ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS
Vol. 329, No. 2, May 15, pp. 145–155, 1996 Article No. 0203
Characterization of the Human Plasmalemmal Carnitine Transporter in Cultured Skin Fibroblasts Ingrid Tein,*,†,1 Scott W. Bukovac,† and Zhong-Wei Xie* *Division of Neurology, The Hospital for Sick Children, and †Department of Clinical Biochemistry, University of Toronto, Toronto, Ontario M5G 1X8, Canada
Received November 13, 1995, and in revised form February 26, 1996
Carnitine is an essential cofactor for long-chain fatty acid oxidation. We characterized the human carnitine transporter in vitro in a cultured skin fibroblast model both at the previously established Km concentration of carnitine uptake in fibroblasts (5 mmol/liter) and at 0.05% Km (0.25 mmol/liter). A rapid exponential dosedependent decrease in mean percentage of carnitine uptake was demonstrated with increasing concentrations of nigericin, but no significant decrease was found with equimolar amounts of valinomycin. This would suggest that the Na/ gradient is integral to carnitine transport function. Interference of the Na/ (out–in) gradient by nigericin may be secondary to cytoplasmic acidification by this K/ proton ionophore. The rate of uptake was fully saturated at an extracellular Na/ concentration of 150 mmol/liter. Replacement of 150 mmol/liter extracellular Na/ with Li/ resulted in an 80 and a 50% reduction, and replacement with K/ and Rb/ ions resulted in a 100 and an 85 to 90% reduction in carnitine uptake, respectively, at carnitine concentrations of 0.25 and 5 mmol/liter, underlining the specific requirement for the Na/ ion. The effects of different site-specific respiratory chain toxins, namely, rotenone (complex I), antimycin A (complex III), and potassium cyanide (KCN) (complex IV) on carnitine uptake was also examined. A rapid exponential dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of inhibitors was demonstrated. These data suggest either a metabolic energy requirement of the carnitine transporter or interference of the Na/ (out–in) gradient by a proton gradient (in–out) secondary to the accumulation of intracellular H/ ions, due to the action of the respiratory chain toxins, further suggesting that the transporter is sensitive to and inhibited by intracellu1 To whom correspondence and reprint requests should be addressed at Division of Neurology, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario M5G 1X8, Canada. Fax: 1416-966-0066.
lar H/ ions. The effects of several sulfhydryl-binding agents, namely 2,4-dinitrofluorobenzene, N-ethylmaleimide, and mersalyl acid, were examined, and a significant inhibition of carnitine uptake was demonstrated, suggesting that free sulfhydryl groups are also integral to the import function of the human fibroblast transporter. q 1996 Academic Press, Inc. Key Words: plasmalemmal carnitine transporter; cultured skin fibroblasts; fatty acid oxidation.
Carnitine (b-hydroxy-g-trimethylaminobutyric acid) is a small water-soluble quaternary amine that serves as an essential cofactor for the transport of long-chain fatty acids as acylcarnitine esters across the inner mitochondrial membrane and modulates the intramitochondrial acyl CoA/CoA sulfhydryl ratio in mammalian cells, thereby providing the cell with a critical source of free CoA (1). Carnitine is therefore of pivotal importance in tissues such as muscle, heart, kidney, and liver which rely heavily on efficient fatty acid oxidation as a major source of ATP production. In nonvegetarians, approximately 75% of body L-carnitine sources comes from the diet and 25% from de novo biosynthesis (2). The carnitine concentrations in tissues are normally 20- to 50-fold higher than in serum. Human tissue concentrations (nmol/g) are heart (3500–6000) ú muscle (2000–4600) ú liver (1000–1900) ú brain (200–500) (3). The highest tissue concentration is found in the epididymis which has 2000 times the serum concentration (4). Uptake into tissues therefore occurs across a large concentration gradient which is maintained by an active transport system (1). Kinetic studies of the plasmalemmal carnitine transporter have demonstrated similar Km values of 2–6 mmol/liter for carnitine transport in cultured muscle (5), heart (6), and fibroblasts (5, 7, 8), suggesting that they share a common transporter, and very different Km values for human liver (500 mmol/liter) and brain (ú1000 mmol/liter) 145
0003-9861/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ 6b18$$9412
04-12-96 10:39:20
arca
AP: Archives
146
TEIN, BUKOVAC, AND XIE
(9). In addition, studies of children with a genetic defect of the plasmalemmal carnitine transporter suggest that the same transport defect is expressed in fibroblasts, muscle, heart, and kidney (7, 8, 10). To characterize the human plasmalemmal carnitine transporter we studied the effects of different substrates and conditions in vitro in a cultured human skin fibroblast model, which appears to share a similar transporter with human muscle, kidney, and heart (7, 8). To investigate the requirement of the transporter for a Na/ gradient, we studied the effects of nigericin, which interferes with both the Na/ and the K/ gradients, and valinomycin, which specifically disrupts the K/ gradient (11). To quantify the extracellular Na/ threshold concentration for maximal carnitine uptake, we also examined the effect of increasing extracellular Na/ concentrations on carnitine uptake. In order to further ascertain the requirement of the transporter specifically for the Na/ ion, a series of physiological buffers was developed in which the Na/ ion was replaced with equimolar amounts of Li/, K/, and Rb/ ions. The energy requirement of the transporter was studied by examination of the effects on carnitine uptake of several metabolic toxins, specific for different sites of the respiratory chain, namely, rotenone which inhibits NADH–ubiquinone reductase or complex I activity, antimycin A which inhibits ubiquinol–cytochrome c reductase or complex III activity, and potassium cyanide (KCN) which inhibits cytochrome oxidase or complex IV activity (11). The requirement of the transporter for free sulfhydryl groups was studied by examination of the effects of several sulfhydryl-binding agents, namely 2,4-dinitrofluorobenzene, N-ethylmaleimide, and mersalyl acid on carnitine uptake. The effects of all of the abovementioned compounds on carnitine uptake were studied at the previously established fibroblast transporter Km concentration of 5 mmol/liter (8), as well as at 0.05% of the Km concentration, namely 0.25 mmol/liter, to increase the sensitivity of detection of effects on the carnitine uptake system. MATERIALS AND METHODS All studies were performed with the approval of the Institutional Review Board of the Hospital for Sick Children, Toronto. Carnitine uptake studies were performed in cultured control human skin fibroblasts in which we had previously documented normal carnitine uptake. Control fibroblasts were obtained from the Human Genetic Mutant Cell Repository, Coriell Institute for Medical Research, Camden, New Jersey. The uptake of carnitine using [3H]L-carnitine was investigated by the method outlined in Tein et al. (8). L-[Methyl3 H]carnitine hydrochloride was purchased from Amersham (Arlington Heights, IL) and L-carnitine was a gift from Sigma-Tau Pharmaceuticals (Pomezia, Italy). Cell-bound radioactivity was determined in 800 ml of the final fibroblast hydrolysate using Aquasol-2 with a counting efficiency of 60%. Cell protein was measured in the remaining 200 ml of the hydrolysate for each individual plate by the method of Lowry et al. (12). Nigericin (sodium salt) (MW 747.0) and valinomycin (MW 1111.3) were used as supplied from Sigma Chemi-
/ 6b18$$9412
04-12-96 10:39:20
arca
cal Co. (St. Louis, MO). Rotenone, antimycin A, 2,4-dinitrofluorobenzene, N-ethylmaleimide, and mersalyl acid were used as supplied from Sigma Chemical Co. Potassium cyanide was used as supplied from Fisher Scientific Co. Experiments with above-mentioned compounds were run in parallel with a standard control carnitine uptake study without the specific compound. Nonspecific uptake was determined at 10 mmol/liter carnitine. Control fibroblasts (passages 6 to 12) were plated onto 9.5-cm2 6well plates (Gibco Laboratories, St. Lawrence, MA) and allowed to grow to confluence in a-MEM2 medium supplemented with 10% fetal calf serum (medium total carnitine, 2 mmol/liter). Following confluence of the fibroblasts, the medium was changed to RPMI with 10% fetal calf serum 24 h prior to the uptake study. On the day of the study, the individual wells were thoroughly washed to remove all fetal calf serum and were then incubated in RPMI (without fetal calf serum) and with fixed carnitine concentrations of 0.25 or 5 mmol/liter. In the first set of experiments, different concentrations of nigericin or valinomycin were added to the incubation media and fixed at concentrations of 0, 1, 2.5, 5, 7.5, 10, 12.5, 15, 17.5, and 20 mmol/ liter. In the second set of experiments, preliminary studies were performed with a wide range of concentrations of the metabolic inhibitors (rotenone, antimycin A, and potassium cyanide) based on previously documented concentrations known to produce metabolic inhibition in vitro. The range of concentrations producing the most significant inhibition of carnitine uptake was determined, in turn, for each of the inhibitors and studied in greater detail as indicated: rotenone (1–10 mmol/liter), antimycin A (10–50 mg/ml), and KCN (0.5–50 mmol/liter). These inhibitors were added to the incubation media at fixed concentrations within the established range of sensitivity. The viability of the fibroblasts was determined in parallel cultures by histological examination. In the third set of experiments, preliminary studies were performed with a wide range of concentrations of the different sulfhydryl-binding agents. The range of concentrations producing the most significant degree of inhibition of carnitine uptake was determined, in turn, for each of the sulfhydryl-binding agents as indicated: 2,4-dinitrofluorobenzene (10-100 mmol/liter), mersalyl acid (1 1 1008 to 5 1 1004 mol/liter), and N-ethylmaleimide (20–100 mmol/liter). In all of these studies, [3H]L-carnitine uptake was measured in vitro following a 4-h incubation period. The specific uptake of radioactivity was used to calculate the rates of total carnitine uptake. Carnitine uptake was calculated in mol/min/mg of fibroblast protein. Carnitine uptake was measured at increasing reagent concentrations for the two carnitine substrate concentrations of 0.25 and 5 mmol/liter. The residual carnitine uptake was expressed as percentage of control carnitine uptake in the normal control cell line not exposed to the specific reagent. All experiments were performed in triplicate and on separate days. The mean percentages of control carnitine uptake were then plotted against increasing concentrations of the specific compound, for each of the two carnitine incubation substrate concentrations of 0.25 and 5 mmol/liter. In the final set of experiments, we developed a series of physiological buffers: one set in which there were different concentrations of extracellular Na/ as its chloride salt (0–150 mmol/liter) and three sets in which the Na/ ion was replaced with equimolar amounts of the Li/, K/, or Rb/ ions as their respective chloride salts, at concentrations of 0 to 150 mmol/liter. The basal medium consisted of 1.4 mmol/liter Ca2/, 0.5 mmol/liter Mg2/, 2.7 mmol/liter K/, 6.5 mmol/ liter Cl0, 6.5 mmol/liter HC030, and 5.5 mmol/liter glucose. Mannitol was used to maintain constant osmolarity within the physiological range. Following confluence of the fibroblasts and on the day of the study, the individual wells were thoroughly washed with the given physiological buffer we had prepared, to remove all of the RPMI and fetal calf serum. The fibroblasts were then incubated with the same
2
Abbreviations used: MEM, minimal essential medium; COX, cytochrome oxidase.
AP: Archives
CHARACTERIZATION OF HUMAN PLASMALEMMAL CARNITINE TRANSPORTER physiological buffer, which had been made up with different concentrations of the given ion (0–150 mmol/liter). [3H]L-Carnitine uptake was measured in vitro following a 4-h incubation period as outlined in Tein et al. (8). The specific uptake of radioactivity was used to calculate the rates of total carnitine uptake, and carnitine uptake was calculated in mol/min/mg of fibroblast protein. Carnitine uptake was measured at different ion concentrations for the two carnitine substrate concentrations of 0.25 and 5 mmol/liter. The residual carnitine uptake was expressed as percentage of control carnitine uptake measured at a physiological extracellular Na/ concentration of 150 mmol/liter. All experiments were performed in triplicate and on separate days. The mean percentages of control carnitine uptake were then plotted against increasing concentrations of the specific ion, for each of the two carnitine incubation substrate concentrations of 0.25 and 5 mmol/liter.
RESULTS
We demonstrated a rapid exponential dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of nigericin ranging from 100% of control carnitine uptake at 0% of nigericin and tapering toward õ20% of carnitine uptake at a concentration of 20 mmol/liter of nigericin (Fig. 1). The relative rate of decrease was similar at both carnitine incubation concentrations, namely, at the previously established Km concentration of carnitine uptake in fibroblasts (5 mmol/liter) as well as at the 0.05% of Km concentration of carnitine (0.25 mmol/liter). In contrast, there was no significant decrease in mean percentage of carnitine uptake, which remained around 100%, with equimolar amounts of valinomycin up to 20 mmol/liter, for both carnitine incubation concentrations. In our studies examining the effect of increasing extracellular Na/ concentrations on carnitine uptake, we demonstrated an exquisite sensitivity to the Na/ (out– in) gradient (Fig. 2). The rate of uptake was saturated at an extracellular Na/ concentration of 150 mmol/liter. At carnitine concentrations of 0.25 and 5 mmol/liter, approximately 80% of maximal uptake was achieved at Na/ concentrations of 30 and 5 mmol/liter, respectively. In further investigations involving the replacement of the extracellular Na/ ion with equimolar amounts of other cations, we demonstrated that the transporter had a specific requirement and/or binding site for the Na/ ion which was only weakly fulfilled by the Li/ ion and not met by the K/ and Rb/ ions (Fig. 2). Replacement of extracellular Na/ with Li/ at a concentration of 150 mmol/ liter resulted in an 80 and 50% reduction, respectively, in carnitine uptake (when expressed as the mean percentage of control carnitine uptake measured at an extracellular Na/ concentration of 150 mmol/liter) at the two carnitine incubation concentrations of 0.25 and 5 mmol/liter. Replacement of extracellular Na/ with the K/ and Rb/ ions at concentrations of 150 mmol/liter resulted in a 100 and an 85 to 90% reduction, respectively, in carnitine uptake at the carnitine incubation concentrations of 0.25 and 5 mmol/liter. In the studies of carnitine uptake in the presence of
/ 6b18$$9412
04-12-96 10:39:20
arca
147
different metabolic inhibitors, we demonstrated a rapid exponential dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of these inhibitors (Fig. 3). The mean percentage of carnitine uptake at carnitine concentrations of 0.25 and 5 mmol/liter, respectively, was reduced as follows: by 80 and 65% at a rotenone concentration of 10 mmol/ liter, by 75 and 50% at an antimycin A concentration of 50 mg/ml, and by 60 and 45% at a KCN concentration of 50 mmol/liter. The relative rate of decrease was similar at both carnitine concentrations, though somewhat more pronounced at the lower carnitine concentration of 0.25 mmol/liter. In the studies of carnitine uptake in the presence of several sulfhydryl-binding agents, we again demonstrated a rapid dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of the different reagents (Fig. 4). The mean percentage of carnitine uptake at carnitine concentrations of 0.25 and 5 mmol/liter, respectively, was reduced as follows: by 75 and 60% at a 2,4-dinitrofluorobenzene concentration of 50 mmol/liter, by 97 and 90% at a mersalyl acid concentration of 50 mmol/liter, and by 85 and 55% at an N-ethylmaleimide concentration of 50 mmol/ liter. The relative rate of decrease was similar at both carnitine substrate concentrations, though somewhat more pronounced at the lower carnitine concentration of 0.25 mmol/liter. DISCUSSION
It is generally held that the plasmalemmal carnitine transporter is Na/ gradient and energy dependent and requires the presence of free sulfhydryl goups. However, review of the literature reveals conflicting results which are summarized in Table I (13–28). This may be attributable to interspecies variation, tissue-specific kinetic differences, differences in methods of tissue preparation (e.g., cultured cells versus perfused intact organs or tissue slices), differences in experimental conditions (e.g., concentration of substrates, pH, temperature), as well as differences in interpretation regarding the significance of results. Also, most previous work has been conducted in animal tissues, primarily in the rat, with little work done in human tissues or cultured cells. The finding of Na/ gradient-dependent carnitine uptake has been previously demonstrated in rat brain slices (13), mouse brain synaptosomes (14), perfused adult rat heart (19), isolated rat muscle (21), rat kidney cortex slices (24), rat renal brush border membrane vesicles (25), perfused rat liver (27), and proximal small intestine (28) (Table I). This gradient dependence appears to be relatively specific for the cation Na/, as replacement of Na/ with Li/ has resulted in significant decreases in carnitine uptake in rat brain slices (13),
AP: Archives
148
TEIN, BUKOVAC, AND XIE
FIG. 1. Effects of increasing nigericin and valinomycin concentrations on mean percentage of control carnitine uptake at carnitine incubation concentrations of 0.25 and 5 mmol/liter.
isolated rat muscle (21), rat kidney cortex slices (24), and perfused rat liver (27). Consistent with previously demonstrated Na/ dependence in certain tissues, ouabain, which abolishes Na/K/ ATPase activity and thereby decreases the Na/ gradient, has been demonstrated to result in a significant reduction in carnitine uptake in rat brain slices (13), mouse brain synaptosomes (14), and rat kidney cortex slices (24) and to a lesser extent in isolated rat muscle (21). Based on the demonstrated inhibition of carnitine uptake by ouabain, an absolute requirement for Na/ and a need for K/ for optimal uptake in rat kidney cortex slices, Huth and Shug (24) have suggested that the carnitine transport in kidney is driven by a Na/ gradient and may involve Na/ –carnitine cotransport by a system for
/ 6b18$$9412
04-12-96 10:39:20
arca
which the energy requirement is derived from the ionic gradient established by the plasma membrane Na/K/ ATPase (29). A similar system has been proposed for the transport of glucose and amino acids (30). However, in contrast to the demonstrated ouabain inhibition of carnitine uptake in the aforementioned studies, other workers have demonstrated no significant ouabain effect in rat muscle (22), isolated rat liver cells (26), or perfused rat liver (27). Similarly, despite a demonstrated Na/ gradient dependence of carnitine uptake in perfused adult rat heart, no significant inhibition of carnitine uptake has been noted with ouabain (19). It has been argued that the lack of effect with ouabain may be because carnitine uses the already built-up Na/ gradient for transport (27).
AP: Archives
CHARACTERIZATION OF HUMAN PLASMALEMMAL CARNITINE TRANSPORTER
149
FIG. 2. Effect of extracellular ion substitution on mean percentage of control carnitine uptake at carnitine incubation concentrations of 0.25 and 5 mmol/liter.
Furthermore, despite the similar kinetic characteristics of the plasmalemmal carnitine transporter in fibroblasts, muscle, kidney, and heart, Bahl et al. (17) demonstrated no significant Na/ dependence of carnitine uptake in isolated rat heart myocytes and no effect of Li/ substitution on uptake. Consistent with this finding is a lack of significant inhibition of carnitine uptake by ouabain (17%). Interestingly, extracellular Ca2/ was found to decrease carnitine uptake by 50%, though Mg2/ had no effect, nor did the slow channel blocker verapamil. Decreasing the Ca2/ concentrations to 0.1 mmol/liter restored carnitine uptake to control values. Further, carnitine uptake in cultured fetal human heart cells (CCL 27) was also not linked to amino acid transport (16). These findings are in contrast to those of Na/ gradient dependence demonstrated in perfused adult rat heart by Vary and Neely (19). These workers found that the effect of Na/ on carnitine transport appeared to be dependent on the Na/ electrochemical gradient and not dependent directly on Na/K/ ATPase or on membrane depolarization. The inconsistent findings may relate to differences in tissue preparation, measurement techniques, or tissue specificity. Tissue specificity has also been underlined by workers who demonstrated that carnitine uptake in rat kidney cortex slices required both Ca2/ and K/ (24) in contrast to uptake in rat brain slices which required neither ion (13). To further characterize the Na/ gradient dependence of the carnitine transporter in a cultured human skin fibroblast model, we examined the individual effects of nigericin and valinomycin on carnitine uptake. As significant inhibition of carnitine uptake was demonstrated with nigericin, which interferes with both the
/ 6b18$$9412
04-12-96 10:39:20
arca
Na/ and the K/ gradient, but not with valinomycin, which highly selectively disrupts the K/ gradient (11, 31–33), this would suggest that the Na/ gradient is integral to carnitine transport function in cultured human skin fibroblasts. Nigericin prefers K/ over Na/ by a factor of 100:1, which is a much greater selectivity for Na/ than that of valinomycin, which has an equilibrium selectivity for K/ over Na/ of Ç10,000:1 (31, 33). The extreme ion selectivity of valinomycin for K/ versus Na/ has been well established by experiments in which Na/ proved a totally ineffective substitute for K/ (32). The finding of sodium-gradient dependence in fibroblasts parallels the transport of certain amino acids in rat kidney cortex in which the sodium gradient (out–in) serves as the driving force for active transport (34, 35). The mechanism for the interference of the Na/ (out–in) gradient by nigericin may relate to cytoplasmic acidification by this K//proton antiporter. Nigericin has been documented to cause cytoplasmic acidification in a variety of different cells (36–39). As a K// H/ antiporter, nigericin is inserted into the cytoplasmic membrane. Under physiological conditions, nigericin transfers H/ to the cytoplasm in exchange for K/ because the transmembrane K/ gradient is 2 orders higher than the H/ gradient. Thus, there may be interference of the Na/ (out–in) gradient-dependent carnitine transport by a proton gradient (in–out), secondary to the accumulation of these intracellular H/ ions. Further, we demonstrated an exquisite dependence of the transporter on the Na/ (out–in) gradient which was only weakly fulfilled by substitution with equimolar amounts of the Li/ ion and strongly inhibited by substitution with the K/ and Rb/ ions. This suggests that the human cultured skin fibroblast transporter has a
AP: Archives
150
TEIN, BUKOVAC, AND XIE
FIG. 3. Effects of increasing concentrations of rotenone, antimycin A, and KCN on mean percentage of control carnitine uptake at carnitine incubation concentrations of 0.25 and 5 mmol/liter.
/ 6b18$$9412
04-12-96 10:39:20
arca
AP: Archives
CHARACTERIZATION OF HUMAN PLASMALEMMAL CARNITINE TRANSPORTER
151
FIG. 4. Effects of increasing concentrations of sulfhydryl-binding agents on mean percentage of control carnitine uptake at carnitine incubation concentrations of 0.25 and 5 mmol/liter. 2,4-DNB, 2,4-dinitrofluorobenzene.
/ 6b18$$9412
04-12-96 10:39:20
arca
AP: Archives
/ 6b18$$9412
04-12-96 10:39:20
arca
AP: Archives 0
/ /
Perfused rat liver (27)
Proximal small intestine (28)
/
/
0 30%
0 20%
0 9%
0
0 87% 0 36%a
KCN
0 41% 0 30% 0 85% 0 81%
025% Anoxia
0 86%
Nitrous oxide
0 50% 0 50%
0 80%b 0 82%
0 42%
0 93% 0 7% 0 20% 0 25% 0 50%
Uncoupler ox/phosph 2,4-dinitrophenol
0
0 87%
0
0 95%
0 95%
Sulfhydryl binding agent
60 min
5 – 15 min
5 – 90 min
6 s – 120 min
60 min
30 – 60 min
3h
3 – 90 min 1 – 4 hr
15 min
10 s – 60 min
5 – 10 – 60 min
2h
1–4 h
5 min
30 min – 3 h
Time of incubation
Note. /, activation of carnitine uptake; 0, inhibition of carnitine uptake; 0, no significant effect. All figure % signify the % of inhibition of control carnitine uptake. a Use of NaCN. b Use of carbonyl cyanide-m-chlorophenylhydrazone as the uncoupler of oxidative phosphorylation.
0 80%
0
/
Rat renal brush border membrane vesicles (25) Isolated rat liver cells (26)
/
/
0 40%
/
0 /
/
0 23% 0
0
/
/
/
Rat kidney cortex slices (24) 0 75%
0 78%
0
0 17%
0
0 89% 0 59%
Ouabain
Energy dependence
/
/
/
0
0 ú90%
Li/ substitution
Rat kidney cortex slices (23)
Isolated rat muscle (22)
/
Mouse brain synaptosomes (14) Fetal human heart cells in culture (CCL 27) (15) Fetal human heart cells in culture (CCL 27) (16) Isolated rat heart myocytes (17) Isolated perfused adult rat heart (18) Isolated perfused adult rat heart (19) Rat heart slices (20) Isolated rat muscle (21) 0
/
Na/ dependence
Rat brain slices (13)
Reagent or condition (Tissue, cell, or preparation)
TABLE I
Effects of Different Reagents and Conditions on Carnitine Uptake
152 TEIN, BUKOVAC, AND XIE
CHARACTERIZATION OF HUMAN PLASMALEMMAL CARNITINE TRANSPORTER
specific requirement for the Na/ gradient and/or binding site for the Na/ ion, as has been demonstrated for rat brain slices (13), isolated rat muscle (21), rat kidney cortex slices (24), and perfused rat liver (27). This also suggests that the nonhydrated ion diameter is important. These results in human skin fibroblasts most closely parallel those in rat kidney. Energy dependent carnitine uptake has been previously demonstrated in rat brain slices (13), mouse brain synaptosomes (14), isolated rat heart myocytes (17), isolated rat muscle (21, 22), rat kidney cortex slices (23, 24), isolated rat liver cells (26), and perfused rat liver (27) (Table I). Energy dependence has been suggested by a significant inhibition of carnitine uptake in vitro in different tissues in the presence of various metabolic inhibitors including the cytochrome oxidase toxin potassium cyanide, uncouplers of oxidative phosphorylation such as 2,4-dinitrophenol and carbonyl cyanide-m-chlorophenylhydrazone, and conditions of anoxia. Significant inhibition of uptake has also been demonstrated in isolated rat muscle incubated with sodium azide, a cytochrome system inhibitor (21). The relative inhibitory effects of KCN or NaCN on carnitine uptake are variable depending upon tissue. The greatest inhibition was noted in rat brain slices with 1 mmol/liter KCN (13) and in mouse brain synaptosomes with 0.5 mmol/liter NaCN (14). Less significant inhibition has been noted in rat kidney cortex slices with 1 mmol/liter (24) and perfused rat liver with 5 mmol/liter KCN (27). By contrast, minimal inhibition has been documented in isolated rat muscle with 10 mmol/liter KCN (21) and no inhibition has been documented in cultured fetal human heart cells with 1 mmol/liter KCN (16). In previous work, we had demonstrated a significant 53 to 80% reduction in the maximal velocity of carnitine uptake and normal Km values in the cultured skin fibroblasts from four children with severe genetic cytochrome oxidase (COX) deficiency and secondary carnitine deficiency (40). This suggested that the binding of carnitine to the transporter was normal but that the transport process did not function efficiently. As the expected metabolic consequence of decreased COX activity is a net decrease of intracellular energy, the decreased carnitine uptake in vitro may suggest decreased function of an energy-dependent active transporter. Replacement of oxygen with 95% nitrous oxide leading to anoxic conditions has also produced selective inhibitory effects in different tissues. Nitrous oxide decreased carnitine uptake in isolated perfused adult rat heart (18) and in isolated rat muscle (21, 22) with more significant inhibition in rat brain slices (13) and rat kidney cortex slices (23, 24). Glucose depletion also inhibited carnitine uptake in rat brain slices by 73% (13). Uncouplers of oxidative phosphorylation (e.g., 0.5–1 mmol/liter 2,4-dinitrophenol or 2 mmol/liter carbonyl-
/ 6b18$$9412
04-12-96 10:39:20
arca
153
m-chlorophenylhydrazone) produced significant reductions in carnitine uptake in isolated rat muscle (21), isolated rat liver cells and perfused rat liver (26, 27), isolated rat heart myocytes (17), rat brain slices (13), and rat kidney cortex slices (23, 24). Less significant reduction was noted in cultured fetal human heart cells (15, 16) with no significant reduction in mouse brain synaptosomes (14). Carnitine uptake in cardiac tissue appears to be least affected by metabolic inhibition. It has been suggested that for high-affinity transport in perfused adult rat heart (18), as well as in cultured fetal human heart cells (16), it would be difficult to deplete the myocardium of ATP to such an extent as to make it rate limiting for carnitine transport, as the required energy per unit time would probably be small. In order to further characterize the energy requirement of the human plasmalemmal carnitine transporter, we examined the individual effects of inhibitors specific for different sites of the respiratory chain complex on carnitine uptake in a cultured human skin fibroblast model. We demonstrated a rapid exponential dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of rotenone, antimycin A, and KCN. The relative rate of decrease was similar at both carnitine concentrations, though somewhat more pronounced at the lower carnitine incubation concentration of 0.25 mmol/liter which is not unexpected given the higher transmembrane gradient at the lower concentration which is 0.05% of the Km value of the transporter. On comparison of the degree of KCN inhibition of carnitine uptake with the previous studies in rat tissues, 1 mmol/liter KCN reduced carnitine uptake by approximately 20% which is similar to the results reported in rat kidney cortex slices (24). These results may suggest that the carnitine transporter in cultured human skin fibroblasts has a minimum metabolic energy requirement, which bears some similarity to the requirements of the transporter in rat kidney cortex and differs from the apparent lack of KCN inhibition of carnitine uptake demonstrated in cultured fetal human heart cells. There appears to be a significant range in the vulnerability of the transporter to a variety of metabolic toxins, suggesting possible tissue-specific differences in rate-limiting energy requirements. Further work needs to be done to elucidate the individual tissue-specific metabolic energy requirements of the carnitine transporter, particularly in human tissues highly dependent upon efficient longchain fatty acid oxidation. On the other hand, as we used only respiratory chain toxins whose metabolic consequence would be an increase in intracellular lactic acid with an accumulation of intracellular H/ ions, an alternative explanation could be that the resultant proton gradient (in–out) may interfere with the Na/ (out– in) gradient. This explanation may be more plausible, given the sustained viability of the cultured skin fibro-
AP: Archives
154
TEIN, BUKOVAC, AND XIE
blasts in our in vitro system, which likely would be decreased by rate-limiting ATP stores. This would further support the concept that the transporter is sensitive to and inhibited by intracellular H/ ions which may represent a common mechanism for inhibition of carnitine transport by respiratory chain toxins, genetic cytochrome oxidase deficiency, as well as nigericin exposure. Finally, sulfhydryl-binding agents have been documented to consistently inhibit carnitine uptake by approximately 90% in rat brain slices (13), cultured fetal human heart cells (16), and rat kidney cortex slices (24), suggesting that the presence of free sulfhydryl groups is integral to transporter function (Table I). We have similarly demonstrated, in a cultured human skin fibroblast model, a rapid dose-dependent decrease in mean percentage of carnitine uptake with increasing concentrations of several sulfhydryl-binding agents, namely 2,4-dinitrofluorobenzene, mersalyl acid, and Nethylmaleimide. This suggests that free sulfhydryl groups are not only integral to carnitine import function in cultured fetal human heart cells, but are also essential for the human skin fibroblast transporter. In contrast, in a study of the effect of mersalyl on perfused rat liver (27), mersalyl had no effect on carnitine uptake but practically abolished the release of carnitine, suggesting that there may be different sites of carnitine import and export on the same transporter protein in liver. As liver is capable of the complete de novo synthesis of carnitine, and since other tissues must send the precursor g-butyrobetaine to liver for final conversion to carnitine, the export of carnitine may be of greater metabolic significance in liver compared to other tissues. Of interest, Indiveri et al. (41) have demonstrated that carnitine acylcarnitine translocase, which is a mitochondrial carnitine carrier, is dependent upon two different classes of free sulfhydryl groups. In conclusion, we have endeavored to further characterize the human plasmalemmal carnitine transporter in a cultured skin fibroblast model in order to gain further insight into the structure and function of this membrane transporter which is essential for the efficient importation of carnitine into the cell to fulfill its vital intracellular functions. Though the carnitine transporter serves a key role in fatty acid metabolism, it has not yet been isolated, purified, or cloned. On the basis of its Na/ gradient dependence and its specificity for carnitine and acylcarnitine substrates, it is possible that the plasmalemmal transporter shares some structural homologies with the Na/ symporter family and the carnitine acyltransferase families. Increased understanding of the integral components of the transporter may also prove to be useful in the future therapeutic management of individuals who are homozygous for the potentially lethal autosomal recessive plasmalemmal carnitine transporter defect. These individ-
/ 6b18$$9412
04-12-96 10:39:20
arca
uals present with infantile-onset carnitine-responsive cardiomyopathy, weakness, hypoglycemic hypoketotic encephalopathy, and failure to thrive, with very low plasma and tissue concentrations of carnitine; lipid storage in muscle, heart, and liver; and severe renal leak of carnitine (7, 8). Formerly, this was a lethal disorder, which is now eminently treatable, provided the diagnosis is made and there is early institution of very high dose oral carnitine supplementation which is life saving and leads to rapid clinical improvement in cardiac function, strength, and growth (8). Further understanding of the human transporter may also contribute to the long-term therapeutic strategies in these individuals. ACKNOWLEDGMENTS This work was supported by an Operating Grant from the Medical Research Council of Canada and by a grant from the University of Toronto Dean’s Fund. I.T. is a recipient of a Medical Research Council of Canada Scholarship. We thank Dr. Brian H. Robinson for his advice and support.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8.
9. 10.
11. 12. 13. 14. 15. 16. 17. 18.
Bremer, J. (1983) Physiol. Rev. 63, 1420–1480. De Vivo, D. C., and Tein, I. (1990) Int. Pediatr. 5, 134–141. Stanley, C. A. (1987) Adv. Pediatr. 34, 59–88. Hinton, B. T., Dott, H. M., and Setchell, B. P. (1979) J. Reprod. Fertil. 56, 167–172. Rebouche, C. J., and Engel, A. G. (1982) In Vitro 18, 495–500. Bahl, J. J., and Bressler, R. (1987) Annu. Rev. Pharmacol. Toxicol. 27, 257–277. Treem, W. R., Stanley, C. A., Finegold, D. N., Hale, D. E., and Coates, P. M. (1988) N. Engl. J. Med. 319, 1331–1336. Tein, I., De Vivo, D. C., Bierman, F., Pulver, P., De Meirleir, L. J., Cvitanovic-Sojat, L., Pagon, R. A., Bertini, E., Dionisi-Vici, C., Servidei, S., and DiMauro, S. (1990) Pediatr. Res. 28, 247– 255. Bieber, L. L. (1988) Annu. Rev. Biochem. 57, 261–283. Stanley, C. A., DeLeeuw, S., Coates, P. M., Vianey-Liaud, C., Divry, P., Bonnefont, J-P., Saudubray, J-M., Haymond, M., Trefz, F. K., Breningstall, G. N., Wappner, R. S., Byrd, D. J., Sansaricq, C., Tein, I., Grover, W., Valle, D., Rutledge, S. L., and Treem, W. R. (1991) Ann. Neurol. 30, 709–716. Lehninger, A. L. (1982) in Principles of Biochemistry (Lehninger, A. L., Ed.), p. 489. Worth Publ., New York. Lowry, O. H., Passonneau, J. V., Hasselberger, F. X., and Schulz, D. W. (1964) J. Biol. Chem. 239, 18–30. Huth, P., Schmidt, M. G., Hall, P. V., Fariello, R. G., and Shug, A. L. (1981) J. Neurochem. 36, 715–723. Hannuniemi, R., and Kontro, P. (1988) Neurochem. Res. 13, 317– 323. Bohmer, T., Eiklid, K., and Jonsen, J. (1977) Biochim. Biophys. Acta 465, 627–633. Molstad, P., Bohmer, T., and Eiklid, K. (1977) Biochim. Biophys. Acta 471, 296–304. Bahl, J., Navin, T., Manian, A., and Bressler, R. (1981) Circ. Res. 48, 378–385. Vary, T. C., and Neely, J. R. (1982) Am. J. Physiol. 242, H585– H592.
AP: Archives
CHARACTERIZATION OF HUMAN PLASMALEMMAL CARNITINE TRANSPORTER 19. Vary, T. C., and Neely, J. R. (1983) Am. J. Physiol. 244, H247– H252. 20. Sartorelli, L., Ciman, M., Rizzoli, V., and Siliprandi, N. (1982) Ital. J. Biochem. 31, 261–268. 21. Rebouche, C. R. (1977) Biochim. Biophys. Acta. 471, 145–155. 22. Willner, J. H., Ginsburg, S., and DiMauro, S. (1978) Neurology 28, 721–724. 23. Huth, P. J., Thomsen, J. H., and Shug, A. L. (1978) Life Sci. 23, 715–722. 24. Huth, P. J., and Shug, A. (1980) Biochim. Biophys. Acta 602, 621–634. 25. Rebouche, C. J., and Mack, D. L. (1984) Arch. Biochem. Biophys. 235, 393–402. 26. Christiansen, R. Z., and Bremer, J. (1976) Biochim. Biophys. Acta 448, 562–577. 27. Kispal, G., Melegh, B., Alkonyi, I., and Sandor, A. (1987) Biochim. Biophys. Acta 896, 96–102. 28. Hamilton, J. W., Li, B. U. K., Shug, A., and Olsen, W. A. (1983) Gastroenterology 84, 1180. 29. Crane, R. K. (1977) Rev. Physiol. Biochem. Pharmacol. 78, 99– 159.
/ 6b18$$9412
04-12-96 10:39:20
arca
155
30. Kinne, R., and Schwarz, I. (1978) Kidney Int. 14, 547–556. 31. Moore, C., and Pressman, B. C. (1964) Biochem. Biophys. Res. Commun. 15, 562–567. 32. Pressman, B. C. (1965) Proc. Natl. Acad. Sci. USA 53, 1076– 1083. 33. Pressman, B. C. (1968) Fed. Proc. 27, 1283–1290. 34. Samarzija, I., and Fromter, E. (1976) Pfluegers Arch. 365, R15. 35. Kinne, R., and Murer, H. (1978) Proc. VII Int. Congr. Nephrol. Montreal, pp. 601–608. 36. Grenstein, S., and Rothstein, A. (1986) J. Membr. Biol. 90, 1– 12. 37. Margolis, L. B., Rosovskaja, I. A., and Skulachev, V. P. (1987) FEBS Lett. 220, 288–290. 38. Graham, C. S., and Tashjian, A. H., Jr. (1992) Biochem. J. 288, 137–143. 39. Budunova, I. V., and Mittelman, L. A. (1992) Cell Biol. Toxicol. 8, 63–73. 40. Tein, I., De Vivo, D. C., Ranucci, D., and DiMauro, S. (1993) J. Inherit. Metab. Dis. 16, 135–146. 41. Indiveri, C., Tonazzi, A., Dierks, T., Kramer, R., and Palmieri, F. (1992) Biochim. Biophys. Acta 1140, 53–58.
AP: Archives