Download Long-chain fatty acids increase basal metabolism

Am J Physiol Heart Circ Physiol 282: H1495–H1501, 2002.
First published January 3, 2002; 10.1152/ajpheart.00696.2001.
Long-chain fatty acids increase basal metabolism
and depolarize mitochondria in cardiac muscle cells
JOHN RAY,1 FRANK NOLL,2 JÜRGEN DAUT,1 AND PETER J. HANLEY1
1
Institut für Normale und Pathologische Physiologie, Universität Marburg, 35037 Marburg;
and 2Fachbereich Chemie, Universität Marburg, 35032 Marburg, Germany
Received 3 August 2001; accepted in final form 10 December 2001
Ray, John, Frank Noll, Jürgen Daut and Peter J.
Hanley. Long-chain fatty acids increase basal metabolism
and depolarize mitochondria in cardiac muscle cells. Am J
Physiol Heart Circ Physiol 282: H1495–H1501, 2002. First
published January 3, 2002; 10.1152/ajpheart.00696.2001.—
The effects of long-chain (LC) fatty acids on rate of heat
production (heat rate) and mitochondrial membrane potential (⌬⌿) of intact guinea pig cardiac muscle were investigated at 37°C. Heat rate of ventricular trabeculae was measured with microcalorimetry, and ⌬⌿ was monitored in
isolated ventricular myocytes with either JC-1 or tetramethylrhodamine ethyl ester (TMRE). Methyl-␤-cyclodextrin was
used as fatty acid carrier. Application of 400 ␮M oleate or
linoleate increased resting heat rate by ⬃30% and ⬃25%,
respectively. When LC fatty acid was supplied as sole metabolic substrate, resting heat rate was decreased by 3-mercaptopropionic acid. In TMRE-loaded myocytes, neither 40–80
␮M oleate nor 40 ␮M linoleate affected ⌬⌿. At a higher
concentration (400 ␮M) both oleate and linoleate increased
TMRE fluorescence by ⬃20% of maximum, obtained using
2,4-dinitrophenol (100 ␮M), indicating a depolarization of the
inner mitochondrial membrane. We conclude that LC fatty
acids, at sufficiently high concentration, increase heat rate
and decrease ⌬⌿ in intact cardiac muscle, consistent with a
protonophoric uncoupling action. These effects may contribute to the high metabolic rate after reperfusion of postischemic myocardium.
important because LC fatty acids are the major metabolic substrates of the heart and, furthermore, they are
known to accumulate in high concentrations during
ischemia (6, 33). The aims of the present study were to
determine the effects of LC fatty acids on both resting
heat rate and mitochondrial membrane potential (⌬⌿)
of intact cardiac muscle.
METHODS
IN CARDIAC MUSCLE, the resting (basal) level of metabolism is extraordinarily high, accounting for 25–30% of
the metabolic rate of mechanically active muscle (3, 4,
9, 12, 22, 29). Although the origin of this high resting
metabolic rate is unknown, it has been shown to be
sensitive to metabolic substrate (4, 9). This is well
illustrated with pyruvate, which has been shown to
increase resting rate of heat production (4) and oxygen
consumption (9) of isolated cardiac muscle preparations.
In isolated mitochondrial preparations, long-chain
(LC) fatty acids have been shown convincingly to uncouple oxidative phosphorylation (27, 31). Whether
this uncoupling action manifests in intact cardiac muscle is not known (14, 35). This unresolved issue is
Isolation of trabeculae and microcalorimetry. Guinea pigs
(200–300 g) were anesthetized with 3–4% isoflurane in
oxygen and then decapitated with a guillotine. The heart
was rapidly excised and perfused with dissecting solution
containing (mM) 105 NaCl, 15 KCl, 2 CaCl2, 1 MgCl2, 1
NaH2PO4, 24 NaHCO3, 20 2,3-butanedione monoxime, and
10 glucose. The solution was equilibrated with 95% O2-5%
CO2, and the pH was 7.4. The right ventricle was opened, and
free running trabeculae (diameter ⬍400 ␮m) were excised
and subsequently transferred to the calorimeter. These preparations were shown recently to be composed chiefly of myocytes accompanied by parallel perimysial collagen fibers (13).
After a trabecula was mounted in the calorimeter, the solution was then changed to a standard physiological salt solution containing (mM) 121 NaCl, 5 KCl, 2 CaCl2, 0.8 MgCl2,
1 NaH2PO4, 24 NaHCO3, and 10 glucose (pH 7.4). The temperature of the calorimeter was maintained at 37°C.
The rate of heat production (heat rate) of trabeculae was
measured with the system schematically illustrated in Fig. 1
and described in detail by Daut and Elzinga (3, 4). Drift in
the baseline (⬍1 ␮W/h) was checked by repeatedly transferring the trabecula out of the recording chamber and was duly
accounted for. In more recent experiments, the mounting
procedure was slightly modified. Trabeculae were tied in situ
using nylon monofilaments that had a terminal preformed
loop of 200-␮m diameter. The loops were placed over hooks
(Fig. 1), which were connected to micromanipulators, and the
preparation was positioned in the center of the recording
chamber.
Isolation of ventricular myocytes. A cannula was attached
to the aorta of the isolated heart, and the coronary arteries
were perfused with physiological salt solution containing
(mM) 115 NaCl, 5.4 KCl, 1.5 MgCl2, 0.5 NaH2PO4, 5 HEPES,
16 taurine, 5 sodium pyruvate, 15 NaHCO3, 1 CaCl2, and 5
glucose (pH 7.4). After 5 min, the heart was perfused for 5
min with nominally Ca2⫹-free solution, followed by solution
containing collagenase type I (Sigma), 0.1% bovine serum
Address for reprint requests and other correspondence: J. Daut,
Institut für Normale und Pathologische Physiologie, Universität
Marburg, Deutschhausstrasse 2, 35037 Marburg, Germany (E-mail:
[email protected]).
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.
mitochondrial membrane potential; microcalorimetry
http://www.ajpheart.org
0363-6135/02 $5.00 Copyright © 2002 the American Physiological Society
H1495
H1496
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
Fig. 1. Recording system of the calorimeter. Schematic diagram of a
trabecula mounted in the recording chamber of the calorimeter. The
preparation was attached to 2 platinum stimulating electrodes by
nylon monofilaments and positioned between 6 sets of chromelconstantan thermocouples connected in series (for clarity, only 1 set
is shown). The chamber is perfused at a constant rate of 1 ␮l/s, and
the voltage difference between positions T1 and T2 (T2 ⫺ T1) is
measured. The measured voltage difference is proportional to the
rate of heat production by the preparation (70 nV/␮W for chromelconstantan couples).
albumin, and 40–60 ␮M Ca2⫹. After enzymatic digestion
(5–7 min), ventricular myocytes were separated by gentle
trituration via a wide-bore pipette in solution containing
(mM) 45 KCl, 70 K glutamate, 3 MgSO4, 15 KH2PO4, 16
taurine, 10 HEPES, 0.5 EGTA, and 10 glucose (pH, 7.4).
After 60 min, myocytes were resuspended in Dulbecco’s modified Eagle’s medium (GIBCO-BRL).
Myocytes were placed in a Perspex bath (volume 0.25 ml)
located on the stage of an inverted microscope (Diaphot 300;
Nikon) and superfused at a rate of 1 ml/min via gravity flow.
Solutions were stored in inverted 50-ml plastic syringes, the
ends of which marginally protruded through the floor of a
custom-built Perspex water jacket heated to 37.5–38°C.
Switching of solutions was executed by means of five miniature three-way solenoid valves (LFAA series; Lee). The Perspex bath was placed on an electrically heated aluminum
plate that, in turn, was attached to a Perspex microscope
stage insert. The temperature of the metal plate was monitored with an embedded thermistor and was maintained at
37°C with a feedback circuit (TC-324A heater controller;
Warner Instrument, Hamden, CT). Bath temperature was
continuously monitored via a second thermistor. Immediately before entering the bath, solutions passed through an
in-line heat exchanger. To minimize heat loss via the oilimmersion objective lens, a major thermal sink, a custommade and tightly fitting brass water jacket was placed over
the objective. The in-line heat exchanger, brass water jacket,
and Perspex water jacket were perfused with warmed water
(37.5–38°C) supplied by a temperature-controlled water bath
(B. Braun Melsungen, Melsungen, Germany).
Measurement of ⌬⌿ with fluorescent indicators JC-1 and
tetramethylrhodamine ethyl ester. Ventricular myocytes were
loaded with JC-1 by incubation at 37°C with solution containing 10 ␮g/ml indicator for 10 min (32). Fluorescence
experiments were subsequently performed with a Deltascan
4000 fluorescence system (Photon Technology International;
Photomed, Seefeld, Germany) that was coupled to the microscope and software controlled. JC-1 is a positively charged
carbocyanine derivative that is driven into mitochondria by
⌬⌿, and when it reaches a critical concentration, J aggregates are formed (24). When excited at 490 nm, the fluorescence emission of JC-1 can be split and simultaneously measured at wavelengths corresponding to its monomer (530 ⫾
15 nm) and J aggregate (⬎ 590 nm) forms (Fig. 2). In preliminary experiments, we noted that there was a delay between the fall in J aggregate fluorescence and the onset of the
AJP-Heart Circ Physiol • VOL
rise in the monomer signal when the ⌬⌿ of a JC-1-loaded
myocyte was dissipated (Fig. 2). This observation is consistent with the previous report that the monomer form of JC-1
is sensitive to a lower range of ⌬⌿ than the J aggregate form
(7). Hence, because JC-1 does not behave as a typical ratioable dye, we used J aggregate fluorescence to monitor ⌬⌿.
Myocytes were loaded with tetramethylrhodamine ethyl
ester (TMRE) by superfusion with a solution containing 1.2
␮M TMRE for 10–15 min. TMRE was excited at 555 nm, and
fluorescence was detected at ⬎590 nm. Because fluorescent
indicators of ⌬⌿ cannot be readily calibrated in intact cells
(7, 8), we used relative fluorescence changes to monitor ⌬⌿.
The uncoupler 2,4-dinitrophenol (DNP) was used to collapse
⌬⌿ and scale JC-1 and TMRE signals (7, 23).
Digitonin-permeabilized myocytes. Myocytes were superfused with Ca2⫹-free solution including 1 mM EGTA for 5
min before being permeabilized by 1- to 2-min exposure to
digitonin (15–20 ␮M) (18). Digitonin was directly added to an
“intracellular” solution composed of (mM) 10 NaCl, 105 KCl,
5 HEPES, 2 K2ATP, 6 EGTA, 4 Ca-EGTA, 12.7 MgCl2, 5
pyruvate, and 2 malate (pH adjusted to 7.2 with KOH). The
calculated free Mg2⫹ concentration ([Mg2⫹]) was 1 mM, and
the free [Ca2⫹] was 100 nM. The solution also contained 300
nM JC-1. Mitochondria were rendered orange-red fluorescent
by the presence of JC-1, indicating that the mitochondria
were well polarized.
Chemicals. LC fatty acids were solubilized with methyl-␤cyclodextrin at a ⬃1:6 molar ratio (Sigma). A major advantage of using methyl-␤-cyclodextrin was that LC fatty acidcontaining solutions did not foam when bubbled with gas.
However, at high concentrations (⬎10 mM), cyclodextrins are
capable of removing cholesterol and other lipid components
from membranes (16). This property, however, is diminished
when the cavities of cyclodextrin molecules are occupied (16).
In some cases, indicated accordingly, the Na salt form of fatty
acids was used. The Na salt form was directly dissolved in
solutions and sonicated for 10 min. Sodium hexanoate was
added directly.
Spectroscopy. Absorption and emission spectra were obtained with aqueous solutions of JC-1 (4 ␮M; 0.3% DMSO)
and TMRE (1 ␮M; 0.1% DMSO). The buffer solution contained 150 mM KCl and 10 mM HEPES (pH 8.2 with KOH).
To minimize bleaching, a moderately fast scan time was
used.
Fig. 2. Effect of membrane depolarization on JC-1 fluorescence in an
isolated myocyte. The uncoupler 2,4-dinitrophenol (DNP) was used
to dissipate the mitochondrial membrane potential while monomer
(530 nm) and J aggregate (590 nm) fluorescence was simultaneously
measured.
282 • APRIL 2002 •
www.ajpheart.org
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
H1497
Fig. 3. Stimulatory effect of oleate on
the rate of heat production of a trabecula. Resting heat rate (Hr) was determined by moving the trabecula out of
the recording chamber (as shown at
the beginning of the recording). Contraction-related heat production was
elicited by stimulating the trabecula
electrically at 2 Hz (as indicated).
When oleate (400 ␮M) was introduced
to the preparation, an increase in resting rate of heat production ensued.
Statistics. All results are expressed as means ⫾ SE. Statistical significance was determined with ANOVA. A P ⬍ 0.05
was considered significant. The number of preparations (n)
from which the data are obtained is indicated in parentheses.
RESULTS
LC fatty acids increase resting heat rate. The rate of
heat production of small isolated cardiac muscle preparations, an indicator of basal metabolism, was measured at high resolution with a microcalorimetric technique. Figure 3 shows a representative example of the
effect of oleate (C18:1) on resting heat rate of an isolated trabecula. Resting heat rate was determined by
transferring the preparation out of the recording chamber (Fig. 1), whereas contraction-related heat production was elicited by stimulating the trabecula at a rate
of 2 Hz (Fig. 3). Application of 400 ␮M oleate, solubilized with methyl-␤-cyclodextrin, increased resting
heat rate, whereas the same concentration of methyl␤-cyclodextrin alone had no effect on resting heat rate
(n ⫽ 10), as illustrated in Fig. 4. On average, 400 ␮M
oleate increased heat rate by 29.3 ⫾ 0.9% (n ⫽ 26).
When oleate was supplied as sole metabolic substrate,
the introduction of the ␤-oxidation blocker 3-mercaptopropionic acid (2 mM; Ref. 26) reduced resting heat
rate by 27 ⫾ 2.6% (n ⫽ 6; not shown). Resting heat rate
was unaffected by 3-mercaptopropionic acid when glucose was provided as metabolic substrate.
The effect of various concentrations of the LC fatty
acids oleate and linoleate (C18:2) on resting heat rate
are summarized in Fig. 5. Trabeculae were superfused
with fatty acid solutions for at least 10 min. At concentrations of 100 and 200 ␮M, neither oleate nor linoleate
increased resting heat rate. At 300 ␮M, oleate and
linoleate increased resting heat rate by 5.1 ⫾ 0.8% (n ⫽
20) and 4.8 ⫾ 0.6% (n ⫽ 10), respectively. At the
highest concentration tested (400 ␮M), oleate increased resting heat rate by ⬃30% and linoleate produced a 25.8 ⫾ 0.9% (n ⫽ 20) increase. For comparison
with LC fatty acids, we examined the effect of the
short-chain fatty acid hexanoate (C6:0) on heat rate.
Hexanoate (1 mM) reversibly increased heat rate by
39.9 ⫾ 7.3% (n ⫽ 4; not shown).
In light of previous work with isolated mitochondrial
preparations, the most likely mechanism for the increase in resting heat rate evoked by LC fatty acids is
a protonophoric uncoupling action. We therefore examined whether oleate and linoleate depolarized the mitochondrial membrane of isolated myocytes under similar experimental conditions (37°C).
LC fatty acids depolarize inner mitochondrial membrane. ⌬⌿ was initially measured in ventricular myocytes with the fluorescent indicator JC-1. To reduce
photobleaching, JC-1 was intermittently excited (as
illustrated in Fig. 6). Cyclodextrin had no significant
effect on ⌬⌿. However, when the superfusate was
switched to a solution containing oleate (solubilized
with the same concentration of cyclodextrin), a decrease in ⌬⌿ was observed (Fig. 6A). Similar results
were observed in four other myocytes. At the end of
experiments, we used DNP to dissipate ⌬⌿ and scale
the JC-1 fluorescence signals. On average, oleate decreased JC-1 fluorescence to 45.3 ⫾ 5.6% (n ⫽ 5) of the
maximal response produced by DNP. Hexanoate (1
mM; C6:0) had no effect on ⌬␺ (Fig. 6B).
Fig. 4. Effect of cyclodextrin on resting
heat rate. The trabecula was superfused with cyclodextrin for 30 min. No
effect of 2.4 mM cyclodextrin on resting
heat rate was observed. For clarity,
only the first 8 min of exposure are
shown. After washout of cyclodextrin,
the amplitude of contraction-related
heat rate was similar to the preexposure value.
AJP-Heart Circ Physiol • VOL
282 • APRIL 2002 •
www.ajpheart.org
H1498
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
Fig. 7. Effects of oleoyl-CoA on J aggregate fluorescence in a digitonin-permeabilized ventricular myocyte. Application of oleoyl-CoA (10
␮M) had no effect on J aggregate fluorescence, an index of ⌬⌿.
Fig. 5. Summary of dose-dependent effects of oleate (E) and linoleate
(‚) on resting heat rate. An increase in resting heat rate was only
observed at concentrations of 300 and 400 ␮M.
Digitonin-permeabilized myocytes. Although the
most likely explanation for the decrease in ⌬⌿ produced by oleate is that it exerts a protonophoric uncoupling action, activation of mitochondrial ATP-sensitive
K⫹ (KATP) channels by its CoA derivative (oleoyl-CoA)
could also play a role. We recently showed (21) that
oleoyl-CoA potently activates the cardiac sarcolemmal
Fig. 6. Effects of oleate and hexanoate on J aggregate fluorescence in
JC-1-loaded myocytes. A: oleate (400 ␮M), solubilized with methyl␤-cyclodextrin, produced a decrease in J aggregate fluorescence,
whereas cyclodextrin alone had no effect. DNP (100 ␮M) was used to
collapse mitochondrial membrane potential (⌬⌿). B: the short-chain
fatty acid hexanoate had no effect on J aggregate fluorescence.
AJP-Heart Circ Physiol • VOL
KATP channel. Because oleoyl-CoA is membrane impermeant, we examined its effects on JC-1 fluorescence in
permeabilized myocytes. Figure 7 shows that a high
concentration of oleoyl-CoA (10 ␮M) did not depolarize
the mitochondria. Comparable results were obtained in
four other permeabilized myocytes. To confirm that
metabolism of oleoyl-CoA was not masking a decrease
in ⌬⌿ (14), we also found that oleoyl-CoA did not
decrease ⌬⌿ in the presence of 3-mercaptopropionic
acid. Hence, it is unlikely that oleoyl-CoA-mediated
activation of mitochondrial KATP channels contributes
to the observed membrane depolarization evoked by
oleate.
In vitro effects of LC fatty acids on indicators JC-1
and TMRE. We tested whether the large decrease in J
aggregate fluorescence produced by oleate could reflect,
at least in part, a direct interaction of oleate with the
indicator. Figure 8A shows an absorption scan of JC-1
(4 ␮M; 0.3% DMSO) in 0.15 M KCl solution (pH 8.2).
The absorption spectrum shows two peaks that correspond to the monomer (peak 500 nm) and J aggregate
(peak 595 nm) forms of JC-1. Addition of 40 ␮M Na
oleate produced a dramatic change in the fluorescence
spectrum (excitation wavelength 488 nm). The loss of
distinct peaks and resulting broad-based spectrum are
probably due to the disruption of J aggregates and the
formation of various JC-1 oligomers. The corresponding emission spectra are shown in Fig. 8B. Under
control conditions, a peak is seen at 591 nm, which
corresponds to J aggregate fluorescence. When 40 ␮M
Na oleate was added, J aggregate fluorescence intensity was decreased to 7% of the control value and a
peak occurred at 534 nm, probably due to either JC-1
monomers or a predominant oligomer. Similar to
oleate, linoleate (40 ␮M) decreased J aggregate fluorescence to 20% of control (not shown). The carboxyl
group of LC fatty acids is unlikely to be responsible for
this effect because addition of acetic acid (40 ␮M) had
no effect on JC-1 fluorescence. When Na oleate was
added to JC-1 at a ratio of 1:1, the decrease in fluorescence was ⬍10%. Hence, between a oleate-to-JC-1 ratio
of 1:1 and a ratio of 10:1 a large decrease in J aggregate
fluorescence occurs. Although we do not know the mitochondrial concentration of JC-1 in myocytes, the critical concentration for J aggregate formation was pre-
282 • APRIL 2002 •
www.ajpheart.org
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
H1499
Fig. 8. In vitro effects of oleate on JC-1
(A and B) and tetramethylrhodamine
ethyl ester (TMRE; C and D) spectra.
The ordinate is expressed in arbitrary
units (a.u.). Addition of 40 ␮M oleate
(⫹oleate) modified the JC-1 absorption
spectrum (A) and decreased J aggregate fluorescence (B). TMRE peak absorption was red-shifted by 8 nm by
addition of 40 ␮M oleate (C), and its
fluorescence emission intensity was reduced (D) (excitation wavelength 550
nm). Cyclodextrin alone (⫹cyclo) and
cyclodextrin-oleate (400 ␮M; ⫹cyclooleate) reduced TMRE fluorescence
emission to the same extent (D).
viously estimated to be 0.26 ␮M under in vitro
conditions (pH 7.2; 37°C) and 0.16 ␮M for suspended
cardiac mitochondria (24).
Because our in vitro observations complicated the
interpretation of in vivo JC-1 experiments, we tested
whether oleate affected the spectral properties of the
rhodamine-based indicator TMRE. Figure 8C shows
the absorption spectrum of 1 ␮M TMRE. Addition of 40
␮M oleate shifted the absorption peak from 552 to 560
nm. The corresponding emission spectrum (excitation
wavelength 550 nm) shows that oleate also reduced
peak fluorescence (Fig. 8D). Linoleate (40 ␮M) had
effects comparable to those of oleate. We also found
that cyclodextrin-solubilized oleate (400 ␮M) decreased
TMRE fluorescence; however, this effect could be accounted for by cyclodextrin alone (Fig. 8D). In conclusion, LC fatty acids are capable of decreasing the fluorescence intensity of the ⌬⌿ indicators JC-1 and
TMRE.
Effects of LC fatty acids on ⌬⌿ in myocytes assessed
with TMRE. In myocytes loaded with TMRE, fluorescence has been reported either to increase or to decrease after membrane depolarization, the difference
probably reflecting the extent of dye loading (10). We
found that depolarization of the mitochondrial membrane by DNP consistently elicited a large and reversible increase in fluorescence after cells had been loaded
by 10- to 15-min incubation with 1.2 ␮M TMRE. Thus
any changes in membrane potential induced by LC
fatty acids may, in principle, be underestimated because of direct interaction of matrix fatty acids with
the dye (Fig. 8). However, this confounding effect
would be minimal because the matrix free concentration of fatty acid is probably in the low micromolar
range.
AJP-Heart Circ Physiol • VOL
Neither oleate (40–80 ␮M) nor linoleate (40 ␮M),
whether supplied as the Na salt (n ⫽ 4) or in the
presence of methyl-␤-cyclodextrin (n ⫽ 3), affected ⌬⌿
in myocytes. At higher concentration, however, cyclodextrin-solubilized oleate (400 ␮M) increased TMRE
fluorescence by 22.1 ⫾ 2% (n ⫽ 5) of maximum, consistent with a decrease in ⌬⌿. A representative example
is shown in Fig. 9A. Cyclodextrin alone had no effect
(not shown), whereas it decreased TMRE fluorescence
in vitro (Fig. 8). This may be explained by the fact that
cyclodextrin is membrane impermeant. Similar to
oleate, application of linoleate (400 ␮M; Fig. 9B) increased TMRE fluorescence by 17.8 ⫾ 1.8% of maximum (n ⫽ 5).
DISCUSSION
We have shown that LC fatty acids increase basal
metabolism (resting heat rate) and depolarize the
inner mitochondrial membrane of intact cardiac
muscle. These stimulatory effects were observed at
total fatty acid concentrations of 300–400 ␮M,
within the range of 200–1,000 ␮M reported for biological fluids (19, 33). In agreement with our calorimetric results, Challoner and Steinberg (2) previously reported that palmitate (C16:0), solubilized
with albumin, enhances the rate of oxygen consumption of isolated rat hearts arrested by an increase in
extracellular potassium concentration. When methyl-␤-cyclodextrin is used as carrier, we do not know
the rate of cellular uptake of LC fatty acids, which is
thought to depend largely on the unbound concentration (1). However, when cyclodextrin-solubilized
oleate was the sole source of substrate for ventricular trabeculae, addition of the ␤-oxidation inhibitor
282 • APRIL 2002 •
www.ajpheart.org
H1500
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
Fig. 9. Measurement of ⌬⌿ in a TMRE-loaded myocyte. A: depolarization of ⌬⌿ with DNP produced a reversible increase in
TMRE fluorescence. Application of oleate (400 ␮M) evoked an
increase in TMRE fluorescence, consistent with a decrease in ⌬⌿.
B: linoleate (400 ␮M) similarly increased TMRE fluorescence. The
membrane could be further depolarized by a second application of
DNP.
3-mercaptopropionic acid decreased heat rate, indicating that the fatty acid was indeed taken up.
During prolonged ischemia (⬎20 min) and after
reperfusion, the intracellular concentration of LC fatty
acids is known to increase significantly (6, 33), probably because of lipase-catalyzed release of acyl groups
from phospholipids and endogenous triacylglycerols.
For example, in the glucose-perfused heart, tissue fatty
acid content has been shown to increase from ⬃70
nmol/g dry wt (preischemia) to ⬃290 nmol/g dry wt
after 30-min ischemia and to remain elevated during
reperfusion (6). When fatty acid content exceeded a
threshold value of ⬃400 nmol/g dry wt, reperfused
hearts did not recover any contractile function. Our
results suggest that the high rate of energy metabolism that follows reperfusion of postischemic myocardium (20) is due, at least in part, to the accompanying increase in intracellular LC fatty acid
concentration.
The ability of LC fatty acids to stimulate resting
heat rate of intact cardiac muscle is probably due to
a protonophoric uncoupling mechanism, as suggested by work with isolated mitochondria. Hütter
and Soboll (15) have also speculated that the increase in oxygen consumption evoked by LC fatty
acids in the intact heart is related to futile cycles
such as extramitochondrial ␤-oxidation or uncoupling of oxidative phosphorylation. The latter effect,
uncoupling, is thought to involve cyclic mitochondrial influx of fatty acid (R-COOH) and efflux of its
anionic (R-COO⫺) form (27, 31). Translocation of the
less membrane-permeant fatty acid anion across the
AJP-Heart Circ Physiol • VOL
inner membrane is rate limiting and has been deduced to be catalyzed by the ADP/ATP carrier and/or
other inner mitochondrial membrane proteins such
as uncoupling protein (5, 17, 30). The LC fatty acids
probably flip-flop between the inner and outer face of
the bilayer, delivering protons to the matrix (11), as
schematically illustrated in Fig. 10. Whether such an
uncoupling mechanism occurs in the intact cell has
been a source of controversy (14, 35). However, consistent with uncoupling, we have shown that LC
fatty acids decrease ⌬⌿ in intact cardiac muscle
when supplied at sufficiently high concentration.
The mechanism by which hexanoate increases heat
rate without affecting ⌬⌿ is probably via a futile
intramitochondrial ATP-consuming cycle of hexanoyl-CoA synthesis and hydrolysis (28).
We also revealed that LC fatty acids directly affect
the fluorescence properties of the commonly used
indicators JC-1 and TMRE. If not recognized, this
interaction might lead to misinterpretation of their
fluorescence signals. The mechanism by which LC
fatty acids disrupt J aggregates in vitro (Fig. 8) is
probably comparable to the recently described effects
of cationic and anionic surfactants on the carbocyanine derivative C803 (34). It should also be noted
that Rottenberg and Hashimoto (25) have speculated
that positively charged lipophilic dyes, which include
JC-1, may directly complex with fatty acid anions
(R-COO⫺) and thereby catalyze their cyclic uncoupling action. This is unlikely to be the case, because
it is improbable that the carboxyl group of fatty acids
complex with carbocyanine derivatives, and, moreover, we were unable to demonstrate any in vitro
effect of acetic acid (CH3-COOH) on JC-1 spectra.
In conclusion, our calorimetric and photometric
measurements suggest that LC fatty acids, at sufficiently high concentration, depolarize the inner mitochondrial membrane and stimulate basal metabolism of intact cardiac muscle. These observations
may partially explain the increased energy expenditure of reperfused postischemic myocardium because
LC fatty acid concentration increases markedly during ischemia.
Fig. 10. Schematic diagram showing the proposed mechanism by
which long-chain fatty acids dissipate the proton gradient of the
inner mitochondrial membrane and thereby produce heat. Translocation of the fatty acid anion is thought to be facilitated by uncoupling protein (UCP; as shown) and/or the ADP/ATP carrier (not
shown).
282 • APRIL 2002 •
www.ajpheart.org
LONG-CHAIN FATTY ACIDS INDUCE MITOCHONDRIAL UNCOUPLING
We thank R. Graf, L. Krapp, E. Reisinger, G. Schlichthörl, and K.
Schneider for technical help.
This study was supported by the Deutsche Forschungsgemeinschaft (Grant Da177/7–3) and the P. E. Kempkes Stiftung.
19.
REFERENCES
20.
1. Berk PD and Stump DD. Mechanisms of cellular uptake of
long chain free fatty acids. Mol Cell Biochem 192: 17–31, 1999.
2. Challoner DR and Steinberg D. Effect of free fatty acid on the
oxygen consumption of perfused rat heart. Am J Physiol 210:
280–286, 1966.
3. Daut J and Elzinga G. Heat production of quiescent ventricular trabeculae isolated from guinea-pig heart. J Physiol (Lond)
398: 259–275, 1988.
4. Daut J and Elzinga G. Substrate dependence of energy metabolism in isolated guinea-pig cardiac muscle: a microcalorimetric
study. J Physiol (Lond) 413: 379–397, 1989.
5. Dedukhova VI, Mokhova EN, Skulachev VP, Starkov AA,
Arrigoni-Martelli E, and Bobyleva VA. Uncoupling effect of
fatty acids on heart muscle mitochondria and submitochondrial
particles. FEBS Lett 295: 51–54, 1991.
6. De Groot MJM, Coumans WA, Willemsen PHM, and van
der Vusse GJ. Substrate-induced changes in the lipid content of
ischemic and reperfused myocardium. Its relation to hemodynamic recovery. Circ Res 72: 176–186, 1993.
7. Di Lisa F, Blank PS, Colonna R, Gambassi G, Silverman
HS, Stern MD, and Hansford RG. Mitochondrial membrane
potential in single living adult rat cardiac myocytes exposed to
anoxia or metabolic inhibition. J Physiol (Lond) 486: 1–13, 1995.
8. Farkas DL, Wei M, Febbroriello P, Carson JH, and Loew
LM. Simultaneous imaging of cell and mitochondrial membrane
potentials. Biophys J 56: 1053–1069, 1989.
9. Gibbs CL. Cardiac energetics. Physiol Rev 58: 174–254, 1978.
10. Griffiths EJ. Mitochondria—potential role in cell life and death.
Cardiovasc Res 46: 24–27, 2000.
11. Hamilton JA. Fatty acid transport: difficult or easy? J Lipid
Res 39: 467–481, 1998.
12. Hanley PJ, Cooper PJ, and Loiselle DS. Effect of hyperosmotic perfusion on rate of oxygen consumption of isolated guinea
pig and rat hearts during cardioplegia. Cardiovasc Res 28: 485–
493, 1994.
13. Hanley PJ, Young AA, LeGrice IJ, Edgar SG, and Loiselle
DS. 3-Dimensional configuration of perimysial collagen fibres in
rat cardiac muscle at resting and extended sarcomere lengths.
J Physiol (Lond) 517: 831–837, 1999.
14. Hermesh O, Kalderon B, and Bar-Tana J. Mitochondria
uncoupling by a long chain fatty acyl analogue. J Biol Chem 273:
3937–3942, 1998.
15. Hütter JF and Soboll S. Role of fatty acid metabolites in the
development of myocardial ischemic damage. Int J Biochem 24:
399–403, 1992.
16. Irie T and Uekama K. Pharmaceutical applications of cyclodextrins. III. Toxicological issues and safety evaluation. J Pharm
Sci 86: 147–162, 1997.
17. Jezek P, Modrianský M, and Garlid KD. A structure-activity
study of fatty acid interaction with mitochondrial uncoupling
protein. FEBS Lett 408: 166–170, 1997.
18. Köhnke D, Schramm M, and Daut J. Oxidative phosphorylation in myocardial mitochondria “in situ”: a calorimetric study on
21.
AJP-Heart Circ Physiol • VOL
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
H1501
permeabilized cardiac muscle preparations. Mol Cell Biochem
174: 101–113, 1997.
Kotani A, Fuse T, and Kusu F. Determination of plasma free
fatty acids by high-performance liquid chromatography with
electrochemical detection. Anal Biochem 284: 65–69, 2000.
Lerch R. Oxidative substrate metabolism during postischemic
reperfusion. Basic Res Cardiol 88: 525–544, 1993.
Liu GX, Hanley PJ, Ray J, and Daut J. Long-chain acylcoenzyme A esters and fatty acids directly link metabolism to
KATP channels in the heart. Circ Res 88: 918–924, 2001.
Loiselle DS and Gibbs CL. Species differences in cardiac
energetics. Am J Physiol Heart Circ Physiol 237: H90–H98,
1979.
Minezaki KK, Suleiman MS, and Chapman RA. Changes in
mitochondrial function induced in isolated guinea-pig ventricular myocytes by calcium overload. J Physiol (Lond) 476: 459–
471, 1994.
Reers M, Smith TW, and Chen LB. J-aggregate formation of a
carbocyanine as a quantitative fluorescent indicator of membrane potential. Biochemistry 30: 4480–4486, 1991.
Rottenberg H and Hashimoto K. Fatty acid uncoupling of
oxidative phosphorylation in rat liver mitochondria. Biochemistry 25: 1747–1755, 1986.
Sabbagh E, Cuebas D, and Schulz H. 3-Mercaptopropionic
acid, a potent inhibitor of fatty acid oxidation in rat heart
mitochondria. J Biol Chem 260: 7337–7342, 1985.
Schönfeld P, Wieckowski MR, and Wojtczak L. Long-chain
fatty acid-promoted swelling of mitochondria: further evidence
for the protonophoric effect of fatty acids in the inner mitochondrial membrane. FEBS Lett 471: 108–112, 2000.
Schönfeld P, Wojtczak AB, Geelen MJH, Kunz W, and
Wojtczak L. On the mechanism of the so-called uncoupling
effect of medium- and short-chain fatty acids. Biochim Biophys
Acta 936: 280–288, 1988.
Schramm M, Klieber HG, and Daut J. The energy expenditure of actomyosin-ATPase, Ca2⫹-ATPase and Na⫹,K⫹-ATPase
in guinea-pig cardiac ventricular muscle. J Physiol (Lond) 481:
647–662, 1994.
Simonyan RA and Skulachev VP. Thermoregulatory uncoupling in heart muscle mitochondria: involvement of the ATP/
ADP antiporter and uncoupling protein. FEBS Lett 436: 81–84,
1998.
Skulachev VP. Uncoupling: new approaches to an old problem
of bioenergetics. Biochim Biophys Acta 1363: 100–124, 1998.
Smiley ST, Reers M, Mottola-Hartshorn C, Lin M, Chen A,
Smith TW, Steele GD, and Chen LB. Intracellular heterogeneity in mitochondrial membrane potentials revealed by a Jaggregate-forming lipophilic cation JC-1. Proc Natl Acad Sci
USA 88: 3671–3675, 1991.
Van der Vusse GJ, Glatz JFC, Stam HCG, and Reneman
RS. Fatty acid homeostasis in the normoxic and ischemic heart.
Physiol Rev 72: 881–940, 1992.
Von Berlepsch H, Böttcher C, Ouart A, Regenbrecht M,
Akari S, Keiderling U, Schnablegger H, Dähne S, and
Kirstein S. Surfactant-induced changes of morphology of Jaggregates: superhelix-to-tubule transformation. Langmuir 16:
5908–5916, 2000.
Wojtczak L and Schönfeld P. Effect of fatty acids on energy
coupling processes in mitochondria. Biochim Biophys Acta 1183:
41–57, 1993.
282 • APRIL 2002 •
www.ajpheart.org