Biochimica et Biophysica Acta. 990 (1989) 25-30Elsevier

BBA 23021

Compartmentation of dicarboxylic acid ,B-oxidation in rat liver:importance of peroxisomes in the metabolism

of dicarboxylic acids

Hisashi Suzuki, Junji Yamada, Takafumi Watanabe and Tetsuya SugaDepartment of Clinical Biochemistry, Tokyo College 0/Pharmacy, Tokyo (Japan)

(Received 26 August 1988)

Key words: Dicarboxylic acid beta oxidation; Beta oxidation; Peroxisome; Mitochondrion; (Rat liver)

25

Peroxisomal and mitochondrial ,B-oxidation of dicarboxylic acids (DCAs) were investigated and compared. Whenisolated hepatocytes were incubated with DCAs of various chain lengths, H 202 was derived from peroxisomal,B-oxidation, the rates of its generation being comparable to those seen with monocarboxylic acids (MCAs), whereas therates of ketone body production, a measure of mitochondrial ,B-oxidation, were much lower than those with MCAs.Peroxisomal ,B-oxidation measured by cyanide-insensitive NAD reduction exhibited similar chain-length specificities forboth dicarboxylyl-CoAs (DC-CoAs) and monocarboxylyl-CoAs (MC-CoAs), except that the activities for DC-CoAs with10-16 carbon atoms were about half of those of the corresponding MC-CoAs. In contrast, mitochondrial ,B-oxidationmeasured by antimycin A-sensitive Oz consumption had no activity for DCAs. In the study with purified enzymes, thereactivities of mitochondrial camitine palmitoyltransferase and acyl-CoA dehydrogenase for DC-CoAs were much lowerthan those for MC-CoAs, while the reactivity of peroxisomal acyl-CoA oxidase for DC-CoAs was comparable to that forthe corresponding MC-CoAs. Accordingly, the properties of carnitine palmitoyltransferase and acyl-CoA dehydrogenasemust be the rate-limiting factors for mitochondrial ,B-oxidation, with the result that DCAs might hardly be oxidized inmitochondria. Comparative study of ,a-oxidation capacities of peroxlsomes and mitochondria in the liver showed thatDC u-CoA was hardly subjected to mitochondrial ,B-oxidation, and that the ,B-oxidation of DCAs in rat liver, therefore,must be carried out exclusively in peroxisomes.

Introduction

In recent years, several inherent diseases associatedwith peroxisomal dysfunction have been reported suchas Zellweger's cerebro-hepato-renal syndrome andadrenoleukodystrophy [1,2]. The patients with such dis­eases show, as some of their clinical abnormalities,deficiency of plasmalogen and accumulations of very­long-chain fatty acids, trihydroxycholestanoic acid anddicarboxylic acids (DCAs) in their tissues [1-4]. Todate, all of the above components except the last havebeen demonstrated to be metabolised exclusively inperoxisomes; the key enzyme of plasmalogen synthesis,dihydroxyacetonephosphate acyltransferase, exists onlyin peroxisomes [5J and very-long-chain fatty acids [6J

Abbreviation: DCA., dicarboxylic acid with n carbon atoms.

Correspondence: H. Suzuki, Department of Clinical Biochemistry,Tokyo College of Pharmacy, Horinouchi, Hachioji, Tokyo 192-03,Japan.

and trihydroxycholestanoic acid [7J are oxidized exclu­sively in peroxisomes. Thus, the accumulation of DCAsmight suggest that DCAs are also preferentiallymetabolized by peroxisomal ,B-oxidation. Moreover,some authors have reported increased urinary excretionof short-chain DCAs (especially adipic acid and sub ericacid) under such conditions as riboflavin deficiency [8],systemic carnitine deficiency [9J, hypoglycine A intoxi­cation [1OJ and congenital acyl-CoA dehydrogenationdefects [11-13], all being conditions in whichmitochondrial ,B-oxidation activity is lowered.

From these observations, we supposed that the per­oxisomal fJ-oxidation system plays an important role inthe degradation of DCAs which are formed from long­chain monocarboxylic acids (MCAs) through PASO­mediated e-oxidation reactions in the cell. In order toexamine this idea, we have investigated peroxisomal andmitochondrial ,a-oxidation of DCAs. Kalvraa et al. [14]have briefly reported the p-oxidation of DCAs in bothorganelles, but no information on the properties of theenzymes has been published. In the present work, we

0304-4165/89/503.50 «:>1989 Elsevier Science Publishers B.V. (Biomedical Division)

26

purified the enzymes related to peroxisomal andmitochondrial ,a-oxidation and investigated their cata­lytic properties for the fJ-oxidation of DCAs.

Materials and Methods

MaterialsDicarboxylyl-CoA esters (DC-CoAs) were prepared

as described [15] and purified by high-performanceliquid chromatography (HPLC). on a JLBondapak C I8column (Waters Associates, U.S.A.). Monocarboxylyl­CoA esters (MC-CoAs) were purchased from SigmaChemical Co., U.S.A. The concentrations of these acyl­CoA esters were determined by the method of Ellman[16]. L-Carnitine was a gift from Earth PharmaceuticalCo., Japan. Other chemicals obtained from commercialsources were of analytical grade.

ExperimentalAnimals. Male Wistar rats weighing about 250 g were

used. The rats were fed a standard diet (CE-2: CleaJapan Inc., Japan).

Preparation of isolated hepatocytes. Parenchymalhepatocytes were isolated by the collagenase perfusionmethod according to Moldeus et aI. [17] and the pre­parations showing more than 90% cell viability in termsof lactate dehydrogenase latency and trypan blue exclu­sion were used. They were preincubated with 1 mML-camitine at 37° C for 20 min before use [18].

Preparation of peroxisomes and mitochondria. Per­oxisomes and mitochondria were prepared by a combi­nation of differential centrifugation and sucrose densitygradient centrifugation as described previously [19].These preparations were respectively used as 'isolatedperoxisomes' and 'isolated mitochondira' for enzymeassay, except for the assay of mitochondrial O2 con­sumption in which mitochondria were prepared by dif­ferential centrifugation as described [20] using 0.3 Mmannitol containing 1 mM EDTA and 10 mM Hepes(pH 7.4) as the isolation medium.

Purification of enzymes. Acyl-CoA oxidase [21],carnitine palmitoyltransferase [22] and long-, medium­and short-chain acyl-CoA dehydrogenases [23,24] werepurified from the liver of rats fed the standard dietcontaining 2% (w/w) di-(2-ethylhexyl)phthalate for 2-6weeks. The homogeneity of each enzyme was confirmedby polyacrylamide gel electrophoresis with or withoutsodium dodecyl sulfate.

AssaysPeroxisomal and mitochondrial {3-oxidation. The assay

methods for these activities were described in detail inour previous report [25]. The activity of peroxisomal{3-oxidation in isolated hepatocytes was measured byperoxidatic generation of formaldehyde from methanolbased on substrate (free acid form) dependent H 20 2

generation. When isolated peroxisomes were used as anenzyme source, cyanide-insensitive substrate (CoA-esterform) dependent NAD reduction was followed spectra­photometrically. The activity of mitochondrial ,a-oxida­tion in isolated hepatocytes was measured by substrate(free acid form) dependent ketone body production(3-hydroxybutyrate and acetoacetate). When using themitochondrial preparation as an enzyme source, anti­mycin A-sensitive substrate (free acid or CoA-esterform) dependent O2 consumption was followed bypolarography.

Other enzymes. Acyl-CoA oxidase was assayed asdescribed previously [25]. Carnitine acyl transferase wasassayed by the method of Markwell et al. [26] in mediumcontaining, in a volume of 1 mI, 50 p.M acyl-CoA, 2.5mM t-camitine, 0.125 mM 5,5'-dithiobis(2-nitrobenzoicacid), 1.25 mM EDTA, 58 mM Tris-HCl (pH 8.0) and0.1 mI enzyme. When purified carnitine palmitoyltrans­ferase was used, 0.2% (vjv) Tween 20 was added to theassay medium. Acyl-CoA dehydrogenase was assayedby the method of Hryb and Hogg [27] with a slightmodification. The assay medium contained, in a volumeof 1 ml, 50 p.M acyl-CoA, 50 JLM FAD, 1 mM phe­nazine methosulfate, 60 p.M 2,6-dichlorophenol­indophenol, 1 mM N-ethylmaleimide, I mM KCN,0.067% (w/w) Triton x-ioo, 0.2 mgy'ml fatty acid-freebovine serum albumin, 60 mM potassium phosphate(pH 7.4) and 0.1 m1 enzyme. When purified acyl-CoAdehydrogenase was used, the activity was measured asdescribed previously [25]. Catalase and glutamate dehy­drogenase were assayed by the methods of Luck [28]and Beaufay et al. [29], respectively.

Protein. Protein was determined by the method ofLowry et al. [30] with bovine serum albumin as astandard.

Calculation. The activities of peroxisomal andmitochondrial enzymes per gram of liver were calcu­lated based on the recoveries of the activities of acyl-CoAoxidase (palmitoyl-CoA as a substrate) and glutamatedehydrogenase, respectively, using peroxisomes andmitochondria as enzyme source.

Results

fJ-Oxidation of DCAs in isolated hepatocytesFig. 1 shows the fJ-oxidation activities for DCAs of

various chain lengths in isolated hepatocytes. The per­oxisomal and mitochondrial ,a-oXidation activities weremeasured by H 20 2 generation and ketone body produc­tion, respectively. The rates of H 202 generation withDCA lO _16 were high and comparable to those withMCAs, but with DCA 4 _ 8 were relatively low. In con­trast, the rates of ketone body production with DCAswere much lower than those with MCAs. Consideringthe detection limit in the ketone body assay, the valueswith DCAs were hardly significant.

27

Fig. 2. Sucrose density gradient centrifugation of light mitochondrialfraction from rat liver. (A) Catalase; (B) cyanide-insensitive DC 12-CoAdependent NAD reduction; (C) DC12-CoA dependent H202 genera­tion; (D) glutamate dehydrogenase; (E) cyanide-insensitive palmitoyl-

CoA dependent NAD reduction; (F) protein.

Fig. 1. Peroxisomal and mitochondrial ,B-oxidation of DCAs(e---e) and MCAs (0.----·0) in isolated hepatocytes. Theactivities were measured by substrate-dependent H202 generation(peroxisomal) and ketone body production (mitochondrial). (A) Per­oxisomal activity. (B) Mitochondrial activity. Results are expressed as

the means ± S.D. of more than four rats.

B~\~

,\

'5--o-~..t~\\ ,is

1500

3000A

2800,..;:- ..- >~=

u '"lO-r:::~ E 1400,..::::N 0r::: EW.s

e.--!I \, \, ,

" ~I \I \I \I \

I ~t \,,~~' \

o cf •••• ,.,........ b o. I .~4 8 12 18 4 8 12 16

Chain length Chain length

Fig. 4. Chain-length specificities of camitine acyltransferase (A) andacyl-CoA dehydrogenase (B). The activities for DC-CoAs (e---e)and MC-CoAs (0·-----0) were measured using mitochondrial frac­tion obtained {rom sucrose density gradient centrifugation as de­scribed under Materials and Methods. Results are expressed as the

means±S.D. of more than three rats.

6-~1\-: 600 / '\

6 ~ I \;:; = I \

s '" " 's

n,,: ~\\4 8 12 16

Chain langth

Fig. 3. ,B-Oxidation of DCAs (e--e) and MCAs (0------0) inisolated peroxisomes. The activities were measured by cyanide-insen­sitive acyl-CoA-dependent NAD reduction. Results are expressed as

the means ± S.D. of more than three rats.

and cyanide-insensitive palmitoyl-CoA dependent NADreduction.

U sing isolated peroxisomes as an enzyme source, theperoxisomal fJ-oxidation of DC-CoAs of various chainlengths was measured by cyanide-insensitive NAD reoduction (Fig. 3). High activities were found with DC­CoAs, while there was low activity with DC6-CoA andno activity with DC4-CoA. The maximum activity wasfound with DC12-CoA. There was no appreciable dif­ference in substrate specificity between DC-CoAs andMC-CoAs, except that the activities for DC-CoAs with10-16 carbon atoms were about half of those for thecorresponding MC-CoAs.

The activity of mitochondrial fJ-oxidation was mea­sured by following 0z consumption. However, substratedependent 02 consumption was not found using anyDCAs or DC-CoAs examined. Under the same condi­tion, definite activities with MCAs and MC-CoAs wereobserved and the intensities were consistent with thosein our previous work [25] (data not shown).

Fig. 4 shows the activities of carnitine acyl transferaseand acyl-CoA dehydrogenase in isolated mitochondriaobtained by sucrose density gradient centrifugation. Incontrast with MC-CoAs, DC-CoAs were apparently

BOD

1BOOo

ESO

45

~a ~

F "~l;,

B .§

-.::l

3l::Q

"l::'il0t:

A

120

4

E21.3?:';;l::IIo

fJ-Oxidation of DCAs in subcellular fractionsFig. 2 shows the subcellular distribution of DC 12-CoA

dependent H 202 generation and cyanide-insensitiveNAD reduction in the light mitochondrial fraction ofrat liver. Catalase, a marker enzyme of peroxisomes,was located mainly in fraction 3 and glutamate dehy­drogenase, a mitochondrial enzyme, existed in fraction7. The H z02-generation and cyanide-insensitive NAD­reduction activities measured with DC12-CoA displayedthe same subcellular distribution as those of catalase

A 4000

l'lB

100 l::

l~.2

e :!7j .. : ..~I , " 'i'lla Ii I a " ,/ \l_~., " I \ ~ ..

.. lD

i \~f s~ "II ~~ C 2000

H '{~ C 50 a -01-.D ~I E

N=:, .. - ,~a a \ l:: aN E \ a E

:l: .5 , ~ l::.. -'&. ~

0 ''0 0 .. • • .I I I

8 12 16 4 8 12 16

Chll" length Chain Iingth

28

TABLE I

Kinetic constants of enzymes related to {J -oxidation

Substrate

MC4-CoA

MC6-CoA

MCs-CoAMClO-CoAMCI2-CoA

MCI4-CoA

MCwCoAMC1S-CoA

DC4-CoADC6-CoA

DCa-CoADC10-CoADCI2-CoADC I4-CoA

DC I6-CoA

Acyl-CoA Carnitine palmitoyl- Long-chain Medium-chain Short-chainoxidase transferase acyl-CoA acyl-CoA acyl-CoA

dehydrogenase dehydrogenase dehydrogenase

K", a Vmaxb Km Vmax Km Vmax Km Vrnax K m Vmax

99.1 0.11 trace C 0 trace 11.2 20.0124 0.92 101 7.94 0 4.87 3.30 34.0 2.8052.4 1.94 83.0 20.4 0 2.73 2.57 017.2 1.70 29.7 36.4 19.0 0.13 2.63 2.57 0

5.29 1.01 12.1 31.3 5.n 0.14 4.45 1.24 09.09 1.74 17.0 34.1 12.5 1.90 4.15 0.12 0

11.3 2.01 16.7 20.3 11.1 2.82 5.21 0.Q7 013.3 0.55 23.5 9.81 11.2 2.13 3.08 0.02 0

0 0 0 0 a113 0.54 0 0 trace 0

19.9 1.08 0 0 trace 010.1 1.27 trace 0 289 0.48 015.6 1.70 61.9 0.62 0 273 1.41 013.7 1.41 25.6 1.22 0 157 0.73 010.7 0.85 20.7 0.57 0 330 0.09 0

• }AM.b I'mol/min per mg protein.c The value could not be evaluated because of its low activity.

TABLE II

poor substrates for both enzymes. The patterns ofchain-length specificities for DC-CoAs and MC-CoAswere dissimilar.

Capacities of peroxisomal and mitochondrial f1.oxidotion

The activities of peroxisomal and mitochondrial f1-oxidation weremeasured by acyl-Co.A-dependent NAD reduction and 02 consump­tion, respectively. Results are expressed as the means ± S.D. of fourrats.

Characteristics of the enzymes involved in DCA ,B-oxida­tion

The kinetic constants of the enzymes related to the,B-oxidation of DCAs and MCAs are summarized inTable 1. For DCS_ 16-CoA the K m values of acyl-CoAoxidase, considered to be the rate-limiting enzyme ofperoxisomal ,B-oxidation, were low (10.1-19.9 JLM), butfor DC6-CoA the K m was relatively high (113 JLM) andfor DCcCoA no significant activity was observed, Thus,the reactivities of the enzyme for DC-CoAs were similarto those for MC-CoAs. Carnitine palmitoyltransferase,responsible for acyl-CoA transportation through themitochondrial membrane, showed activity towardDClO _ 16-CoA. However, the Vmax values were muchlower than those for MC-CoAs, although the K m values

were slightly lower or almost similar. With DC-CoAs,medium-chain acyl-CoA dehydrogenase, which cata­lyzes the first step of mitochondrial ,a-oxidation, hadVmax values which were less than 40% of that forMC6-CoA, and the K m values were much higher thanthose for MC-CoA!!. Long- and short-chain acyl-CoAdehydrogenases showed no activities toward DC-CoAsof various chain lengths.

Discussion

Capacities of peroxisomal and mitochondrial ,B-oxidationof DCA in the liver

Hepatic peroxisomal and mitochondrial ,B-oxidationcapacities for DC l2-CoA were compared (Table II).With the use of peroxisomes and mitochondria isolatedfrom rat liver, the activities of DC12-CoA dependentNAD reduction and 02 consumption, respectively, weremeasured. The activity of peroxisomal ,a-oxidation forpalrnitoyl-CoA was about 25% of the mitochondrialone. However, when DC 12-CoA was used as a substrate,the peroxisomal activity was about 70% of that forpalrnitoyl-CoA whereas no significant activity was ob­served in mitochondria.

Our present results from the experiments using iso­lated hepatocytes, subcellular fractions and purified en­zymes clearly indicate that the ,B-oxidation of DCAsmust be carried out exclusively in peroxisomes. In theexperiment with isolated hepatocytes, high activities ofperoxisomal ,B-oxidation of DCAs were observed,

mitochondria

1l± 161973±150

Activity (nmol/min per g liver)

321 ± 18468±36

peroxisomes

DC 12-CoAPalmitoyl-CoA

Substrate

whereas those of mitochondrial ,a-oxidation were notsignificant. Similar results were obtained with isolatedperoxisomes and mitochondria. The lower reactivity ofthe mitochondrial ,a-oxidation system for DCAs isaccounted for by the kinetic properties of the enzymes.The Vm ax values of carnitine palrnitoyltransferase forDC-CoAs are very low. It is likely that the lowerreactivity of carnitine palmitoyltransferase for DC-CoAswould lead to a restriction of the transport of DC-CoAsinto the mitochondrial matrix. Moreover, even if thesubstrates are transported, dehydrogenation of them,the first reaction of mitochondrial ,a-oxidation, couldnot occur because of low reactivities of acyl-CoA dehy­drogenase for DC-CoAs. These facts indicate that theproperties of carnitine palmitoyltransferase and acyl­CoA dehydrogenase are the rate-limiting factors formitochondrial ,a-oxidation of DC-CoAs. In contrastwith the above mitochondrial enzymes, the kinetic prop­erties of peroxisomal acyl-CoA oxidase for DC-CoAswere similar to those for MC-CoAs. Furthermore, it isconsidered that DCAs may be mainly chain-shortenedin peroxisomes to a length of about six carbon atoms(adipic acid). This coincides with the fact that adipicacid is abundant in the urine of patients with dicarbox­ylic aciduria.

Concerning the mitochondrial ,a-oxidation of MCAsin liver cells, it is considered that long-chain MCAs(more than C lO ) are activated to CoA-ester form byacyl-CoA synthetase, and then transferred across themitochondrial membrane in a carnitine palmitoyltrans­ferase dependent reaction [31]. On the other hand,short-chain MCAs (CeCa) are directly transferredacross the membrane and activated by acyl-CoA syn­thetase in the mitochondrial matrix [32]. Consideringthese two processes for DCAs, we investigated chain­length specificity of mitochondrial ,a-oxidation usingthe free acid form of DCAs. However, no mitochondrial,a-oxidation activities were observed when DCAs wereused as substrates. These results indicate that thechain-shortening reaction of DCAs occurred exclusivelyin peroxisomes.

A difference in peroxisomal .B-oxidation activitieswhen using DC-CoAs and MC-CoAs as substrates wasobserved in the experiment using isolated peroxisomes(Fig. 3), whereas the activities for DCAs were compara­ble to those for MCAs when isolated hepatocytes wereused (Fig. 1A). Although the reason for this differenceis not clear, the peroxisomal ,a-oxidation activities ob­tained when isolated peroxisomes were used must repre­sent the real capacity of ,a-oxidation itself. However, inthe case of isolated hepatocytes, there are various fac­tors, for example, activation to CoA-ester form, lipidsynthesis and other metabolism, which might modulatethe peroxisomal ,a-oxidation activity.

The physiological significance of peroxisomal,a-oxidation is considered to be relatively low in fatty

29

acid oxidation associated with energy-generation;Osmundsen [33] reported that peroxisomal ,a-oxidationwould operate as an auxiliary system in a situationwhere the mitochondrial process operates close to itsmaximum capacity. In a previous study on xenobioticmetabolism by ,a-oxidation we also have suggested thatperoxisomal ,a-oxidation has an important role in thedegradation of acyl compounds [25,34]. On the otherhand, it is generally understood that in the cell DCAsare formed form MCAs through P-450-mediated w-hy­droxylation and subsequent oxidation in the cytosol,and that the DCAs are degraded by ,a-oxidation. How­ever, the physiological significance of such DCAmetabolism is unknown. In this work, some suggestiveobservations have been made. As described at the be­ginning of this paper, DCA metabolism is enhancedunder certain pathological conditions [8-13] wheremitochondrial ,8-oxidation activity is lowered. More­over, this fact is also observed in starvation and diabetes[35-37] where excessive influx of MCAs into the liveroccurs and peroxisomal ,a-oxidation increases [38,39].Thus, we propose that the formation of DCAs and theirdegradation by ,a-oxidation might serve as a pathway tomoderate the accumulation of MCAs in the cell, andthat the peroxisomal ,a-oxidation system would play amain role in the degradation process. Sharma et al.[40,41] reported co-induction of microsomal "'-oxida­tion and peroxisomal ,a-oxidation by hypolipidemicagents. We also found similar co-induction with phen­oxyacetate derivatives and perfluorinated fatty acids(unpublished data). If the expression of the enzymesinvolved in both systems were also synchronously regu­lated under physiological conditions, this would behighly significant in the light of our proposal regardingDCA metabolism.

References

1 Wanders, R.J.A., Schutgens, R.B.H., Heymans, H.S.A., Collins, 1.,Goldfischer, S., Hashimoto, T., Schrakamp, G., Van den Bosch,H., Tager, J.M. and Schram, A.W. (1987) Peroxisomes in Biologyand Medicine (Fahimi, RD. and Sics, H., eds.), pp. 341-352,Springer-Verlag, Heidelberg.

2 Schutgens, R.B.H., Heymans, H.S.A., Wanders, R.J.A., Van denBosch, H. and Tager, J.M. (1986) Eur. J. Pediatr. 144,430-440.

3 Bjorkhen, I., Blomstrand, S., HAgA, P., KWIC, B.F., Palanek, E.,Pedersen, J.I., Strandvik, B. and Wikstrom. S. (1984) Biochim.Biophys. Acta 795, 15-19.

4 Rocchiccioli, F., Aubourg, P. and Bougneres, P.F. (1986) Pediatr,Res. 20, 62-66.

5 Datta, N.S., Wilson, G.N. and Hajra, A.K. (1984) New Engl. J.Med. 311, 1080-1083. .

6 Singh, I., Moser, A.E., Goldfischer, S. and Moser, H.W. (1984)Proc. Nat!. Acad. Sci. USA 81,4203-4207.

7 Kase, B.F., Prydz, K., Bjokhem, I. and Pedersen, J.1. (1986)Biochim. Biophys, Acta 817, 37-42.

8 Gregersen, N. and Kalvraa, S. (1982) J. Inherit. Metab. Dis. 5,17-18.

9 Karpati, G., Carpenter, S., Engel, A.G.• Watters. G., Allen. J.,

30

Rothman, S., Klassen, G. and Mamer, a.A. (1975) Neurology 25,16-24.

10 Tanaka, K. (1972) J. Bio\. Chern. 247, 7465-7478.11 Gregersen, N., Kalvraa, S., Rasmussen, K., Mortensen, P.B., Di­

vry, P., David, M. and Hobolth, N. (1983) Clin. Chim. Acta 132,181-191.

12 Przyrembel, H., Wendel, V., Becker, K., Bremer, HJ., Bruinvis, L.,Ketting, D. and Wadman, S.K. (1976) Clin. Chim. Acta 66,227-239.

13 Mantagos, S., Genel, M. and Tanaka, K. (1979) 1. Clin. Invest. 64,1580-1589.

14 Kelvraa, S. and Gregersen, N. (1986) Biochim. Biophys. Acta 876,515-525.

15 Seubert, W. (1959) Biochem. Prep. 7, 80-83.16 Ellman, G.L. (1959) Arch. Biochem. Biophys. 82, 70-77.17 Moldeus, P., Hogberg, 1. and Orrenius, S. (1978) Methods En­

zymo\. 52, 66-71.18 Christiansen, R.Z. and Bremer, J. (1976) Biochim. Biophys. Acta

448, 562-577.19 Suga, T., Watanabe, T., Matsumoto, Y. and Horie, S. (1984)

Biochim. Biophys. Acta 794, 218-224.20 Hoppel, C., DiMarco, J.P. and Tandler, B. (1979) 1. BioI. Chern.

254, 4164-4170.21 Osumi, T., Hashimoto, T. and Vi, N. (1980) 1. Biochem, 87,

1735-1746.22 Miyazawa, S., Ozasa, H., Osumi, T. and Hashimoto, T. (1983) J.

Biochem. 94, 529-542.23 Furuta, S., Miyazawa, S. and Hashimoto, T. (1981) J. Biochem. 90,

1739-1750.24 Ikeda, Y., Dabrowski, C. and Tanaka, K. (1983) 1. BioI. Chern.

258, 1066-1076.

25 Yamada, J., Ogawa, S., Horie, S., Watanabe, T. and Suga, T.(1987) Biochim. Biophys. Acta 921, 292-301.

26 Markwell, M.A.K., McGroarty, E.l., Bieber, L.L. and Tolbert,N.E. (1973) 1. BioI. Chern. 248, 3426-3432.

27 Hryb, D.l. and Hogg, J.F. (1979) Biochem. Biophys. Res. Com­mun. 87, 1200-1206.

28 LUck, H. (1963) Methods of Enzymatic Analysis (Bergmeyer,H.-V., ed.), pp. 885-894, Academic Press, New York.

29 Beaufay, H., Bendall, D.S., Baydhuin, P. and De Duve, C. (1959)Biochem. J. 73, 623-628.

30 Lowry, o.n., Rosebrough, N.J., Farr, A.L. and Randall, R.J.(1951) J. BioI. Chern. 193, 265-275.

31 Van Tal, A. and HUlsmann, W.e. (1970) Biochim. Biophys. Acta223, 416-428.

32 Aas, M. and Bremer, J. (1968) Biochim. Biophys. Acta 164,157-166.

33 Osmundsen, H. (1982) Ann. N.Y. Acad. Sci. 386, 13-29.34 Yamada, J., Harle, S., Watanabe, T. and Suga, T. (1984) Biochem.

Biophys. Res. Commun. 125,123-128.35 Pettersen, J.E., Jellum, E. and Eljarn, L. (1972) Clin. Chim. Acta

38,17-24.36 Pettersen, J.E. (1972) Clin. Chim. Acta 41,231-237.37 Mortensen, P.B. and Gregersen, N. (1981) Biochim. Biophys. Acta

666, 394-404.38 Ishii, H., Horie, S. and Suga, T. (1980) J. Biochem. 87, 1855-1858.39 Horie, S., Ishii, H. and Suga, T. (1981) 1. Biochem. 90, 1691-1696.40 Sharma, R., Lake, B.G., Foster, 1. and Gibson, G.G. (1988)

Biochem. Pharmacal. 37, 1193-1201.41 Sharma, R., Lake, B.G. and Gibson, G.G. (1988) Biochem.

Pharmacal. 37, 1203-1206.