321

Biochimica et Biophysics Acta, 574 (1979) @ ElsevierlNorth-Holland Biomedical Press

321-333

BBA 57426

PURIFICATION AND CHARACTERIZATION OF A LONG-CHAIN ACYL-CoA HYDROLASE FROM RAT LIVER MICROSOMES

ROLF

KRISTIAN

E;ERGE

Laboratory of Clinical Biochemistry, Bergen (Norway) (Received

February

University of Bergen, Haukeland Sykehus, N-5016

8th, 1979)

Key words: Long-chain acyl-CoA hydrolase; (Characterization,

Rat liver microsome)

Summary A longchain acyl-CoA hydrolase from rat liver microsomes has been purified by solvent extraction and gel chromatography to homogeneity as judged by polyacrylamide gel electrophoresis in the presence and absence of sodium dodecyl sulfate. The enzyme was a monomer of molecular weight 59 000. In a sucrose gradient it sedimented at 4.3 S. The isoelectric point, pl was 6.9, and the Stokes radius was approx. 31 A. The enzyme hydrolyzed long-chain fatty acyl-CoA (C,-C,,) with maximum activity for palmitoyl-CoA. Bovine serum albumin activation of the enzyme was related to the ratio acylCoA/bovine serum albumin, and at high ratios, acyl-CoA inhibited the enzyme activity. Disregarding the substrate inhibition, an apparent K, of 65 nmol/mg protein or 1 * 10m7 M and a V of 750 nmol/mg protein per min were calculated. The enzyme was inhibited by p-hydroxymercuribenzoate and N-ethylmaleimide. Reactivation by means of dithiothreitol was not complete.

Introduction The hydrolyses of acyl-CoA esters by thioesterase I and II, and palmitoylCoA hydrolase (EC 3.1.2.2) take place in both prokaryotic and eukaryotic cells [l-9]. These enzymes seem to be important in lipid metabolism [ 3,101, in controling the chain length of synthesized fatty acids [ 111 and modifying the product specificity of fatty acid synthetase [l]. Abbreviations: HEPES, N-2-hydroxyethylpiperazine-N’-2ethanesulpho~c acid; PIPES, piperazine-N,N’bis(2+thanesulphonic acid); TAPS, tris(hydroxymethyl)methylaminopropanesulphonic acid; EPPS, 4-(2hydroxyethyl)-l-piperazinepropanesulphonic acid: SDS, sodium dodecyl sulphate.

322

When studying subcellular fractions of rat liver, we discovered that longchain acyl-CoA hydrolase (trivial name palmitoyl-CoA hydrolase, EC 3.1.2.2) was located both in the mitochondrial matrix and in the microsomal fraction [ 12,131. Several lines of evidence strongly suggested that hydrolysis of palmitoyl-CoA by crude microsomes or mitochondrial matrix was mediated by different enzymes as the two activities were differently inactivated by heating, differently affected by addition of cations and anions, had different K, values for palmitoyl-CoA and were differently inactivated by sulphydryl reagents [ 121. However, solubilization and purification of the enzyme(s) to homogeneity would provide conclusive evidence that they were different enzymes. Recently, we purified the mitochondrial matrix enzyme with a molecular weight of 19 000 and a Stokes radius of approx. 19 A [ 141. The purification and some properties of the long-chain acyl-CoA hydrolase from rat liver microsomes are reported in this paper. Materials and Methods Assay of hydroluse actiuity. Hydrolyses of long-chain acyl-CoA derivatives were done by two methods. The most convenient spectrophotometric method uses 5,5’dithiobis-(2-nitrobenzoic acid) (DTNB) to detect the release of free thiol groups. The molar extinction coefficient, c412 = 13 600 M-l * cm-’ for the nitromercaptobenzoate anion [ 151. Unless otherwise stated, the assay mixture contained 30 mM HEPES buffer (pH 7.5), 1 mM EDTA, 0.3 mM 5.5’-dithiobis(2-nitrobenzoic acid), 45 /IM acyl-CoA and enzyme protein (see legends to figures or tables). The hydrolase activity was also determined by a radiochemical method where the release of (1-i4C)-labelled fatty acids from (1-14C)labelled acyl-CoA was measured. Unless otherwise stated the assay system contained 30 mM HEPES buffer (pH 7.5). 45 PM (1-14C)-labelled acyl-CoA (8-10 nCi) and enzyme in a final volume of 0.25 ml. Reaction was started by addition of substrate. Incubation temperature was 35”C, and incubation time 2 min. The reaction was stopped by adding 2.0 ml Dole’s reagent [16], and unesterified 14C-labelled fatty acid were extracted by the method of Bar-Tana et al. [17] modified as described [ 131. Sometimes, non equivalent results were obtained by the two methods because exposure of the enzyme to glass surfaces appears to cause some denaturation. Palmitoyl-L-carnitine hydrolase (EC 3.1.1.28) activity was measured under identical conditions as described above for acyl-CoA hydrolase using 45 PM (l-‘4C)palmitoyl-L-carnitine (0.03 PCi) as substrate. One unit of activity is the amount of enzyme catalyzing the hydrolysis of 1 nmol palmitoyl-CoA per min at 35°C. The extraction of palmitate was checked by thin-layer chromatography [ 181. Determination of molecular weight. The molecular weight of the purified hydrolase was determined by two methods. The subunit molecular weight was determined by SDS-polyacrylamide gel electrophoresis using transferrin (M, = 76 000), bovine serum albumin (M, = 65 000), ovalbumin (M, = 45 000), pepsin (M, = 35 000) and chymotrypsinogen (M, = 25 000) as standards. The molecular weight of the native hydrolase was determined by gel filtration

323

chromatography on an Ultrogel AcA column (1.6 X 90 cm) and a Sephadex G-100-120 column (1.6 X 90 cm). The enzyme was eluted by buffer C * with 0.15 M NaCl. Blue Dextran was used to determine the void volume and the standard proteins were bovine serum albumin, ovalbumin, chymotrypsinogen and cytochrome c (M, = 12 400). Sucrose gradient centrifugation. Linear sucrose gradients (lo-25%, w/v) were prepared in buffer C with 0.15 M NaCl. Bovine serum albumin (4.5 S), ovalbumin (3.9 S), pepsin (3.3 S) and myoglobin (2.0 S) were used as markers. Centrifugation was performed for 48 h at 32 000 rev./min (4°C) in 12-ml tubes in a swing-out rotor SW 41 in a Beckman ultracentrifuge L-50. 0.44 ml fractions were assayed for palmitoyl-CoA hydrolase activity. Isoelectro focusing. This was carried out using a 110 ml column (LKB Instrument). The amount of carrier ampholyte used (pH 3-10) was 3% (w/v) in O-48% (w/v) sucrose gradient. Electrofocusing was initiated at 200 V; as the current decreased, voltage was increased to and maintained at 700 V. Electrofocusing was done for 48 h. Other methods. SDS-gel electrophoresis in 7.5, 9 and 10% polyacrylamide gels was done essentially as described by Weber et al. [19] with some modifications [ 201. Polyacrylamide gel electrophoresis of the hydrolase was performed at different pH and gel concentrations (5, 7.5, and 9%) and 3% crosslinking in tubes (0.5 X 0.9 cm). 5-50 yg protein were applied to the gel in 50% glycerol. The electrophoresis proceeded at 4°C for 2-5 h (depending on gel concentration) at 0.5-l mA/gel. The following buffer systems were used: A, 188 mM glycine/acetic acid (pH 3.6); B, 15 mM Tris/glycine (pH 8.9). The gels were stained in Coomassie blue and destained in 7% acetic acid. Protein concentration was determined by the method of Lowry et al. [21]. Chemicals. [ 1-14C]PaImitoyl-L-carnitine, 14C-labelled palmitoyl-, stearoyland oleoyl-CoA esters were obtained from New England Nuclear Corp., Boston, MA, U.S.A. Acyl-CoA esters, bovine serum albumin, 5,5’-dithiobis(2_nitrobenzoic acid), dithiothreitol, HEPES, PIPES, EPPS, TAPS, and Sephadex G-100-120 were from Sigma Chemical Co. (St. Louis, MO). SDS, ammonium persulfate, N,N,N’,N’-tetramethylethylenediamine, and hydroxyapatite were purchased from Bio-Rad Laboratories, Richmond, CA. DEAE-cellulose and CM-cellulose were from Whatman Biochemicals, Maidstone, Kent, U.K. Ampholine (pH 3-10) was from LKB, Bromma, Sweden. Proteins used as molecular weight standards were obtained from Boehringer, Mannheim, F.R.G. All other reagents were of analytical purity. Results Purification

of microsomal

long-chain

acyl-CoA

hydrolase

Unless otherwise noted the following steps were carried out at 0-4”C, and all centrifugation procedures were performed at 15 000 rev./min for 30 min using a SS-34 rotor in a Sorvall RC-5 centrifuge. Both spectrophotometric and * See under Results.

324

radiochemical assay methods were used throughout the purification procedure. 12-h-fasted rats of the Miill-Wistar strain were used. Rat liver microsomes were prepared according to de Duve et al. [ 22 3. The liver microsomes were stored at -8O*C for several months without notable loss of hydrolase activity. Liver microsomes from 4 -6 fasted rats of 200-300 g body weight were prepared and diluted by cold 15 mM HEPES buffer (pH 7.4), 1.5 mM MgC12,

2 ? 022 012

10

20

30 Fraction

Fraction

number

number

i c, E

325

Fraction

Fraction

number

number

Fig. lose (50 tein the

1. Elution profile of longshain acyl-CoA hydrolase. A, DEAE-cellulose chromatogram. B, CM-ceiluchromatogram. C. hydroxyapatite chromatogram; and D. Ultrogel AcA-54 chromatogram. Fractions ~1) were assayed for palmitoyl-CoA hydrolysis (Od) using the radiochemical assay method. Prowas determined by measuring per cent transmission or absorbance at 280 and 260 nm. The slope of NaCl and phosphate gradient (X---X) was determined by a 11 300 Ultra Gradient Mixer (LKB. (Sweden).

0.1 mM dithiothreitol and 0.1 mM EDTA (buffer A) to give a protein content of approx. 20 mg/ml. The diluted microsomal fraction was mixed with an equal amount of cold, water-saturated n-butanol under constant stirring for about 2 min. The solution was then centrifuged, and the upper butanol layer was

326

dicarded as it contained no hydrolase activity. The water phase and the interphase were collected and dialyzed for 12 h against buffer A. The dialysis tubing used has a molecular weight cut off of approx. 10 000. The diffusate was applied to a DEAE-cellulose column (1.5 X 30 cm) (Fig. 1A) equilibrated with buffer A and eluted with 15 ml of the same buffer followed by a NaCl gradient (0.05-0.3 M. The hydrolase activity eluted (flow rate, 0.5 ml/min) as two peaks centred near 0.05 and 0.075 M NaCl, respectively. Addition of (NH4)#04 eluted only protein without any hydrolase activity. Fractions confining approx. 70% of the eluted hydrolase (fractions 12 --22) were collected and concentrated by polyethyleneglycol 6000 followed by dialysis for 3 h against buffer A. After dialysis the preparations were applied to a CM-cellulose column (2.6 X 48 cm) (Fig. 1B) equilibrated with buffer A and eluted with 100 ml buffer A followed by 0.5 M NaCl in the same buffer (5-ml fractions). The hydrolase was resolved as a distinct protein peak. Peak enzyme-containing fractions (recovery approx. 80% not regarding specific activity) (fractions 27.-33) were collected and dialyzed for 3 h against a buffer containing 10 mM potassium phosphate (pH 7.4), 1.5 mM MgCL, 0.1 mM dithiothreitol and 0.1 mM EDTA (buffer B). The diffusate was applied to a hydroxyapatite column (1.5 X 30 cm) (Fig. 1C) equilibrated with buffer B. The column was eluted with 20 ml of the same buffer, followed by 200 ml phosphate gradient (0.03-0.4 M) in the same buffer. The hydrolase activity eluted as a peak with a shoulder centred near 0.12 and 0.16 M phosphate, respectively. Fractions containing about 75% of the eluted hydrolase activity (fractions 18-23), were collected and concentrated by polyethyieneglycol 6000. The concentrate was dialyzed for 5 h against 4 1 of a buffer containing 15 mM HEPES buffer (pH 7.4) and 0.1 M dithiothreitol (buffer C). The diffusate was applied to a column of Ultrogel-AcA-54 (1.6 X 90 cm) equilibrated with buffer C. The flow rate was kept at 0.7 ml/min and the column eluted with the same buffer (0.6-ml fractions). The palmitoyl~CoA hydrolase activity was eluted mainly as one main peak (peak I). Sometimes two additional peaks (peaks II and III) were obtained (see Fig. 1D). However, the TABLE

I

PURIFICATION

OF LONG-CHAIN

ACYCCoA

HYDROLASE

FROM

The hydrolase activity was determined by the radioche-nical approx. 120 nmolfmg protein of palmitoyl-CoA. Fraction

Volume (ml)

Microsomes (4 rats) Butanol fraction DEAE-cellulose peak fractions CM-cellulose peak fractions Hydroxyapatite peak fractions Ultrogel-AcA-54 (peak I)

9.0 13.0 45.0 22.0 7.5 3.3

method

RAT

LIVER

fmg)

Total activity (units)

Specific activity (units~mg protein)

172.0 31.7 21.6 4.8 1.2 0.83

6566 3506 2830 1880 1332 949

38 93 131 290 1110 1150

Total protein

MICROSOMES

at a substrate

concentration

m,)

Purification factor

100 53 43 29 20 14

1 2.5 3.6 10.0 29.0 30.0

Recov-

ery

of

321

specific activity of peaks II and III was only 25 units/mg protein, respectively. In the case illustrated, fractions 33.-41 were of essentially equivalent specific activity (1080-1130 units/mg protein) and were combined (peak I). Peak I was concentrated by polyethyleneglycol 6000 followed by dialysis against buffer C and stored at -20°C for maximally 2 months. Table I outlines the results of a typical purification experiment. Hydrolysis of palmitoyl-L-carnitine of rat liver microsomes has been reported [ 231. HOWever, hydrolytic activity was lost by the butanol fractionation. The butanol fractionation procedure also solubilized the hydrolase from the microsomal membrane, and removed most of lipids and protein without excessive loss of palmitoyl-CoA hydrolase activity. Criteria of purity and some physiochemical properties That the long-chain acyl-CoA hydrolase was of high purity was indicated by several observations including a single symmetrical peak of constant specific activity after rechromatography on a Ultrogel-AcA-44 column (not shown). In addition, polyacrylamide gel electrophoresis in the presence of SDS gave a single major band with minor components, constituting less than 8% of the stainable protein (Fig. 2A). At pH 8.9 (Fig. 2A, a and b) a few minor components were observed, the proportion of which appears to depend on the age of the purified enzyme. Polyacrylamide gel electrophoresis at pH 8.9 and 3.6 also gave a single major protein band (Fig. 2A, c and d). The gels were cut into slices and the single protein band was eluted from the gel in 10 mM HEPES buffer (pH 7.4) and the hydrolase activity assayed. SDS-polyacrylamide gel electrophoresis

IO-

86-

d

02

04

06

08

IO

R~

Fig. 2A. Polyacrylamide gel electrophoresis (10% gels) of purified microsomal hydrolase. SDS was included for gels a and b. which received 5 and 50 /Ig protein, respectively. Gels c and d were run in the absence of SDS at PH 8.9 and 3.6, respectively. Each of these gels received approx. 8 pg protein. B. Estimation of molecular weight of long-chain acyl-CoA hydrolase subunit by SDS gel electrophoresis (9%. w/v, gels). The RF values of standard subunit markers (see Materials and Methods) were compared to that of purified hydrolase. The open circles represent the hydrolase band.

Stokes radus

(A)

Fig. 3. Estimation of Stokes radius of long-chain acyl-CoA hydrolase protein by Sephadex G-100-120 chromatography. Samples (1.3 ml) were chromatographed. The elution positions of the standard molecular weight markers were determined by absorbance while the elution of the hydrolase protein was determined radiochemically. Treatment of data f34.351 indicates a Stokes radius of the hydrofase of 31 A. The standard protein markers used were: bovine serum albumin (35 A), ovalbumin (27.4 a), chymotrypsinogen (21 a) and cytochrome e (17 a).

of this active hydrolase in 10% polyacrylamide gels gave a molecular weight of approx . 59 000 (Fig, 2B). The molecular weight was 57 000 f 3000 by Sephadex G-l 00-l 20 chromato~aphy . The purified hydrolase sedimented as a single boundary on analytical ultracentrifugation, indicating a high degree of purity (unpublished results). The pl was 6.9 as determined in three separate experiments (range pH 6.87 6.93) (not shown). The enzyme sedimented as 4.3 S by sucrose density gradient centrifugation (not shown) and eluted from a Sephadex G-100-120 column corresponding to a Stokes radius of about 31 8, (Fig. 3). The purified enzyme can be stored in plastic vessels at -ZO”C, in 30 mM HEPES buffer (pH 7.5) and 1 mM dithiothreitol for at least 2 months without any loss of activity. However, on storing in glass vessels at -20°C for only 2 weeks a precipitate formed accompanied by loss of enzyme activity. Thus, exposure to glass surfaces appears to cause some denaturation. Kinetic properties The effect of the chain-length on the ability of the purified enzyme to hydrolyze acyl-CoA esters in the presence of bovine serum albumin is illustrated in Fig, 4. In each case, activity reached a maximum at an acyl-CoA con-

Acyl -CoA

concentratton (nmol /mg protein

)

Fig. 4. Acyf specificity of the hydrolase on my1 thioesters of CoA. A, results obtained using the radiochemical assay with 30 mM HEPES buffer (PH 7.5). 0.2 mg/mI bovine serum albumin and 1.4 gg protein. The incubation temperature was 3O’C. B, results obtained using the spectrophotometric assay with 10 mM HEPES buffer (pH 7.4), 1 mM EDTA, 0.2 mg/ml bovine serum albumin, 0.3 mM 5_5’-dithiobis(2nitrobenzoic acid) and 2.4 peg hydrolase protein. The incubation temperature was 37°C. The acyl-CoA concentration

added in both assays had a range of 3-40

MM.

centration which seemed to depend on the ratio of acyl-CoA/protein concentrations. Higher substrate concentrations resulted in decreased activity. Highest activity was found with palmitoyl-CoA as the substrate. Maximum activity levels of lauroyl-, myristoyl- and stearoylthioesters of CoA were quite similar, while that of oleoyl-CoA was distinctly lower, No activity was obtained with thioesters of fatty acids with less than 7 carbon atoms. Although the data obtained in Fig. 4 are based on different incubation temperature, the spectrophotometric method showed highest specific activity with palmitoyl-CoA as the substrate. The loss of activity using glass tubes by the radiochemical method could be prevented by addition of a nonionic detergent such as Triton X-100 (O.OOOZ~, v/v). These results are consistent with other reports [ 123. The ascending portions of some of the substrate-velocity curves were slightly sigmoidal (particularly for C14:0, C,,:, and CISEO) (Fig. 4A). Thus, these results cannot be plotted satisfactorily in a Lineweaver-Burk plot to obtain kinetics constants for the model substrates. A latent period of about 30-45 s was seen when palmitoyl-CoA was the last component in the spe~trophotometric assay. This effect was not observed when bovine serum albumin or Triton X-100 was added 1-2 mm before the reaction was started with palmitoyl-CoA. These results (data not shown) are consistent with the results obtained using purified mitochondrial matrix enzyme of rat liver [ 13,141. Increasing bovine serum albumin eoncentrat~ons increased the rate of

nm

Bovine serum

albumhn (mglml)

Fig. 5. Effect of bovine serum albumin on the activity of the hydrolase protein with palmitoyl-CoA as the substrate. The radiochemical assay was performed as described in Materials and Methods. Bovine serum albumin was added as shown. The substrate concentration was 23 ,uhf paimitoyl-Cob fo---0) or 46 MMof,palmitoyl-CoA (a -----a) and 2.8 ug hydrolase protein.

hydrolysis of palmitoyl-CoA to a maximum level, then a further increase in albumin concen~ation ~hibited the hydrolysis (Fig. 5). However, as the palmitoyl-CoA concentration was increased, higher concentrations of bovine serum albumin were required for m~imum activity. Thus, the m~imum activity is related to the ratio of amount of palmitoyl-CoA and protein with maximum level of substrate of approx. 115 nmol/mg protein (Fig. 5). At low concentrations of palmitoyl-CoA (up to 75 nmol/mg protein), the rate of hydrolysis was not proportional to the concentration of substrate (Fig. 4A) and small increases over 75 nmol/mg protein showed large increases in the rate of hydrolysis. Subsequent increases of palmitoyl-Coil, concentration resulted in linear increase in the rate of hydrolysis until maximal rate was reached at approx. 120 nmol palmitoyl-CoA/mg protein. Although the saturation pattern of this substrate appears to be similar to that observed with enzymes exhibiting allosteric activation, it is likely that in the present system albumin might bind to the substrate. Such conclusions can therefore not be drawn. Probably at very low substrate concentrations the tight-binding sites of bovine serum albumin compete effectively for the substrate, and when these sites are saturated the substrate becomes available to the hydrolase protein. Consequently, linear increases in rates are obtained with increasing substrate condensations. In support of this explanation was the finding that without bovine serum albumin, increasing concentrations of palmitoyl-CoA showed linear increases in the rate of hydrolysis of palmitoyl-CoA up to the critical micelle concentration of palmitoyl-CoA. A further increase in substrate concentration caused severe inhibition (not shown), This phenomenon indicates that the monomer form of palmitoyl-CoA is the active substrate. A double-reeiprocal plot was linear except for the inhibitory concentration of the substrate. Disregarding the nonlinear region an apparent I& of 65 nmolfmg protein or 1 *

331

low7 M and a V of 750 nmol/mg protein per min were calculated; this is in good agreement with results obtained in Fig. 4A. The kinetic data obtained for the enzyme is compli~a~d by the fact that palmitoyl-CoA is a detergent. With the complication caused by substrate inhibition these parameters do not have the usual meaning of the kinetic parameters but have practical significance for the enzyme assay, Thus, since inclusion of bovine albumin gives higher specific activities, optimal levels of albumin to p~mitoyl~CoA concentration were used for routine assays. Effect of pH, ionic strength, cations and temperature The effect of pH on the hydrolase activity was determined using PIPES, HEPES, TAPS and EPPS buffer to span the pH range 6-9.5. The hydrolase activity was constant over a pH range from 7.8 to 8.2, but dropped off above and below this range (not shown). When the buffer concentration was held constant and the ionic strength was increased by NaCl and KC1 in 50 mM HEPES (pH 7.5, I = 0.025), addition of KC1 had no effect on enzyme activity (27.8 ,uM palmitoyl-CoA), unless ionic strength of the medium exceeded 0.175, whereupon inhibition was observed. NaCl also strongly inhibited enzyme activity when the ionic strength of the medium was over 0.125. At all concentrations ~hibition by NaCl was greater than by KCl. MgZf and Mn” inhibited the hydrolase activity at all concentrations tested. However, the hydrolase activity was somewhat more sensitive to Mn” than Mg*‘; about 0.5 mM Mg2* gave 40% inhibition, while about 0.2 mM Mn*’ gave approx. 60% inhibition. Changing the temperature from 19 to 40°C hydrolysis of palmitoyl-CoA increased almost lineary and the apparent energy of activation was calculated to be 39.8 kJ/mol, using the temperature range 19-40°C (not shown). The Arrhenius plot, gave a Q 10 of about 1.8, representing a 7.5% change in the rate of palmitate production per “C. Palmitoyl-CoA hydrolase activity was sensitive to reagents which primarily attack sulphyd~l groups, The inhibitor p-hydroxymercuribenzoate (50% inhibition at 5 FM) was considerably more potent than ~-ethylm~eimide (50% inhibition at 10 mM) in this regard (not shown), Dithiothreitol slightly stimulated the hydrolase activity, and in an experiment with p-hydroxymercuribenzoate and N-ethylmaleimide as inhibitors reactivation by dithiothreitol was not complete. Discussion This study reports that purified long-chain acyl-CoA hydrolase from rat liver microsomes behaves as a single molecular species with a molecular weight of 57 006-59 000. A similar molecular weight (60 000) was calculated from the equation hf, = 67 77A&/( 1 - Vp) where q = viscosity of medium, p = density of medium, and N = Avogardo’s number, a = experimentally determined Stokes radius, S = Svedberg units of sedimentation coefficient, and D = partial specific colume (0.74 cm3/g) (unpublished results).

332

The molecular weight of mitochondrial matrix hydrolase (19 000) [ 12,141, being similar to the low molecular weight acyl-CoA hydrolase from Escherichia coli [5] and the lowest form of the long-chain acyl thioester hydrolase (17 500), is about one-third of the enzyme studied here. Also the sedimentation coefficient (2.1 S), Stokes radius (19 a) and isoelectric point (PI= 5.9) for the matrix enzyme [ 121 differed from the respective physical constants of the microsomal long-chain acyl-CoA hydrolase. As the microsomal hydrolase catalyzed the hydrolysis of saturated acyl-CoA from C, to Cl8 the two enzymes are different in structure and to some extent also in specificity [ 141, The concentration of detergents above which micelles are formed is known to be influenced by solvent conditions [24,25]. However, since Michaelis-Menten kinetics were observed at a palmitoyl-CoA concentration below critical micelle concentration [ 131, it is apparent that the palmitoylCoA monomer is an acceptable substrate for the hydrolase. Far above the critical micellar concentration of palmitoyl-CoA, bovine serum albumin could have prevented enzyme inhibition by substrate in the micellar form [ 14,24-261. Bovine serum albumin increased the hydrolase activity of the mitochondrial matrix hydrolase over as well as under the critical micelle concentration of palmitoyl-CoA [ 131. According to Arvidsson [ 27,281 palmitoyl-CoA can either bind to the hydrophobic regions of bovine serum albumin at low molar ratio (w) or arrange on bovine serum albumin in a micellar fashion when N is high. Substrates inhibition was observed for all acyl-CoA esters used. The slopes of both ascending and descending portions of the substrate velocity profiles (Fig. 4) suggest that hydrophobic binding is important in each case which is in agreement with other reports [ 1,2]. Inhibition of enzyme activity in high ionic strength media might be due to the tendency of palmitoyl-CoA to form inhibitory micelles at lower concentrations in high ionic strength media. Whether Mg’+ and Mn*’ exerted a direct effect on the enzyme or an indirect effect on the substrate is at present unknown. The effects of thiol-directed reagents such as p-hydroxymercuribenzoate and Nethylmaleimide and reactivation by dithiothreitol (about 40%) indicate that sulphydryl groups are important for the enzyme action. However, it is impossible to say at this time whether the affected residue(s) are in the vicinity of the active site(s). Papain, a cysteine active site protease [ 291, is inhibited by phenylmethanesulfonylfluoride and is reversed by dithiothreitol. The microsomal acyl-CoA was also inhibited by this ‘active serine’-directed reagent (unpublished). Most mammalian tissues [l-4] and microorganisms [5,6] were found to contain some acyl-CoA hydrolase activity with a very wide range of molecular weight estimations [ 11,20--321. However, the isolated thioesterase [ 2,30,31] were obtained by limited digestion with trypsin, thus, extensive changes in threedimensional structure of the thioesterase might have taken place. Thioesterase II is believed to be concerned with the metabolism of acylthioester, capable of shifting the chain length of the fatty acid synthetase products when malonyl-CoA is generated in rate-limiting quantities by acetylCoA carboxylase [ 11,331. The function of long-chain acyl-CoA hydrolase from rat liver mitochondria and microsomes is still unclear at this time.

333

Acknowledgements The technical assistance of Mrs. 3. Dossland and Mr. L.E. acknowledged. The author is indebted to Professor M. Farstad, and criticism during the preparation of the manuscript. The ported in part by the Norwegian Research Council of Science ities and by the Norwegian Council of Cardiovascular Diseases.

Hagen is greatly M.D., for advice study was supand the Human-

References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Libertini, L.J. and Smith, S, (1978) J. Bioi. Chem. 253. 1393-1401 Lin, C.Y. and Smith, S. (1978) J. Biol. Chem. 253,1954-1962 Kurooka, 8, Hosoki. K. and Yoshimura, Y. (1972) J. Biochem. 71,625-634 Anderson, A.D. and Erwin, V.G. (1971) J. Neurochem. 18,1179-1186 Barnes. EM. and Wakil. S.J. (1968) J. Biol. Chem. 243, 2955-2962 Barnes, E.M., SwindelI, A.C. and WakiI, S.J. (1970) J. Biol. Chem. 245.3122-3128 Smith, 8, Agradi, E., Libertini, L. and Dileepen, K.N. (1976) Proc. Natl. Acad. Sci. U.S. 73,11841188 Agradi, E., Libertini, L. and Smith, S. (1976) Biochem. Biophys. Res. Commun. 68. 894-900 Berge, R.K. and Farstad, M. (1978) Stand. J. Clin. Lab. Invest. 38,699-706 Porter, J.W. and Long, R.W. (1956) J. Biol. Chem. 243,20-28 Knudsen, J,, Clark, S. and Dils, R. (1975) Biochem, Biophys. Res. Commun. 65,921~926 Beige, R.K. and Far&ad, M. (1977) I lth FEBS Meeting, Copenhagen. Abstr. A-5-2-766. PP. Berge, R.K. and Far&ad, M. (1979) Eur. J. Biochem., in the press Berge, R.K. and Farstad, M. (1978) 12th FEBS Meeting, Dresden, Abstr. 3443. PP. Means, G.E. and Feeney, R.E. (1971) Chemical Modification of Proteins, Holden-Day, San Francisco,

16 Dole, V.P. (1965) J. Clin. Invest. 35,150-154 17 Bar-Tana, J., Rose, G. and Shapiro. B. (1971) Biochem. J. 122.253-362 18 Devkin, D. and Desser, R.K. (1968) J. Clin. Invest. 47, 1590-1602 19 Weber. K., Pringie, J.R. and Osborn, M. (1972) Methods Enzymoi. 26C, 3-28 20 Berge, R.K., Haarr, L. and Nygaard. A.P. (1976) Nucleic Acids Res. 3,1937-1945 21 Lowry, O.H., Rosebrough, N.J.. Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193,265-275 22 De Duve. C., Pressmann, B.C., Gianetto, R., Wattiaux, R. and Applemans, F. (1955) Biochem. J. 60, 604-607 23 Mahadevan. S. and Sauer. F. (1969) J. Biol. Chem. 244.4448-4453 24 Zahler, W.L., Barden, R.E. and Cleland. W.W. (1968) Biochim. Biophys. Acta 164, l-11 25 Barden. R.E. and Cleland, W.W. (1969) J. Biol. Chem. 244, 3677-3684 26 Flick, P.K. and Bloch, K. (1974) J. Biol. Chem. 249, 1031-1036 27 Arvidsson, E-0. and Belfage, P. (1969) Acta. Chem. Stand. 23,232-236 28 Arvidsson, E.O. (1965) Small Molecule-Protein Interaction, Thesis, Lund, Sweden 29 Whitaker, J.R. and Perez-Viilasenor. J. (1968) Arch. Biochem. Biophys. 124, 70-78 30 Bedord. C.J.. Kolattukudy. P.E. and Rogers, L. (1978) .4rch. Biochem. Biophys. 186,139--151 31 Deleepen, K.N. and Lin, C.Y. (1978) Biochem. J. 175,199.-206 32 Banner, W.M. and Bloch. K. (1972) J. Biol. Chem. 247, 3123-3133 33 Knudsen, J., Clark, S. and DiIs, R. (1976) Biochem. J. 160,683-691 34 Laurent, T. and KiBander. J. (1964) J. Chromatogr. 14.317-333 35 Brewer, M., Pesce, A.J. and Ashworth, R.B. (1974) Experimentai Techniques in Biochemistry, PP. 46-169. Prentice-Hall Inc., Enghwood Cliffs, NJ