Download The malonyl-CoA insensitive fraction of muscle CPT1 is greater in

AJP-Endo Articles in PresS. Published on December 11, 2001 as DOI 10.1152/ajpendo.00233.2001
E-0233-2001-R2
Evidence of a malonyl-CoA-insensitive carnitine
palmitoyltransferase 1 activity in red skeletal muscle
Jong-Yeon Kim*†, Timothy R. Koves*†, Geng-Sheng Yu§ , Tod Gulick§, Ronald N. Cortright*, G.
Lynis Dohm* and Deborah M. Muoio*‡†
*
Departments of Biochemistry and Physiology, East Carolina University, Greenville, NC, 27858
‡
Department of Medicine, Duke University Medical School, Durham NC, 27710
§
Diabetes Research Laboratory, Massachusetts General Hospital and Department of Medicine,
Harvard Medical School, Charlestown MA, 02129
†
authors contributed equally to the manuscript
Short title: Malonyl-CoA resistance in red muscle
Corresponding author:
Deborah M. Muoio, PhD
PO Box 3327
Duke University Medical Center
Durham, NC 27710
Tel. 919-684-3644
FAX 919-684-8907
Email: [email protected]
Copyright 2001 by the American Physiological Society.
E-0233-2001-R2
2
ABSTRACT
CPT1, which is expressed as two distinct isoforms in liver (α) and muscle (β), catalyzes
the rate-limiting step in the transport of fatty acid into the mitochondria. Malonyl-CoA, a potent
inhibitor CPT1, is considered a key regulator of fatty acid oxidation in both tissues. Still
unanswered is how muscle β-oxidation proceeds despite malonyl-CoA concentrations that
exceed the IC50 for CPT1β. We evaluated malonyl-CoA suppressible [14C]palmitate oxidation
and CPT1 activity in homogenates of red (RG) and white (WG) gastrocnemius, soleus (SOL)
and extensor digitorum longus (EDL) muscles. Adding 10 µM malonyl-CoA inhibited palmitate
oxidation 29%, 39%, 60%, and 89% in RG, SOL, EDL and WG, respectively. Thus, malonylCoA
resistance,
which
correlated
strongly
(0.678)
with
absolute
oxidation
rates
(RG>SOL>EDL>WG), was greater in red than in white muscles. Similarly, malonyl-CoAresistant palmitate oxidation and CPT1 activity were greater in mitochondria from RG compared
to WG. Ribonuclease protection assays were performed to evaluate whether our data might be
explained by differential expression of CPT1 splice variants. We detected the presence of two
CPT1β splice variants that were more abundant in red compared to white muscle, but the relative
expression of the two mRNA species was unrelated to malonyl-CoA resistance. These results
provide evidence of a malonyl-CoA-insensitive CPT1 activity in red muscle, suggesting fibertype specific expression of distinct CPT1 isoforms and/or post-translational modulations that
have yet to be elucidated.
Key words: fatty acid oxidation, fiber-type specificity
E-0233-2001-R2
3
Abbreviations: ACC, acyl-CoA carboxylase; ACS, acyl-CoA synthetase; ASM, acid soluble
metabolites; CPT1, carnitine palmitoyltransferase 1; EPI, epitrochleas; EDL, extensor digitorum
longus; RG, red gastrocnemius; SR, sarcoplasmic reticulum; SOL, soleus; WG, white
gastrocnemius.
E-0233-2001-R2
4
INTRODUCTION
After entering cells, long-chain fatty acid is converted to acyl-CoA by acyl-CoA
synthetase (ACS), a family of integral membrane proteins that are present in various subcellular
organelles, including mitochondria (8). Because long chain acyl-CoAs cannot diffuse freely
across membranes, the mitochondrial carnitine shuttle system plays an obligatory role in βoxidation by permitting acyl-CoA translocation from the cytosol into the mitochondria in all
mammalian cells (10).
Carnitine palmitoyltransferase 1 (CPT1), which spans the outer
mitochondrial membrane, catalyzes the initial step in this process by transferring acyl groups
from CoA to carnitine. The acylcarnitines formed by CPT1 traverse the inner membrane via a
specific translocase that is coupled to carnitine palmitoyltransferase 2 (CPT2), which regenerates
acyl-CoA upon transporting the fatty acyl groups into the mitochondrial matrix. Because CPT1
represents the pace-setting reaction in the carnitine shuttle system, it is widely considered the
most critical step in controlling fatty acid flux through the β-oxidative pathway (13).
CPT1 is expressed as at least two isoforms that are the products of different genes; a liver
enzyme (CPT1α) (~88 kDa) and its smaller counterpart in cardiac and skeletal muscle (CPT1β)
(~82 kDa), each having distinct kinetic properties (13). A distinguishing regulatory property of
these isoenzymes is that both are inhibited by malonyl-CoA (15), which is produced in the
cytosol by acetyl-CoA carboxylase (ACC) (26).
In both liver and muscle, physiological
alterations in malonyl-CoA concentrations correlate inversely with changes in β-oxidation. For
example, starvation (15) and exercise (28) decrease tissue levels of malonyl-CoA, which
presumably relieves inhibition of CPT1 and increases fatty acyl-CoA entry into mitochondria
(3).
Conversely, carbohydrate feedings stimulate ACC activity and increase production of
E-0233-2001-R2
5
malonyl-CoA, which corresponds with a decrease in fatty acid oxidation and an increase in longchain acyl-CoA accumulation (3). In some reports, physiological regulation of malonyl-CoA
concentration appears to differ between red and white muscle (3;29), suggesting that malonylCoA-mediated control of fatty acid metabolism might depend on muscle fiber-type. However,
the role of fiber-type in modulating the malonyl-CoA/CPT1 system has not been addressed.
Paradoxically, concentrations of malonyl-CoA that are measured in both red and white muscle
(1-4 µM) (3;15) should completely inhibit CPT1 activity at all times. This is because the muscle
isoform of CPT1 is approximately 100 times more sensitive to malonyl-CoA (IC50~ 0.03 µM)
than the liver isoform (IC50 ~2.7 µM) (7;15). Thus, the question of how fatty acid oxidation
proceeds in muscle despite constitutively high malonyl-CoA levels has remained an enigma.
Importantly, several investigators have reported that a significant portion of CPT activity
measured in muscle mitochondria is uninhibited by malonyl-CoA (7;15;23).
Historically,
investigators have attributed this non-suppressible fraction to CPT2, the inner mitochondrial
membrane enzyme that can be expressed if membranes are damaged during mitochondrial
preparation (7;15). We considered an alternative view that muscle might contain a malonyl-CoA
insensitive CPT1 fraction, and hypothesized that malonyl-CoA inhibition of muscle fatty acid
oxidation might depend on the fiber type of the muscle. Thus, discrepancies among studies
reporting varying degrees of malonyl-CoA insensitivity might have arisen because the studies
were performed using muscles composed of dissimilar fiber compositions. To address this
hypothesis we compared malonyl-CoA inhibition of fatty acid oxidation and CPT1 activity in
whole homogenates and isolated mitochondria that were prepared from red and white skeletal
E-0233-2001-R2
6
muscles. The results provide evidence of a malonyl-CoA insensitive CPT1 fraction that is
predominately active in red muscle.
METHODS
Materials. The 3H-carnitine and the chemiluminescent detection kit were from Amersham,
(Piscataway, NJ). The calnexin antibody was from Santa Cruz (Santa Cruz, CA) and the
polyvinylidene fluoride membrane (PVDF) was from Bio-Rad (Hercules, CA).
All other
reagents were obtained from Sigma (St. Louis, MO).
Animals. Male Sprague Dawley rats from Harlan (Indianopolis, IN) were fed a chow diet and
water ad libitum. Rats weighing ~300 g were used for all experiments.
Muscle homogenization.
Rats were anesthestized by an intraperitoneal injection of
ketamine/xylazine (90/10 mg/kg) and skeletal muscles were removed. Muscle homogenates and
isolated mitochondria were prepared using white (WG) and red (RG) gastrocnemius muscles and
from soleus (SOL), extensor digitorum longus muscles (EDL) and epitrochlears (EPI).
Approximately 50-70 mg of tissue was minced thoroughly with scissors in 300 µL of a modified
sucrose-EDTA medium (SET) containing 250 mM sucrose, 1 mM EDTA, 10 mM Tris-HCl, pH
7.4 (21).
SET buffer was added to a 20-fold diluted (m:v) suspension and samples were
homogenized in 3.0 mL Potter-Elvehjem glass homogenizer at 10 passes across 30 seconds at
1200 RPM with a Teflon pestle. Protein concentrations in muscle homogenates ranged from 0.9
to 2.3 mg/ml.
E-0233-2001-R2
7
Isolation of Mitochondria and Microsomes. Muscles were excised and immediately placed in
ice-cold modified Chapell-Perry buffer (100 mM KCL, 40 mM Tris-HCl, 10 mM Tris-Base,
5mM MgSO4, 1mM EDTA, 1mM ATP, pH 7.5) and separated into red, white, or mixed
gastrocnemius; only RG and WG were used for these experiments. Muscles were blotted,
weighed, and placed into 2.0 ml (RG) or 4 ml (WG) of Chapell-Perry buffer. Samples were
minced thoroughly on ice, diluted 10-fold (w:v) with Chapell-Perry buffer and then
homogenized 2 x 15 sec using an Ultra-turrax at approximately 9,500 rpm. Tissue homogenates
were centrifuged at 650 x g for 10 min at 4°C. The supernatant was gravity filtered through 4
layers of surgical gauze and centrifuged at 8,500 x g for 10 min at 4°C. Microsomes were
isolated from the supernatant by ultracentrifugation at 100,000 x g for 1 h at 4°C.
The
mitochondrial pellet from the 8,500 x g spin was washed to remove erythrocytes, resuspended in
1.3 ml Chapell-Perry buffer, and centrifuged at 8,500 x g for 10 min. Both the microsomal and
mitochondrial pellets were suspended in 1.0 ml SET buffer and used immediately for assaying
CPT1 specific activity and palmitate oxidation. Proteins were determined by the BCA method.
For Western blots, 15 µg protein from mitochondrial and microsomal subfractions were
separated by 8-12% gradient SDS-PAGE, transferred onto PVDF membranes and probed 2 h
with calnexin antibody per the manufacturer’s protocol. Proteins were visualized by horseradish
peroxidase-conjugated goat anti-rabbit immunoglobulin G with a chemiluminescent Western
blotting detection kit.
Fatty acid oxidation.
14
Palmitate oxidation rates were determined by measuring production of
C-labeled acid soluble metabolites (ASM), a measure of TCA cycle intermediates and acetyl
esters (incomplete oxidation) (27), and [14C]CO2 (complete oxidation) using a modified 48-well
E-0233-2001-R2
8
microtiter plate (Costar, Cambridge, MA) as previously described (11). Reactions were initiated
by adding 40 µl whole homogenates or isolated mitochondria to 160 ul of the incubation buffer
(pH 7.4), yielding final concentrations of 0.2 mM palmitate ([1-14C]palmitate at 0.5 µCi/ml), 100
mM sucrose, 10 mM Tris.HCl, 5 mM potassium phosphate, 80 mM potassium chloride, 1 mM
magnesium chloride, 2 mM L-carnitine, 0.1 mM malate, 2 mM ATP, 0.05 mM coenzyme A, 1
mM dithiothreitol, 0.2 mM EDTA and 0.5% bovine serum albumin. After incubating 60 min at
30°C, reactions were terminated by adding100 µl of 4 N sulfuric acid and the CO2 produced
during the incubation was trapped in 200 µl of 1N sodium hydroxide that had been added to
adjacent wells (11). The acidified media was stored at 4°C overnight and then ASM were
assayed in supernatants of the acid precipitate (27). Radioactivity of CO2 and ASM were
determined by liquid scintillation counting using 4 ml Uniscint BD (National Diagnostics).
Carnitine palmitoyltransferase I Assay.
CPT1 activity was measured using 10-20 µg
mitochondrial or microsomal protein, or 40-60 µg protein from whole homogenates. The assays
were carried out with 50 µM palmitoyl CoA and 0.2 mM 3H-carnitine (0.5 µCi) as previously
described (11). After 10 min at 30°C, the assay was terminated by adding 60 µl of 1.2 mM ice
cold HCl. The [3H]palmitoyl-carnitine formed was extracted with water-saturated butanol and
quantified by liquid scintillation counting.
RPA analysis. RNA was isolated from skeletal muscle using TRIzol (Gibco-BRL) reagent as
previously described (9). Ribonuclease protection assays were performed as described in (31)
using a complementary RNA probe generated by T7 polymerase from a linearized template
E-0233-2001-R2
9
consisting of the rat CPT-Iβ2 cDNA fragment extending from NcoI at nucleotide 238 (relative to
the initiation AUG) to EcoRI at nucleotide 423 subcloned into pBluescipt.
Statistics. Statistical analyses were performed using JMP Statistical Software (SAS, Cary NC).
The correlation between malonyl-CoA resistant activity and total oxidation rates was evaluated
using a bivariate linear regression analysis and significance was determined by ANOVA. Twoand three-way ANOVA were performed using a Standard Least Squares model to test both the
main and interaction effects of muscle type x incubation time x malonyl-CoA concentration
(where appropriate) on palmitate oxidation, CPT1 activity or percent inhibition. In experiments
consisting of a single time point and/or malonyl-CoA concentration, differences between red and
white muscle were performed using a one-way ANOVA.
RESULTS
Malonyl-CoA inhibition of palmate oxidation in red and white muscle. To address the question
of whether malonyl-CoA inhibits fatty acid oxidation equally in muscles of varied fiber
compositions, we first examined the effects of a high but still physiological malonyl-CoA
concentration (10 µM) on [14C]palmitate oxidation in homogenates of red and white muscles.
Predictably, palmitate oxidation (ASM plus CO2) was greater in homogenates of red muscle (RG
and SOL) than in homogenates of white muscle (EDL and WG) (Figure 1A). Adding 10 µM
malonyl-CoA, which exceeds the IC50 of rCPT1 by ~100-fold, inhibited palmitate oxidation
29%, 39%, 60% and 89% in RG, SOL, EDL and WG, respectively (Figure 1B). Relative
inhibition by malonyl-CoA was similar between measures of complete oxidation (CO2) and total
oxidation (CO2 plus ASM). Linear regression analyses indicated that the degree of malonyl-CoA
E-0233-2001-R2
10
resistance correlated positively (R2=0.678, P<0.005) with absolute rates of palmitate oxidation
(Figure 1C), thus highly oxidative red muscles were more resistant to malonyl-CoA than highly
glycolytic white muscles. Adding protease inhibitors to the homogeneization buffer did not
eliminate the differences between red and white muscles (data not shown), suggesting that fibertype specific malonyl-CoA resistance was not due to in vitro protease modification of CPT1 that
may have occurred during homogenate preparation and/or incubation. Because differences in
malonyl-CoA resistance were most marked between RG and WG, these muscles were chosen for
subsequent experiments.
Figure 2 shows the results of separate experiments that were conducted to determine
whether malonyl-CoA inhibition of palmitate oxidation remains linear during the course of a
one-hour incubation. In this experiment, 10 µM malonyl-CoA inhibited palmitate oxidation 67%
and 92% (P<0.001) in homogenates of RG and WG, respectively, and the relative inhibition was
similar (P=0.71) at 15, 30, 45 and 60-minute time points.
Peroxisomal oxidation in red and white muscle. Others have shown that both mitochondria and
peroxisomes contribute to total palmitate oxidation in muscle homogenate systems (18). To
quantify the contribution of peroxisomes to palmitate oxidation in red and white muscle, we
blocked mitochondrial respiration by incubating muscle homogenates in the presence of the
electron transport chain inhibitors, KCN and rotenone. In both red and white gastrocnemius,
peroxisomes accounted for approximately 20% of total palmitate oxidation (Table 1), similar to
previous reports (18). These results suggest that fiber-type-dependent differences in malonylCoA resistance are unrelated to differences in peroxisomal oxidation. Table 1 also shows that
palmitate oxidation to CO2 depended fully on the presence of carnitine, indicating that
E-0233-2001-R2
11
mitochondrial membrane integrity was not compromised during homogenate preparations.
Similarly, subtracting carnitine from the buffer diminished production of ASM 90%, however,
since a small portion of the [14C]-labeled acid soluble products appeared to be carnitineindependent, oxidation results from subsequent experiments are presented as CO2 only.
Malonyl-CoA sensitivity in homogenates of red and white gastrocnemius. To better evaluate
fiber-type dependent differences in malonyl-CoA sensitivity, palmitate oxidation to CO2 was
studied in the presence of increasing malonyl-CoA in whole homogenates of red and white
gastrocnemius (Figure 3A). In the presence of 1 and 5 µM malonyl-CoA, concentrations that
fall within the range previously reported in rat muscle (3), palmitate oxidation was inhibited only
32%-51% in RG, and 52-72% in WG (P<0.001) (Figure 3B). When the malonyl-CoA resistant
fraction was subtracted from the total activity, as previously described (15), the IC50 values
appeared similar between red and white muscle (Figure 3C), suggesting differences in efficacy
but not the potency of malonyl-CoA inhibition. Figure 4 shows that CPT1 activity, measured in
whole homogenates, was 1.7-fold greater (P<0.01) in SOL (red) than in EPI (white) muscle. In
EPI, 10 µM malonyl-CoA inhibited CPT1 activity 62% compared to only 34% in SOL (P<0.01).
These data recapitulate the results from oxidation experiments and demonstrate that fiber-type
selective malonyl-CoA inhibition of CPT1 activity is consistent with inhibition of palmitate
oxidation, lending further support to our hypothesis that malonyl-CoA inhibition of muscle
CPT1 depends on fiber composition.
Malonyl-CoA sensitivity and palmitate oxidation in isolated mitochondria. Previous reports
have suggested that the non-suppressible component of CPT1 activity might be related to
E-0233-2001-R2
12
mitochondrial damage. To address this possibility we tested the integrity of mitochondria that
were isolated from our homogenate preparations. Similar to the results in whole homogenates
(Figure 1), we found that oxidation rates were greater and the inhibitory effect of malonyl-CoA
was less (P<0.001) in mitochondria from RG compared to WG (Figure 5). Importantly, in the
isolated mitochondrial preparations, palmitate oxidation to CO2 fully required the presence of
both CoA and carnitine, respective substrates for the outer mitochondrial enzymes, ACS and
CPT1, which catalyze the first two reactions of palmitate oxidation. These data support results
obtained from whole homogenates (Table 1) and confirm that mitochondrial membranes were
intact.
Additionally, others have suggested that contamination of mitochondria with
sarcoplasmic reticulum (SR) might contribute to changes in malonyl-CoA sensitivity (17). We
evaluated this possibility by assessing the presence of the SR-specific marker protein, calnexin,
in mitochondrial compared to microsomal subfractions. Western blot analyses indicated that SR
contamination was similarly low in mitochondria from red and white muscles (Figure 6a).
Taken together, the data shown in Figures 5 and 6 provide evidence that mitochondrial CPT1
contributes to malonyl-CoA resistance in red muscle and that this fiber type-dependent property
is unlikely to be an artifact due to SR contamination. Interestingly, we found that the CPT1
specific activity in microsomes was similar to that in mitochondria (Figure 6b). Palmitate
oxidation to CO2 was undetectable in our microsomal preparations, indicating negligible levels
of mitochondrial contamination (data not shown).
The high CPT1 activities in muscle
microsomes, which is consistent with previous observations in liver microsomes, indicate that
both mitochondrial and SR CPT1 could have contributed to malonyl-CoA resistance in our
E-0233-2001-R2
13
homogenate system. How acylcarnitines that are synthesized in extramitochondrial organelles
might be subsequently shuttled to the mitochondria for β-oxidation remains uncertain.
Expression of CPT1 splice variants in red and white muscle. Next, we investigated the
possibility that our observations might be related to fiber-type selective mRNA abundance of a
recently identified CPT1β splice variant that is preferentially expressed in rat heart and skeletal
muscle (30;31). Relative expression of alternatively spliced mRNAs in different muscles was
quantified by RPA (Figure 7). Expression of the predominant splice variant, CPT1β1, was
approximately 3-fold greater in red muscles (SOL and RG) than in white muscles (EDL and
WG) (Figure 7B). Likewise, expression of the alternatively spliced mRNA species, CPT1β2,
was also more abundant in red muscles (Figure 7C). Thus, CPT1 gene expression across fibertypes is consistent with the higher enzyme activity and fatty acid oxidative capacity measured in
red compared to white muscles. However, since the ratio of CPT1β1 to CPT1β2 abundance was
similar between red and white muscles, mRNA expression of CPT1β2 appeared to be unrelated
to fiber-type selective differences in malonyl-CoA resistance. In separate experiments using a
cRNA probe for CPT1α, this mRNA showed very low abundance in skeletal muscle (data not
shown), as previously reported (7).
DISCUSSION
Malonyl-CoA is presumed to be a key regulator of muscle fatty acid oxidation by virtue
of its potent inhibition of CPT1, and because muscle malonyl-CoA content changes reciprocally
with β-oxidation (3).
Unexplained, however, is how fatty acid oxidation proceeds despite
E-0233-2001-R2
14
muscle concentrations of malonyl-CoA that should completely inhibit CPT1β. This enigma
might be at least partly reconciled by our data, which provide evidence of a malonyl-CoA
insensitive CPT1 activity in red skeletal muscle. Similar to previous reports, we found that in
homogenates of WG, physiological concentrations of malonyl-CoA inhibited palmitate oxidation
75-90%. Conversely, in RG, these same concentrations inhibited palmitate oxidation only 3554%. Further, 100 µM malonyl-CoA, which exceeds the reported IC50 by several orders of
magnitude, inhibited palmitate oxidation only 62%, suggesting that CPT1 in red muscle is
resistant to full inhibition by malonyl-CoA. Additionally, a comparison across muscle types
showed that malonyl-CoA resistance correlated positively with the muscle’s fatty acid oxidative
capacity. These results provide the first reported evidence of fiber-type specific differences in
malonyl-CoA-mediated regulation of muscle lipid oxidation.
Although the present investigation did not evaluate fiber type by parameters other than
fatty acid oxidative capacity, a recent study of rat hindlimb muscles provided detailed analyses
of fiber composition based on the protein expression profile of four distinct myosin heavy chain
isoforms (22). Investigators found that RG, SOL, EDL and WG consisted of 24%, 84%, 5% and
0.8 % type I fibers, respectively. Type II fibers were further classified into IIA, IIB, IIC, IID,
IIAD and IIDB subtypes. We combined their data on fiber type with our results presented in
Figure 1 to evaluate the relationship between histochemical fiber type and malonyl-CoA
resistance. The only significant correlation detected by linear regression (R2=0.63, P<0.01) was
that between malonyl-resistance and the proportion of type IIA fibers, which was 18%, 9%, 15%
and <1% in RG, SOL, EDL and WG, respectively (22). Surprisingly, these correlative findings
across studies suggest that expression of the malonyl-CoA resistant CPT1 subfraction might be
E-0233-2001-R2
15
more closely linked to type IIA than type I fibers, although we acknowledge that this result
should be interpreted cautiously due to the small number of animals that were used for the
analysis.
Still, in experiments using larger populations (data not shown), both fatty acid
oxidative capacity and malonyl-CoA resistance were consistently greater in RG than in SOL
(other muscle types were not evaluated), again implying that these metabolic properties are
unrelated to the proportion of type I fibers. This finding is reflective of the poor association that
is often reported between histochemical fiber type, delineated by expression of specific myosin
isoforms, and the metabolic fiber type of a muscle.
Consistent with our findings, previous studies, including those that first described the
kinetic properties of muscle CPT1 (15), have also reported residual CPT activity in muscle
mitochondria exposed to high concentrations of malonyl-CoA. Investigators have attributed this
activity to inner mitochondrial membrane CPT2, which can be exposed during mitochondrial
isolation (7). However, exposure of CPT2 cannot explain our data because, unlike previous
investigations, the present study evaluated malonyl-CoA inhibition of palmitate oxidation. We
consider it unlikely that mitochondrial damage occurred in a manner that differed systematically
between red and white muscle, and in such a way as to permit CPT1-independent acyl-CoA entry
into the matrix without accompanying disruption the β-oxidative and TCA pathways. Thus, our
observation that palmitate oxidation to CO2 was retained indicates that mitochondria were
physiologically intact. Additionally, demonstration that palmitate oxidation was fully dependent
on the presence of both carnitine and CoA further confirms the integrity of mitochondrial
membranes.
E-0233-2001-R2
16
Importantly, the present investigation evaluated the inhibitory effects of malonyl-CoA in
whole muscle homogenates. This represents another key distinction between our data and
previous studies because skeletal muscle possesses two mitochondrial subpopulations,
intermyofibrillar and subsarcolemmal, which exhibit distinct biochemical properties and respond
differently to physiological stimuli (12). Thus, in contrast to previous studies, our experiments
using whole homogenates eliminated potential problems associated with poor mitochondrial
yield and disproportionate recovery of mitochondrial subpopulations, either of which might
result in mischaracterization of muscle mitochondrial enzymes. In a homogenate system both
mitochondria and peroxisomes contribute to total oxidative capacity (18).
We found that
peroxisomes contributed equally (approximately 19%) to total oxidation in red compared to
white muscle, suggesting that differences in peroxisomal oxidation cannot explain fiber-type
selective malonyl-CoA resistance.
Taken together, our findings provide strong evidence that red muscle expresses a
malonyl-CoA-resistant CPT1 subfraction. CPT1β has not been isolated in a catalytically active
form that would allow kinetic characterization of the purified enzyme; however, the full-length
cDNA has been expressed in yeast (20) and mammalian COS cells (25). The recombinant
enzyme is completely inhibited by 1.0-10 µM malonyl-CoA (20;25), which appears to contradict
our finding that in RG, 10-100 µM malonyl-CoA inhibited palmitate only 35-62%.
This
discrepancy suggests that red muscle might express a novel CPT1 isoform that confers a
modified malonyl-CoA regulatory site, a possibility that is supported by evidence indicating that
both rat (31) and human (30) muscle expresses multiple mRNA splice variants of the CPT1β
transcript.
The deduced polypeptide sequence of CPT1β2, a novel mRNA species that is
E-0233-2001-R2
17
expressed in rat muscle (31), predicts an isoform of the enzyme that omits putative malonyl-CoA
regulatory regions. In the present investigation we found that both the CPT1β1 and CPT1β2
transcripts were ~3 times more abundant in red than in white muscles. To our knowledge, these
data are the first to show that differences in CPT1 gene expression across muscle fiber types are
consistent with similar differences observed in enzyme activity and fatty acid oxidation rates.
However, relative expression of the two CPT1 mRNA species was similar in red and white
muscles and thus appeared to be unrelated to fiber-type specific differences in malonyl-CoA
sensitivity. These results do not exclude the possibility that protein expression of β2 relative to
β1 might differ among muscle fiber-types, although direct demonstration that the CPT1β2
transcript is in fact translated into a distinct isoenzyme is still lacking. Protein expression and
characterization of the catalytic and regulatory properties of novel CPT1 splice variants should
provide further insight into their potential role in conferring malonyl-CoA insensitivity.
The emerging model of CPT1 topology predicts that the catalytic site resides in the large
carboxy terminal domain facing the cytosol, and that the smaller cytosolic amino terminal
domain is crucial for maintaining a confirmation that permits optimal malonyl-CoA binding and
inhibition (10). When CPT1β is expressed in yeast, deletion of the conserved first 28 N-terminal
amino acids abolishes malonyl-CoA sensitivity and increases catalytic activity 2.5-fold,
indicating that an intact N-terminal domain is required for malonyl-CoA inhibition (20). Further,
several investigators have proposed that it is the interaction between the C- and N-domains that
determines malonyl-CoA sensitivity (25).
According to this model, post-translational
modifications of either domain and/or other factors that alter CPT1 confirmation might
contribute to physiological modulation of malonyl-CoA sensitivity in vivo. Consistent with this
E-0233-2001-R2
18
premise, malonyl-CoA sensitivity of hepatic CPT1 decreases in response to physiological
stresses that increase β-oxidation (e.g. starvation and diabetes) (5). The precise mechanism of
this desensitization is unknown, but several laboratories have implicated changes in the lipid
environment of the mitochondrial membrane (2;14). Although only limited data suggest that a
similar phenomenon might occur in muscle (10), membrane phospholipid composition does
differ markedly between red and white fibers (1), thus leaving open the possibility that
distinctions in mitochondrial membrane properties might contribute to fiber-specific variations in
CPT1 kinetics.
Voluminous evidence indicates that skeletal muscle, by virtue of its highly heterogenous
composition, can vary considerably with respect to its metabolic properties. Compared to white
muscle, red muscle exhibits greater insulin responsiveness and a higher capacity to oxidize fatty
acids (4;6;16). Clinical relevance of these fiber-specific properties is strongly suggested by
correlative data linking a disproportionately high number of white muscle fibers to metabolic
disorders such as cardiovascular disease, obesity and type II diabetes (4;24). Interestingly, these
diseases have also been associated with malonyl-CoA/CPT1-mediated alterations in muscle lipid
homeostasis (11;19). By inhibiting CPT1 and preventing acyl-CoA entry into mitochondria,
malonyl-CoA not only decreases β-oxidation but also increases muscle accumulation of longchain acyl-CoA (3). Because long-chain acyl-CoA and its derivatives can serve as signaling
and/or gene regulatory molecules, this putative malonyl-CoA-acyl-CoA axis is thought to
mediate several physiological and pathophysiological processes (19;32). Thus, when taken
together with the aforementioned reports, our observation of fiber-type specific malonyl-CoA
E-0233-2001-R2
19
resistance not only presents implications for muscle cell functioning, but also suggests a novel
link between muscle fiber composition and disorders of energy homeostasis.
E-0233-2001-R2
ACKNOWLEDGEMENTS
This work was supported by DK 46121-06 (GLD) and F32DK 10017-01 (DMM) from the
National Institutes of Health, and the North Carolina Institute of Nutrition (TMK). We thank
Donghai Zheng for assistance with RNA isolations.
20
E-0233-2001-R2
21
REFERENCES
1. Blackard, W.G., J. Li, J.N. Clore, and W.B. Rizzo. Phospholipid fatty acid composition
in type I and type II rat muscle. Lipids 32: 193-198, 1997.
2. Broadway, N.M. and E.D. Saggerson. Effect of membrane environment on the activity
and inhabitability by malonyl-CoA of the carnitine acyltransferase of hepatic microsomal
membranes. Biochem. J. 322: 435-440, 1997.
3. Chien, D., D. Dean, A.K. Saha, J.P. Flatt, and N.B. Ruderman. Malonyl-CoA content
and fatty acid oxidation in rat muscle and liver in vivo. Am. J. Physiol. 279: E259-E655,
2000.
4. Cortright, R.N., D.M. Muoio, and G.L. Dohm. Skeletal muscle lipid metabolism: A
fronteir for new insights into fuel homeostasis. Nutr. Biochem. 8: 228-245, 1997.
5. Drynan, L., P.A. Quant, and V.A. Zammit. The role of changes in the sensitivity of
hepatic mitochondrial overt carnitine palmitoyltransferase in determining the onset of the
ketosis of starvation in the rat. Biochem. J. 318: 767-770, 1996.
6. Dyck, D.J., S.J. Peters, J. Glatz, J. Gorski, H. Keizer, B. Kiens, S. Liu, E.A. Richter,
L.L. Spriet, d. van, V, and A. Bonen. Functional differences in lipid metabolism in
resting skeletal muscle of various fiber types. Am. J. Physiol. 272: E340-E351, 1997.
7. Esser, V., N.F. Brown, A.T. Cowan, D.W. Foster, and J.D. McGarry. Expression of
cDNA isolated from rat brown adipose tissue and heart identifies the product as the muscle
isoform of carnitine palmitoyltransferase I (M-CPT I). M-CPT I is the predominant CPT I
E-0233-2001-R2
22
isoform expressed in both white (epididymal) and brown adipocytes. J. Biol. Chem. 271:
6972-6977, 1996.
8. Iijima, H., T. Fujino, H. Minekura, H. Suzuki, M.J. Kang, and T. Yamamoto.
Biochemical studies of two rat acyl-CoA synthetases, ACS1 and ACS2. Eur. J. Biochem.
242: 186-190, 1996.
9. Jones, J.P. and G.L. Dohm. Regulation of glucose transporter GLUT-4 and hexokinase II
gene transcription by insulin and epinephrine. Am. J. Physiol. 273: E682-E687, 1997.
10. Kerner, J. and C. Hoppel. Fatty acid import into mitochondria. Biochim. Biophys. Acta.
1486: 1-17, 2000.
11. Kim, J.-Y., R.C. Hickner, R.N. Cortright, G.L. Dohm, and J.A. Houmard. Lipid
oxidation is reduced in obese human skeletal muscle. Am. J. Physiol. 297: E1039-E1044,
2000.
12. Lombardi, A., M. Damon, A. Vincent, F. Goglia, and P. Herpin. Characterization of
oxidative phosphorylation in skeletal muscle mitochondria subpopulations in pig: a study
using top-down elasticity analysis. FEBS Lett. 475: 84-88, 2000.
13. McGarry, J.D. and N.F. Brown. The mitochondrial carnitine palmitoyltransferase system.
From concept to molecular analysis. Eur. J. Biochem. 244: 1-14, 1997.
14. McGarry, J.D. and N.F. Brown. Reconstitution of purified, active and malonyl-CoAsensitive rat liver carnitine palmitoyltransferase I: relationship between membrane
environment and malonyl-CoA sensitivity. Biochem. J. 349: 179-187, 2000.
E-0233-2001-R2
23
15. McGarry, J.D., S.E. Mills, C.S. Long, and D.W. Foster. Observations on the affinity for
carnitine, and malonyl-CoA sensitivity, of carnitine palmitoyltransferase I in animal and
human tissues. Biochem. J. 214: 21-28, 1983.
16. Megeney, L.A., P.D. Neufer, G.L. Dohm, M.H. Tan, C.A. Blewett, G.C.B. Elder, and
A. Bonen. Effects of muscle activity and fiber composition on glucose transport and
GLUT-4. Am. J. Physiol. 264: E583-E5931993.
17. Niot, I., F. Pacot, P. Bouchard, J. Gresti, A. Bernard, J. Bezard, and P. Clouet.
Involvement
of
microsomal
vesicles
in
part
of
the
sensitivity
of
carnitine
palmitoyltransferase I to malonyl-CoA inhibition in mitochondrial fractions of rat liver.
Biochem. J. 304: 577-584, 1994.
18. Piot, C., J.H. Veerkamp, C. Bauchart, and J. Hocquette. Contribution of mitochondria
and peroxisomes to palmitate oxidation in rat and bovine tissues. Comp. Biochem. Physiol.
121: 185-194, 1998.
19. Ruderman, N.B., A.K. Saha, D. Vavvas, and L.A. Witters. Malonyl-CoA, fuel sensing,
and insulin resistance. Am. J. Physiol. E1-E18, 1998.
20. Shi, J., H. Zhu, D.N. Arvidson, and G. Woldegiorgis. The first 28 N-terminal amino acid
residues of human heart muscle carnitine palmityoltransferase I are essential for malonylCoA sensitivity and high-affinity binding. Biochemistry 39: 712-717, 2000.
21. Sholte, H., Y. Yu, J. Ross, H. Oosterkamp, A. Boonman, and H. Busch. Rapid isolation
of muscle and heart mitochondria, the liability of oxidative phosphorylation and attempts to
E-0233-2001-R2
24
stabilize the process in vitro by taurine, carnitine and other compounds. Mol. Cell.
Biochem. 174: 61-66, 1997.
22. Staron, R.S., W.J. Kraemer, R.S. Hikida, A.C. Fry, J.D. Murray, and G.E. Campos.
Fiber type composition of four hindlimb muscles of adult Fisher 344 rats. Histochem. Cel.l
Biol. 111: 117-123, 1999.
23. Starritt, E.C., R.A. Howlett, G.J. Heigenhauser, and L.L. Spriet. Sensitivity of CPT I to
malonyl-CoA in trained and untrained human skeletal muscle. Am. J. Physiol. 278: E462E468, 2000.
24. Storlien, L.H., D.A. Pan, A.D. Kriketos, J. O'Connor, I.D. Caterson, G.J. Cooney,
A.B. Jenkins, and L.A. Baur. Skeletal muscle membrane lipids and insulin resistance.
Lipids Suppl: S261-S265, 1996.
25. Swanson, S.T., D.W. Foster, J.D. McGarry, and N.F. Brown. Roles of the N- and Cterminal domains of carnitine palmitoyltransferase I isoforms in malonyl-CoA sensitivity
of the enzymes: insights from expression of chimaeric proteins and mutation of conserved
histidine residues. Biochem. J. 335: 513-519, 1998.
26. Trumble, G.E., M.A. Smith, and W.W. Winder. Purification and characterization of rat
skeletal muscle acetyl-CoA carboxylase. Eur. J. Biochem. 231: 192-198, 1995.
27. Veerkamp, J.H., H. van Moerkerk, J.F. Glatz, J. Zuurveld, A. Jacobs, and A.J.M.
Wagenmakers. 14CO2 production is no adequate measure of [14C]fatty acid oxidation.
Biochem. Med. Met. Biol. 35: 248-259, 1984.
E-0233-2001-R2
25
28. Winder, W.W. and D.G. Hardie. Activation of acetyl-CoA carboxylase and activation of
AMP-activated protein kinase in muscle during exercise. Am. J. Physiol. 270: E299-E304.
1996.
29. Winder, W.W., B.F. Holmes, D.S. Rubink, J.O. Holloszy, and M. Chen. Activation of
AMP-activated protein kinase increases mitochondrial enzymes in skeletal muscle. J.
Appl. Physiol. 88: 2219-2226, 2000.
30. Yu, G., Y. Lu, and T. Gulick. Expression of novel isoforms of carnitine
palmitoyltransferase I (CPT-I) generated by alternative splicing of the CPT-Iβ gene.
Biochem. J. 334: 225-231, 1998.
31. Yu, G., Y. Lu, and T. Gulick. Rat carnitine palmitoyltransferase I β mRNA splicing
isoforms. Biochim. Biophys. Acta. 1393: 166-172, 1998.
32. Zammit, V.A. The malonyl-CoA-long-chain acyl-CoA axis in the maintenance of
mammalian cell function. Biochem. J. 343: 505-515, 1999.
E-0233-2001-R2
26
Figure Legends
Figure 1. Fiber-type dependent palmitate oxidation and malonyl-CoA resistance. Whole
muscle homogenates were incubated 1 h at 29oC in a media containing 200 µM [14C]palmitic
acid (0.1µCi/well) in the presence or absence of 10µM malonyl-CoA. Palmitate oxidation
(nmol/g protein/min) was determined by measuring production of 14C-labeled CO2 and ASM as
described in METHODS. Data are means ± SEM from 2-3 animals assayed in quadruplicates
and presented as (A) Total oxidation (CO2 plus ASM) (B) Malonyl-CoA resistant palmitate
oxidation (total and CO2) in the presence of 10 µM malonyl-CoA, expressed relative to the
activity at 0 µM malonyl-CoA, and (C) relationship between malonyl-CoA resistance and total
oxidation.
Figure 2. Time course of malonyl-CoA resistance in red and white muscle. Whole muscle
homogenates from red (RG) and white (WG) gastrocnemius were incubated 15-60 min at 29oC
in a media containing 200 µM [14C]palmitic acid (0.1µCi/well) in the presence or absence of
10µM malonyl-CoA.
production of
14
Palmitate oxidation (nmol/g protein) was determined by measuring
C-labeled CO2 and ASM as described in METHODS. Data are means ± SEM
from 3 animals, assayed in quadruplicates, and were analyzed by three-way ANOVA. Time
points at which relative malonyl-CoA inhibition is significantly different (P<0.001) between red
and white muscles are indicated by *.
Figure 3. Malonyl-CoA inhibition of palmitate oxidation in red and white muscles. Whole
muscle homogenates were incubated 1 h at 29oC in a media containing 200 µM [14C]palmitic
E-0233-2001-R2
27
acid (0.1µCi/well) in the presence of 0-100 µM malonyl-CoA. Palmitate oxidation (nmol/g
protein/min) was determined by measuring production of
14
C-labeled CO2 as described in
METHODS and is presented as (A) Absolute rates of palmitate oxidized to CO2, (B) Relative
inhibition expressed as a percent of the total activity at 0 uM malonyl-CoA, and (C) Relative
inhibition expressed as a percent of the total activity after subtracting the residual activity at 100
µM malonyl-CoA. Data are means ± SEM from 10 animals, assayed in quadruplicates, and were
analyzed by two-way ANOVA.
Significant differences (P<0.001) between red and white
muscles are indicated by *.
Figure 4. Malonyl-CoA inhibition of carnitine palmitoyltransferase I activity in red and white
muscles. Malonyl-CoA inhibition of CPT1 activity (nmol/g protein/min) was measured in
muscle homogenates of red soleus (SOL) and white epitrochlears (EPI) muscles as described in
the METHODS. Data are means ± SEM from 10 animals assayed in triplicate. Values that are
significantly different compared to controls (without malonyl-CoA) * and/or between red and
white muscles § (P<0.01) were analyzed by one-way ANOVA.
Figure 5. Malonyl-CoA inhibition of palmitate oxidation in mitochondria from red and white
muscles.
Mitochondria isolated from homogenates of red and white gastrocnemius were
incubate 10 min at 29oC in a in a modified Chappell-Perry buffer containing 200 µM
[14C]palmitic acid (0.1µCi/well) in the presence or absence of 10 µM malonyl-CoA, 1 mM LCarnitine or 50 µM coenzyme A. Palmitate oxidation (nmol/g protein/min) was determined by
measuring production of 14C-labeled CO2 as described in Methods. Data are means ± SEM from
E-0233-2001-R2
28
5 animals assayed in quadruplicates. Values that are significantly different compared to controls
* and/or between red and white muscles § (P<0.01) were analyzed by two-way ANOVA.
Figure 6. Microsomal carnitine palmitoyltransferase I activity in red and white muscles. A)
Microsomal and mitochondrial protein (15 µg) from red (RG) and white (WG) gastrocnemius
was separated by SDS-PAGE, transferred onto polyvinylidene fluoride membranes, probed with
an antibody against the SR marker protein, calnexin, and visualized using a chemiluminescence
Western-blot detections kit. B) Mitochondrial and microsomal subfractions were prepared by
differential centrifugation and then used immediately for assaying CPT1 activity (µmol/g
protein/min) as described in METHODS. Data are means ± SEM from 3 animals, assayed in
triplicate, and were analyzed by two-way ANOVA. Values that are significantly different
(P<0.05) between red and white muscles are indicated by §.
Figure 7. Fiber-type dependent mRNA expression of CPT1β splice variants. Alternatively
spliced CPT1β mRNAs in red and white rat skeletal muscle. (A) Location of the cRNA probe
used in RPA analyses is superimposed on a schematic of a partial CPT1 mRNA sequence. White
lines indicate splice junctions and lengths of protected mRNA fragments are shown. (B)
Expression of CPT1β1 shown from short exposure RPAs performed with total RNA from rat
extensor digitorum longus (EDL), red gastrocnemius (RG), soleus (SOL) and white
gastrocnemius (WG), and mixed skeletal muscle from Clonetics (SkM) using the cRNA probe
shown in A. (C). Panel 3 represents expression of CPT1β2 shown from a longer exposed RPA
performed with total RNA described in B using the cRNA probe shown in A. Protected
fragments corresponding to each isoform are indicated to the right of each panel.
E-0233-2001-R2
29
E-0233-2001-R2
30
Red Gastrocnemius
White Gastrocnemius
palmitate oxidation (nmol/g/h)
ASM
CO2
ASM
CO2
2615 ± 270
969 ± 79
1137 ± 110
682 ± 60
ETS inhibitors
686 ± 43
ND
350 ± 56
ND
no carnitine
281 ± 31
ND
175 ±25
ND
control
Table 1 . Peroxisomal and carnitine-dependent palmitate oxidation in muscle homogenates.
Whole muscle homogenates were incubated 1 h at 29oC in a media containing 200 µM
[14C]palmitic acid (0.1µCi/well) in the presence or absence of mitochondrial electron transport
sytem (ETS) inhibitors (2.2mM KCN and 40mg/L rotenone) and 1 mM L-carnitine. Production
of
14
C-labeled CO2 and ASM were assayed as described in Methods. Data are means ± SEM
from 3 animals assayed in quadruplicates.
background could not be detected.
ND indicates that [14C]CO2 production over
(Percent activity at 10 µM malonyl-CoA)
Malonyl-CoA-resistant palmitate oxidation
Palmitate oxidation
(nmol/g/min)
E-0233-2001-R2
31
Figure 1.
A.
60
50
40
30
20
10
0
RG
RG
SOL
SOL
Muscle
EDL
100
EDL
WG
Muscle
B.
Total
CO2
80
60
40
20
0
WG
E-0233-2001-R2
32
Figure 1.
Malonyl-CoA-resistant palmitate oxidation
(Percent activity at 10 µM malonyl-CoA)
C.
100
R2= 0.678
80
60
40
20
0
10
20
30
40
50
Total Palmitate oxidation (nmol/g/min)
60
E-0233-2001-R2
33
Figure 2.
300
RG control
RG 10 µM mal-CoA
Palmitate oxidation
(nmol/g)
250
WG control
WG 10 µM mal-CoA
200
150
100
50
*
0
15
*
*
*
30
45
60
Time (min)
E-0233-2001-R2
34
Figure 3.
red
white
15
(nmol/g/min)
Palmitate Oxidation to CO2
A.
10
*
*
*
*
5
0
20
40
*
60
80
100
Malonyl-CoA (µM)
100
(%of control)
Palmitate Oxidation to CO2
B.
red
white
75
*
*
50
*
*
*
25
0
20
40
60
80
Malonyl CoA (µM)
100
E-0233-2001-R2
35
Figure 3.
C.
Palmitate oxidation to CO2
(% of control)
100
red
white
80
60
40
20
0
0
10
20
30
40
Malonyl-CoA (µM)
50
E-0233-2001-R2
36
Carnitine palmitoyltransferase I activity
(nmo/g/min)
Figure 4.
70
control
60
10 µM malonyl-CoA
50
*
§
40
30
20
*§
10
0
SOL
EPI
E-0233-2001-R2
37
Figure 5.
red
(nmol/mg/min
Palmitate oxidation to CO2
1.60
1.20
0.80
white
§
*
0.40
*§
0
control
10 µM
malonylCoA
* *
minus
carnitine
*
*
minus CoA
E-0233-2001-R2
38
Figure 6.
A.
RG
mt mcs
WG
mt mcs
Carnitine palmitoyltransferase I activity
(µmol/g/min)
B.
7.0
mitochondria
6.0
microsomes
5.0
4.0
§
3.0
§
2.0
1.0
0
RG
WG
E-0233-2001-R2
39
Figure 7.
A
EcoRI
NcoI
174
β1
β2
ATG
2
3
B
5
6
C
RPA β 2
EDL
RG
SOL
WG
Sk M
RPA β 2
EDL
RG
SOL
WG
rCPT1 β 2
RPA β 2 2
185
Probe
β1
β2
β1