only. Twelve
hours later, heart and soleus muscles were removed. As cytokine
administration decreases rodent food intake (which can potentially
affect gene expression), rats were injected at 07.00 h. During the
following 12 h period in the light, rodent ingestion is minimal.
To ensure standardized experimental conditions, food was withdrawn from
all animals at the time of injection.
PPAR-/- mice
Isolated RNA from hearts of age matched wild-type and
PPAR-/- mice was a kind gift from Dr L. Nagy
(The Salk Institute for Biological Studies, Gene Expression Laboratory,
La Jolla, Calif.).
RNA extraction and quantitative reverse transcription-polymerase
chain reaction (RT-PCR)
RNA extraction and quantitative RT-PCR of samples was performed
using previously described methods well established in our laboratory
(14
, 16
17
18)
. Specific quantitative assays were designed
from the rat sequences available in GenBank (Table 1
). Primers and probes were designed from unconserved sequences of the
genes (allowing for isoform specificity). The correlation between
Ct (the number of PCR cycles required for the
fluorescent signal to reach a detection threshold) and the amount of
standard was linear over a 5 log range of RNA for all assays
(Fig. 1
illustrates the values for rat/mouse UCP-2 and rat UCP-3). The level of
transcripts for the constitutive housekeeping gene product cyclophilin
was quantitatively measured in each sample to control for
sample-to-sample differences in RNA concentration. PCR data are
reported as the number of transcripts per number of cyclophilin
molecules.
|
|
Statistical analysis
Data are presented as the mean ± SE.
Statistically significant differences between groups were calculated by
the Student’s t test. A value of P<0.05 was
considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. The expression of UCP-2
and UCP-3 were decreased in unloaded hearts (Fig. 2A
, B
). Pressure overload resulted in cardiac hypertrophy
(after 1 wk the heart weight to body weight ratios were 3.42±0.06 vs.
2.98±0.05 in the experimental and control groups; P<0.05).
Hypertrophy also decreased the expression of UCP-2 and UCP-3 (Fig. 2C
, D
). In all hearts, levels of UCP-2 expression were
100-fold higher compared to UCP-3 (Fig. 2)
.
|
Mechanical unloading decreases cardiac PPAR expression
A possible mechanism by which reduced cardiac workload results in
decreased UCP expression is a decrease in the level of PPAR
expression. Unloading reduced cardiac PPAR expression by half
(0.137±0.027 vs. 0.282±0.018 in experimental and control groups;
P<0.05).
Pressure overload induced changes in UCP expression are dependent
on dietary fatty acids
The involvement of PPAR in decreased UCP expression during
pressure overload was investigated in two ways. First, we measured the
expression of PPAR during overloading. Second, we determined whether
the availability of the natural ligand for PPAR (fatty acids)
affected pressure overload induced alterations in UCP expression.
Therefore, rats were fed one of two diets: either a
high-carbohydrate/low-fat (HC/LF) diet or a low-carbohydrate/high-fat
(LC/HF) diet. On day 7 after the initiation of the special diets, half
of the rats underwent aortic banding, whereas the other half were
operated without banding. Rats were maintained on their specific diets
for 9 days after surgery, then the hearts were isolated. Aortic banding
resulted in similar degrees of cardiac hypertrophy in both groups
(Table 2
).
|
Hearts isolated from sham-operated rats fed the LC/HF diet compared to
sham-operated rats fed the HC/LF diet possessed significantly higher
levels of both UCP-2 and UCP-3 expression (Table 2)
. However, the fold
induction of UCP-3 was far greater than that of UCP-2. As observed
during standard laboratory chow feeding, banding significantly reduced
the expression of both UCP-2 and UCP-3 when rats were fed the LC/HF
diet (Table 2)
. In contrast, pressure overload did not decrease the
cardiac expression of either UCP-2 or UCP-3 for rats fed the HC/LF diet
(Table 2)
.
Banding resulted in a significant decrease in PPAR expression for
rats fed either diet (Table 2)
. Furthermore, rats fed the HC/LF diet
showed significantly higher cardiac PPAR expression compared to rats
fed the LC/HF diet (Table 2)
.
Increased dietary fat increases cardiac UCP-3, but not UCP-2,
expression; heart vs. skeletal muscle
To investigate the time course over which cardiac UCP expression
changes, rats were fed the LC/HF diet for various lengths of time (1,
2, 4, or 8 days), after which the expression of cardiac UCP-2 and –3
as well as PPAR were measured (Fig. 3
). At all time points, feeding with the LC/HF diet had no effect on
cardiac UCP-2 expression (Fig. 3A
). In contrast, UCP-3
expression was rapidly induced by the LC/HF diet (within 24 h),
and this induction was maintained (Fig. 3B
). In the case of
PPAR, the LC/HF diet repressed expression, an effect that was
partially normalized after 8 days of continuous feeding (Fig. 3C
).
|
We also measured UCP and PPAR expression in soleus muscle (Fig. 3D
Fig. 3E
Fig. 3F
). Feeding of rats with the LC/HF diet resulted in a
rapid increase in soleus muscle UCP-2 and UCP-3 expression (Fig. 3D
, E
). This increase was maintained for the full duration
of the feeding study. PPAR expression was not affected by the LC/HF
diet (Fig. 3F
).
Fasting and refeeding modulates cardiac UCP-3, but not UCP-2,
expression; heart vs. skeletal muscle
Rats were fasted for either 1 or 2 days, after which a subset was
refed for an additional 4 days on the HC/LF diet. Fasting had no effect
on cardiac UCP-2 expression, although refeeding with the HC/LF diet
resulted in a significant decrease in expression (Fig. 4A
. In contrast, fasting significantly increased UCP-3
expression by threefold within 24 h (Fig. 4B
). This
increased UCP-3 expression was severely blunted when fasting continued
for 2 days (Fig. 4B
). Refeeding returned cardiac UCP-3
expression to normal levels (Fig. 4B
). Fasting for either 1
or 2 days significantly lowered the expression of cardiac PPAR, an
effect that was reversed on refeeding (Fig. 4C
).
|
To investigate further differences in UCP regulation between cardiac
and skeletal muscle, we also determined the effects of fasting and
refeeding on soleus muscle UCP expression. Fasting resulted in
significant increases in UCP-2 and -3 expression (Fig. 4D
, E
, respectively). In the case of UCP-3, this increase observed
after 1 day was severely reduced at the second day of fasting, a result
similar to that observed in the heart (Fig. 4E
, B
). Refeeding normalized soleus muscle UCP (both -2 and -3) expression
(Fig. 4D
, E
, respectively). Soleus muscle PPAR expression was not affected by either fasting or refeeding (Fig. 4F
).
STZ-induced diabetes increases cardiac UCP-3 expression
STZ-induction of diabetes through pancreatic ß-cell destruction
resulted in significant elevations in plasma glucose levels at 5
(3.12-fold; P<0.001), 7 (4.45-fold; P<0.001),
14 (3.18-fold; P<0.001), and 182 (5.00-fold;
P<0.001) days after STZ injection (compared to age-matched
controls). UCP-2 expression in the heart was not affected by
STZ-induced diabetes (Fig. 5A
).However, cardiac UCP-3 expression increased rapidly in STZ-induced
diabetes and remained elevated at all time points investigated (Fig. 5B
). The highest fold induction in cardiac UCP-3 expression
occurred 14 days after the initial STZ injection (6.4-fold). Despite a
maintenance in the induction of UCP-3 at day 182 (4.6-fold), there was
a substantial decrease in the absolute level of cardiac UCP-3
expression in both control and STZ diabetic animals (Fig. 5B
). Although there was no significant difference in cardiac
UCP-2 expression with respect to time, there was a trend for older
animals to show reduced expression (Fig. 5A
). Thus, cardiac
UCP expression declined with age.
|
We also investigated the effects of STZ-induced diabetes on cardiac
PPAR gene expression. After either 5 or 7 days of diabetes, there is
no effect on PPAR expression (Fig. 5C
). After 14 days of
diabetes there is a decrease in PPAR expression, although this
effect is not significant (Fig. 5C
). However, there is a
significant decrease in PPAR expression at 182 days after the
induction of diabetes (Fig. 5C
). Age alone had no effect on
cardiac PPAR expression (Fig. 5C
).
The PPAR agonist WY-14,643 alters heart and soleus muscle gene
expression
Rats were fed either control diet or diet containing the PPAR
agonist WY-14,643 (0.01% w/w) for 4 days, after which cardiac and
soleus muscle UCP-2, UCP-3, PPAR, PDK-2, PDK-4, and muscle CPT-I
expression were measured. WY-14,643 caused an increase in soleus muscle
UCP-2 expression, with no effect on cardiac UCP-2 expression
(Fig. 6A
). Both cardiac and soleus muscle UCP-3 expression was
increased by WY-14,643 (3.5- and 13.4-fold, respectively; Fig. 6B
). PPAR expression was not affected by WY-14,643 in
either heart or soleus muscle (Fig. 6C
). To determine the
effectiveness of WY-14,643 feeding, we measured pyruvate dehydrogenase
kinase-4 (PDK-4) and muscle-specific carnitine palmitoyltransferase-I
(CPT-I) expression, genes known to be induced by PPAR activation
(19
, 20)
. PDK-4 and muscle CPT-I expression increased in
response to WY-14,643 feeding in both heart and soleus muscle (Fig. 6C
, F
). PDK-2, whose expression is not believed to be
regulated by PPAR (19)
, was not affected by WY-14,643
(Fig. 6D
).
|
TNF- decreases cardiac UCP-2, UCP-3, and PPAR expression
We investigated whether cytokine exposure alters cardiac gene
expression by injecting rats with TNF-. TNF- administration (30
µg/kg; i.v.) resulted in a significant decrease in cardiac PPAR
expression after 12 h (Fig. 7C
). Reduced PPAR expression in response to TNF- was
accompanied by decreased cardiac UCP-2 and UCP-3 expression (Figs. 7A
, B
). Similarly, TNF- administration resulted in
decreased soleus muscle UCP-3 and PPAR expression (Fig. 7B
, C
). In contrast, soleus muscle UCP-2 expression increased (Fig. 7A
). To determine the effectiveness of TNF-
administration, cardiac and soleus muscle inducible nitric oxide
synthase (iNOS) expression was also measured. TNF- increased both
cardiac and soleus muscle iNOS expression to similar extents (Fig. 7D
).
|
Genetic mutation of the PPAR gene decreases expression of
cardiac UCP-3, but not UCP-2
We investigated whether PPAR was essential for the expression
of cardiac UCP-2 and UCP-3 by comparing wild-type and
PPAR-/- mouse hearts. The level of cardiac
UCP-2 and UCP-3 in wild-type mice were relatively similar (Table 3
). There was a 20-fold lower level of cardiac UCP-3 expression in
PPAR-/- mice compared to wild-type hearts
(Table 3)
. No differences were observed in the level of UCP-2
expression in hearts isolated from PPAR-/-
mice compared to those isolated from wild-type mice (Table 3)
.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. In contrast to
soleus muscle, cardiac UCP-2 expression does not significantly change
in response to high-fat feeding, fasting, diabetes, or WY-14,643
treatment, all of which lead to PPAR activation. However, UCP-3
expression increases in cardiac and soleus muscle under the same
conditions. Expression of both UCP-3 and PPAR decrease during
cardiac mechanical unloading and pressure overloading. Mutation of the
PPAR gene results in a near complete inhibition of cardiac UCP-3
expression, with no effect on UCP-2 expression. These results suggest
that in the adult rodent heart, UCP-3 expression is regulated by
PPAR whereas UCP-2 expression is not.
Cardiac UCP-2 expression decreased during mechanical unloading and
pressure overloading. Furthermore, decreased UCP-2 expression during
pressure overload was dependent on the presence of dietary fatty acids.
Decreased dietary fatty acid intake alone reduced cardiac UCP-2 and
UCP-3 expression. These results suggest that UCP-2 expression in the
heart is regulated, at least in part, by a fatty acid-dependent,
PPAR-independent mechanism. Last, we show that TNF- reduces the
expression of both UCP-2 and UCP-3 in the heart, suggesting that
TNF- induced UCP expression in the failing heart is not involved in
energetic dysfunction.
The results of the present study must be interpreted in a wider
context. The uncoupling of mitochondrial electron transport from ADP
phosphorylation causes a collapse of the proton gradient across the
inner mitochondrial membrane and thereby limits ATP generation through
oxidative phosphorylation. Instead, the potential energy ‘stored’ by
the gradient is liberated as heat, an essential process in nonshivering
thermogenesis (1
, 2)
. Such a proton leak is catalyzed by
UCPs. Potential roles for UCP-2 and UCP-3 have been suggested for
skeletal muscle and adipose tissue, primarily from studies
investigating regulation of their expression in response to different
stimuli. For example, cold exposure, thyroid hormone, TNF-, elevated
dietary fat composition, insulin-dependent diabetes mellitus, and
specific PPAR agonists have all been shown to increase skeletal muscle
UCP expression (8
9
10
, 21
22
23)
. In contrast, exercise
training lowers skeletal muscle UCP expression (23)
. These
results, plus consideration of the sheer mass of skeletal muscle in the
body, have implicated skeletal muscle UCPs in the processes of heat
generation, obesity, and perhaps maintenance of insulin sensitivity
(9
, 24
, 25)
. However, the role(s) of UCPs in the adult
heart have not been addressed and are more difficult to rationalize,
which gives rise to the following considerations.
Altered cardiac workload affects both UCP-2 and UCP-3 expression
In response to increased or decreased workload, the heart
increases glucose utilization and decreases fatty acid utilization
(26
27
28)
. This change is mirrored by changes in the
expression of several genes encoding metabolic enzymes, including
decreases in the expression of PDK-4 (M. E. Young et al.,
unpublished observation), muscle-specific CPT-I, and medium chain acyl
CoA dehydrogenase (MCAD) in the heart (14
, 29)
. All three
genes are regulated by PPAR in the heart (19
, 20
, 30
, 31)
. The present study has found that PPAR expression
decreases during both unloading and overloading (see Results section
and Table 2
), which is responsible for the observed changes in PDK-4,
CPT-I, and MCAD. Likewise, cardiac expression of both UCP-2 and UCP-3
decreased during unloading and overloading (Fig. 2)
. This decrease in
UCP expression depended on the presence of fatty acids, as pressure
overloaded hearts isolated from animals fed a low-fat diet (HC/LF) did
not possess decreased UCP expression compared to diet-matched controls,
despite a decrease in PPAR expression (Table 2)
. These results
initially suggested to us that PPAR acts in the decreased UCP
expression during hypertrophy only in the presence of its ligand (fatty
acids). Similarly, the decreased PPAR expression during unloading
could potentially result in decreased UCP expression. Consistent with
the decrease observed in UCP expression during pressure overload,
exercise training, which is associated with cardiac hypertrophy,
reduces UCP expression in the heart (as well as in skeletal muscle)
(23)
.
Fatty acid availability influences both cardiac and skeletal muscle
UCP expression
High-fat feeding, fasting, and insulin-dependent diabetes mellitus
would be expected to activate PPAR (30
, 32
, 33)
. In
certain experiments, a direct comparison was made between expression of
UCPs in heart and a slow-twitch skeletal muscle (soleus) that is
relatively similar to the heart (34
, 35)
. Still, there
were differences between the muscles. High-fat feeding, fasting, and
diabetes all increased the expression of cardiac UCP-3 expression but
had no effect on UCP-2 expression (Figs. 3
4
5)
. These results agree
with a recent study that investigated cardiac UCP expression during
diabetes and fasting (36)
. In contrast to the heart,
high-fat feeding and fasting caused induction of both UCP-2 and UCP-3
in soleus muscle (Figs. 3
and 4)
; previously reported studies have
shown both UCP-2 and UCP-3 expression to increase in skeletal muscle
during STZ-induced diabetes (21)
. The fold induction of
UCP-3 in soleus muscle was greater than that of the heart during
high-fat feeding and fasting. Specific activation of PPAR through
feeding with WY-14,643 resulted in a similar pattern of UCP expression
observed with elevation of plasma fatty acid levels: increased UCP-3
expression in both heart and soleus muscles, with the largest induction
in the latter, whereas UCP-2 was induced only in soleus, not heart,
muscle (Fig. 6)
.
Cardiac UCP-3, but not UCP-2, expression is dependent on PPAR
signaling in the adult rodent heart
Table 4
summarizes the relationship between PPAR and the expression of
cardiac UCP-2 and UCP-3, as observed in the present study. Decreased
PPAR expression during cardiac unloading and overloading occurs in
concert with decreased expression of UCP-2 and UCP-3. Furthermore, the
decreased UCP expression during hypertrophy is dependent on the
presence of fatty acids in the diet, providing evidence for the
hypothesis that this decrease is related to the reduced PPAR
expression. Similarly, on low-fat feeding (either postfasting or
comparing the HC/LF and LC/HF feeding effects), when PPAR activation
should be reduced, UCP-2 and UCP-3 expression both decrease (Fig. 4
and
Table 2
). However, situations in which PPAR is stimulated (high-fat
feeding, fasting, diabetes, and WY-14,643 feeding) increase the
expression of cardiac UCP-3, with no effect on cardiac UCP-2
expression. These observations are consistent with the hypothesis that
UCP-3 expression is regulated by PPAR. One possible explanation for
these observations was that in the adult rat heart, UCP-2, but not
UCP-3, expression is maximal through PPAR signaling. Thus, a further
stimulation of PPAR would have no effect on UCP-2 expression, yet
reduced PPAR signaling would reduce UCP-2 expression. To investigate
this hypothesis further, we determined the level of expression of
cardiac UCP-2 and UCP-3 in PPAR-/- mice. In
this model, expression of various PPAR-regulated genes is decreased
(37)
. Because we find reduced expression of UCP-3, but not
UCP-2, in PPAR-/- mice, these results provide
evidence that UCP-2 expression in the adult rodent heart is not
regulated by PPAR. Instead, there must be a fatty acid-dependent,
PPAR-independent mechanism modulating cardiac UCP-2 expression,
which is maximally activated under normal physiological circumstances.
PPAR
, another member of the family of PPAR transcription factors,
does not appear to regulate UCP-3 expression in the same way as PPAR
in the heart. Treatment of adult rats with the PPAR agonist
troglitazone (0.1% w/w in the diet for 4 days), has no effect on UCP-3
(nor UCP-2) expression in the heart (M. E. Young et al.,
unpublished observation). This lack of an effect by troglitazone
treatment (which did induce known PPAR-regulated genes in the
skeletal muscle) on cardiac UCP gene expression is most likely due to
the very low abundance of PPAR in the heart, as compared with both
adipose tissue and skeletal muscle.
|
TNF- reduces cardiac UCP expression
TNF- is known to cause contractile dysfunction
(38)
, and the expression of TNF- increases in the
failing heart (11)
. Furthermore, TNF- antagonism
appears to have beneficial effects in subjects with heart failure
(39)
. TNF- has been shown to induce UCP expression in
skeletal muscle (10)
. As overexpression of UCPs might
limit mitochondrial ATP production, we hypothesized that
TNF--induced cardiac UCP expression could result in the contractile
dysfunction observed previously. However, administration of TNF- to
adult rats caused a significant reduction in cardiac UCP expression
(Fig. 7)
. Whether reduced cardiac UCP expression in response to TNF-
administration was due to a direct effect of TNF- on the
cardiomyocyte or to a systemic effect cannot be determined in the
present study. It is also still possible that local TNF- generation
by the cardiomyocyte, as observed in failing myocardium, induces UCP
expression.
Complexity of cardiac and skeletal muscle UCP gene expression
regulation
There are obvious differences in the regulation of UCP-2 and UCP-3
expression, some of which are summarized in Table 4
. Under
physiological conditions, cardiac UCP-2 expression is 
100-fold
greater than that of UCP-3. Expression of UCP-3 is regulated by PPAR
in the adult rodent heart, whereas UCP-2 is not. On the second day of
exposure to elevated fatty acid levels (for both the high-fat feeding
and fasting experiments), both cardiac and soleus muscle UCP-3
expression fall transiently (Figs. 3
and 4)
. As fatty acid exposure
continues, UCP-3 expression increases again (Fig. 3)
. Increased age was
associated with decreased cardiac UCP expression. In addition, TNF-
increases soleus muscle UCP-2 expression while decreasing cardiac UCP-2
expression (Fig. 7)
. Thus, multiple mechanisms for the regulation of
UCP expression must operate in heart and skeletal muscle. These might
include altered expression and/or activity in PPAR (as observed in
the heart for the present study), PPAR dimerization partners (e.g.,
RXR) or coactivators (e.g., PGC-1), which confer specificity between
UCP-2 and UCP-3 promoter regions, or even as yet unidentified
PPAR-independent mechanisms (40
41
42)
. A study published
during the preparation of this manuscript suggests that cardiac UCP-2
expression is regulated by PPAR (43)
. In this study,
neonatal cardiomyocytes were cultured in the presence of
triiodothyronine, fatty acids or WY-14,643, resulting in increased
UCP-2 expression. From these observations, it was concluded that
increased cardiac UCP-2 expression on birth was due to combined
stimulation of the thyroid hormone receptor and PPAR. However, at
birth there is a substantial increase in mitochondrial biogenesis,
which might explain the increased expression of a mitochondrial protein
such as UCP-2 (44)
. It should be noted that the results
observed in the present study are not consistent with altered
mitochondrial biosynthesis (e.g., acute pressure overload results in
increased mitochondrial biogenesis whereas UCP expression decreases;
45
). It is possible that neonatal cardiomyocytes in
culture possess a specific factor (e.g., a PPAR dimerization partner
or coactivator) that is not present in the adult heart.
Increased fatty acid availability reduces cardiac PPAR
expression
The present results suggest that conditions associated with
increased cardiac fatty acid utilization (high-fat feeding, fasting,
and diabetes) result in decreased PPAR expression, a mechanism not
acutely observed in skeletal muscle (Figs. 3
4
5)
. Two hypotheses can be
drawn regarding the potential mechanism by which this phenomenon
occurs. First, increased signaling through PPAR might cause the
altered expression of a protein that affects the expression of PPAR.
This autoregulation mechanism would therefore be PPAR-dependent. A
second, PPAR-independent mechanism can be postulated wherein fatty
acids activate a specific pathway that regulates the expression of
PPAR. Thus, the first mechanism is fatty acid independent and
PPAR dependent whereas the latter mechanism is fatty acid dependent
and PPAR independent. To investigate which mechanism was responsible
for the observed results, we used a specific PPAR agonist
(WY-14,643) that would lead to PPAR activation in the absence of
elevated fatty acid levels. WY-14,643 increased cardiac and skeletal
muscle UCP-3, PDK-4, and muscle CPT-I expression, suggesting PPAR
was activated. In contrast to fatty acids, WY-14,643 feeding had no
effect on cardiac PPAR expression (or soleus muscle PPAR
expression). The results are consistent with the hypothesis that a
fatty acid-dependent mechanism is responsible for the observed
down-regulation of cardiac PPAR. Further evidence toward this
hypothesis is the observation that hearts isolated from the obese
Zucker rat, a model of insulin resistance in which plasma fatty acid
levels are elevated, have severely reduced PPAR expression
(46)
. It could be hypothesized that repression of PPAR
expression by fatty acids serves as a mechanism to prevent excessive
fluctuations in PPAR signaling in this tissue during altered fatty
acid availability.
Potential roles of UCPs in the adult rodent heart
The present study has focused on uncovering the mechanisms by
which gene expression of the uncoupling proteins is regulated in the
adult heart during both physiological and pathophysiological
conditions. Determination of UCP protein levels (requiring antibody
generation) and altered mitochondrial function is beyond the scope of
the present study. It is believed that elucidating the mechanisms by
which the heart alters the expression of the UCPs might aid in the
understanding of their function. Possible major functions for UCPs in
the heart include thermogenesis, regulation of fatty acid oxidation,
regulation of ATP synthesis, and reduction of ROS formation. Due to its
size (compared with the body), the heart is unlikely to play a role in
global thermogenesis. UCP-2 expression is relatively insensitive to
substrate availability, suggesting it plays little function in cardiac
fatty acid utilization (which increases, for example, in diabetes).
However, UCP-3 is highly responsive to fatty acids. Whether UCP-3
induction helps prevent lipotoxicity when fatty acid levels are high or
acts in an antioxidant capacity when oxidative metabolism increases
(fatty acids cannot be metabolized anaerobically) are distinct
possibilities. When the heart requires increased efficiency, for
example, during increased or decreased workload or substrate limitation
(as observed during low dietary fatty acid availability, as fatty acids
are the primary fuel for the heart under physiological conditions),
cardiac UCP expression decreases. If cardiac UCPs affect mitochondrial
ATP synthesis, decreased UCP expression would be expected to increase
mitochondrial efficiency. Last, evidence suggests that uncoupling
proteins can act as antioxidants (47
, 48)
. The heart is a
continuously contracting organ, with high oxidative metabolism. The
observation that cardiac UCP-2 expression is relatively high (even
higher than skeletal muscle) and relatively constant during various
conditions suggests that it has an essential, constitutive role in the
heart, such as prevention of ROS formation. Indirect evidence for this
hypothesis comes from the observations that TNF- administration,
which causes a rapid decrease in cardiac UCP-2 and UCP-3 expression, is
known to increase oxidative stress (49)
.
Limitations of the study
Whether changes in UCP gene expression result in changes in either
UCP protein or activity or ultimately lead to altered cardiac function
has not been determined. Future studies are required to address these
issues. For example, the expression of UCP-2 in the heart is
100-fold greater than that of UCP-3. Therefore, if both UCP-2 and
UCP-3 possess uncoupling activity, what is the physiological
significance of UCP-3 induction during increased fatty acid
availability? It is possible that UCP-2 and UCP-3 possess different
intrinsic activities, are differentially regulated
post-transcriptionally, or are located within different regions of the
inner mitochondrial membrane, akin to the light harvesting complexes
and photosystems of chloroplasts. Whether allosteric factors other than
guanosine nucleotides and fatty acids differentially affect the
activity of UCP-2 and UCP-3 is unknown. Another concern is that each
intervention investigated in the present study could potentially affect
multiple factors in these complex in vivo models. For
example, nervous activity, workload, and multiple growth factor and
cytokine signaling cascades are all altered in both the unloaded and
pressure overloaded heart. In addition, various hormonal alterations
occur during nutritional manipulation and diabetes. Even in knockout
mice, compensatory mechanisms become activated, allowing for adaptation
in the absence of a specific gene. All these factors could potentially
affect cardiac UCP gene expression. However, one common factor that
links cardiac UCP-3 (but not UCP-2) expression with these diverse
animal models is the transcription factor PPAR (see Table 4
).
Although the present study has not directly measured PPAR protein or
activity, previous studies have shown that increased fatty acid
availability results in PPAR activation (30
, 32
, 33)
and that reduced PPAR transcription—for example, during cardiac
hypertrophy—is associated with reduced activity and expression of
PPAR-regulated genes (50)
.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. In contrast to skeletal muscle, in which UCP
expression alters dramatically in response to substrate availability,
heart UCP expression does not fluctuate very much. The expression of
UCP-2, the major UCP isoform in the rat heart (100-fold higher
expression compared to UCP-3), changes relatively little in response to
dramatic alterations in substrate availability. Uncoupling of skeletal
muscle mitochondria may be important for the utilization of fatty
acids, whereas the uncoupling of cardiac mitochondria might play more
of a role in an antioxidant capacity, preventing over-reduction of the
electron transport chain. In situations in which increased cardiac
efficiency is required, such as unloading, overloading, and decreased
fatty acid availability, cardiac UCP expression is decreased. Whether
other factors are able to significantly increase the expression of
uncoupling proteins in the adult heart is unknown.
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 24, 2000.
Revision received July 27, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
