Myocardial carnitine palmitoyltransferase of the mitochondrial outer membrane is not altered by fasting

Biochbnica et Biophysica Acta, 1]28 (1992) 11)5-i !1 ~:~ 1992 Elsevier Science Publishers B.V. All rights reserved 1)111)5-2701)/92/$1)5.1)0

105

BBALIP 54006

Myocardial carnitine palmitoyltransferase of the mitochondrial outer membrane is not altered by fasting Randall L. Mynatt, Michael D. Lapp( ~ and George A. Cook Department of Pharmacoloh~.', CoUegeof Medicine, The Unirersity 01"Tennessee, Men~phis, Tennessee (USA) (Received 19 February 1992)

Key words: Carnitine palmitoyltranst'erase; Fany acid oxidation; Mitocholldrion; Enzyme regulation; Fasting: (Heart)

The regulation of heart carnitine palmitoyltransfcras¢ was studied during the transition to the fasting state. Using dccanoyI-CoA or palmitoyI-CoA as substrates, we found IlO differences in earn(line palmitoyltransferase activity ,n in its sensitivity to i,lhibition by malonyl-CoA between led and fasted states. No cooperativity was seen with either substrate, and the malonyi-CoA-induccd shift to sigmoid kinetics normally observed with liver mitochondria was not obvious with heart mitochondria. Analysis of malonyI-CoA inhibition data revealed that mitochondria from rat heart exhibited incomplete maximum inhibition of earn(fine palmitoyltransferase (partial inhibition). Homogenization of intact liver mitochondria resulted in a similar pattern of incomplete inhibition and suggested that the malonyI-CoA-insensitive carnitine pahnitoyltransferase of the inner membrane was also being assayed. Carnitine palmitoyltransferase in mitochondrial outer membranes, isolated from the heart, proved to be extremely sensitive l.o malonyl-CoA inhibition and had maximum inhibition values of 90-101)% with either decanoyi-CoA or palmitoyI-CoA as substrates, but fasting had no effect. Fasling produced no change in the K i for malonyi-CoA (0.10 + 0.04 and I).14 + 0.02 #M lbr the fed and fasted groups, respectively). AcyI-CoA chain length specificity was CI0 > C16 > C14 > CI2 > CI8 = C8 fl~r carnitine pahnitoyltransferase in heart mitochondrial outer membranes, it is concluded that the regulation of carnitinc palmitoyltransferase of heart mitochondrial outer membranes differs from regulation of the liver enzyme in three characteristics - the heart enzyme (a) has greater sensitivity to malonyI-CoA inhibition, (b) is resistant to the effects of fi~s!ing and (c) has somewhat different acyI-CoA substrate specificity.

Introduction Cacnitine palmitoyltransferase (palmitoyl-CoA: L-carnitine O-palmitoyltransferase, EC 2.3.1.21) of the heart is not well understood in terms of either its kinetic characteristics or its physiological regulation. Similar to liver, heart contains a carnitine palmitoyltransferase activity localized in the mitochondriai outer membrane and another associated with the mitochondrial inner membrane [1]. The outer membrane carnitine palmitoyltransferase of the heart is apparently slightly smaller than its counterpart in the liver with apparent molecular sizes of approx. 86 kDa and 90 kDa, respectively [1,2]. The inner membrane enzyme has an apparent molecular size of approx. 68 kDa [1-3] and appears to be the same protein in both the heart and liver [2]. In the hepatic system, the outer mem-

Correspondence to: G.A. Cook, Department of Pharmacology, The University of Tennessee, 874 Union Avenue, Memphis, TN 38163,

USA. Present address: Department of Physiology,Ohio State University, Columbus, OH, USA.

brane enzyme regulates fatty acid oxidation through malonyI-CoA inhibition [4] and through changes in activity and sensitivity of the enzyme to inhibition by malonyI-CoA [5-7]. Starvation, diabetes and hyperthyroidism result in an increase in hepatic mitochondrial outer membrane carnitine palmitoyltransferase activity [5-7]. During these states of increased fatty acid oxidation, the sensitivity of the hepatic enzyme to malonyI-CoA inhibition decreases due to an increase in the K~ for malonyI-CoA of about 10 fold [5,6]. Cardiac carnitine palmitoyltransferz, se is much more sensitive to malonyl-CoA inhibition [5,8,9] with a K, for malonyl-CoA of about 0.2 ~ M [5]. However, there are contradictory reports regarding the effects of fasting on carnitine palmitoyltransferase activity and malonyl-CoA inhibition in the heart. Two previous studies found no effect of fasting on carnitine palmitoyitransfcrase activity or its sensitivity to inhibition by malonylCoA [5,9]. A later study reported both an increase in activity and a decrease in the sensitivity of carnitine palmitoyltransferase to malonyI-CoA inhibition after fasting [10]. All three reports used isolated heart mitochondria as the source of earnitine palmitoyltransferase, but the two earlier reports used a radioisotope-

106 based assay and the later study used a spectrophotometric method involving dithionitrobenzoic acid to measure carnitine palmitoyltransferase activity. Since isolated mitochondria contain both inner and outer membrane carnitine palmitoyltransferases, the possibility exists for the contamination of the assay with inner carnitine palmitoyltransferase by disruption of membrane integrity during the isolation process. The exposure of malonyl.CoA-insensitive inner membrane carnitine palmitoyltransferase would increase overall activity, but would decrease the extent to which malonylCoA could inhibit the total activity measured. In addition, there may be inherent differences in the two carnitine paimitoyltransferase assay procedures that could have led to different conclusions. The purpose of the current study was to determine whether the method of assaying carnitine paimitoyltransfcrase activity was responsible t'or the discrepancies reported and to eliminate ,'he possibility of interference from the inner carnittne palmitoyltransferase by studying the effects of fasting using mitochondrial outer membranes isolated from the heart. Materials and Methods Male Sprague-Dawley rats, 150-2511 g, (Harlan Industries, Indianapolis. IN., USA)were fed Purina Rodent chow (Ralston Purina Richmond, IN., U S A ) a d libitum or fasted for 48 h, The rats were decapitated: livers and hearts were promptly removed and placed in ice-cold isolation buffer. Hepatic mitochondria were isolated by the method of Johnson and Lardy [! i] with previously described modifications [6]. Outer membranes were prepared from isolated hepatic mitochondria by the method of Parsons et al. [12]. Heart mitochondria were isolated as previously described [51. Some heart mitochondriai outer membranes were prepared by the method used for liver [I 2], but, in order to improve the yield of outer membranes, another procedure was also used [13]. in the latter procedure, heart mitochondria were resuspeaded in 21) mM potassium phosphate buffer (pH 7.2) and allowed to incubate on ice for 21) rain. Trypsin was added (101) #g trypsin/mg mitochondrial protein) to the suspension of swollen mitochondria and the suspension was incubated on ice for an additional 15 rain, BSA (10%) was added to the suspension to yield a ~ ,~,"~z :,~ BSA solution and mito'1'1 chondria were centrifuged for 20 rain at 31111111}× g at 4°C, This treatment with BSA is necessary to inhibit and remove trypsin, The pellet was resuspended in 20 mM phosphate buffer containing 0.02% BSA and subjected to sucrose density gradient separation [ 12]. The band containing the outer membrane fraction was removed and diluted to 40 ml with deionized water and centrifuged at 30000 x g at 4°C. The pellet was resuspended in 20 mM phosphate buffer to yield approx. 2

mg protein/ml, Carnitine palmitoyltransferase-specific activity in these outer membranes was decreased by only 20% by the action of trypsin, but the yield of enzyme activity and of total protein in the outer membrane fraction was doubled. Isolation of outer membranes typically resulted in a 4- to 5-fold increase in the specific activity of carnitine palmitoyltransferase when assays were conducted under identical conditions at 2 mM carnitine. Malonyi-CoA inhibition was not affected by trypsin treatment (76% inhibition without trypsin treatment and 75% with trypsin treatment when assayed with 411 ~M palmitoyl-CoA and 20/zM maionyI-CoA). Ar identicai effect of trypsin to decrease activity of c~,mitine palmitoyltransferase without changing its se~,sitivity to inhibition by malonyl-CoA has been rep~rted previously with hepatic mitochondria [14]. in experiment~, involving heart mitochondria, earn(fine palmitoyltrausferase activity was estimated spectrophotometrically using the method of Fiol et al. [10]. Each assay contained 5(1 mM imidazole, 15(1mM HCI, ll.2 mM dithionitrobenzoic acid (DTNB), 0.86 mM EDTA, 10-100 ,ttM acyl-CoA, 2,0 mM u-carnitine and 200 ~g heart mitochondrial protein, in assays containing palmitoyl-CoA, 2 mg of BSA were also added. Assays were performed at 37°C at pH 7.0 in a total vol. of 2.0 ml. in experiments involving outer membranes, a modified procedure of Bremer [15] was employed as described earlier [5]. Each assay contained 82 mM sucrose, 70 mM KCi, 70 mM imidazole, i #g antimycin A, 0.5 mM L-carnitine (0.4 mCi of L-[methyl.'~H]carni tine), II1- I(10 ~ M acyI-CoA, 0.05-50/.t M malonyi-CoA and 2 mg bovine serum albumin in a total volume of 1.0 ml. Assays were initiated by adding carnitine to the remaining components and were carried out at 37°C for l0 rain at pH 7.0. In experiments in which acylCpA's of differing chain length were examined, assays were stopped by the addition of HCI and extracted by the saturated butanol.water procedure since acyI-CoA's of shorter chain length are not precipitated quantitatively by perchloric acid. To test the effect of damaged mitochondrial membranes on the assay of carnitine palmitoyltransferase, intact liver mitochondria were mechanically disrupted by vigorously homogenizing isolated mitochondria for 5 rain with rapid up and down strokes in a Teflon/glass homogenizer with the pestle rotating at 80(} rpm. Protein was determined by the BCA protein kit (Pierce, Rockford, IL, USA). Ultra-pure (99.9%) sucrose was obtained from ICN Biomedicals (Cleveland, OH, USA). L-[methyl- 3H]Carnitine-HCI was purchased from Amersham Corporation (Arlington Heights, IL, USA). All other chemicals were purchased from Sigma Chemical Company (St. Louis, MO, USA).

107

-= 3o

i~ '

~ . , . ~' . ,

'

irj~

|

lo

o o

50

lOO

0

50

100

150

[Decanoyl-CoA], pM

[PalmltoyFCoA], pM

Fig. I. Malonyl-CoA inhibition of carnitine palmitoyltransferase in heart milochondria isolated from fed and fasted rats. Carnitine palmitoyltransferase was assayed under spectrophotometric conditions (Materials and Methods) using pahnitoyI-CoA (left panel) or decanoyI-CoA (right panel) as substrate with (filled symbols) or without (open symbols) the addition of 50 # M malonyl-CoA. Squares in both panels represent assays conducted using mitochondria from led rats. Circles represent assays using mitochondria from fasted rats. Points are means :t S.E. (n = 4).

Results A spectrophotometric determination, similar to the one used by Fiol et ai. [10], did not reveal any changes in carnitine palmitoyltransferase activity or inhibition by malonyl-CoA as a result of fasting (Fig. 1). Whether using palmitoyi-CoA or decanoyl-CoA as the substrate, carnitine palmitoyltransferase activity was identical in mitochondria from fed and fasted rats and inhibition by malonyl-CoA was identical in the two groups. These experiments also gave no indication of cooperativity with respect to substrate. Double-reciprocal replots of these data (Lineweaver-Burk plots) were linear except at the highest concentrations of substrates where we observed substrate inhibition. We have previously shown with liver mitochondria (of. Fig. l of Ref. 5) that there was a shift to a sigmoid-shaped curve in the presence of malonyi-CoA that would indicate positively cooperative competitive inhibition. A similar, malonylCoA-induced shift to sigmoidal kinetics with heart mitochondria was not obvious (Fig. 2), but such effects might have been masked by substrate inhibiticn or the much greater sensitivity to malonyI-CoA inhibition. All these data are consistent with our previously published report in which the radiochemical assay was used [5]. Since these more recent experiments did not reveal a reason for the previous discrepant results, we sought to find the reason by additional analysis of the data (both ours and those of other research groups). We noted from the data of Fiol et al. (Fig. 4 of Ref. 10) that the heart enzyme preparation they used seemed resistant to complete inhibition by malonyl-CoA at the highest concentrations of the inhibitor. In fact, it appears from that figure that only a relatively small fraction of the enzyme could be inhibited. In order to examine this more carefully, we replotted the data from this figure in a double reciprocal manner. A plot of 1/%inhibition vs. l/[malonyl-CoA] will yield 1/150

(where 1~1) equals the inhibitor concentration which gives 1/2 of maximum inhibition) at the intersection of the x-axis and it will yield 1/im,,x (where l,,,~,x is the maximum % inhibition possible at infinite malonyl-CoA concentration) at the intersection of the y-axis. This transformation of data is based on the fact that taking reciprocals of x and y axes of a hyperbolic curve will result in a straight line, as with the transformation of a hyperbolic Michaelis-Menten plot into a linear Lineweaver-Burk plot. A mathematical derivation of the condition where two enzymes catalyzing the same reaction are present but only one is inhibited has been given by Segel [16]. Earlier work by Webb has shown an identical linear plot for partial inhibition [17]. The plot of 1/% inhibition vs. l/[inhibitor] has been used previously for carnitine palmitoyltransferase assays where incomplete inhibition was suspected [18-20]. Such a replot of the data (Fig. 2) revealed 150 values of

--i..-.p

|

0.12

c 0

0.08

D

"

0.04

/ I 0.00 - ~ ~ J -0.6

-0.2

' ~

0.2

-

0~6 1,'Malonyl-CoA (pM")

~

1.0

Fig. 2. Double-reciprocal replot of malonyI-CoA inhibiton curves from the data of Fiol, Kerner, and Bieber (Biochim. Biophys. Acta 916 (1987) 482-492). Data were replotted in a double-reciprocal manner as explained in the text. Squares represent data from fed rats and circles represent data from fasted rats.

108 4.3 and 2.8 #M malonyl-CoA for fed and fasted rats, respectively, indicating essentially no difference in sensitivity to malonyl-CoA inhibition in the two states. If these numbers were significantly different, they would suggest greater sensitivity in the fasting state, not less. Ira,,, values obtained from this plot were 44% for fed rats and 29% for fasted rats, indicating that not all of the enzyme being assayed as carnitine palmitoyltransferase was susceptible to inhibition by malonyl-CoA. Similar replots of data from several experiments in our laboratory also indicate a lack of complete inhibition with lm,,.~ values varying from 60-85%. Analysis of our data from liver mitoehondria showed large differences in Is0 values between fed and fasted rats (6-1(I fold), but Ira,,.~ values varied in a much narrower range be. tween 90--110%. The lower 1,1,,,` values with heart mitochondria suggested that they were more severely damaged than liver mitochondria. These data confirm a previous obse~'ation that there is inherenuy more damage to heart mitochondria during the isolation procedure [21]. Damage to mitochondrial membranes during isolation would be expected to expose the inner carnitine palmitoyltransferase activity that is not inhibited by malonyl-CoA and could result in incomplete maximum inhibition of the total activity assayed. In order to examine this possibility more completely, we isolated intact liver mitochondria and compared i~. and lm,,x values from those mitochondria with values obtained after subjecting the same mitochondria to mechanical damage. Fig. 3 indicates that the artificially damaged liver mitochondria are less sensitive to inhibition by malonyl-CoA. Double reciprocal plots revealed that the damaged mitochondria had an lm,,x value of 38% whereas the intact mitochondria had an Ira,,, value of 96%. These data suggest exposure of the inner carnitine paimitoyltransferase, in cases such as these, the

data can be corrected by subtracting out the non-inhibited enzyme activity as suggested by Saggerson and Carpenter [18]. If this is done, the /5o values obtained are identical. An additional problem with the estimation of maionyl-CoA inhibition in heart mitochondria is the common use of nonspecific proteinases in the isolation procedure. It has been shown that some proteinases such as subtilisin BPN and papain cause a more rapid loss of malonyi-CoA sensitivity than loss of activity [1,14] while trypsin decreases activity without loss of sensitivity [14]. Fiol et al. [10] apparently used a method of mitochondria isolation that included limited proteinase digestion. The extent to which proteolysis may have changed malonyI-CoA sensitivity in this case is not known, but it would be expected that such changes would vary according to the concentration of proteinase, quantity of tissue and incubation time and could vary with each preparation of mitochondria. The best method for overcoming the problems inherent with isolating intact heart mitochondria would be to perform experiments on isolated mitochondrial outer membranes. We found no effect of fasting in these preparations in either activity of the enzyme or in its sensitivity to inhibition by malonyi-CoA. Specific activities were 24 :l: 3 nmol/min per mg protein in isolated outer membranes from heart mitochondria of ted rats and 20 + 3 nmol/min per mg protein in preparations from fasting rats. Inhibition of carnitine palmitoyltransferase activity by malonyI-CoA in mitochondriai outer membranes isolated from rat heart is shown in Fig. 4. These data indicate that there were no differences between fed and fasted rats with respect to malonyi-CoA inhibition at palmitoyI-CoA concentrations of 15, 30, and 50 p.M (the 30 /~M curve lies between the 15 #M and 50 #M curves, but is not shown in order to avoid congestion on the graph.)

C 100

11

l

-]

o09t

80

j

/ .°.°o..,..o ..--

0 "v

I

I

i

a

,

o

20

.o

so

eo

loo

[ Malonyl.C~ ], pM

1.1-

,

.oo~

~"

_l

!

I

ooo

0.02

o.o4

..

14 M ~ d , 4 ~ A ], idil "~

Fig. 3. Effect of mechanical damage to intact hepatic mitochondria on malonyi-CoA inhibition of carnitine pa]mitoyltransferase. Left panel. Carnitine palmitoyltransferase was assayed using the radioisotopic procedure (Materials and Methods)with 40/~M palmitoyl-CoA and 0.5 mM L-carnitine. Right pane]: Double-reciprocal replot of data in the left panel. Intact mitochondria are represented by open symbols and mechanically damaged mitochondria are represented by closed symbols.

109 ~oo

;,--.-,....T.,.-,.-,,,~-.~.T.,-,,-.,.,,-.,,,-.---

~

TABLE 1 T

. ,.

hzhibilion constants for the inhibition by malonyl-C'oA of carnitine pabnitoyltran.~,rase hi isolated heart mitochondrial outer membranes

8O

Carnitine palmito~ltransferase activity was determined radioisotopitally at 0.5 mM L-carnitine and 0-50 p.M malonyI-CoA. For details on plots see text. Values are means+ S.E. (n = 3).

g 6o

lnllibitor Constant

,o

Fed

K~ (gM)

,°I 0

0

15o ( g M): at 15 p. M paimitoyI-CoA .

i

10

.

J

~

i

,

i

i

20 30 40 [Malonyl-CoA], j.LM

50

at 30 p.M palmitoyI-CoA at 50/z M palmitoyI-CoA

60

Fig. 4. Malonyl-CoA inhibition of carnitine palmitoyltransferase in isolat,~d heart n)ilochondrial outer n)embranes from fed and starved rats. Carnitine palmiloyltransferase was assayed using the radioisotopic assay procedure (Materials and M~tht)ds) with palmih)yl-('oA as substrate at 15, 3(), and 50 pM and nocarnitim: at 0.5 raM. Circles represent assays conducted in the presence of 15 p M palnlitoyI-CoA alld squ~,|l'e5 represent assays conducted in the preseilce of 50 /.tM palmitoyl-CoA. Data from assays eonla,lilig 30 /.tM palmitt~yI-CoA were omitted fi)r clarity. Filled sym!~,is represent assays using outer membranes from led rats. Open symbols represent assays using outer membranes from fasted rats, Points are means±S.E. (n = 3).

Dixon plots (l/activity vs. [malonyl-CoA]) of the data give K i values for malonyl-CoA of 0.10 + 0.04 p,M in fed rats and 0.14+ 0.02 /~M in fasted rats, Table 1. Double-reciprocal plots ( 1 / % i n h i b i t i o n vs. l/[malonyl-CoA]) yielded /so values for malonyl-CoA from 0.06-0.08 p M in fed rats and from 0.05-0.07 p M in fasted rats (Table l, no statistically significant differences between fed and fasted rats). Also, there were no significant differences between the fed and fasted rats in terms of Im:~xvalues regardless of the concentration of palmitoyl-CoA. The effects of acyI-CoA chain length on mitochondrial outer membranes from heart and liver are shown

Ima~(~): at 15 ~ M palmitoyI.CoA at 30 p,M palmitoyl-CoA al 50 g M pahnitoyI-CoA

Fasted

0.10+ 0.04

0,14_+ 0.02

0.06 + 0.1)2 0.06±0.01 0.08 +_0.02

0.05 ± 0.02 0.06±0.01 0.07 ± 0.02

83 78 65

±7 ±8 ±9

80 78 64

±4 ±8 ±9

in Fig. 5. Decanoyl-CoA was the preferred substrate for heart mitochondrial outer membrane carnitine palmitoyitransferase, producing at least twice the activity of the other acyl-CoA's from 25-100 ttM. Comparative activity with the remaining acyl-CoA substrates was Cl6 > Cl4 > Cl2 > C18 = C8. Carnitine palmitoyltransferase activity peaked at 50 to 75/.tM with all acyl-CoA's. Decanoyi-CoA was also the best substrate for liver mitochondrial outer membranes. However, myristoyi-CoA produced the second highest activity in liver membranes. PalmitoyI-CoA and stearoyI-CoA produced essentially the same activity in the liver followed by lauroyl-CoA and octanoyl-CoA. Activities with these acyI-CoA substrates were very similar to that of purified carnitine palmitoyltr~qsferase from mitochondriai inner membranes from liver [22]. A curious and still unexplained observation for all three of these enzymes is that activity is much lower with the

120

so )- Hear;

Liver

80

|

40"

RS '

50

75

[AcyI-CoA],pM

100

0-

25

50 75 [A¢yI-CoA], pM

100

Fig. 5. Acyl-CoA chain length specificity for carnitine palmitoyltransferase of isolated heart and liver mitochondrial outer membranes. Carnitine palmitoyltransferase was assayed using the radioisotopic procedure (Materials and Methods). AcyI-CoA concentrations were 25, 50, 75 and I(X) p.M with 0.5 mM L-carnitine. Acyl-CoA chain lengths with each concentration of acyI-CoA were (from left to right) C8, CI0, C12, C14, CI6, and C18.

110 C12 than with either the C14 or the C10 acyl-CoA. In contrast to heart mitochondrial outer membrane carnitine pa!mitoyh~ansferase, the activity of the liver enzyme did n,~t plateau in the 50 ~M range but continued to increase through t!~e entire concentration range tested.

Discussion The experiments reported here demonstrate the risks of drawing conclusions regarding the regulation of the mitochondrial outer carnitine palmitoyltransferase from data obtained with preparations of whole mitochondria. The discrepant conclusions that there is an effect of fasting on the heart enzyme [10] or that there is no effect [5,9] appear to have arisen from differences in the amount of damage sustained by heart mito. chondria during the isolation procedure. The first danger lies in the exposure of the inner carnitine palmitoyltransferase which is latent in the absence of damage to the mitochondrial membranes but contributes quantitatively to the measurement of enzyme activity in damaged mitochondria. The second danger lies in the fact that the inner carnitine palmitoyltransferase is not inhibited by malonyl-CoA; so that not only is the activity in damaged mitochondria increased, but also the maximum possible inhibition of the enzyme is decreased which gives the appearance of decreased sensitivity. This combination of effects mimics the regulatory effects of fasting, but can be distinguished from the fasting effects by the characteristic incomplete (or partial) inhibition as indicated in Fig. 2. Our current results, obtained from a spectrophotometric carnitine paimitoyltransferase assay using whole mitochondria from heart (Fig. !) and from other experiments using isolated mitochondriai outer membranes (Fig. 4 and Table 1), indicate that there is clearly no significant difference between fed and fasted animal hearts in either carnitine palmitoyltransferase activity or in its sensitivity to inhibition by malonyl-CoA. Our analysis of previous data suggests that both the higher activity and altered sensitivity reported by others was due to damage of mitochondriai membranes during the isolation procedure caused by mechanical rupture, proteinase action, or both. The K, value for malonyl-CoA that we obtained in mitochondrial outer membranes isolated from heart was essentially the same as the K~ value we reported earlier (0.2 MM) for intact mitochondria assayed under similar conditic~ns [5]. This suggests that Ki values obtained in a c~mplex system such as whole isolated mitochondria are much less likely than/so values to be affected by artifacts induced by the isolation process and less susceptible to errors arising from the presep,ce of a second noninhibitable enzyme. This should not be surprising since the K~ value in this case is an equilib-

rium constant (the disassociation constant for the enzyme-inhibitor complex since malonyl-CoA is a competitive inhibitor) and would not be expected to be susceptible to problems related to enzyme artifacts such as incomplete inhibition. The use of the Dixon plot to determine K i values also ensures that only the K i value of the most sensitive em~me is measured [23]. The Ki value is certainly a better index to assess the affinity of the enzyme for malonyI-CoA. It must therefore be concluded: (a) that fasting did not change the affinity of the enzyme for malonyI-CoA and (b) that the protein responsible for binding malonyl-CoA is present in both intact mitochondria and isolated mitochondrial outer membranes of the heart and the liver, but must have different properties in the heart compared with the liver. Three important conclusions can be drawn from the current studies regarding the activity and regulation of carnitine palmitoyltransferases located in the mitochondrial outer membranes of the heart and liver. First, there is no effect of fasting on either the activity or malonyl-CoA sensitivity of the heart enzyme. Second, the heart enzyme has inherently greater sensitivity to inhibition by malonyl-CoA. Finally, the heart enzyme of mitochondrial outer membranes demonstrates somewhat different substrate specificity than the enzyme from the liver.

Acknowledgements The authors are indebted to Mr. J. Greenhaw and Ms. L.J. Weakley for expert technical assistance. This work was supported by a Grant-in-Aid (no. 88-1003) Irol~a the American Heart Association with contributions derived in part from the AHA, Tennessee Affiliate. Additional support was received through grant no. HL-40929 from the National Institutes of Health (United States Public Health Service, USA). R.L. Mynatt was supported by National Institutes of Health Postdoctoral Training Grant T32-HL07641.

References I Murthy, M,S. and Pande, S.V. (1990) iJicchem. J. 268, 599-604. 2 WoelUe, K,F., Esser, V., Weis, B.C., Cox, W.F., Schroeder, J.G., Liao, T.-S., Foster, D.W. and McGarv/, J.D. (1990)J. Biol. Chem. 265, 10714-10719. 3 Ghadimincjad, !. and Saggerson, E.D. (1991) Biochem. J. 277, 611-617. 4 McGarr?, J,D., Mannaerts, G.P. and Foster, D.W. (1977) J. Clin. Invest. 60, 265-270. 5 Cook, G.A. (1984) I. Biol. Chem. 259, 12030-12033. 6 Cook, G.A. and Gamble, S. (1987)J. Biol. Chem. 262., 2050-2055. 7 Stakkestad, J.A. and Bremer, J. (l~'g3) Biochim. Biophys. Acta 750, 244-252. 8 Saggerson, E.D. and Carpenter, C.A. (1981) FEBS Lett. 129, 229-232. 9 Paulson, D.J., Ward, K.M. and Shug, A.L. (1984) FEBS Lett. 176, 381-384.

111 10 Fiol, C.J., Kerner, J. and Bieber, L.J. (19871 Biochim. Biophys. Acta 916, 482-492. I I Johnson, D. and Lardy, H. (19671 Methods Enzymol. 10, 94-96. 12 Parsons, D.F., Williams, G.R and Chance, B. (19661 Ann. N.Y. Acad. Sci. USA 137, 643-666. 13 Schoite, H.R. (19731 Biochim. Biophys. Acta 330, 283-293. 14 Kashfi, K. and Cook, G.A. (1991) 12~iochL:m.Biophys. Res. Commun. 178, 600-605. 15 Bremer, J. (1981) Biochim. Biophys Acta 665, 628-631. 16 Segel, I.H. (1975) Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems, pp. 196198, John Wiley & Sons, New York.

17 Webb, J,L. (19631 Enzyme and Metabolic lnhibitors. Vol. 1. pp. 157-160, Academic Press, New York. 18 Saggerson, E.D. and Carpenter, C.A. (19811 FEBS Lett. 129, 225-228, 19 McGarry, J.D., Mills, S.E., Long, S.S. and Foster, D.W. (19831 Biochem. J. 214~ 21-28. 20 Stephens, T.W., Higgins, A.J., Cook, G.A. and Harris, R.A. (19851 Biochem. J. 227, 651-660. 21 Chao, D.L. and Davis, E.J. (19721 Biochemistry 11, 1943-1952. 22 Miyazawa, S., Ozasa, H., Osumi, T. and Hashimoto, T. (19831 J. Biochem. 94, 529-542. 23 Gamble, S. and Cook, G.A. (1985) J. Biol. Chem. 260, 9516-9519.