Carnitine palmitoyltransferase: characterization of a labile detergent-extracted malonyl-CoA-sensitive enzyme from rat liver mitochondria

Leptin and the control of respiratory gene expression in muscle

Leptin plays a central role in the regulation of fatty acid homeostasis, promoting lipid storage in adipose tissue and fatty acid oxidation in peripheral tissues. Loss of leptin signaling leads to accumulation of lipids in muscle and loss of insulin sensitivity secondary to obesity. In this study, we examined the direct and indirect effects of leptin signaling on mitochondrial enzymes including those essential for peripheral fatty acid oxidation. We assessed the impact of leptin using the JCR:LA-cp rat, which lacks functional leptin receptors. The activities of marker mitochondrial enzymes citrate synthase (CS) and cytochrome oxidase (COX) were similar between wild-type (+/?) and corpulent (cp/cp) rats. In contrast, several tissues showed variations in the fatty acid oxidizing enzymes carnitine palmitoyltransferase II (CPT II), long-chain acyl-CoA dehydrogenase (LCAD) and 3-hydroxyacyl-CoA dehydrogenase (HOAD). It was not clear if these changes were due to loss of leptin signaling or to insulin insensitivity. Consequently, we examined the effects of leptin on cultured C(2)C(12) and Sol8 cells. Leptin (3 days at 0, 0.2, or 2.0 nM) had no direct effect on the activities of CS, COX, or fatty acid oxidizing enzymes. Leptin treatment did not affect luciferase-based reporter genes under the control of transcription factors involved in mitochondrial biogenesis (nuclear respiratory factor-1 (NRF-1), nuclear respiratory factor-2 (NRF-2)) or fatty acid enzyme expression (peroxisome proliferator-activated receptors (PPARs)). These studies suggest that leptin exerts only indirect effects on mitochondrial gene expression in muscle, possibly arising from insulin resistance.

Complementation analysis of carnitine palmitoyltransferase I and II defects

Carnitine palmitoyltransferase (CPT) consists of two activities located in the outer (CPT I) and the inner (CPT II) mitochondrial membranes. CPT II deficiency in the adult as well as in the infantile form of the disease has been shown to result from mutations in the CPT II cDNA. Nothing is known regarding the genetic defect in CPT I deficiency. We carried out complementation experiments between CPT I- and infantile CPT II-deficient cell lines. Restoration of 3H2O release from [9,10(n)-3H]-palmitate was chosen as criterion of complementation. As expected, no complementation was observed in heteropolykaryons resulting from fusions between CPT II-deficient cells. Similar results were obtained in fusions between CPT I-deficient cells, suggesting that the enzymatic defect in these cell lines results from mutations in the same gene. Conversely, complementation was observed in fusions between CPT I- and CPT II-deficient cells. These data support that CPT I and CPT II defects result from mutations in distinct genes. Palmitate oxidation by control or CPT I-deficient cell lines was decreased when cocultured with infantile CPT II-deficient cell lines. This effect, not observed in coculture including an adult CPT II-deficient cell line, was carnitine-dependent. The possible mechanism of this effect, suppressed by a high carnitine concentration, is discussed.

Involvement of microsomal vesicles in part of the sensitivity of carnitine palmitoyltransferase I to malonyl-CoA inhibition in mitochondrial fractions of rat liver

Liver mitochondrial fractions as normally isolated contain only 10-20% of total mitochondria and may not be representative of the whole mitochondrial population. This study was designed to evaluate the dependence of the sensitivity of carnitine palmitoyl-transferase I (CPT I) to malonyl-CoA inhibition in mitochondrial fractions that are not normally studied. Four fractions prepared from rat liver were found to be contaminated to different extents by microsome vesicles, on the basis of marker-enzyme activities and micrographic data. Purification of mitochondrial fractions on a Percoll gradient decreased to some extent the microsomal contamination, which was due in part to the existence of close bonds between microsomes and the outer membranes of mitochondria. A greater degree of contamination of mitochondrial fractions by microsomes was correlated with a greater sensitivity of CPT I to malonyl-CoA inhibition. Attempts were made to enhance the sensitivity of CPT I to malonyl-CoA with the use of microsomes. Measurements performed by adding mitochondria and microsomes in the same CPT I assay failed to demonstrate any significant enhancement of malonyl-CoA inhibition. However, addition of ATP to a mixture of mitochondria and microsomes was shown to trigger the binding of both particles, as assessed by enzymic and micrographic data, and to increase the sensitivity of CPT I to malonyl-CoA inhibition. These results demonstrated that the binding of microsomes to mitochondria, unlike the simple mixing of both particles, was capable of altering the sensitivity of CPT I to malonyl-CoA. The data also suggest that this process could be of physiological importance, owing to the frequency of contiguous zones between mitochondria and endoplasmic reticulum observed in sections of intact liver cells.

Restoration of malonyl-CoA sensitivity of soluble rat liver mitochondria carnitine palmitoyltransferase by reconstitution with a partially purified malonyl-CoA binding protein

Solubilization of rat liver mitochondria in 5% Triton X-100 followed by chromatography on a hydroxylapatite column resulted in the identification of malonyl-CoA binding protein(s) distinct from a major carnitine palmitoyltransferase activity peak. Further purification of the malonyl-CoA binding protein(s) on an acyl-CoA affinity column followed by sodium dodecyl sulfate gel electrophoresis indicated proteins with Mr mass of 90 and 45-33 kDa. A purified liver malonyl-CoA binding fraction, which was devoid of carnitine palmitoyltransferase, and a soluble malonyl-CoA-insensitive carnitine palmitoyltransferase were reconstituted by dialysis in a liposome system. The enzyme activity in the reconstituted system was decreased by 50% in the presence of 100 microM malonyl-CoA. Rat liver mitochondria carnitine palmitoyltransferase may be composed of an easily dissociable catalytic unit and a malonyl-CoA sensitivity conferring regulatory component.

Use of mitochondrial inner membrane proteins and phospholipids to facilitate disengagement of the catalytic and malonyl-CoA binding components of carnitine palmitoyltransferase from liver mitochondrial outer membranes

1. It was shown by Ghadiminejad and Saggerson (1991) that the anionic detergent cholate caused disengagement of the malonyl-CoA binding entity from the catalytic entity of outer membrane carnitine palmitoyltransferase (CPT1). 2. This disengagement was only observed if inner membrane material was present. 3. It is now shown that this effect is mimicked by a CPT-free inner membrane protein fraction together with an inner membrane lipid extract or with individual phospholipids (phosphatidylcholine, phosphatidylethanolamine or diphosphatidylglycerol). 4. The lipids alone have no effect but act synergistically with the inner membrane protein fraction.

Regulation of mitochondrial carnitine palmitoyl transferases from liver and extrahepatic tissues

Developments in our understanding of the complex CPT enzyme system over the past ten years have been reviewed. Liver CPT1, which is probably distinct from that in several extrahepatic tissues, is subject to up- or down-regulation of its activity and kinetic properties with changing physiological state. Evidence is now accumulating to support the notion that the catalytic and malonyl-CoA-binding entities of CPT1 are separate polypeptides.

Cholate separates the catalytic and malonyl-CoA-binding components of carnitine palmitoyltransferase from liver outer mitochondrial membranes

Sodium cholate was used as an anionic detergent to discriminate the two components of liver overt carnitine palmitoyltransferase (CPT1); namely a catalytic entity and a regulatory component that bound malonyl-CoA. Cholate solubilized approx. 40% of the malonyl-CoA binding entity from mitochondrial outer membranes without appreciable solubilization of CPT1 activity. Cholate did not interfere with binding of [14C]malonyl-CoA to outer membranes or to crude total mitochondrial membrane fractions. By contrast, the non-ionic detergent Tween-20 was ineffective in solubilizing the malonyl-CoA binding entity and also substantially interfered with the binding of [14C]malonyl-CoA. Both detergents appeared to cause total disengagement of the malonyl-CoA binding entity from the catalytic entity of CPT1 only when some inner membrane material was present. 'Reconstitution' experiments were performed in which a malonyl-CoA sensitivity conferring factor in cholate extracts from outer membranes was associated with CPT derived from inner membranes (CPT2). The IC50 for inhibition of CPT2 by malonyl-CoA in this artificial system was similar to that observed with CPT1 in situ in outer membranes. Extracts containing malonyl-CoA sensitivity conferring factor derived from outer membranes of fed or 48 h fasted rats were associated with CPT2 derived from fed rats. The outer membrane extracts from fasted animals conferred a lower maximum responsiveness to malonyl-CoA, but appeared to have a higher affinity for CPT2 than the extracts from fed rats. These results suggest that physiological state can alter the intrinsic properties of the malonyl-CoA sensitivity confering factor.

Infantile form of carnitine palmitoyltransferase II deficiency with hepatomuscular symptoms and sudden death. Physiopathological approach to carnitine palmitoyltransferase II deficiencies

Reported cases of carnitine palmitoyltransferase II (CPT II) deficiency are characterized only by a muscular symptomatology in young adults although the defect is expressed in extra-muscular tissues as well as in skeletal muscle. We describe here a CPT II deficiency associating hypoketotic hypoglycemia, high plasma creatine kinase level, heart beat disorders, and sudden death in a 3-mo-old boy. CPT II defect (-90%) diagnosed in fibroblasts is qualitatively similar to that (-75%) of two "classical" CPT II-deficient patients previously studied: It resulted from a decreased amount of CPT II probably arising from its reduced biosynthesis. Consequences of CPT II deficiency studied in fibroblasts differed in both sets of patients. An impaired oxidation of long-chain fatty acids was found in the proband but not in patients with the "classical" form of the deficiency. The metabolic and clinical consequences of CPT II deficiency might depend, in part, on the magnitude of residual CPT II activity. With 25% residual activity CPT II would become rate limiting in skeletal muscle but not in liver, heart, and fibroblasts. As observed in the patient described herein, CPT II activity ought to be more reduced to induce an impaired oxidation of long-chain fatty acids in these tissues.

Carnitine palmitoyltransferase (CPT2) from liver mitochondrial inner membrane becomes inhibitable by malonyl-CoA if reconstituted with outer membrane malonyl-CoA binding protein

A soluble extract was obtained on treatment of rat liver mitochondrial outer membranes with cholate which bound [14C]malonyl-CoA but was essentially free of carnitine palmitoyltransferase (CPT) activity. Extraction of mitochondrial inner membranes with cholate readily solubilized a CPT activity which was insensitive to malonyl-CoA. Combination of these two extracts caused the CPT derived from inner membranes to become inhibitable by malonyl-CoA.

Characterization of a solubilized malonyl-CoA-sensitive carnitine palmitoyltransferase from the mitochondrial outer membrane as a protein distinct from the malonyl-CoA-insensitive carnitine palmitoyltransferase of the inner membrane

By using octyl glucoside in the presence of glycerol, it is possible to obtain a solubilized malonyl-CoA-sensitive carnitine palmitoyltransferase (CPTo) from the outer membranes of rat liver mitochondria. H.p.l.c. on hydroxyapatite column has now allowed a clear separation of the CPTo from the malonyl-CoA-insensitive CPT activity of the inner membranes (CPTi). The separated CPTo activity showed inhibition by low micromolar concentrations of malonyl-CoA, 2-tetradecylglycidyl-CoA and etomoxir-CoA. On solubilization and fractionation, the CPTo rapidly lost activity, unlike the relatively stable CPTi activity. Reconstitution into asolectin liposomes enhanced the activity and the malonyl-CoA-sensitivity of the CPTo fractions, whereas it had no such effect on the activity or malonyl-CoA insensitivity of the CPTi fractions. A polyclonal antibody raised against the malonyl-CoA-insensitive enzyme, purified from the inner membranes, precipitated the CPTi activity, but showed no reactivity with the CPTo fractions. In Western blots, the above antibody did not react with any polypeptide of the CPTo fractions. Incubation of the outer-membrane preparations with [3H]etomoxir, in the presence of ATP and CoA, led to labelling of a 90 kDa polypeptide that in the above hydroxyapatite chromatography was eluted in the same region as the CPTo. No such polypeptide labelling was seen in the CPTi fractions. With heart and skeletal-muscle mitochondria, the correspondingly labelled polypeptide was of about 86 kDa. These results show that the CPTo and CPTi are distinct proteins, that a subunit of 90 kDa for liver and 86 kDa for muscle constitutes a component of their respective CPTo systems, and that the 66 kDa subunit of the CPTi does not constitute a part of the CPTo system.

Freeze-thawing causes masking of membrane-bound outer carnitine palmitoyltransferase activity: implications for studies on carnitine palmitoyltransferases deficiency

Carnitine-dependent transport of fatty acids into mitochondria is believed to require participation of two carnitine palmitoyltransferase (CPT) activities, one outer, overt (CPTo) and the other inner, latent (CPTi). For exposing the CPTi and monitoring of the total CPT activity, freeze-thawing and sonication have been frequently employed as membrane-disruptive procedures, particularly when examining for CPT-deficiency diseases. Our evaluations have shown, however, that freeze-thawing and sonication yield misleading data for both the CPT activities owing to their previously unrecognized masking and unmasking effects on CPT activities. Formation of vesicular/sheath structures with mixed membrane orientation that prevents the access of medium substrate to enzymes on both aspects of the membrane at the same time appears responsible for these results. That such procedures can yield inexact data when monitoring the latency and sidedness of other membrane-bound biocatalysts as well needs to be recognized. We show that in muscle mitochondria also, a malonyl-CoA-inhibitable CPTo activity resides in the outer membrane, while a malonyl-CoA-insensitive, CPTi, activity is present in the inner membrane. Our results rationalize why Zierz and Engel ((1987) Neurology 37, 1785) were unable to obtain evidences for a latent CPT activity in mitochondria particularly of muscles. Although simple methods to allow an unambiguous quantitation of the two CPT activities in tissue extracts remain unavailable, evaluation of the possibility that two different CPT deficiencies occur appears justified.

Characteristics of the suppressive effect of nicardipine on peroxisome induction in rat liver

In vivo administration of nicardipine, a known calcium antagonist, suppressed the clofibrate-evoked induction of activities of peroxisomal enzymes, such as catalase, the peroxisomal fatty acyl-CoA oxidizing system, carnitine acetyltransferase and mitochondrial carnitine palmitoyltransferase in rat liver. On a time-course study, the suppression of induction in the activities of the peroxisomal fatty acyl-CoA oxidizing system and carnitine acetyltransferase was found at 5 days after the treatment, whereas the induction by clofibrate was already observed at 1 day after the treatment, suggesting that in the process of peroxisome induction by clofibrate there might be two steps, i.e., a triggering step and an enhancing step, and nicardipine might act as suppressor for the later step. The precursor-incorporation studies with [3H]leucine showed that the rate of the synthesis of the peroxisomal bifunctional enzyme was increased by 4.2-fold after clofibrate-treatment, whereas nicardipine suppressed this enhancement to only 2.2-fold of the control. The rate of degradation of this enzyme was not affected by any treatment. These results show that nicardipine affects the regulation mechanism of the biosynthesis of this enzyme. Nicardipine showed hardly any suppressive-effect on the hepatic peroxisomal enzyme induction observed in high-fat diet fed rat. Furthermore, the suppression of clofibrate-evoked induction of peroxisomal enzymes was observed also in mice. These interesting findings suggest that there is a difference in the mechanism of peroxisome proliferation and/or the induction of peroxisomal enzymes between clofibrate and physiological conditions, such as high-fat diet feeding. The suppression of drug-induced peroxisome proliferation by calcium antagonists may help in dissecting the causal relationship between the multiple effects mediated by peroxisomal proliferators.

Target size analysis by radiation inactivation of carnitine palmitoyltransferase activity and malonyl-CoA binding in outer membranes from rat liver mitochondria

The functional molecular sizes of the protein(s) mediating the carnitine palmitoyltransferase I (CPT I) activity and the [14C]malonyl-CoA binding in purified outer-membrane preparations from rat liver mitochondria were determined by radiation-inactivation analysis. In all preparations tested the dose-dependent decay in [14C]malonyl-CoA binding was less steep than that for CPT I activity, suggesting that the protein involved in malonyl-CoA binding may be smaller than that catalysing the CPT I activity. The respective sizes computed from simultaneous analysis for molecular-size standards exposed under identical conditions were 60,000 and 83,000 DA for malonyl-CoA binding and CPT I activity respectively. In irradiated membranes the sensitivity of CPT activity to malonyl-CoA inhibition was increased, as judged by malonyl-CoA inhibition curves for the activity in control and in irradiated membranes that had received 20 Mrad radiation and in which CPT activity had decayed by 60%. Possible correlations between these data and other recent observations on the CPT system are discussed.

Normal muscle CPT1 and CPT2 activities in hepatic presentation patients with CPT1 deficiency in fibroblasts. Tissue specific isoforms of CPT1?

Human carnitine palmitoyltransferase (CPT) deficiency results in 2 clinical forms: a more common "muscular form" with myoglobinuria with or without delayed or impaired ketogenesis and a rare "hepatic form" with hypoketotic hypoglycemia, encephalopathy and seizures without muscular manifestations. We present 2 patients, a male (patient 1) and a female (patient 2) with infantile "hepatic" CPT deficiency and previously documented CPT1 deficiency in fibroblasts. In patient 2, a deficiency of "total" CPT activity in liver had also been previously documented. We set up an isotope exchange assay system that effectively differentiated CPT1 and CPT2 activities in muscle. We found normal CPT1 and CPT2 activities in our patients under near saturating substrate conditions. The CPT1 and CPT2 activities were suppressed to a strikingly similar degree under different kinetic conditions as compared to control muscle and were found to have similar Km values for carnitine and PCoA. With Km concentrations of carnitine, the mean residual activities of CPT1 for patients 1 and 2 were 49 and 44%, respectively (control range 40-53%); the mean residual activities of CPT2 were 60 and 46%, respectively (control range 49-59%). With Km concentrations of PCoA, the mean residual activities of CPT1 for patients 1 and 2 were 52 and 58%, respectively (control range of 52-59%); mean residual activities of CPT2 were 54% and 56%, respectively (control range of 51-68%). When the Vmax concentration of PCoA was doubled and bovine serum albumin reduced to 0.1%, the mean residual activities of CPT1 for patients 1 and 2 were 69 and 63%, respectively (control range 60-80%). In "muscular" patients, a marked absolute deficiency of CPT2 activity (less than 12% residual) was found with an apparent increased sensitivity to suppression of enzymatic activity when the Km concentration of carnitine was used. We suggest that CPT1 and CPT2 may be separate proteins. Furthermore, CPT1 itself may exist as tissue-specific isoforms being the same protein in liver and fibroblasts and a different protein in muscle. Either could be encoded for by the same or closely related genes.

A case of carnitine palmitoyltransferase II deficiency in human skeletal muscle

A 20-year-old man was shown to have a deficiency of carnitine palmitoyltransferase (CPT) II in skeletal muscle. The evidence was: (i) there was no significant oxidation of [9,10-3H]-palmitate or of [1-14C]palmitate in mitochondrial fractions from fresh skeletal muscle from the patient; (ii) all the CPT activity in a homogenate of fresh muscle from the patient was overt (CPT I) with no increase in activity after the inner membrane was disrupted; (iii) all the CPT activity in the patient's muscle was inhibited by malonyl-CoA; and (iv) an immunoreactive peptide of 67 kDa corresponding to CPT II, present in mitochondria from controls, was absent in those from the patient.

Interactions of inhibitors of carnitine palmitoyltransferase I and fibrates in cultured hepatocytes

Culture of rat hepatocytes with etomoxir, an inhibitor of carnitine palmitoyltransferase I (CPT I), for 48 h, resulted in increased carnitine acetyltransferase (CAT) activity (74%), a marked decrease in CPT activity (82%) measured in detergent extracts, and increased activities of glucose-6-phosphate dehydrogenase (227%) and fructose-1,6-bisphosphatase (65%). Changes in CAT and CPT activities were not observed after 4 h culture with etomoxir. When hepatocytes were cultured with etomoxir and benzafibrate (a hypolipidaemic analogue of clofibrate) for 48 h, etomoxir prevented the 5-fold increase in CAT activity caused by bezafibrate, whereas bezafibrate suppressed the increase in glucose-6-phosphate dehydrogenase and fructose-bisphosphatase caused by etomoxir. However, bezafibrate did not prevent the suppression of CPT activity by etomoxir. Etomoxir inhibited palmitate beta-oxidation and ketogenesis after short-term (0-4 h) and long-term (48 h) exposure, but it caused accumulation of triacylglycerol in hepatocytes only after short-term exposure (0-4 h). These effects of etomoxir on fatty acid metabolism and suppression of CPT (after 48 h) were similar in periportal and perivenous hepatocytes, but the increases in CAT and glucose-6-phosphate dehydrogenase activities were higher in periportal than in perivenous cells. The effects of CPT I inhibitors on CAT activity and long-term suppression of CPT activity are probably mediated by independent mechanisms.

Evidence for distinct functional molecular sizes of carnitine palmitoyltransferases I and II in rat liver mitochondria

1. Estimates of the functional sizes of the molecular species responsible for the overt (I) and latent (II) activities of carnitine palmitoyltransferase (CPT) in 48 h-starved rat liver mitochondria were obtained from radiation inactivation experiments. 2. The decay in the activity of total CPT and that of CPT II only (after inhibition of CPT I) was measured in mitochondrial samples exposed to different doses of high-energy ionizing radiation. 3. The decay curves obtained by plotting residual activity of total CPT as a logarithm function of irradiation dose suggested the contribution of more than one target towards total CPT activity. 4. By contrast, in mitochondria in which CPT I activity was approximately 95% inhibited, the activity of CPT decayed in a simple mono-exponential manner. Target-size analysis yielded an approximate Mr of 69,700 for this component (CPT II). 5. This information, as well as that on the relative non-irradiated activities of CPT I and CPT II, was used in graphical and statistical methods to obtain the parameters of the decay curve for CPT I. These analyses yielded an approximate Mr of 96,700 for CPT I.

Pyrene dodecanoic acid coenzyme A ester: peroxisomal oxidation and chain shortening

Pyrenedodecanoyl-CoA was beta-oxidized by isolated rat liver peroxisomes at a rate which was about 50% of that observed with palmitoyl-CoA. Measurement of the quantity of NADH formed from a limiting amount of pyrenedodecanoyl-CoA suggested that it was subjected to two to three cycles of beta-oxidation. Pyrenedodecanoyl-CoA was a very poor substrate for carnitine palmitoyltransferase, exhibiting less than 1% of the rate obtained with palmitoyl-CoA; it also was a strong inhibitor of this enzyme. With rat liver microsomal alpha-glycerophosphate acyltransferase the rate of reaction with pyrenedodecanoyl-CoA was only 3-4% of that observed with palmitoyl-CoA.

Some differences in the properties of carnitine palmitoyltransferase activities of the mitochondrial outer and inner membranes

Recent evidence has shown that the outer, overt, malonyl-CoA-inhibitable carnitine palmitoyltransferase (CPTo) activity resides in the mitochondrial outer membrane [Murthy & Pande (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 378-382]. A comparison of CPTo activity of rat liver mitochondria with the inner, initially latent, carnitine palmitoyltransferase (CPTi) of the mitochondrial inner membrane has revealed that the presence of digitonin and several other detergents inactivates CPTo activity. The CPTi activity, in contrast, was markedly stimulated by various detergents and phospholipid liposomes. These findings explain why in previous studies, which used digitonin or other detergents to expose, separate and purify the CPT activities, the inferences were drawn that (a) the ratio of latent to overt CPT was quite high, (b) both the CPT activities could be ascribed to one active protein recovered, and (c) the observed lack of malonyl-CoA inhibition indicated possible loss/separation of a putative malonyl-CoA-inhibition-conferring protein. Although both CPTo and CPTi were found to catalyse the forward and the backward reactions, CPTo showed greater capacity for the forward reaction and CPTi for the backward reaction. The easily solubilizable CPT, released on sonication of mitoplasts or of intact mitochondria under hypo-osmotic conditions, resembled CPTi in its properties. When octyl glucoside was used under appropriate conditions, 40-50% of the CPTo of outer membranes became solubilized, but it showed limited stability and decreased malonyl-CoA sensitivity. Malonyl-CoA-inhibitability of CPTo was decreased also on exposure of outer membranes to phospholipase C. When outer membranes that had been exposed to octyl glucoside or to phospholipase C were subjected to a reconstitution procedure using asolectin liposomes, the malonyl-CoA-inhibitability of CPTo was restored. A role of phospholipids in the malonyl-CoA sensitivity of CPTo is thus indicated.