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

Biochimicu Elsevier

et Biophysics

67

Actu 918 (1987) 67-75

BBA 52459

Camitine palmitoyltransferase:

characterization of a labile detergent-extracted

malonyl-CoA-sensitive

enzyme from rat liver mitochondria Henrik Lund

Institute

of Medical Biochemistry,

Univemty

(Received

Key words:

Camitine

palmitoyltransferase;

of Oslo, Blindern, Oslo (Norway)

15 October

Malonyl-CoA;

1986)

Detergent;

Enzyme inhibitor;

(Rat liver mitochondria)

(1) Rat liver mitochondria were preextracted with Triton X-100 in the absence of salts to remove malonyl-CoA-insensitive camitine palmitoyltransferase. (2) From the remaining membrane residues a malonyl-CoA-sensitive enzyme was solubilized with octyl glucopyranoside in the presence of KCI. Significant enzyme activity, [2-14C]malonyl-CoA binding and malonyl-CoA inhibition of this enzyme was present only after removal of detergent by precipitation with poly(ethylene glycol). The enzyme activity was rapidly lost in the solubilized form. High concentrations of glycerol protected the enzyme. (3) The alkylating irreversible inhibitor, S-(4-bromo-2,3-dioxobutyl)-CoA, strongly inhibited the malonyl-CoA-sensitive enzyme in the membrane residues. The enzyme was protected against this inhibitor by malonyl-CoA and palmitoyl-CoA. (4) The more loosely membrane-bound malonyl-CoA-insensitive enzyme failed to bind malonyl-CoA, was stable in the presence of detergents and was not inhibited by S-(4-bromo-2,3-dioxobutyl)CoA. (5) It is suggested that two different carnitine palmitoyltransferase proteins exist in the inner mitochondriaf membrane and that the detergent-labile malonyl-CoA-sensitive enzyme is the less easily extracted of the two.

Introduction Extensive studies on membrane-bound carnitine palmitoyltransferase in mitochondria have shown the existence of an external and an internal latent form of the enzyme. Only the external enzyme is inhibited by malonyl-CoA. However, upon extraction of the enzyme activity from rat liver mitochondria with detergents an almost complete loss of malonyl-CoA inhibition is observed and, irrespective of the methods employed, only one

Abbreviations: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulphonic acid; GSH, glutathione (reduced form); PMSF, phenylmethanesulphonyl fluoride. Correspondence: H. Lund, Institute of Medical Biochemistry, University of Oslo, P.B. 1112, Blindem, 0137 Oslo 3, Norway.

0005-2760/87/$03.50

0 1987 Elsevier Science Publishers

carnitine palmitoyltransferase protein has been purified [1,2]. This suggests either that only one enzyme protein is present in the membrane and that the outer part of the enzyme is associated with a regulatory component, or that the activity of a distinct malonyl-CoA-sensitive enzyme is completely lost during purification procedures. In rat liver the fraction of carnitine palmitoyltransferase remaining in the membrane after detergent extraction is still malonyl-CoA-sensitive [3,4]. We have also demonstrated the chromatographic separation of malonyl-CoA-binding proteins from most of the carnitine palmitoyltransferase activity [5]. We have interpreted these observations to support the existence of a complex of the enzyme and a regulatory unit more difficult to extract than the uncomplexed enzyme. However, numerous attempts to recombine the enzyme

B.V. (Biomedical

Division)

68

with malonyl-CoA-binding components have not been successful. No convincing malonyl-CoA inhibition in such a reconstituted system has been obtained. In the present paper we have studied the properties of the malonyl-CoA enzyme remaining in the membrane after removal of most of the carnitine pahnitoyltransferase by detergent extraction. The conditions for a combined detergent/salt extraction of a labile malonyl-CoA-sensitive enzyme from the membrane residues are given. The properties of this enzyme are compared with the properties of the malonyl-CoA-insensitive carnitine pahnitoyltransferase more easily extracted from intact mitochondria. A preliminary report on some of the results has been presented [6]. Materials and Methods Materials [2-‘4C]Malonyl-CoA (48.7 mCi/ mmol) was from Amersham International, Amersham, Bucks., U.K. [G-3H]Coenzyme A (3.51 Ci/mmol) .was from New England Nuclear, Boston, MA, U.S.A. was (5 mCi/ mmol) ( - )-[ methyl- 3H]Camitine prepared according to Ref. 7. 1,4-Dibromo-2,3butanedione was obtained from Aldrich Chemie, Steinheim, F.R.G. S-(4-Bromo-2,3-dioxobutyl)CoA and S-(4-bromo-2,3-dioxobutyl)[G-3H]CoA (7500-30 000 cpm/nmol) were prepared according to Ref. 8. (+)-Palmitoylcarnitine and (+)-decanoylcamitine were prepared according to Ref. 9. Palmitoyl-CoA was prepared according to Ref. 10. N-Octyl p-D-glucopyranoside, Triton X-100, sodium dodecyl sulphate, PMSF, aprotinin, CoA and malonyl-CoA were from Sigma Chemical Co., St. Louis, MO, U.S.A. Poly(ethylene glycol) 6000 was from Fluka A.G., Buchs, Switzerland. All other reagents were of analytical grade. Rat liver mitochondria were prepared at 0°C from fed male Wistar rats (150-250 g), which had had free access to a standard pellet diet and water, in an isolation medium comprising 75 mM sucrose, 225 mM mannitol, 25 mM Hepes and 1 mM EGTA (pH 7.4) according to Ref. 11. Mitochondrial membrane residues were prepared by twice extracting mitochondria suspended in isolation medium (17-20 mg/ml) with 0.5% (w/v) Triton X-100 as previously described [4]. This procedure removes more than 90% of cami-

tine palmitoyltransferase activity and approx. 80% of the mitochondrial matrix and membrane protein. A lower detergent/protein ratio during preextraction tended to reduce malonyl-CoA sensitivity of the enzyme in the membrane residues due to incomplete removal of insensitive enzyme. A higher detergent/protein ratio removed almost all carnitine palmitoyltransferase activity from the membrane residues. The membrane residues could be washed repeatedly in isolation medium and stored at - 70°C in the same medium for several weeks without deterioration of enzyme activity or malonyl-CoA inhibition. Upon repeated freezing and thawing, however, the enzyme activity and malonyl-CoA sensitivity was gradually lost. Malonyl-CoA-sensitive enzyme was extracted from the membrane residues. To-a 300 ~1 suspension of membrane residues (6-8 mg/ml) in isolation medium were added 75 ~1 200 mM octylglucoside, 5 $100 mM Hepes (pH 7.4), 25 ~12 M KC1 and 100 ~1 water. The mixture was vortexed, left on an ice/water bath for 5 min and then centrifuged for 30 min at 30000 x g. The clear supematant obtained after centrifugation was removed and, if not otherwise stated, immediately precipitated by addition of poly(ethylene glycol) 20% (w/v). The sample was vortexed thoroughly and centrifuged at 30000 x g for 30 min. The poly(ethylene glycol)-containing supematant was subsequently removed by aspiration, the pellet suspended in isolation medium at a protein concentration of 6-8 mg/ml and used immediately for assays. Poly(ethylene glycol) in the range 15-25% completely precipitated the octyl glucopyranosidesolubilized proteins under the experimental conditions used. When resuspended in the isolation medium, the crude malonyl-CoA-sensitive enzyme preparation was insoluble and had the same specific camitine palmitoyltransferase activity as the enzyme in the membrane residues. About half of the membrane residue proteins and all carnitine palmitoyltransferase activity was extracted; thus, apparently about half of the enzyme activity was lost during solubilization. Malonyl-CoA-insensitive enzyme was prepared from intact mitochondria suspended in isolation medium (40 mg/ml). The mitochondria were frozen, thawed and centrifuged at 30000 X g for

IO min to remove matrix proteins. The resulting pellet was resuspended (6-8 mg/ml) in isolation medium. Octyl glucopyranoside was then added to a concentration of 30 mM and the suspension was centrifuged at 30000 x g for 30 min. To clear supematant were added 1 mM Hepes and 100 mM KC1 and, if not stated otherwise, poly(ethylene glycol) (20% (w/v)) to precipitate the proteins. (Salts were added after the solubilization step in order to avoid any extraction of malonyl-CoAsensitive enzyme and to obtain conditions identical to those used for pr~ipitation and assay of the malonyl-CoA-sensitive enzyme.) The pellet was resuspended in isolation medium at 6-8 mg protein/ml and used immediately for assays. The malonyl-CoA-insensitive enzyme was insoluble in the isolation medium. Assay of S-(4-bromo-2,3-dioxobutyl)-CoA inhibition or binding was performed by adding freshly prepared 3H-labelled or unlabelled S-(4bromo-2,3-dioxobutyl)-CoA dissolved in isolation medium to membrane residues (6-8 mg/ml) or precipitated insensitive enzyme (6-8 mg/ml) at 0°C or 3O”C, and the reaction was stopped by the addition of ~t~othreitol at a final concentration of 100 mM. The protein samples were then washed with isolation medium (3 ml) four times to remove unbound reagents and samples were taken for determination of malonyl-CoA binding, carnitine palmitoyltransferase activity or S-(4-bromo-2,3-dioxobutyl)-CoA-bound 3H radioactivity. Measurements of [2-‘4C]malonyl-CoA binding was performed at 0°C for 20 min in a total volume of 0.5 ml containing 0.1-0.5 FM [2r4C]malonyl-CoA (50 ~-60 000 cpm/ nmol), 150 mM KCI, 1% (w/v) albumin, 10 mM Hepes (pH 7.4). Assays were initiated by addition of 0.1 ml membrane protein (0.6-0.8 mg protein). Quantitation of bound radioactivity has been described previously [ 51. Assay of camitine palmitoyltransferase activity was performed at 30°C for 5 min in a total volume of 0.5 ml containing 0.4 mM ( - )-[methyl3H]camitine (5000-7000 cpm/nmol), 150 mM KCI, 1% (w/v) albumin, 50 FM pahnitoyl-CoA and 10 mM Hepes (pH 7.4). Assays were initiated by addition of 0.1 ml membrane protein (0.6-0.8 mg protein). Extraction and quantitation of radioactivity has been described previously 141. Protein

was measured with the method of Lowry et al. 1121. When proteins were determined in the presence of detergents, 1% sodium dodecyl sulphate was included in the assay. Results Fig. 1 shows that detergents strongly inhibited both camitine palmitoyltransferase activity (A) and malonyl-CoA binding (B) in the membrane residues, while poly(ethylene glycol) was virtually inert in this respect. Octyl glucopyr~oside did not inhibit at low concentrations. Fig. 2A shows how the malonyl-CoA-sensitive camitine palmitoyltransferase is extracted from the membrane residues by octyl glucopyronaside and KCl. Assay in the supematant showed apparently a poor recovery of enzyme activity and malonyl-CoA inhibition. However, this was due to the presence of inhibitory concentrations of detergent in the enzyme assay in the supematant. When detergent was removed by poly(ethylene glycol) precipitation a much better recovery of enzyme with malonyl-CoA sensitivity was obtained (Fig. 2B). A biphasic pattern of enzyme recovery was observed. At high detergent concentrations an almost complete extraction of protein was obtained, but camitine palmitoyltransferase activity was gradually lost. Enzyme activity measured in the presence of 5 PM malonyl-CoA (insensitive enzyme) was much less affected. Incomplete precipitation or incomplete removal of detergent from the solubilized enzyme did not cause this loss of enzyme activity, since no protein or camitine p~~toyltransferase activity was detectable in the supematants after poly(ethylene glycol) precipitation. The same biphasic pattern of enzyme recovery was observed if the octyl glucopyranoside concentration was kept constant and the malonyl-CoA-sensitive enzyme was extracted by increasing KC1 concentration, or if Triton X100 was used instead of octyl glucopyranoside (results not shown). The biphasic pattern of enzyme recovery suggested enzyme inactivation at high detergent and salt concentrations. We therefore compared the stability of the extracted malonyl-CoA-sensitive and -insensitive enzymes in the presence of detergents and salts. Fig. 3A shows that when poly(eth-

Detergents

ImM

I

I

12.5 PEG

%

1

Detergents 25.0

(o-1

(mM

I

I

I

25.0

12.5 PEG

%

1

W-0)

Fig. 1. The effect of detergents on malonyl-CoA-sensitive camitine palmitoyltransferase activity (A) and on specific binding of 0.2 KM [1-r4C]malonyl-CoA (B) in membrane residues. Detergents were added to the standard assay mixtures for measurements of camitine palmitoyltransferase activity and malonyl-CoA binding (see Materials and Methods). q, octyl glucopyranoside; V. (+)-decanoylcamitine; 0, (+)-palmitoylcamitine; A, Triton X-100 (M, = 650): l, poly(ethylene glycol) 6000 (PEG).

ylene glycol) precipitation of the octyl glucopyranoside-solubilized malonyl-CoA-sensitive enzyme was delayed for the indicated length of time, activity was rapidly lost *. Again, activity measured in the presence of malonyl-CoA (insensitive enzyme) was only slightly decreased. This was contrasted by the finding that only 10% loss of activity occurred throughout the observation period (22 h) if octyl glucopyranoside was removed from the enzyme by poly(ethylene glycol) precipitation immediately after its extraction from the membrane residues. The malonyl-CoA-insensitive enzyme measured under identical conditions did not show any decrease in activity in the presence of detergent (octyl glucopyranoside or Triton X-100). From Fig. 3B it is evident that the rate of enzyme inactivation was increased if additional octyl glucopyranoside or KC1 was added to the already solubilized enzyme prior to poly(ethylene glycol) precipitation. * The observed Tl,? was approximately 3-4 h. Since about half of the enzyme activity was already lost during centrifugation of the solubilized membrane residues (30 min), the initial loss of enzyme activity is probably very rapid.

In a series of control experiments, several agents were tested for their ability to counteract the loss of enzyme activity. The SH-group protectors GSH and dithiothreitol were without effect, as were the proteolysis inhibitors aprotinin and PMSF. pH was varied between 6.0 and 8.0 without influence on enzyme activity, but pH values below 6.0 and above 8.0 increased the inactivation of the enzyme. Increasing concentrations of glycerol gave a partial protection of the enzyme in the presence of detergents and significantly delayed the rate of inactivation. However, even in the presence of high concentrations of glycerol we did not obtain any further purification of the malonyl-CoA-sensitive enzyme using some conventional column chromatographic procedures (results not shown). After removal of detergents with poly(ethylene glycol), the enzyme extracted under optimized conditions was very sensitive to malonyl-CoA inhibition. An apparent IC,, of 0.1 PM was found both for the extracted enzyme and the enzyme in the membrane residues, although the extracted enzyme showed the presence of some insensitive enzyme (Fig. 4A). This is the lowest IC,, reported

71

Fig. 2. Octyl glucopyranoside extraction of malonyl-CoA-sensitive camitine palmitoyltransferase activity from membrane residues. Effect of detergent removal from the extract. (A) Solubilization was performed at 0°C by exposing membrane residues (6-8 mg protein/ml) to increasing concentrations of octyl glucopyranoside in the presence of 100 mM KC1 and 10 mM Hepes (pH 7.4). After centrifugation (30000X 8 for 30 min), the clear supernatants were removed and camitine palmitoyltransferase activity was measured in the remaining membrane residue and in the octyl glucopyranoside-containing supematants. (Octyl glucopyranoside was diluted 5-fold in the assays of supematant activity.) 0, 0, enzyme activity in the remaining membrane residues with and without 5 pM malonylCoA. q- - - - - -0, W- - - - --¤, enzyme activity in supematants with and without 5 pM malonyl-CoA. (B) Extracts were prepared as in (A) and the supematants immediately precipitated with poly(ethylene glycol) (20%, w/v). The precipitate was resuspended in isolation medium (6-8 mg protein/ml) and carntine palmitoyltransferase activity was measured in the absence (0) and presence (0) of 5 pM malonyl-CoA. Other symbols: q- - - -- -0, protein remaining in membrane residue pellets after octyl glucopyranoside/salt extraction.

for the enzyme in rat liver. The actual IC,, may even be lower, since albumin present in the assay binds malonyl-CoA. The extracted crude enzyme preparation also showed high-affinity specific binding of [2-

“C]malonyl-CoA (Fig. 4B). Approximately the same specific binding was found in membrane residues and in the octyl glucopyranoside-extracted malonyl-CoA-sensitive enzyme. Thus, a reasonable close correlation between malonyl-CoA binding and inhibition was present. The carnitine palmitoyltransferase extracted from whole mitochondria at the same octyl glucopyranoside concentration (30 mM in the absence of KCl) precipitated and assayed under identical conditions had an approx. lOO-fold higher specific activity but did not show any inhibition or binding of malonyl-CoA. The malonyl-CoA-sensitive enzyme in membrane residues and the extracted/reprecipitated malonyl-CoA-sensitive and -insensitive enzyme extracted from whole mitochondria were exposed to the irreversible alkylating inhibitor, S-(4bromo-2,3-dioxobutyl)-CoA. As shown in Fig. 5A, this inhibitor caused an almost complete irreversible inhibition of carnitine palmitoyltransferase activity in the membrane residues. A similar inhibition of the detergent-extracted/reprecipitated enzyme was also obtained (results not shown). Specific binding of malonylCoA to the S-(4-bromo-2,3-dioxobutyl)-CoAtreated membrane residues was almost completely blocked (Fig. 5B), suggesting that the malonylCoA-binding site associated with the malonylCoA-sensitive enzyme was covalently modified through the binding of S-(Cbromo-2,3-dioxobutyl)-CoA. In Fig. 6 the kinetics of S-(4-bromo-2,3-dioxobutyl)-CoA inhibition of malonyl-CoA-sensitive camitine palmitoyltransferase are shown. Inactivation was rapid and dependent on the concentration of the inhibitor. The malonyl-CoA-insensitive enzyme extracted by detergents (octyl glucopyranoside or Triton) was completely insensitive to S-(4-bromo-2,3-dioxobutyl)-CoA, whether tested in the presence or absence of detergents, even at very high concentrations of the inhibitor or prolonged exposure (Fig. 6). The specificity of inhibition of the malonylCoA-sensitive enzyme in the membrane residues was further investigated using 3H-labelled S-(4bromo-2,3-dioxobutyl)-CoA. Fig. 7A shows a fast time-dependent incorporation of radioactivity into the membrane residues

B

5 Time

10

Octyl

(hours)

L

I I 50 100

20 glucopyronoslde

added

J

200 KCI

(mM) 500

added

(mM)

Fig. 3. Inactivation of malonyl-CoA-sensitive camitine palmitoyltransferase in the presence of detergents. (A) Extracts were prepared as in Fig. 2. After centrifugation (30000 X g for 30 min) aliquots of the clear supematant (0.5 ml) was either precipitated immediately (zero time) by poly(ethylene glycol) (20%, w/v), this followed by camitine palmitoyltransferase assay in the precipitates at the indicated time points (0, n) or was kept in the solubilized form for the indicated length of time before poly(ethylene glycol) precipitation, whereupon camitine palmitoyltransferase assays were performed (A, A). Closed symbols represent activity in the presence of 5 gM malonyl-CoA. 0, 0, represent an identical experiment in which extracts of the malonyl-CoA-insensitive enzyme were kept solubilized for the indicated lengths of time before poly(ethylene glycol) precipitation and camitine palmitoyltransferase assays were then performed. (B) Conditions were as in (A) except for the addition of the indicated extra amount of octyl glucopyranoside (0, 0) or KC1 (A, A) to aliquots of the supematant. After 30 min at 0°C poly(ethylene glycol) precipitation and camitine palmitoyltransferase assays were performed. Closed symbols represent activity measured in the presence of 5 PM malonyl-CoA.

B

Malonyl

-CoA

(PM)

73

6

Fig. 5. Effect of S-(~brom~2,3-~oxobutyl)-CoA on camitine palmitoyltransferase activity (A) and specific [2-i4C]malonylCoA binding (B) to membrane residues. Membrane residues (6-8 mg protein/ml) were incubated with 50 pM S-(Cbromo2,3-dioxobutyl)-CoA at 30°C and the reaction was stopped by addition of 100 mM dithiothreitol after 5 min. The S-(4bromo-2,3-dioxobutyl)CoA-treated membrane residues (A, A) and control membrane residues (0,O) were washed four times at O°C with isolation medium (3 ml) before samples were taken for measurements of camitine palmitoyltrausferase activity (A) and malonyi-CoA binding (B). Both total (0, A) and nonspecific (0, A) binding are shown.

using 5 PM of S-(4-bromo-2,3-dioxobutyl)-CoA. 85% saturation of binding was achieved within 60 s. Exposure for up to 5 min increased incorporated radioactivity only by an additional 20%. The presence of malonyl-CoA and palmitoyl-CoA prevented the initial rapid binding of the labelled S-(4-bromo-2,3-dioxobutyl)-CoA (Figs. 7A and 7B). Some variable degree of protection, especially by malonyl-CoA, was obtained, depending on the different membrane residue preparations used, but the protection was clearly reproducible (Fig. 7B). Palmitoyl-CoA apparently gave a more pronounced protection and both CoA esters protected even after 5 min exposure to the alkylating inhibitor. That both the physiological substrate and in-

L

I

15

M

120 I+-----Yw

60 Ttme

(seconds)

Fig. 6. Effect of S-(4-bromo-2,3-dioxobutyl)-CoA concentration and time of exposure on malonyl-CoA-sensitive camitine palmitoyltransferase in membrane residues. (A) Membrane residues (0, A, 0) and solubil~ed/reprecipita~d malonylCoA-insensitive enzyme (0) (6-8 mg protein/ml) were exposed to different concentrations of S-(4-bromu2,3-dioxobutyl)-boa at 30°C as shown. Reactions were stopped by addition of 100 mM dithiothreitol. Washing of the proteins and assay procedures were performed as described in the legend to Fig. 6. Symbols: 0, A, 0, 0; 10, 20, 40 and 100 pM S-(4bromo-2,3dioxobutyl)-CoA, respectively.

hibitor protected against binding of S-(6bromo2,3-dioxobutyl)-CoA suggested that the malonylCoA site was labelled. Palmitoyl-CoA is known to be the most potent displacer of malonyl-CoA binding in membrane residues [5]. Thus, it was expected that both these CoA esters would provide protection against S-(4-bromo-2,3-dioxobutyl)-CoA binding. Apparently, malonyl-CoA provided better protection at lower concentrations (Fig. 7B), protection being more equal at higher concentrations of the two CoA esters.

Fig. 4. Inhibition by malonyl-CoA (A) and specific binding of [2-‘4C]malonyl-CoA (B) to detergent-extracted malonyl-CoA-sensitive camitine palmitoyltransferase. Sensitive (0) and insensitive (0) carnitine palmitoyltransferase activity was solubilized from membrane residues and whole mitochondrial membranes (6-8 mg protein/ml) with 30 mM octyl glucopyranoside and 10 mM Hepes (pH 7.4) in the presence or absence of 100 mM KCl, respectively. After centrifugation (30000~ g for 30 min) the supematants were precipitated by poly(ethylene glycol) (20%. w/v). C amitine pal~toyltr~sfer~e activity and malonyl-CoA binding were measured in the precipitates and compared to activities in intact membrane residues (0). The results are expressed as mean5S.E. from three preparations of membrane proteins from different animals.

Ttme

(seconds)

malonyl-CoA-insesnitive enzyme is easily removed with detergents in the absence of salts. The malonyl-CoA-sensitive enzyme could then be solubilized from the remaining membrane residues only within a narrow range of detergent and salt concentrations used during extraction. Immediate removal of detergents prior to assay was necessary to detect the regulatory properties of this enzyme and to prevent rapid inactivation of the enzyme in the presence of detergents. Only the sensitive enzyme was inhibited by S-(4-bromo-2,3-dioxobutyl)-CoA. This was contrasted to the findings for the more easily extracted malonyl-CoA-insensitive enzyme, which, under identical conditions, showed no deterioration in activity and was completely insensitive to malonyl-CoA and S-(4-bromo-2,3dioxobutyl)-CoA. If the same enzyme protein were present in our two detergent-extracted crude carnitine palmitoyltransferase preparations and their different properties were due only to other membrane components coextracting with the enzyme, one would expect malonyl-CoA inhibition of the sensitive enzyme to disappear in the presence of detergents, not the enzyme activity itself. It is not likely that one enzyme protein can possess these two completely different properties. Thus, our result are most simply explained by two different camitine palmitoyltransferase proteins in the membrane. According to our results, the malonyl-CoA-sensitive enzyme is not identical to the camitine palmitoyltransferase fraction most easily removed when whole mitochondria are exposed to detergents [1,3,13-U]. Because of the quantitative domination of the more easily extracted malonylCoA-insensitive enzyme (more than 90% of total mitochondrial carnitine palmitoyltransferase activity) and the rapid loss of the sensitive enzyme in the presence of detergents, it probably escapes detection during most enzyme fractionation and purification procedures involving the use of detergents. Binding of the irreversible inhibitor S-(4bromo-2,3-dioxobutyl)-CoA to the malonyl-CoAsensitive enzyme was counteracted by the presence of both palmitoyl-CoA and malonyl-CoA. The fact that malonyl-CoA provided protection strongly suggested that the site of S-(rl-bromo2,3-dioxobutyl)-CoA binding to the membrane re-

i VI -“-‘:1-__.... \

1

5

10

Malonyl-CoA

25 (.I

, Pahtoyl-CoA

50 (0)

OIM)

Fig. 7. Inhibition by palmitoyl-CoA and malonyl-CoA on ‘H-labelled S-(4.bromo-2,3-dioxobutyl)(BDB-) CoA binding to membrane residues. (A) Membrane residues (6-8 mg protein/ml) were exposed to 5 PM ‘H-labeled S-(4-bromo-2,3-dioxobutyl)-CoA in the absence (0) or presence of 250 PM palmitoyl-CoA (W) or 250 pM malonyl-CoA (A). The reaction was stopped at different time points by addition of 100 mM dithiothreitol After washing of the membrane residues, samples were taken for determination of bound radioactivity. (B) Radioactivity bound after 15 s exposure to 5 PM ‘H-labelled S-(4-bromo-2.3.dioxobutyl)-CoA in the presence or absence of increasing concentrations of malonyl-CoA (0) or palmitoylCoA (0). The results represent mean + SE. from four different membrane residue preparations.

Discussion

The detergent extraction of a malonyl-CoAsensitive camitine pahnitoyltransferase from rat liver reported for the first time in the present study indicates the existence of two different proteins with camitine pahnitoyltransferase activity in the mitochondrial inner membrane. From whole mitochondria, most of the

75

sidues is in fact identical to the malonyl-CoAbinding site. Although the protective effect of palmitoyl-CoA may be of unspecific nature, since this compound shows a high unspecific binding to mitochondrial protein [16], our present data are consistent with palmitoyl-CoA and malonyl-CoA interacting at the same site. Our observations do not distinguish between a substrate site for palmitoyl-CoA which also binds malonyl-CoA, or a separate substrate site (for palmitoyl-CoA) and a regulatory site (or subunit) with affinity for both malonyl-CoA and palmitoyl-CoA. Acknowledgements The expert technical Haviken and Jon Holten

assistance of June Taje is gratefully appreciated.

References 1 Bergstrom, J.D. and Reitz, D.C. Biophys. (1980) 204, 71-79 2 Miyazawa, S., Ozaza, H., Osumi, (1983) J. B&hem. 94, 529-542

(1980) T. and

Arch.

Biochem.

Hashimoto,

T.

3 McGarry, J.D., Leatherman, G.F. and Foster, D.W. (1978) J. Biol. Chem. 253,4128-4136 4 Bremer, J., Woldegiorgis ,G., Schalinske, K. and Schrago, E. (1985) Biochim. Biophys. Acta 833, 9-l 5 Lund, H. and Woldegiorgis, G. (1986) Biochim. Biophys. Acta 878, 243-249 6 Lund, H. and Bremer, J. (1986) Abstract No. 11, 27th International Congress on the Biochemistry of Lipids, Oslo, Sept. 1986 7 Stokke, 0. and Bremer, J. (1970) Biochim. Biophys. Acta 218, 552-554 8 Barden, R.E., Owens, M.S. and Clements, P.R. (1981) Methods Enzymol. 72, 580-583 9 Bremer, J. (1968) Biochem. Prep. 12, 69-73 10 Kawaguchi, A., Yoshimura, T. and Okuda, S. (1980) J. Biochem. 89, 337-339 11 Johnson, P. and Lardy, H. (1967) Methods Enzymol. 10. 94-96 12 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 13 Hoppel, C.L. and Tomec, R.J. (1972) J. Biol. Chem. 247, 832-841 14 Saggerson, E.D. (1982) Biochem. J. 202, 397-405 15 Zammit, V.A. and Corstorphine, C.G. (1985) Biochem. J. 230, 389-394 16 Wood, J.M. (1973) Biochemistry 12, 5268-5273