230
The Dependence of Electrophysiological
Derangements on Accumulation of Endogenous
Long-Chain Acyl Camitine in Hypoxic Neonatal
Rat Myocytes
Maureen T. Knabb, Jeffrey E. Saffitz, Peter B. Corr, and Burton E. Sobel
From the Cardiovascular Division and Department of Pathology, Washington University School of Medicine, St. Louis, Missouri
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
SUMMARY. To determine whether accumulation of long-chain acyl camitine contributes to
electrophysiological abnormalities induced by hypoxia, we characterized effects of normoxic and
hypoxic perfusion on the subcellular distribution of endogenous long-chain acyl camitine and
transmembrane potentials of cultured rat neonatal myocytes. Hypoxia increased long-chain acyl
camitine more than 5-fold. Sodium 2-[5-(4-chlorophenyl)-pentyl]-oxirane-2-carboxylate (10 JIM),
a camitine acyltransferase inhibitor, precluded accumulation of long-chain acyl camitine induced
by hypoxia. Tissue was processed for electron microscopy by a procedure specifically developed
for selective extraction of endogenous short-chain and free camitine but retention of endogenous
long-chain acyl camitine. In normoxic-perfused cells, long-chain acyl camitine was concentrated
in mitochondria and cytoplasmic membranous components. Only small amounts were present in
sarcolemma. Hypoxia increased mitochondrial long-chain acyl camitine by 10-fold and sarcolemmal long-chain acyl camitine by 70-fold. After 60 minutes of hypoxia, sarcolemma contained 1.4
X 107 long-chain acyl camitine molecules//un3 of membrane volume, a value corresponding to
approximately 3.5% of total sarcolemmal phospholipid. Hypoxia also significantly decreased
maximum diastolic potential, action potential amplitude and maximum upstroke velocity of
phase 0. Sodium 2-[5-(4-chlorophenyI)-pentyl]-oxirane-2-carboxylate inhibited accumulation of
long-chain acyl camitine in each subcellular compartment and prevented the depression of
electrophysiological function induced by hypoxia. These results strongly implicate endogenous
long-chain acyl camitine as a mediator of electrophysiological alterations induced by hypoxia.
(Circ Res 58: 230-240, 1986)
ENDOGENOUS cardiac amphiphiles, such as lysophosphoglycerides and long-chain acyl camitines
(LCA), accumulate in ischemic myocardium (IdellWenger et al., 1978; liedtke et al., 1978; Shug et al.,
1978; Sobel et al., 1978; Opie, 1979) and induce
electrophysiological alterations in vitro resembling
those seen in ischemic myocardium in vivo (Corr et
al., 1979, 1981, 1984; Arnsdorf and Sawicki, 1981;
Clarkson and Ten Eick, 1983). Their amphiphilic
properties may facilitate their incorporation into sarcolemma with consequent perturbation of ion transport (Katz and Messineo, 1981; Corr et al., 1984).
However, little direct information is available regarding the potential redistribution and accumulation of endogenous LCA in sarcolemma of hypoxic
cells. Furthermore, the extent to which electrophysiological derangements induced by hypoxia depend
on accumulation of endogenous amphiphiles in sarcolemma has not been elucidated. Accordingly, we
employed quantitative electron microscopic autoradiography to characterize the subcellular distribution and concentration of endogenous LCA in rat
neonatal myocytes superfused with normoxic or hypoxic solutions. Tissue was processed for electron
microscopy with a novel fixation procedure (Knabb
et al., 1985) designed to extract endogenous shortchain and free camitine selectively, but retain and
spatially fix endogenous LCA. To elucidate the dependence of electrophysiological changes on accumulation of LCA, cells were treated with phenylalkyloxirane carboxylic acid (POCA), an inhibitor of
camitine acyltransferase, to prevent accumulation of
LCA when cells were rendered hypoxic. Both transmembrane action potentials and concentrations of
LCA were measured in myocytes perfused under
normoxic, hypoxic, or hypoxic conditions with concomitant exposure to POCA. Electrophysiological
alterations induced by hypoxia correlated with accumulation of LCA.
Methods
Preparation and Perfusion of Myocytes
Cultured rat neonatal myocytes were chosen for this
study to permit uniform labeling of inrracellular camitine
pools with [3H]carnitine and to achieve a specific activity
of LCA necessary for electron microscopic autoradiography. Uniform labeling would be far more difficult to fulfill
Knabb et al./Long-Chain Acyl Carnitine and Hypoxia
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
in intact adult tissue. In addition, maintaining viable adult
myocytes over several days in culture to permit prelabeling
is difficult.
Monolayer cultures of spontaneously contracting rat
neonatal myocytes were prepared as previously described
(Ahumada et al., 1980). Briefly, hearts from 2-day-old rats
were removed aseptically and placed in sterile KrebsHenseleit solution. After atria and blood clots had been
removed, ventricles were minced and disaggregated during four successive 30-minute intervals in balanced salt
solution containing 0.1% collagenase type II (Worthington). Cells collected from the last three incubations were
cenrrifuged for 5 minutes (275 g) and resuspended in
Ham's F-10 medium with 20% fetal calf serum and 1%
penicillin/streptomycin. After filtration through Nitex 80
mesh to remove undissociated tissue, cells were plated at
1 X 10s cells/cm2 and incubated for 2 hours. Myocyteenriched supernatant fractions were transferred to a tube
containing [3H]d/-carnitine (Amersham, 5 /iCi/ml), mixed
thoroughly, and added to dishes containing glass culture
disks (3-4 X 10* cells/60-mm dish). Experiments were
performed after continuous culture in the presence of
[3H]carnitine for 3 days.
Prior to each experiment, cells were washed extensively
with buffer to remove unincorporated radioactivity. Myocytes were superfused at a flow rate of 5 ml/min for
1 hour with a modified Krebs-Henseleit solution containing (mM):NaCI, 128; KC1, 4; CaCl2, 1.2; MgClj, 1.0;
NaH2PO4, 0.9; NaHCO3, 22; glucose, 5. The solution was
equilibrated with 5% CO2 in room air (normoxia) or 95%
N2:5% CO2 (hypoxia). During hypoxia, the perfusion
chamber was covered with a bell jar so that the 95%
N2:5% CO2 atmosphere was maintained continuously over
the perfusate. Oxygen tension (P02) and pH were measured at the onset and conclusion of each experiment with
a 213 blood gas analyzer (Instrumentation Laboratories).
Temperature and pH in the perfusion chamber were maintained at 36°C and 7.2, respectively, in all experiments.
Under hypoxic conditions, P03 of the solution ranged
between 10 and 18 mm Hg. Cells were field stimulated at
120 beats/min with gold disk electrodes and 6-msec
square wave pulses at twice threshold.
In some experiments, sodium 2-[5-(4-chlorophenyl)pentyl]-oxirane-2-carboxylate (POCA, 10 MM, 97% pure,
kindly provided by Dr. Ludwig, ByK Gulden Pharmazeutika) was used to inhibit carnitine acyltransferase I activity
(Bartlett et al., 1981; Wolf et al., 1982, 1985). Thioesterificarion of this compound is required for expression of its
inhibitory activity (Wolf and Engel, 1985).
Biochemical Procedures
After superfusion for 1 hour with normoxic or hypoxic
solutions, prelabeled cells were scraped from glass disks
and homogenized in 7% (wt/vol) perchloric acid (PCA).
An aliquot was assayed for protein (Markwell et al., 1981).
The remainder was filtered through a 0.22-/im Millipore
filter. Radioactivity in acid-insoluble LCA retained on the
filter and in acid-soluble short-chain and free carnitine in
the filtrate was quantified with liquid scintillation spectrometry.
PCA extracts of unlabeled cells were assayed for longchain, short-chain and free carnitine radioisotopically
(McGarry and Foster, 1976). Assay conditions were modified slightly to account for changes in blank values or
slopes of the standard curve that may occur when LCA is
measured in small tissue samples. Blanks and standards
were prepared in neutralized PCA rather than H2O. In
231
addition, a small amount of free carnitine was added to
one of the triplicate determinations for each sample to
correct for any changes in the slope of the standard curve.
Preparation and Analysis of Autoradlographs
Six disks of myocytes prelabeled with [3H]carnitine were
perfused in either normoxic, hypoxic, or hypoxic + POCA
solutions for 1 hour. Following each perfusion, cells were
scraped from two disks for quantification of the distribution of radiolabeled carnitine. The remaining four disks
were processed for electron microscopy by a novel method
verified to spatially fix endogenous LCA with selective
extraction of short-chain and free carnitine (Knabb et al.,
1985). Two disks were utilized to assess labeled chemical
species removed from and retained by cells during processing for microscopy. The last two disks were embedded
in epoxy resin and sectioned for autoradiography.
Electron microscopic autoradiographs were prepared by
the flat substrate method of Salpeter and Bachmann
(1972). Pale gold sections (1000 A thick) were mounted
on collodian-coated slides, covered with monolayers of
Ilford L-4 emulsion, and incubated for 38 days (hypoxic
cells) or 80 days (normoxic and hypoxic + POCA treated
cells) in a dry N3 atmosphere at 4°C. Sections of unlabeled
cells were mounted on the same slides as labeled cells to
assess background grain densities.
Autoradiographs were developed in 1.1% p-phenylenediamine and 12.6% sodium sulfite and fixed in 30%
sodium thiosulfate as previously described (Neufeld et al.,
1985). Sections were collected on 200 mesh copper grids
and examined with a Phillips 200 electron microscope.
The use of tannic acid and uranyl magnesium acetate
during tissue processing provided adequate contrast so
that further staining of the sections was unnecessary.
Potential removal of developed grains caused by alkaline
lead citrate stains was avoided.
The physical developer p-phenylenediamine yields developed grains that are compact rather than filamentous,
thus minimizing obstruction of underlying fine structure.
This developer has a measured efficiency of approximately
1 grain per 15 disintegrations (Bachmann and Salpeter,
1967). The half-distance for this^utoradiographic system
(3H, Ilford L-4 emulsion, 1000 A thick sections,#p-phenylenediamine developer) is approximately 1500 A (Salpeter et al., 1978).
Grain distributions were analyzed by the 'mask overlay' technique of Salpeter et al. (1978), with the use of a
computer program by Land and Salpeter (1978). A final
magnification of 33,000 was selected to permit adequate
resolution of cellular structures. Because only a small
portion of an individual cell can be photographed in a
single frame, composite photographs of entire cells or large
portions of cells were made to eliminate biases associated
with photographing certain portions of selected cells. A
total of 224 composite photographs depicting 158 randomly selected myocytes (65 normoxic, 41 hypoxic, 52
hypoxic + POCA) containing 2293 grains (545 normoxic,
1208 hypoxic, 540 hypoxic + POCA) were analyzed.
Potential loci of radioactivity in cardiac myocytes included membranous organelles and non-membranous
structures (cytoplasm and nudeoplasm). Because of the
amphiphilic properties of long-chain acyl carnitine,
sources of radioactivity were expected to reside in membranes. Accordingly, membranous structures of the
myocyte were divided into the following compartments:
sarcolemma, mitochondria, and all remaining internal cytoplasmic membranes. The cytoplasmic membrane com-
232
Circulation Research/Vo/. 58, No. 2, February 1986
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
partment comprised predominantly sarcoplasmic reticulum which, in these neonatal cells, consisted of extensive
networks of rough endoplasmic reticulum lined with ribosomes. It included also smooth sarcoplasmic reticulum
and nuclear membranes. Other structures included in the
photographs and analyzed as separate compartments were
'intercellular space" and "degenerated cells." The latter
were encountered occasionally and, when situated adjacent to viable myocytes, were unavoidably included in the
composite photographs.
'Mask overlay' data defining transition probabilities
(the probability of finding a grain over structure y when
the radioactive source is located in structure x) and the
relative sizes of each compartment were obtained separately for normoxic, hypoxic, and hypoxic + POCA autoradiographs. There were no significant differences between these sets of data, indicating that each set of autoradiographs represented an equivalent sampling of cardiac
myocytes and that hypoxia did not induce major ultrastructural changes in the myocyte. This latter conclusion
was supported by careful inspection of the electron micrographs of normoxic and hypoxic myocytes, which were
ulrrastructurally indistinguishable. Grain densities were
therefore determined by analyzing each unique set of
grain locations in autoradiographs from normoxic, hypoxic, and hypoxic + POCA-rreated cells with the combined 'mask overlay' data from all three sets of experiments.
Electrophysiological Measurements
Transmembrane action potentials were recorded with
glass capillary microelectrodes (30-40 Mfi, DC resistance)
filled with 3 M KC1 connected to a WPIM4-A electrometer.
Signals were processed through two stages of amplification before storage on FM analog tape. A WPI precision
millivolt source was used for in series voltage calibration.
Cells were field stimulated at 120 beats/min with gold
disc electrodes placed at opposite sides of the perfusion
bath. Stimulus duration (6 msec) and intensity (twice
diastolic threshold) were not altered during the interval of
hypoxia. Action potential recordings were analyzed
off-line with an automated system described previously
(Witkowski and Corr, 1978). Impalements of several prelabeled cells (3-8 cells/disk) were made in the 30- to 60minute interval during perfusion under each set of conditions. Cells were extracted immediately for analysis of
radiolabeled carnitine after perfusion for 1 hour.
The effects of POCA on intracellular recordings during
normoxic and hypoxic perfusion were tested with single,
maintained impalements in several unlabeled cells. No
biochemical measurements were made in these experiments.
Statistical Methods
Results are reported as means ± SE. Significant differences (P < 0.05) were determined with the use of the nonpaired Student's t test.
Results
Accumulation of Long-Chain Acyl Carnitine in
Hypoxic Myocytes
Myocytes prelabeled with [3H]carnitine for 3 days
prior to perfusion for 1 hour under normoxic conditions contained 4.5% of label in LCA (Fig. 1A).
Hypoxia resulted in more than a 5-fold increase in
A. PRELABELED MYOCYTES
600
400
200
a
-o
NORMOXIC
(n-31)
HYPOXIC
(n=31)
B. UNLABELED MYOCYTES
HYPOXIC + POCA
(n=31)
ZZZ2 S C + F
^
LCA
1200r —Lc
e
2 800
a
e
"5 400
E
a
HYPOXIC + POCA
(n'5)
FIGURE 1. Distribution of carnitine in myocytes prelabeled with
pHJcamitine (panel A, above) or in unlabeled myocytes (panel B,
below) after superfusion for 1 hour with normoxic, hypoxic, or hypoxic
+ POCA (10 HM) solutions. Long-chain acyl carnitine (LCA) accounted
for 4.5% of total carnitine (T) after normoxic perfusion, 24.1% after
hypoxic perfusion, and 1.9% after hypoxic perfusion with phenylalkytoxirane carboxylic acid (POCA) in prelabeled cells. Unlabeled
myocytes contained 5.7% of total carnitine after normoxic perfusion,
35.5% after hypoxia, and 1.8% after hypoxia with POCA. Results are
means ± SE f — Significantly different from normoxic and hypoxic +
POCA short-chain and free carnitine (SC+F) levels; » •= significantly
different from normoxic ayd hypoxic + POCA cells; •• = significantly
different from normoxic and hypoxic cells.
LCA. The increase was blocked completely by
POCA.
Endogenous LCA comprised 5.7% of total carnitine in normoxic, perfused cells (Fig. IB). LCA increased more than 6-fold with hypoxia. POCA completely inhibited the accumulation induced otherwise by hypoxia. Thus, the distribution of radiolabeled LCA in specific chemical species corresponds
to that of endogenous LCA. Both radiolabeled and
endogenous LCA levels change with hypoxia and
are dependent upon the presence or absence of
POCA.
Autoradiographic Analysis
For autoradiographic studies, cells on six disks
were perfused under each set of conditions. Radioactivity was extracted and assayed from two disks
before processing (Fig. 2A) and from two after processing (Fig. 2B). The remaining two disks were processed fully and embedded for autoradiography. No
protein measurements were possible after tissue
Knabb et al./Long-Chain Acyl Carnitine and Hypoxia
233
mately 55% of total photographic area and 68% of
myocyte area, contained only 20% to 25% of total
myocyte radioactivity. The mitochondria and internal cellular membranes had the highest specific
activity and contained the bulk of myocyte radioactivity. In normoxic cells, the sarcolemma contained little LCA. However, in cells subjected to 60
A. BEFORE FIXATION
800
^600
2
x 400
a200
0
NORMOXIC
HYPOXIC
B. AFTER FIXATION
HYPOXIC+POCA
WZA SOF
I M LCA
150r
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
NORMOXIC
HYPOXIC
HYPOXIC+POCA
FIGURE 2. Distribution of radiolabeled carnitine in perfused myocytes
before (panel A, above) and after (panel B, below) processing for
electron microscopic autoradiography. Values represent averages from
two samples for each perfusion. Before processing, LCA comprised
4.9% of total carnitine after normoxic perfusion, 25% after hypoxic
perfusion, and 3.5% after hypoxic perfusion with POCA. After processing, LCA accounted for 88.6% of total dpm/'disk in normoxicperfused cells, 96.6% in hypoxic cells, and 86.4% in hypoxic cells
with POCA. The processing procedure selectively extracts short-chain
and free carnitine (SC + F) with retention of LCA.
processing and results are expressed in total dpm/
disk. Before fixation, the distribution of labeled LCA
was comparable to that in experiments summarized
in Figure 1. Processing resulted in selective extraction of endogenous short-chain and free carnitine
under all experimental conditions with endogenous
LCA comprising 86% to 97% of retained radioactivity (Fig. 2B). Selected autoradiographs of myocytes
perfused under normoxic, hypoxic, and POCAtreated conditions are shown in Figure 3.
Table 1 lists the computer-generated grain densities and the percentage of total photographic area
of each compartment analyzed by the mask overlay
method. The grain densities, expressed as grains/
100 nm2 of photographic area, represent relative
specific activities reflecting the amount of radioactivity per unit area of each structure. Figure 4 illustrates the proportion of total myocyte radioactivity
in each compartment, determined by multiplying
the grain density by the compartment size in each
case. Unlabeled cells demonstrated negligible background grain density (approximately 0.01 grains/
100 M"I2) at the longest exposure interval.
As shown in Table 1 and Figure 4, radioactivity
was concentrated in membranous structures of cardiac myocytes. The non-membranous myocyte cytoplasm and nucleoplasm, comprising approxi-
FIGURE 3. Autoradiographs of cultured neonatal rat myocytes labeled
with [3H]carnitine for 3 days followed by perfusion for 60 minutes
under hypoxic (panel A), normoxic (panel B), or hypoxic + POCA
(panel O conditions. The cells are identified as myocytes by the
presence of myofibrils (MF). The cells subjected to hypoxic perfusion
conditions are ultrastructurally normal and indistinguishable from
the normoxic cell. Autoradiographic grains (dense black particles) are
concentrated over mitochondria (M). In the hypoxic cell (panel A),
some grains are noted in close proximity to the sarcolemma (arroivs).
The nonmembranous cytoplasm and nucleus (N) contain few grains.
Panel A, x 15,500, bar = 1.0 urn; panel B, x 16,700, bar = 1.0 urn;
panel C x 15,000, bar = 1.0 urn.
234
Circulation Research/Vol. 58, No. 2, February 1986
TABLE 1
Computer-Calculated Grain Densities in Each Tissue Compartment
Grain density (mean ± SEM)*
(grains/100 MHI2)
Compartment
Photographic
area (%)
Sarcolemma
Cytoplasmic membranes
Mitochondria
Nucleus
Cytoplasm
Degenerated cells
Intercellular space
1.9
10.0
14.3
14.4
40.4
8.6
10.4
Total photographic area
100.0
Normoxia
9.2
54.7
36.0
8.1
5.2
3.7
0.2
± 17.5
±8.5
± 3.9
± 1.5
± 2.0
± 1.7
±1.6
14.3f
Hypoxia
638.3 ± 123.6
244.1 ± 38.8
331.4 ± 20.4
29.3 ±5.6
43.6 ± 10.3
12.3 ± 8.1
0
105.9J
Hypoxia +
POCA
16.6
43.3
18.4
7.9
5.0
4.1
±15.1
± 6.6
± 2.6
± 1.3
±1.6
±1.5
0
10.8f
• Error values shown represent the "standard error' of the xJ minimization routine of the analysis
program (Land and Salpeter, 1978).
t These numbers represent the overall grain densities for each perfusion condition.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
minutes of hypoxia, the specific activity of LCA in
the sarcolemma increased 70-fold from 9.2 to 638
grains/100 /an2 and the proportion of total LCA in
sarcolemma rose from 1.5% to 11.5%. Mitochondria
exhibited a nearly 10-fold increase and the internal
cellular membranes a 4-fold increase of LCA with
hypoxia. The overall grain density of hypoxic my-
z
100
SL
8 80
CYTO
MEMBRANES
CYTO
MEMBRANES
CYTO
MEMBRANES
ocytes was approximately 7.4 times that of normoxic
cells. This value, derived from autoradiographic
measurements, corresponds closely to the 5-fold
increase of labeled LCA measured chemically in cells
before tissue processing (Fig. 2). The substantial
increases of LCA induced by hypoxia in membranous structures were blocked completely by POCA.
Figure 5 shows the number of endogenous LCA
molecules contained per unit volume of each compartment calculated from the computer-generated
grain densities according to the formula: M = (GdA)/
(ES), where M = total number of molecules/Mm3;
G = grains/^m3 calculated by multiplying computergenerated grain densities (grains/10 jim2) by the
section thickness (0.1 nm); d = the inverse of emul-
150
60
<
MITO
LU
MITO
MITO
40
E
a. 100
U
NUC
g 20k
NUC
NUC
50
CYTO
0
CYTO
CYTO
d
L
NORMOXIC
HYPOXIC
HYPOXIC
POCA
FIGURE 4. The percentage of total grains located in each source
compartment under conditions of normoxia, hypoxia, and hypoxia
plus POCA perfusion. Grains are found predominantly in myocyte
membranous compartments: the mitochondria (MITO), cytoplasmic
membrane structures (CYTO MEMBRANES), and sarcolemma (SI).
Some grains are also found in nonmentbranous cytoplasm (CYTO)
and nucleus (NUC). A redistribution of grains induced by hypoxia is
reflected by the large accumulation of LCA in sarcolemma (1.5%
under normoxic conditions compared with 11.5% under hypoxic
conditions).
N HP
CYTO
N H P
N H P N H P
NUC
MITO
CMS
COMPARTMENT
N H P
SL
FIGURE 5. LCA molecules/pm* of each source structure following
normoxic (N), hypoxic (H), or hypoxic plus POCA (P) perfusion.
Hypoxia resulted in an increased number of LCA molecules/urn3 in
each source compartment. POCA CIO JIM> prevented accumulation of
LCA. Source compartments are cytoplasm (CYTO), nucleus (NUC),
mitochondria (MITO), cytoplasmic membrane structures (CMS), and
sarcolemma (SL).
Knabb et al. /Long-Chain Acyl Carnitine and Hypoxia
235
3
endogenous LCA molecules/jtm (obtained from
autoradiographic measurements) by 4 X 108 phospholipid molecules/Vm3. The latter value was obtained by assuming that a pair of phospholipids in
a bilayer forms a cylinder 8 A in diameter and 100
A in length (Hauser et al., 1981; Smaby et al., 1983)
and, hence, occupies a volume of approximately 5
X 103 A3 which corresponds to 4 X 108 molecules/
nm3. LCA concentrations expressed in nmols/mg
protein were calculated based on measurements of
the ratio of phospholipid to protein in purified adult
canine sarcolemma [2.7 jtmols phospholipid/mg
protein (Gross, 1984)] and sarcoplasmic reticulum
[1.4 ^mols phospholipid/mg protein (Gross, 1985)].
As shown in Table 2, sarcolemma from hypoxic
myocytes contained approximately 95 nmol of endogenous LCA/mg protein, corresponding to approximately 3.5% of total sarcolemmal phospholipid. It has been shown previously that incorporation of only approximately 1% of exogenous lysophosphatidylcholine into sarcolemmal phospholipids is sufficient to induce electrophysiological effects (Gross et al., 1982; Saffitz et al., 1984). Thus,
it is likely that the incorporation of 3.5% of endogenous LCA into sarcolemma observed in hypoxic
myocytes is capable of inducing electrophysiological
derangements.
TABLE 2
Concentrations of Long-Chain Acyl Carnitine in Sarcolemma
and Cytoplasmic Membranes
Sarcolemma
Cytoplasmic
membranes
1.9 X 105
1.2 X 10*
Normoxia
Molecules LCA/MM 3
% of membrane phospholipids
nmol LCA/mg protein
Hypoxia
Molecules LCA/Mm3
% of membrane phospholipids
nmol LCA/mg protein
0.05
1.3
0.3
4.2
1.4 x 107
3.5
95
5.3 X 106
1.3
19
2.0 x 10*
5.2 x 10*
0.13
1.8
Hypoxia + POCA
Molecules LCA/MM 3
% of membrane phospholipids
nmol LCA/mg protein
0.05
1.4
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
sion efficiency (15 decays/grain); A = Avogadro's
number; E = exposure time in minutes; S = specific
activity of the radioactive probe (normoxic cells =
373, hypoxic cells = 359, POCA treated cells = 647
dpm/pmol). As shown in Figure 5, with hypoxia the
LCA content of the myocyte sarcolemma increased
from 1.9 X 105 to 1.4 X 107 molecules/^m3 of
membrane volume.
Table 2 lists the concentrations of endogenous
LCA in the myocyte sarcolemma and internal cytoplasmic membranes. The concentration of LCA, expressed as a percentage of total membrane phospholipids, was calculated by dividing the number of
Electrophysiological Alterations with POCA and
Hypoxia
POCA added during normoxic perfusion significantly decreased LCA radioactivity from 22.7 to 9.4
TABLE 3
Electrophysiological and Biochemical Effects of POCA in Normoxic and Hypoxic Myocytes
Electrophysiological parameters
Group Condition (»)
1
1
2
2
2
3
3
3
Normoxia
Normoxia
+ POCA
Normoxia
Hypoxia
Hypoxia
+ POCA
Normoxia
Hypoxia
Hypoxia
+ POCA
MDP
(-mV)
4 78.9 ± 0.9
4 76.4 ± 2.3
AMP
(mV)
95.7 ±5,.8
96.3 4,.1
APD
(msec)
(msec)
86.1 ±9.5 118.0 ±8.3
82.9 ±7.8 115.7 ±5.9
Biochemical parameters
APD«
(msec)
(V/S)
Short-chain + Long-chain
free camitine acylcarnitine
(dpm//ig
(dpm/Mg
protein)
protein)
194.3 ± 5.5 104.8 ±5.9 768.9 ± 8.1
189.1 ± 4.3 108.3 ± 5.3 816.3 ± 75.3
22.7 ± 5.0
9.4 ± 0.8
4 82.8 ±2.0 104.4 ±3..2 107.3 ±7.3 132.9 ±9.6 232.3 ± 28.5 88.0 ± 15.9 991.4 ± 147.1 57.6 ± 3.1
6 55.1 ± 2.8* 58.9 ± 5.9* 131.9 ± 10.8 168.4 ± 12.2 237.9 ± 18.4 19.7 ± 5.6* 859.9 ± 91.8 89.5 ± 19.9
4 79.8 ±4.2 91.2 3..5 117.3 ±8.4 143.0 ± 12.0 209.8 ± 17.9 80.8 ± 15.5 865.7 ± 3.3
17.8±2.0t
5 77.5 2.2 94.8 2.0 100.8 ±4.1 125.3 ±6.1 212.5 ± 19.4 83.8 ± 9.1 785.2 ± 86.4 56.1 ± 6.4
7 28.0 ±1.1*
0*
0*
0*
1
0*
728.0 ± 54.9 132.0 ± 24.5*
6 66.8 ± 6.0 69.5 ± 7.•2t 135.7 ± 18.2 165.4 ± 22.2 213.1 ± 19.7 31.7 ± 9.9| 746.0 ± 76.5 16.7±1.6f
Data expressed as mean ± SE; MDP = maximum diastolic potential; AMP = amplitude; APDJO,7O,« •• action potential duration at 50,
70, or 95% repolarization; V ^ = maximum upstroke velocity. In group 1, electrophysiological measurements were made in the same
cell before and after administration of POCA (10 FIU) (n =• number of cells). Biochemical measurements were made in separate
experiments. In groups 2 and 3, electrophysiological parameters were measured in 3-8 cells from each disk (n = number of disks).
Biochemical parameters were measured in each disk. Group 2 included cells in which action potentials were generated after the hypoxic
interval and Group 3 included cells in which the resting membrane potential depolarized to —28 mV and electrical activity ceased.
* Significantly different (P < 0.05) from normoxic and hypoxic + POCA control values.
f Significantly different (P < 0.05) from normoxic and hypoxic values.
236
Circulation Research/Vol. 58, No. 2, February 1986
NORMOXIA + POCA
control 5'
10'
15'
20'
-80-»J
200 mj
HYPOXIA
control 5'
membrane potential was reduced to —28 mV and
the cells became unresponsive to stimulation concurrent with marked elevation of endogenous LCA
(Table 3; Fig. 7). Exposure of cells to POCA under
conditions of hypoxic perfusion prevented accumulation of LCA and depression of maximum diastolic potential, amplitude, and Vmax of phase 0.
Action potential parameters remained at control values in the presence of POCA when only moderate
elevations in LCA were measured in hypoxic cells
(Fig. 7A). However, with higher concentrations of
LCA induced by hypoxia, POCA did not completely
prevent depression of the amplitude or Vmix of phase
0, although the maximum diastolic potential did
-8fP
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
HYPOXIA + POCA
5'
110'
control
tl
0-
-80^
\
15'
\
u
V V V V. V
FIGURE 6. The effects of POCA 00 ntf on cardiac action potentials
during normoxic and hypoxic perfusion. A single impalement was
maintained for a 10-minute control period followed by a 20-minute
intervention period. Recordings from both hypoxic and hypoxic +
POCA-treated cells were made on the same group of cells.
dpm/Vg protein after 20 minutes in prelabeled cells,
but did not alter electrophysiologic parameters
(group 1, Table 3, and Fig. 6). In these experiments,
electrophysiologic measurements were made in the
same cell (n = 4) before and after POCA administration. Biochemical measurements were made in a
separate series of experiments, with prelabeled cells.
Transmembrane potentials of prelabeled myocytes perfused under normoxic, hypoxic, or hypoxic plus POCA conditions were measured in numerous cells during the last 30 minutes of perfusion.
After 1 hour, carnitine derivatives were extracted
and assayed for radioactivity. Hypoxia significantly
increased LCA radioactivity from 56.8 ± 3.4 to 112.5
± 16.9 dpm/jtg protein. LCA increased only 2- to 3fold with hypoxia, rather than 5-fold, because it was
difficult to maintain low P02 with microelectrodes
in the perfusion chamber (P02 range 16-20 mm Hg).
Results were stratified into two groups based on the
severity of electrophysiological depression induced
during hypoxia. In one group, hypoxia led to depolarization of the resting membrane potential but
action potentials were generated. In these cells there
was a moderate, though not significant, increase in
LCA (group 2, Table 3). In the other group, resting
membrane potential depolarized to —28 mV, electrical activity ceased, and a statistically significant increase in LCA was seen (group 3, Table 3).
Hypoxia resulted in a significant decrease in maximum diastolic potential, total amplitude, and V m ^
of phase 0 (Table 3; Fig. 7). In group 3, the resting
ti^
-100r
E.°
11
0)
-D
J
"a
E
<
>. 60"
O
20•
H+P
FIGURE 7. Biochemical and electrophysiological characteristics of
prelabeled myocytes perfused under the three conditions of normoxia
(N), hypoxia (H) and hypoxia plus POCA (H + P). Experiments were
divided into two groups based on the severity of electrical depression
with hypoxia. In the first group, moderate accumulation of LCA was
seen (panel A), and in the second group, more marked accumulation
of LCA was seen (panel B). A significant decrease in maximum
diastolic potential, amplitude, and V.,, of the action potential was
associated with moderate increases of LCA during hypoxia. POCA
prevented depression of these parameters. When cells became depolarized and electrical activity ceased, higher levels of LCA were
measured during hypoxia (panel B). POCA prevented severe electrophysiological derangements, but did not entirely normalize action
potentials; • = significantly different from normoxic and hypoxic +
POCA control values. •• = significantly different from normoxic and
hypoxic values.
Knabb et a/. /Long-Chain Acyl Carnitine and Hypoxia
237
remain at control, prehypoxic values. Thus, factors
other than accumulation of LCA appear to be involved in the electrophysiological depression associated with hypoxia.
tially partition into cytoplasmic membranes. (2) Mitochondrial LCA loosely bound to the outer mitochondrial surface (Idell-Wenger et al., 1978) may
partition into cytoplasmic membrane lipids or bind
to cytoplasmic membrane proteins. (3) LCA may be
formed in the cytoplasmic membrane compartment.
This possibility is unlikely, however, because longchain carnitine acyltransferase is believed to be associated only with mitochondria. Only short- and
medium-chain carnitine acyltransferase activities
have been associated with microsomes (Bieber et al.,
1982).
LCA levels increase more than 5-fold in hypoxic
myocytes. The greatest increases occur in the sarcolemma (70-fold) and mitochondria (10-fold). Our
autoradiographic results correspond to the calculated 9-fold elevation in mitochondrial and 5-fold
increase in nonmitochondrial LCA in severely ischemic perfused rat hearts (Idell-Wenger et al.,
1978) assayed after differential centrifugation of
subcellular components. Thus, LCA appears to be
synthesized and accumulate first in mitochondria,
and then is transported or diffuses out of mitochondria to other subcellular compartments. The sarcolemma appears to accumulate LCA preferentially,
judging from the marked elevation of LCA seen in
the sarcolemmal compartment.
When hypoxic cells are treated with POCA, the
concentration of LCA in all compartments remains
at or below normoxic control levels. The proportion
of total LCA declines in mitochondria (25%) and
increases in cytoplasmic membranes (42%) of
POCA-treated cells compared with normoxic or hypoxic myocytes. These differences probably reflect
inhibition of mitochondrial carnitine acyltransferase
by POCA.
In order to compare results from in vitro studies
on the effects of palmitoyl carnitine on membranebound enzymes in sarcolemma and sarcoplasmic
reticulum with those in this study, we calculated
corresponding values (Table 2) from the autoradiography data. Sarcolemmal LCA reached a concentration of 95 nmol/mg protein with hypoxia. This
concentration is not sufficient to inhibit Na+,K+ATPase in vitro (Wood et al., 1977; Adams et al.,
1979; Lamers et al., 1984; Kramer and Weglicki,
1985) but may be sufficient to inhibit Na+-Ca++
exchange (Lamers et al., 1984). Enzymes in sarcoplasmic reticulum appear to be more sensitive to
inhibition by exogenous palmitoyl carnitine than
those in sarcolemma, but the calculated concentration of LCA in sarcoplasmic reticulum with hypoxia
(19 nmol/mg protein) is not sufficient to inhibit
Ca++-ATPase (Pitts et al., 1978; Adams et al., 1979)
or Ca++ release (Adams et al., 1979). Interpretations
of comparisons between results from studies in vitro
with exogenous amphiphiles and those obtained in
this study must be cautious, because the kinetics
and extent of incorporation of exogenous LCA into
membrane preparations are not well characterized.
Furthermore, LCA may bind selectively to mem-
Discussion
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
The concentration of LCA increases in ischemic
myocardium in vivo (Idell-Wenger et al., 1978;
Liedtke et al., 1978; Shug et al., 1978; Opie, 1979).
Administration of exogenous LCA induces electrophysiological alterations in normoxic ventricular
muscle in vitro resembling changes seen in ischemic
cells in vivo (Corr et al., 1981, 1984). The present
observations demonstrate a 70-fold increase of endogenous LCA in sarcolemma of hypoxic myocytes.
Furthermore, they indicate that the increased LCA
in sarcolemma is associated with electrophysiological alterations analogous to those seen during ischemia in vivo. Both sarcolemmal accumulation of LCA
and the electrophysiological effects of hypoxia are
attenuated by POCA, an inhibitor of carnitine acyltransferase. Thus, our observations provide direct
evidence that accumulation of endogenous LCA
contributes to electrophysiological derangements induced by hypoxia.
Autoradiography requires retention and spatial
fixation of a chemically defined radioactive probe.
We have shown previously in myocyte suspensions
labeled with exogenous [14C]palmitoyl-L-carnitine
that the fixation procedure developed preserves 85%
of tissue radioactivity (Knabb et al., 1985). The procedure selectivity preserves endogenous LCA while
99% of short-chain and free carnitine is extracted
from myocytes prelabeled with [3H]carnitine. After
processing for electron microscopy, short-chain and
free carnitine accounted for less than 15% of total
radioactivity under any of the experimental conditions utilized.
Autoradiographic analysis of normoxic tissue
showed that most of the LCA was localized in
mitochondria and cytoplasmic membrane components. Little was in nucleus and nonmembranous
cytoplasm. Idell-Wenger et al. (1978) found no LCA
in mitochondria from freshly excised nonperfused
hearts, although they did detect 37% of total LCA
in mitochondria in control, perfused hearts. The
difference may be related to characteristics of perfused compared with nonperfused isolated hearts.
We localized a substantial proportion (39%) of
endogenous LCA in the internal cytoplasmic membranes of normoxic neonatal myocytes. Idell-Wenger et al. (1978) also found a substantial proportion
(63%) of endogenous LCA in nonmitochondrial
compartments, although their methods were not
designed to differentiate cytoplasmic compartments.
Several potential explanations may pertain. (1) Mitochondrial LCA can be exchanged for cytosolic
carnitine through the mitochondrial carnitine/acyl
carnitine translocase (Pande, 1975; Ramsey and
Tubbs, 1975). Cytosolic LCA would then preferen-
238
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
brane-bound proteins rather than distribute randomly throughout the bilayer.
Incorporation of endogenous LCA into sarcolemma (3.5% of phospholipid) appears to be sufficient to induce biophysical changes (Fink and Gross,
1984). As little as 1% incorporation of exogenous
lysophosphatidylcholine into sarcolemmal phospholipids induces electrophysiological effects resembling those seen in ischemic tissue in vivo (Gross et
al., 1982; Saffitz et al., 1984). Thus, the concentration of endogenous sarcolemmal LCA that we observed autoradiographically appears to be sufficient
to contribute to the electrophysiological abnormalities induced by the underlying hypoxia.
Administration of exogenous palmitoyl camitine
(100-200 /XM) decreases maximum diastolic potential, amplitude, Vm*x, and action potential duration
in canine Purkinje fibers (Corr et al., 1981). Hypoxia
or ischemia elicits similar decreases of resting potential, amplitude, and V m> of cardiac action potentials
(McDonald and MacLeod, 1973; Downar et al.,
1977; Carmeliet, 1978). Metabolic inhibition of glycolysis and oxidative phosphorylation in cultured
myocytes results in gradual loss of diastolic membrane potential and failure of depolarization (Clusin,
1983; Hasin and Barry, 1984).
The severity of electrophysiological depression
observed in this study was proportional to the magnitude of increase of LCA induced by hypoxia.
Modest decreases in maximum diastolic potential,
amplitude, and V ^ of phase 0 of the action potential were associated with a modest increase in LCA.
The insignificant rise of LCA associated with significant changes in the action potential is difficult to
interpret because we were not able to monitor electrophysiological and biochemical changes in the
same cell. Membrane depolarization and loss of a
generated action potential was associated with significant increases of LCA.
Although the precise mechanisms responsible for
the electrophysiological effects of LCA are unknown, the effects are similar to those seen with a
structurally similar amphiphile, lysophosphatidylcholine (Corr et al., 1981). Both amphiphiles decrease Vmax of phase 0 independent of a change in
membrane potential (Corr et al., 1979; Arnsdorf and
Sawicki, 1981). Thus, the decrease in V ^ in response to increasing LCA during hypoxia may be
due to a direct decrease in the magnitude of the
rapid sodium inward current and membrane depolarization. Since the extent of membrane depolarization induced by exogenous lysophosphatidylcholine and LCA is nearly identical (Corr et al., 1981),
it is possible that the mechanism of decreased outward K+ conductance demonstrated with exogenous
lysophosphatidylcholine in ventricular muscle
(Clarkson and Ten Eick, 1983) also occurs with
sarcolemmal accumulation of LCA. Although lysophosphatidylcholine appears to decrease the mag-
Circulation Research/VoJ. 58, No. 2, February 1986
nitude of the slow inward current (Lj) in ventricular
muscle (Clarkson and Ten Eick, 1983), palmitoyl
carnitine actually increases the magnitude of the Lj
in chick ventricular muscle depolarized to —60 mV
(Inoue and Pappano, 1983). Thus, despite the fact
that several of the electrophysiological effects of the
two amphiphiles are similar, major differences also
exist.
Carnitine acyltransferase inhibitors, such as 2tetraglycidic acid and oxfenicine, have been used to
assess potentially deleterious effects of fatty acid
oxidation in hypoxic hearts. Pearce et al. (1979)
found that 2-tetraglycidic acid decreased the size
and number of anoxic areas in hypoxic perfused rat
hearts. Apparently by inhibiting fatty acid oxidation,
oxfenicine improved mechanical function in ischemic rat (Higgins et al., 1980) and pig (Liedtke et
al., 1984) hearts. Carnitine itself improved mechanical and electrophysiological function in ischemic
(Folts et al., 1978) and hypoxic (Hazashi et al., 1984)
myocardium.
As shown in the present study, POCA completely
prevents the hypoxia-induced electrophysiological
alterations of cells otherwise exhibiting a modest
accumulation of LCA. This carnitine acyltransferase
inhibitor significantly attenuates but does not completely prevent electrophysiological impairment in
hypoxic myocytes otherwise markedly accumulating
LCA. Analogous, beneficial effects of POCA on
mechanical function of ischemic hearts have been
reported (Paulson et al., 1984).
Although electrophysiologically beneficial effects
of POCA may reflect inhibition of accumulation of
LCA in sarcolemma, other possibilities must be considered. POCA may have a direct effect on the
cardiac action potential. This is unlikely, since
POCA does not alter electrophysiological parameters in normoxic myocytes or change regional function in normal ventricular myocardium (Seitelberg
et al., 1985). By inhibiting fatty acid oxidation,
POCA may diminish metabolism through 'oxygenwasting' pathways that may accelerate membrane
damage. Alternatively, POCA may prevent the electrophysiological alterations by increasing uptake and
metabolism of glucose (Wolf et al., 1982, 1985).
Several observations (McDonald and MacLeod,
1973; Higgins et al., 1981; Hasin and Barry, 1984)
suggest that ATP produced through glycolysis preferentially subserves maintenance of membrane
function. Nevertheless, our results indicate that accumulation of endogenous LCA induced by hypoxia
is one factor mediating electrophysiological abnormalities. The concentration of endogenous LCA in
sarcolemma increases 70-fold with hypoxia to levels
sufficient to induce electrophysiological effects.
POCA prevents the accumulation of LCA in sarcolemma and attenuates the electrophysiological alterations induced by hypoxia. Thus, endogenous
amphiphiles appear to comprise one class of ar-
Knabb et a/./Long-Chain Acyl Carnitine and Hypoxia
rhythmogenic metabolites contributing to impaired
electrophysiological function secondary to hypoxia.
We thank Dr. Gail Ahumada, Dave Scherrer, and Amy Grace for
preparation of cultured myocytes; Charles Krueger for preparation of
electron microscopic autoradiographs; Dr. Kathryn Yamada and Dr.
Michael Creer for helpful discussions; Dr. Miriam Salpeter for the
automated autoradiography analysis program; and Barbara Donnelly
and Kelly Ragsdale for preparation of the typescript.
Supported in part by National Institutes of Health NHLB1 Grants
HL 1 7646, SCOR in lschemic Heart Disease, and HL-28995. Dr. Corr
is supported by an Established Investigatorship from the American
Heart Association.
Address for reprints: Jeffrey E Saffitz, M.D., Department of Pathology, Washington University School of Medicine, 660 South Euclid
Avenue, Box 8118, St. Louis, Missouri 63110.
Received July 5, 1985; accepted for publication December 5, 1985.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
References
Adams RJ, Cohen DW, Gupte S, Johnson JD, Wallick ET, Wang
T, Schwartz A (1979) In vitro effects of palmitylcarnitine on
cardiac plasma membrane Na+,K+-ATPase, and sarcoplasmic
reticulum Ca2+-ATPase and Ca2+ transport. J Biol Chem 264:
12404-12410
Ahumada GG, Sobel BE, Needleman P (1980) Synthesis of prostaglandins by cultured rat heart myocytes and cardiac mesenchymal cells. J Mol Cell Cardiol 12: 685-700
Amsdorf MF, Sawicki GJ (1981) The effects of lysophosphatidylcholine, a toxic metabolite of ischemia, on the components of
cardiac excitability in sheep Purkinje fibers. Circ Res 49: 16-30
Bachmann L, Salpeter MM (1967) Absolute sensitivity of electron
microscope autoradiography. J Cell Biol 33: 299-305
Bartlett K, Bone AJ, Koudakjian PP,' Meredith E, Turnbull DM,
Sherratt HSA (1981) Inhibition of mitochondrial ^-oxidation at
the stage of camitine palmitoyl transferase 1 by the coenzyme
A esters of some substituted hypoglycaemic oxiran-2-carboxylic
acids. Biochem Soc Trans 9: 574-575
Bieber LL, Emaus R, Valkner K, Farrell S (1982) Possible functions
of short-chain and medium-chain carnitine acyltransferases.
Fed Proc 41: 2858-2862
Carmeliet E (1978) Cardiac transmembrane potentials and metabolism. Circ Res 42: 577-587
Clarkson CW, Ten Eick RE (1983) On the mechanism of lysophospharidylcholine-induced depolarization of cat ventricular
myocardium. Circ Res 52: 543-556
Clusin WT (1983) Mechanisms by which metabolic inhibitors
depolarize cultured cardiac cells. Proc Natl Acad Sci USA 80:
3865-3869
Corr PB, Cain ME, Witkowski FX, Price DA, Sobel BE (1979)
Potential arrhythmogenic electrophysiological derangements in
canine Purkinje fibers induced by lysophosphoglycerides. Circ
Res 44: 822-832
Corr PB, Snyder DW, Cain ME, Crafford WA, Gross RW, Sobel
BE (1981) Electrophysiological effects of amphiphiles on canine
Purkinje fibers: Implications for dysrhythmia secondary to
ischemia. Circ Res 49: 354-363
Corr PB, Gross RW, Sobel BE (1984) Amphipathic metabolites
and membrane dysfunction in ischemic myocardium. Circ Res
55: 135-154
Downar E, Janse MJ, Durrer D (1977) The effect of "ischemic"
blood on transmembrane potentials of normal porcine ventricular myocardium. Circulation 55: 455-462
Fink KL, Gross RW (1984) Modulation of canine myocardial
sarcolemmal membrane fluidity by amphiphilic compounds.
Circ Res 55: 585-594
Folts JD, Shug AL, Koke JR, Bittar N (1978) Protection of the
ischemic dog myocardium with carnitine. Am J Cardiol 41:
1209-1214
239
Gross RW (1984) High plasmalogen and arachidonic acid content
of canine myocardial sarcolemma: A fast atom bombardment
mass spectroscopic and gas chromatography-mass spectroscopic characterization. Biochemistry 23: 158-165
Gross RW (1985) Identification of plasmalogen as the major
phospholipid constituent of cardiac sarcoplasmic reticulum.
Biochemistry 24: 1662-1668
Gross RW, Corr PB, Lee BI, Saffitz JE, Crafford WA, Sobel BE
(1982) Incorporation of radiolabeled lysophosphatidylcholine
into canine Purkinje fibers and ventricular muscle: Electrophysiological, biochemical and autoradiographic correlations.
Circ Res 51: 27-36
Hasin Y, Barry WH (1984) Myocardial metabolic inhibition and
membrane potential, contraction, and potassium uptake. Am J
Physiol 247: H322-H329
Hauser H, Pascher I, Pearson RH, Sundell S (1981) Preferred
conformation and molecular packing of phosphatidylethanolamine and phosphatidylcholine. Biochim Biophys Acta 650:
21-51
Hayashi H, Suzuki Y, Abe M, Nishihara K, Masumura Y, Kamikawa T, Kobayashi A, Yamazaki N (1984) Effects of L-camitine
on action potential of canine papillary muscle during hypoxic
perfusion. J Electrocardiol 17: 85-90
Higgins AJ, Morville M, Burges RA, Gardiner DG, Page MG,
Blackburn KJ (1980) Oxfenicine diverts rat muscle metabolism
from fatty acid to carbohydrate oxidation and protects the
ischaemic rat heart. Life Sci 27: 963-970
Higgins TJC, Allsopp D, Bailey PJ, D'Souza EDA (1981) The
relationship between glycolysis, fatty acid metabolism and
membrane integrity in neonatal myocytes. J Mol Cell Cardiol
13: 599-615
Idell-Wenger JA, Grotyohann LW, Neely JR (1978) Coenzyme A
and camitine distribution in normal and ischemic hearts. J Biol
Chem 253: 4310-4318
Inoue D, Pappano AJ (1983) L-palmitoylcarnitine and calcium
ions act similarly on excitatory ionic currents in avian ventricular muscle. Circ Res 52: 625-634
Katz AM, Messineo FC (1981) Upid-membrane interactions and
the pathogenesis of ischemic damage in the myocardium. Circ
Res 48: 1-16
Knabb MT, Ahumada GG, Sobel BE, Saffitz JE (1985) A fixation
procedure suitable for autoradiogTaphy of endogenous longchain acyl carnitine. J Histochem Cytochem 33: 744-748
Kramer JH, Weglicki WB (1985) Inhibition of sarcolemmal Na+K+-ATPase by palmitoyl carnitine: Potentiation by propranolol.
Am J Physiol 248: H75-H81
Lamers JMJ, Srinis HT, Montfoort A, Hulsmann WC (1984) The
effect of lipid intermediates on Ca2+ and Na+ permeability and
(Na+ + K+)-ATPase of cardiac sarcolemma: A possible role in
myocardial ischemia. Biochim Biophys Acta 774: 127-137
Land B, Salpeter EE (1978) Basis for computer program for mask
analysis of EM autoradiographs. J Cell Biol 76: 142-145
liedtke AJ, Nellis SH, Neely JR (1978) Effects of excess free fatty
acids on mechanical and metabolic function in normal and
ischemic myocardium in swine. Circ Res 43: 652-661
Liedtke AJ, Nellis SH, Mjos OD (1984) Effects of reducing fatty
acid metabolism on mechanical function in regionally ischemic
hearts. Am J Physiol 247: H387-H394
Markwell MK, Haas SM, Tolbert NE, Bieber LL (1981) Protein
determination in membrane and lipoprotein samples: Manual
and automated procedures. Methods Enzymol 72: 296-303
McDonald TF, MacLeod DP (1973) Metabolism and the electrical
activity of anoxic ventricular muscle. J Physiol (Lond) 229: 559582
McGarry JD, Foster DW (1976) An improved and simplified
radioisotopic assay for the determination of free and esterified
camitine. J Lipid Res 17: 277-281
Neufeld EJ, Majerus PW, Krueger CM, Saffitz JE (1985) Uptake
and subcellular distribution of ['HJarachidonic acid in murine
fibrosarcoma cells measured by electron microscopic autoradiography. J Cell Biol 101: 573-581
Opie LH (1979) Role of camitine in fatty acid metabolism of
normal and ischemic myocardium. Am Heart J 97: 375-383
240
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Oram JF, Bennetch SL, Neely JR (1973) Regulation of fatty acid
utilization in isolated perfused rat hearts. J Biol Chem 248:
5299-5309
Owens K, Kennett FF, Weglicki WB (1982) Effects of fatty acid
intermediates on Na+-K+-ATPase activity of cardiac sarcolemma. Am J Physiol 11: H456-H461
Pande SV (1975) A mitochondrial carnitine acylcarnitine translocase system: Camitine acylcarnitine transport/exchange diffusion/acyl (+) camitine inhibition/fatty acyl transport. Proc
Natl Acad Sci USA 72: 883-887
Paulson DJ, Ward K, Sherratt HSA, Shug AL (1984) Effects of
sodium
2-[5-(4-chlorophenyl)-pentyl]-oxirane-2-carboxylate
on the metabolism and function of the ischemic myocardium
(abstr). Fed Proc 43: 902
Pearce FJ, Forster J, DeLeeuw G, Williamson JR, Tutwiler GF
(1979) Inhibition of fatty acid oxidation in normal and hypoxic
perfused rat hearts by 2-tetradecylglycidic acid. J Mol Cell
Cardiol 11: 893-915
Pitts BJR, Tate CA, Van Winkle B, McMillan-Wood JM, Entman
ML (1978) Palmitoyl camitine inhibition of the calcium pump
in cardiac sarcoplasmic reticulum: A possible role in myocardial
ischemic. Life Sci 23: 391-402
Ramsay RR, Tubbs PK (1975) The mechanism of fatty acid uptake
by heart mitochondria: An acylcarnitine-carnitine exchange.
FEBS Lett 54: 21-25
Saffirz JE, Corr PB, Lee BI, Gross RW, Williamson EK, Sobel BE
(1984) Pathophysiologic concentrations of lysophosphoglycerides quantified by electron microscopic autoradiography. Lab
Invest 50: 278-286
Salpeter MM, Bachmann L (1972) Autoradiography. In Principles
and Techniques of Electron Microscopy: Biological Applications, vol 2, edited by MA Hayat. New York, Van Nostrand
Co., pp 221-279
Salpeter MM, McHenry FA, Salpeter EE (1978) Resolution in
electron microscope autoradiography. IV. Application to analysis of autoradiographs. J Cell Biol 76: 127-145
Seitelberger R, Kraupp O, Winkler M, Brugger G, Raberger G
(1985) Effects of the acylcarnirine-transferase blocking agent
Circulation Research/Vol. 58, No. 2, February 1986
sodium 2[5-(4-chlorophenyl)-pentyl]-oxirane-2-carboxylate
(POCA) on metabolism and regional function in the underperfused canine myocardium. J Cardiovasc Pharmacol 7: 273-280
Shug AL, Thomsen JH, Folts JD, Bittar N, Klein MI, Koke JR,
Huth PJ (1978) Changes in tissue levels of camitine and other
metabolites during myocardial ischemia and anoxia. Arch
Biochem Biophys 187: 25-33
Smaby JM, Hermetter A, Schmid PC, Paltauf F, Brockman HL
(1983) Packing of ether and ester phospholipids in monolayeis.
Evidence for hydrogen-bonded water at the sn-1 acyl group of
phosphatidylcholine. Biochemistry 22: 5808-5813
Sobel BE, Corr PB, Robison AK, Goldstein RA, Witkowski FX,
Klein MS (1978) Accumulation of lysophosphoglycerides with
arrhythmogenic properties in ischemic myocardium. J Clin
Invest 62: 546-553
Whirmer JT, Wenger JI, Rovetto MJ, Neely JR (1978) Control of
fatty acid metabolism in hypoxic and ischemic hearts. J Biol
Chem 253: 4305-4309
Witkowski FX, Corr PB (1978) Automated analysis of cardiac
inrracellular transmembrane action potentials. Comput Cardiol
IEEE, pp 315-318
Wolf HPO, Engel DW (1985) Decrease of fatty acid oxidation,
ketogenesis and gluconeogenesis in isolated perfused rat liver
by phenylalkyl oxirane carboxylate (B 807-27) due to inhibition
of CPT I (EC 2.3.1.21) EurJ Biochem 146: 359-363
Wolf HPO, Eistetter K, Ludwig G (1982) Phenylalklyoxirane
carboxylic acids, a new class of hypoglycaemic substances:
Hypoglycaemic and hypoketonaemic effects of sodium 2-[5-(4chlorophenyl)-penryl]-oxirane-2-carboxylate (B 807-27) in
fasted animals. Diabetologia 22: 456-463
Wood JM, Bush B, Pitts BJR, Schwartz A (1977) Inhibition of
bovine heart Na+, K+-ATPase by palmitoyl carnitine and palmitoyl CoA. Biochem Biophys Res Commun 74: 677-684
INDEX TERMS: Arrhythmogenesis • Autoradiography • Camitine acyltransferase inhibition • Hypoxia • Long-chain acyl carnitine • Sarcolemma
The dependence of electrophysiological derangements on accumulation of endogenous
long-chain acyl carnitine in hypoxic neonatal rat myocytes.
M T Knabb, J E Saffitz, P B Corr and B E Sobel
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Circ Res. 1986;58:230-240
doi: 10.1161/01.RES.58.2.230
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/58/2/230
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/