Interrelationships between ornithine, glutamate and GABA-III. an ornithine aminotransferase activity that is resistant to inactivation by 5-fluoromethylornithine

Neurocbem. Int. Vol. 13, No. 3, pp. 383-391, 1988 Printed in Great Britain. All fights reserved

0197-0186/88 $3.00+ 0.00 Copyright © 1988 Perlpmmn Press pic

INTERRELATIONSHIPS BETWEEN ORNITHINE, GLUTAMATE AND GABA--III. AN ORNITHINE AMINOTRANSFERASE ACTIVITY THAT IS RESISTANT TO INACTIVATION BY 5-FLUOROMETHYLORNITHINE G ~ D^UNE and NIKOLAUS S~Lmt* Merrell Dow Research Institute, Strasbourg Center, 16 rue d'Ankara, 67084 Strasbourg Cedex, France (Received 14 March 1988; accepted 2 May 1988) Almtrmet--5-Fluoromethylornithine (5-FMOrn) is a specific inactivator of L-ornithine: 2-oxoacid aminotransferase (OAT). However, a certain proportion of the OAT activity in mouse brain, fiver and kidney is not inactivated by this compound. In the present work, the occurrence, distribution and subceHular localization of this 5-FMOrn-resistant OAT is reported. It was shown that the 5-FMOrn-rcsistant brain enzyme is kineticaily different from the corresponding liver enzyme, and it also differs from the 5-FMOro-semitive OAT. The most conspicuous difference between the 5-FMOru-rcsistant OAT of liver and brain is the sensitivity of the latter against excessive concentrations of its substrate 2-oxoglutarate. 5-FMOru and GABA are reversible inhibitors of the 5-FMOrn-rcsistant enzyme. Both compounds compete with Orn for the enzymes active site. A number of known inactivators of GABA-T which are at the same time inactivators of OAT, and canaiine, a natural inhibitor of OAT, inactivate both the 5-FMOrn-sensitive and the 5-FMOrn-resistant enzyme. Cmbaculine is the most potent inhibitor of the 5-FMOrn-resistant enzyme that is presently known. Our results are compatible with the suggestion that the 5-FMOrn-resistant OAT is an isoenzyme. From the fact that this form of OAT prevails in the brain, and its occurrence in the nerve ending fraction of brain homogenates supports the view that 5-FMOrn-rcsistant OAT may be involved in the intraneuronai generation of neurotranmmiUer glutamate and/or GABA from Orn as precursor. Further support in favour of this notion are previous fmdlngs which suggest feedback inhibition of OAT by GABA in GABAergic nerve endings.

L-Ornithine: 2-oxoacid aminotransferase (OAT) (EC 2.6.1.13) is present in many tissues. It is a mitochondrial enzyme, although a small proportion of O A T activity has been found in the nuclear fraction of rat fiver (Peraino and Pitot, 1963). The purified enzyme from rat fiver seemed uniform (Peraino et al., 1969), however, the enzyme forms aggregates, and with aggregation it changes it kinetic properties (Boernke et aL, 1981). Recently, the primary structure of the enzyme was established (Mueckler and Pitot, 1985; Simmaco et aL, 1986). The analysis of the geue structure suggests the existence o f a single gene for O A T in the rat (Shull and Pitot, 1986). There are, however, a number of observations which indicate the existence of more than one form of OAT: In ascites bepatoma cells (AH 130) and human cancer cells on OAT activity with properties different *To whom correspondence should be addressed.

from those of the enzyme from normal tissues was observed (Matsuzawa et al., 1982). Drejer and Schousboe (1984) reported different kinetic properties for O A T in cultured astrocytes, cerebral cortex interneurons and cerebellar granule veils. In contrast with the fiver enzyme (Peraino and Pitot, 1963), O A T purified from rat brain is, according to Deshmuk and Srivastava (1984), capable of transaminating lysiue (Lys) at a rate 40% of that achieved with ornithine (Orn) as substrate. A further perturbation of the picture of a single O A T activity comes from our recent observation that a certain proportion of brain and fiver O A T activity was resistant to inactivation by 5-fluoromethylornithine (5-FMOrn). This compound has been demonstrated to be a specific inactivator of O A T in vitro and in vivo. But, 5-20% of the enzyme activity which transforms O m into At-pyrroline-5-carboxylic acid, could not be inactivated even by incubation with 5 mM 5-FMOrn (Daune et al., 1988). 383

384

GENEVIEVE DAUNE and NIKOLAUS SELLER

We have studied some properties o f the 5 - F M O r n - r e s i s t a n t O A T activity in mouse tissues. These are described in the following. EXPERIMENTAL PROCEDURES

Chemicals If not stated otherwise chemicals and solvents were from Merck (Darmstadt, Germany) or Baker Chemicals (Derenter, The Netherlands). The following compounds were from Sigma Chemical Co. (St Louis, Mo., U.S.A.):2-oxoglutarate, dithiothreitol, oaminobenzaldehyde, L-ornithine, L-lysine, aminooxyacetic acid and L-canaline. o-Phthalaldehyde was from Roth (Karlsruhe, Germany), [l-~4C]D,L-ornithine (specific radioactivity 61 Ci/mol) from Amersham International (Amersham, U.K.). and L-2,3-[3H]ornithine (specific radioactivity 50.4ci/mmol) from NEN Research Products (Boston, Mass.). 5-Fluororomethylornithine.HC1 (MDL 72912) (Daune et al., 1988); gabaculine •HC1 (5-amino-l,3-cyclohexadienyl carboxylic acid) (Francois and Gittos, 1979) and ~'-acetylenic GABA (4-amino-hex-5-ynoic acid) (Metcalf and Casara, 1975) were synthesized in our Institute according to published methods. 2-Difluoromethylornithine (OrnidylTM) (Bey, 1978) is a product of Merrell Dow Pharmaceutical Co. Laboratory animals Female CDI albino mice (weighing 22 + 2 g) were from Charles River (St Aubin-les-Elbeuf, France). They were kept at standardized conditions (22°C, 60% rel. humidity, 12h light, 12h dark cycle; water and standard diet ad libitum), and were adapted to laboratory conditions.

Enzyme assays L-ornithine :2-oxoacid aminotrans)erase (EC 2,0.1.13) (OAT). OAT activity was determined as described in detail previously (Seiler et al., 1987). The incubation mixture was composed as follows: 100#10rn solution (variable conc.), 100pl 2-oxoglutarate solution (variable conc.) (both substrates were dissolved in 50mM potassium phosphate buffer, pH 8.0), 300/~1 enzyme preparation (0.1 Units) or tissue homogenate in 0.5 mM potassium phosphate buffer, pH 8.0, containing 80/~M pyridoxalphosphate, 5 mM dithiothreitol, 1 mM EDTA and 1% Triton X-100. The mixture was incubated at 37°C for 30 min under agitation. Reaction was stopped by addition of 100/ll 40% trichloroacetic acid and colour reaction was developed by addition of 0.4 ml of an aqueous 0.125% solution of o-aminobenzaldehyde, and further incubation for 10min. After centrifugation, 200 #1 aliquots of the supernatants were separated by HPLC (Varian Microcap column MCH 5 cap; eluent: 0.2 M acetic acid + methanol = 6 + 4; flow rate 1 ml/min). The coloured benzoquinazolinium compound was determined by recording absorbance at 440 nm. Assay for time-depending inactivation One ml of OAT preparation (I + 10 tissue homogenate in the above mentioned buffer) was mixed with 2 ml of 0.05 M potassium phosphate (pH 8.0) buffer, containing the inactivator, and incubated at 37°C. At certain time intervals between 0 and 25 min, 300 #1 aliquots were assayed for the residual enzyme activity under standardized conditions (Corn = 175 mM; C2.oxo~.ut~at, = 35.2 raM). The experimental results were evaluated according to Kitz and Wilson (1962).

Partial purification of O A T and 5-FMOrn-resistant OA T from mouse brain and liver Brains and livers were homogenized in 5vol 0.5 mM potassium phosphate buffer, pH 8.0, containing 1% Triton X-100, 0.1 mM pyridoxalphosphate, 5 m M dithiothreitol and 1 mM EDTA. The homogenate was sonicated (3 x 10 s) and then centrifuged at 1000g (30min). The pellet was rehomogenized in the same volume of buffer and the supernatant combined with the first extract. The extract was centrifuged at 25,000g (45 min), followed by ammonium sulphate fractionation. The precipitates obtained with 20-50% (w/v) ammonium sulphate were resuspended in 50mM potassium phosphate buffer pH 8.0 (containing 0.1 mM pyridoxal phosphate) and dialysed against the same buffer. The soluble proteins of this preparation of total OAT had a specific activity of 1.1/~mol Al-pyrroline-5-earboxylic acid formed/rag protein/h ( = 1.1 Units/mg protein) in brain and 0.75 Units/mg protein in liver. The 5-FMOrn-resistant enzymes were obtained by the same procedure, using the organs of mice 5 h after i.p. administration of 100mg/kg 5-FMOrn. Specific activity: brain, 0.13 Units/rag protein; liver, 0.11 Units/mg protein.

Determination of the reaction products of total OAT and 5-FMOrn-resistant O A T with L-[2,3)H]ornithine as substrate The incubation conditions were similar as those described above. The Orn conc. was 60 mM, and the incubation mixture contained 26/~Ci L-[2,3-3H]Orn. In order to obtain about the same amount of reaction product 0.2 Units of each enzyme preparation were used per assay. In the case of total OAT 35.2 mM 2-oxoglutarate were used, in the case of 5-FMOrn resistant enzyme only I mM. Reaction was stopped after 30min by addition of 0.5ml of 8% trichloroacetic acid; 850 ~1 of the acid extract were submitted to ion exchange column chromatography. Columns (5 × 12 mm) were filled with Dowex 50 Wx8 (200--400 mesh, H+-form). After sample application the columns were washed with 30 ml of water, and then eluted using 1 N HCI, whereby 1 ml fractions were collected. Radioactivity was counted in 0.5ml aliquots and A~-pyrroline-5-carboxylic acid was determined in 300#1 fractions by reaction with o-aminobenzaldehyde (for details see above). Under these elution conditions Orn was completely separated from N-pyrroline-5-carboxylic acid. In the L-[2,3:H]Orn preparation no significant radioactive impurities were found that co-eluted with the reaction product of transamination.

Preparation o f subcellular fractions For the subcellular fractionation of mouse brain homogenates, the procedure of sucrose density gradient centrifugation by Doddet al. (1981) was used. Nuclei were purified according to Kato and Kurokawa (1967). Proteins in the fractions were determined by the method of Lowry et al. (1951).

Determination of 14CO2-formation.from [ l-14C]D,L-ornithine in vivo The method was the same that has previously been used for the determination of respiratory ~4CO2 from labelled putreseine (Seiler and Eicbentopf, 1975). Groups of 4 mice were injected 1 #Ci, i.p., (2 #g) [1-14C]D,L-Orn and 14CO2 respiration was determined for the following 2 h. 5-FMOrn was administered 2 h before the labelled Orn was injected.

385

Interrelationships between ornithine, glutamate and GABA--III Table I. Affinities of substrates for total and 5-fluoromethylornithine (5-FMOrn)-resistant OAT activity of mouse liver and brain 5-FMOrn-resistant

Total OAT

OAT

K.

K=

Tissue

Subetrate

(raM)

(mM) .

Brain

Ornithine A B 2-Oxoglutarate Ornithine A 2-Oxoglutarate

I.I +0.1

3.9 + 1.1" 24±9 0.2 ± 0.01" 3.4 ± 1.5" 0.6 ± 0.2*

Liver

--

2.6 1.2 ± 0.11 2 ± 0.5

For the determination o f / ~ values the partially purified enzyme preparations were used (see Experimental Procedures). / ~ values for Orn were determined in the presence of either 35.2 raM (A) or ! raM (B) 2-oxoglutarate. Km values for 2-oxoglutarate were determined at 175raM L-ornithine. The data are mean values + SD (N = 2 or 3); the Km values of 2-oxoglutarate are the means of two independent experiments. *Statistically significant (P ~ 0.05) difference between the Km of the 5-FMOrn-resistant OAT and that of total OAT.

Calculations The experimental data of the enzyme kinetic experiments were evaluated by finear regression analyses RESULTS

Effect of 5-fluoromethylornithine on [l-14C]ornithine respiration If l # C i D,L-[l-m4C]Orn (specific radioactivity 61Ci/mol) was administered intraperitoneally to mice, 23% of the injected radioactivity was detected in the respiration within 60 rain. Pretreatment of the mice with 50 mg/kg 5-FMOrn, 2 h prior administration of the labelled Orn, reduced 14CO2 in the respiration by 87%. This value is comparable with the reduction of OAT activity in the liver of mice after the same dose of 5-FMOrn (Daune et al., 1988), however, the reduction of the 14CO2 respiration rate is the result of both, inhibition of OAT, and dilution of the labelled Orn due to the enhancement of the endogenous Orn pool. After administration of 100mg/kg of 5-FMOm the Orn concentration in liver increases at a rate of 20-30 nmol/g/min (Daune et al., 1988).

The transaminase activity which is resistant to 5-FMOrn had a higher affinity for 2-oxoglutarate than total OAT both in brain and liver. Its affinity for Orn was about the same in brain and liver. Another difference between the two enzyme activities is that the 5-FMOrn-resistant brain enzyme is more sensitive to changes of 2-oxoglutarate concentration than the fiver enzyme (Fig. 1). Maximum transamination rates were observed at about I mM 2-oxoglutarate. At 35 mM the transamination rate was reduced to about

~2oo e~ ®

..A,N

Io0

\

"\

""X

/

/ z o

"

re

z (n z

o

E

Susbstrate properties of ornithine and 2-oxoglutarate Mice received 100 mg/kg 5-FMOrn i.p., in order to achieve maximum inhibition of OAT. After 5 h, brain and liver were removed. If homogenates of these tissues were incubated for 30 min in the presence of 5 mM 5-FMOrn, the Orn transaminating activity was not decreased below the level that had been achieved in vivo. Using partially purified enzyme preparations (see Experimental Procedures) the substrate properties of Orn and 2-oxoglutarate were determined, and Km values were calculated from Lineweaver-Burk plots. They are summarized in Table 1.

/

I00 r

,~....~......=~s ~

° t ..f,.ov 50

laJ I<( re

oL!

0.1

i 02

t! (18 I

ER

I 2

2-OXOOLUTARAT(

Fig.

1.

Rate

of

ornithine

' ' 8 IO (mM

= 35.Z

)

transamination

by

5-FMOrn-resistant OAT activity of mouse liver and brain as a function of 2-oxoglutasate concentration. Mice were pretreated with 100 mg/kg 5.FMOrn, and organ homogehates were pt~tred 5 h after administration of the drug. The a=,3saymixtures contained 9.1 m8 of tissue and 175 mM Orn. Each point represents a separate determination.

386

GENEVIEVE DAUNE and NIKOLAUS SELLER

H

3H

0 L-[2,3- 3HI ORNITHINE

H2N3 " ~ ~ ~ H NH2 OH TRANSAMINATtON

0.~-T R A N S A M I N A T ~

0

H 3H 0 H ~ O H ;~H NH2

HaH 0 HzN " ~ , ~ ' ~ O H 0

1

O

3H

H2N~OH O-

3H OOH

l

A1-PYRROLINE-5-CARBOXYLIC ACID

~ C O O H

A1.PYRROLINE.2,CARBOXYLIC ACID Fig. 2. Scheme o f &- and ~-transamination o f L-[2,3-3H]omithine.

half the maximum rate. In contrast, the activity of the liver enzyme was not reduced by high 2-oxoglutarate concentrations (Fig. 1). With increasing 2-oxogtutarate concentration the affinity of Om for the brain enzyme was increasing. The Lineweaver-Burk plots demonstrated an increase of the / ~ of Orn from 1.1mM at 35.2mM 2-oxoglutarate to 24mM at

1 mM 2-oxoglutarate for the 5-FMOrn-resistant brain enzyme (Table 1). Homogenates of organs from normal and 5-FMOrn-pretreated mice apparently do not metabolize Lys as was reported for a partially purified OAT preparation of rat brain (Deshmuk and Srivastava, 1984). We did not observe the coloured product that

Table 2. Specific radioactivities of L-[2,3-3H]Orn, the substrate and of the reaction products of total OAT and 5-FMOrn-resi~tant OAT Reaction product L-[2,3-3H]Orn

Total OAT

5-FMOrn-resistant OAT

Specific radioactivity 0.73 + 0.09 0.87 ± 0.07 0.74 + 0.05 L-[2,3)H]Orn (60raM) was incubated with 35.2mM (total OAT) or 1 mM (5-FMOrn-re~ltant OAT) 2-oxoglutarate with partially purified enzymes (0.21 Units per assay) for 30 rain at 37°C. The reaction products were separated by ion exchange chromatography (for details, see Experimental Procedures). The values are the means of 7 determinations _+SD.

Interrelationships between ornithine, glutamate and GABA--III

5mM

I j K i , 1.graM O

2

lOmM

5

[5-

'lOmM

FMOrn3 (mM)

Fig. 3. Inhibition of 5-FMOrn resistant brain OAT activity by 5-FMOrn in the presence of I mM 2-oxoglutarate. Dixon plot. Each point is the mean of two experiments (and duplicate determinations). is expected to be formed from ALpiperidine-6 carboxylic acid and o-aminobenzaldehyde, nor did Lys competitively inhibit Orn transamination.

387

qulnazolinium salt. In order to exclude ,,-transamination by the 5-FMOrn-resistant OAT, L-[2,3-3I-I]Orn was incubated with partially purified preparations of both, total OAT and 5-FMOrnresistant OAT from mouse brain under the usual reaction conditions. ALPyrroline carboxylate was separated from Orn by ion exchange column chromatography, and in the fractious eluting from Dowex 50 Wx8, with 1 N HCI the specific radioactivity of the product of transamination was determined by assaying from aliquots radioactivity and dihydroquinozolinium salt formation. (For details of the experimental protocol, see Experimental Procedures.) It appears from Table 2 that the specific radioactivity of the reaction products of total OAT, 5-FMOrn-resistant OAT, and that of [2,3-3H]Orn, which served as substrate, were identical within the precision of the method. As appears from Fig. 2, ~-transamination of [2,3-3H]Orn would have caused a near to complete loss of tritium.

Inhibitors of the 5-FMOrn-resistant OAT activity Reaction product of 5-FM Orn -resistant OAT Based on the formation of a coloured dihydroquinazolinium salt with o-aminobenzaldehyde (Vogel and Davis, 1952), the reaction product of 5-FMOrnresistant OAT seemed identical with that of total OAT. However, one could not exclude that instead of ALpyrroline-5-carboxylic acid, the isomeric ALpyrroline-2-carboxylic acid was formed by ~-transamination, which may also react with oaminobenzaldehyde and form a similar dihydro-

Incubation of tissue preparations from 5-FMOrn-treated mice did not show any further reduction of OAT activity by 5 mM 5-FMOrn, if OAT activity was determined in the presence of 175 mM Orn and 35 mM 2-oxoglutarate. If, however, 2-oxoglutarate concentration was reduced to 1 mM the enhanced OAT activity measured under these conditions (see Fig. 1) could be inhibited by 5 mM 5-FMOrn. From the Dixon plot (Fig. 3) a /~ = 1.9 mM was calculated for 5-FMOrn. This plot

A

B

5(3

~

_~ 25

-~o6

.~5o

-o.e~o,,I-~.ooa 0.02 -o~z~,

IOC

!

i

005

O.I

I/[Orn]

( r a M ) "1

K,-ao

I [GARA] (raM)

Fig. 4. Inhibition of 5-FMOrn-resistant brain OAT by GABA in the presence of 1 mM 2-oxoglutarate. (A) Lineweaver-Burk plot. (B) Determination of/~- from the apparent K, values. Each point is the result of duplicate determinations. NC.I. 13/3--H

~88

GENtVIliVt! DAUNF

BRAIN

©

I00


I

NIKOLAUS SEIt£R

and

E~ E

50 i

IZ FZ 0 (D

LIVER

IOO-

50-

o -

Fig. 5. Inhibition of the 5-FMOrn-resistant OAT activity by known inactivators of GABA-T at 5 mM concentration in the presence of 175 mM Orn and 35.2 mM 2-oxoglutarate. The bars indicate SD (N = 3).

indicates competition between Orn and 5-FM()rn l\~r the enzymes active site. Incubation of 10 m M 5-FMOrn for 30 min at 37 C with 0.05 Units of 5-FMOrn-resistant O A T from mouse brain, in the presence of I m M 2-oxoglutaratc, and subsequent reaction with o-aminobenzaldehyde under conditions which are usual for the deterruination of ALpyrroline-5-carboxylic acid, did not yield measurable amounts of a coloured reaction product. This finding, although not conclusive, seems to suggest that 5 - F M O r n is not a good substrate of the 5-FMOrn-resistant OAT. In order to obtain a definitive answer to this question, radioactively labelled 5-FMOrn will be required. In contrast with published work (Grillo and Colombatto, 1986) neither inhibition nor inactivation of total OAT, or of 5-FMOrn-resistant O A T activity could be observed with 2-difluoromethylornithine, indicating that this compound is neither a competitive inhibitor nor an inactivator of OAT. It has previously been shown that G A B A is a competitive inhibitor of synaptosomal O A T of the brain (Yoneda et al., 1982) and of the liver enzyme ( K ~ = 3 . 4 m M ) (Seiler et al., 1987). The 5-FMOrn-resistant enzyme of brain is also competitively inhibited by G A B A (Fig. 4); a K, = 0.8 m M was found under the experimental conditions. In contrast with 7-vinylGABA, a number of other inactivators of G A B A - T (gabaculine, 7-acetylenic G A B A , L-canaline, aminooxyacetic acid) that have

B

A IO0 5o o u 3O 2O

2raM

2GABACUUNE 5

~

S

m

M GABACULI

I

I

6

I

i

12

I

I

18

I

I

I

24

0

i

i

6

i

12

i

I

IB

NE I

L 24

TIME OF INCUBATION (rain)

Fig. 6. Time-dependent inactivation of total OAT of mouse liver (A) and 5-FMOrn-resistant brain OAT (B) by gabaculine. After preincubation with gabaculine for various length of time, OAT activity was determined in the presence of 175 mM Orn and 35.2 mM (A) and 1 mM (B) 2-oxoglutarate. From the incubations with total OAT (A) 30/~1 have been used per standard OAT assay whereas 300 #1 have been used in the case of 5-FMOrn-resistant enzyme (B). The concentrations of gabaculine in the figure represent the values during the preincubation period. Each point is the result of an individual experiment.

Interrelationships between omithine, glutamate and GABA--III previously been established as inactivators of OAT (Jung and Seller, 1978; Rosenthal, 1978; Murphy and Brosnan, 1976) are capable of inhibiting the 5-FMOrn-resistant OAT activity (Fig. 5). Among these compounds D,L-gabaculine was most potent, followed by L-canaline. As appears from Fig. 6, gabaculine is a potent competitive inhibitor of the 5-FMOrn-resistant OAT, but it also inactivates this enzyme in a timedependent manner: in the case of total OAT the inactivation curves intercepted close to the origin of the control curve, whereas inhibition was observed in the case of 5-FMOrn-resistant enzyme even at time zero of incubation. From competition experiments with Orn a/~. = 0.045 #M for gabaculine was determined.

Organ distribution of 5-FMOrn-resistant O A T activity Groups of 4 mice received a 0.036% solution in tap water of 5-FMOrn for up to 13 days. The drug intake corresponded to 45--60mg/kg 5-FMOrn per day. In order to minimize changes of OAT activity due to variations of the time elapsing between water intake and isolation of the organs, the animals were always killed by decapitation between 9 and 10a.m. The activity of OAT was determined in the tissue homogehates in the presence of a saturating concentration of Orn. The other substrate, 2-oxoglutarate, was present either at 35.2 mM, or at I mM. The results of these

389

measurements are summarized in Table 3. Repeated administration of the drug did not change significantly the activity of 5-FMOrn-resistant OAT, but there was a tendency to decreased values in all four brain parts. Mouse brain contains higher activities of this enzyme than the other tissues that have been studied so far. Brainstem and medulla showed considerably higher activities than hemispheres and cerebellum.

Subcellular distribution of 5-FMOrn-resistant O A T activity Mitochondria and synaptosomes were prepared from homogenates of whole mouse brain by differential centrifugation, followed by sucrose density gradient centrifugation (Dodd et al., 1981), and the crude nuclear fraction was purified according to Kato and Kurokawa (1967). Aliquots of these fractions were incubated with 10mM 5-FMOrn for 20m in at 37°C. Subsequently OAT activity was determined in the presence of saturating concentrations of Orn and 1 or 35.2 mM 2-oxoglutarate. The high speed supernatant fraction and the purified nuclei did not contain significant OAT activities. The highest specific activity of total (1.5#mol/mg protein/h) and 5-FMOrn-resistant OAT (0.5/zmol/mg protein/h in the presence of I mM 2-oxoglutarate) was found, in agreement with the expectations, in the mitochondria. Synaptosomes also exhibited a significant OAT activity:

Table 3. Total and 5-FMOrn-resistant OAT activity in organs of mice after oral administration of 40 mg/kg of the drug per day Duration of treatment with 5-FMOrn (Days) Tissue Hemispheres Brainstem Cerebellum Medulla Liver Kidney Spleen Lung Muscle

0 A

1 A

B

22.9 -i- 1.6 (100) 29.9 + 1.4 (100) 33.4 + 1.2 (100) 29.0+2.7 (100) 151 + 4 6 (100) 155 ± 15 (100) 27 ± 2 (100) 26 ± 3 (100) 7.9 ± 0,6

1.5 + 0.4 (6.5) 6.4 + 0.3 (21.4) 3.4 ± 0.6 (10.2) 4.0±0.3 (13.8) 9±3 (6) 4.8 ± 0.2 (3.1) b.d.I,

7.5 +_0.3 (32.8) 14.6 + 0.9 (48.8) 9.7 ± 0.7 (29.0) 17.7±3.1 (61.0) 8.8± 1.5 (6) 5.6 _+0.5 (3.6) b.d.l.

1.1 ± 0.4 (4.2) b.d.l,

b.d.l.

b.d.I,

4 A

B 6.6 + 0.6 (28.8) 14.2 + 1.2 (47.5) 9,0 ± 0.8 (27.0) 13.5±1.9 (46.6) 6.0±0.5 (4) 6.1 ± 0.5 (3.9) b,d.I.

b.d.l,

1.9 + 0.2 (8.3) 5.5 _ 1.5 ( 18.4) 3.5 _ 0.3 (10.5) 4.4_+0.5 (15.2) 9 . 9 ± 1.5 (7) 5.4 ± 0.3 (3.5) 0.6 ± 0.4 (2.2) 1.1 ± 0.3 (4.2) b.d.I,

b.d.I,

b.d.l,

13 A

B 5.1 ± 0.4 (22.4) 13.3 + 0.4 (44.5) 8.6 + 1.4 (25.8) 12.7±0.6 (43.8) 8.7_+ 1.6 (6) 5.1 ± 1.2 (3.3) b.d.I.

b.d.I,

1.2 + 0.9 (5.3) 2.4 + 0. I (8.0) 3.3 ± 0.2 (9.9) 4.5+0.2 (15.5) 17.8 ± 4.5 (l 2) 4.2 ± 0.4 (2.7) 0.2 ± 0.1 (0.7) 1.2 ± 0.2 (4.6) b.d.l,

b.d.I,

b.d.l,

b.d.I.

b.d.l.

b.d.L b.d.I.

(100) Eye

15.6 ± 1.3

Mice received a 0.036% solution of 5-FMOrn in tap water as drinking fluid. After certain time intervals the organs were isolated, and OAT activity was determined either in the presence of 35.2 mM (A) or 1 mM (B) 2-oxoglutarate and 175 mM Ore. Mean values + SD (N ffi 4) in/Jmol of ALpyrroline-5-caboxylate formed/h/g tissue; percent of control in parentheses. b.d.I, ffi below the detection limit of the method.

390

(JENt~\;ItVt DAI~N[ and NIKOLAUSStlLER

0.41tmol/mg protein/h (total OAT activity): 0.12/~mol/mg protein/h (5-FMOrn resistant OAT at I mM 2-oxoglutarate). The ratio o1" 5-FMOrn-resistant and total O A T was not significantly different in these fractions. DISCUSSION

From our experiments it is evident that in addition to the well known OAT another form of Orn-o)aminotransferase exists, which exhibits different kinetic properties. A major characteristic of this new form of OAT is its resistance to inactivation by 5-FMOrn, whereas OAT undergoes time-dependent inactivation by this Orn analogue. Assumedly, the 5-FMOrn-resistant OAT has not been detected previously, because it represents a small part of the total OAT activity and it is inhibited by several other known inhibitors of OAT. These include canaline (Rosenthal, 1978) aminooxyacetic acid (Murphy and Brosnan, 1976), 7-acetylenic GABA and gabaculine (Jung and Seiler, 1978). For gabaculine it was shown that it is not only a potent competitive inhibitor of this new enzyme activity, but also an inactivator. A thorough comparison of other OAT inhibitors will presumably reveal more differences between the 5-FMOrn-sensitive OAT activity and the 5-FMOrn-resistant enzyme, but it appears already from the present work that the differences are not profound as far as substrate and inhibitor characteristics are concerned. Since we have no indication for a natural substrate other than Orn, we consider presently the 5-FMOrn-resistant activity an isoenzyme. The characterization of the kinetic properties of the new OAT activity is necessarily preliminary, because relatively crude enzyme preparations have been used. Nevertheless, the observed differences between the 5-FMOrn-resistant activity of liver and brain are obvious. The most conspicuous distinction is the sensitivity of the brain enzyme against 2-oxoglutarate at high concentrations, whereas the liver enzyme shows normal Michaelis-Menten type kinetics. 5-FMOrn-sensitive OAT of liver and brain also show normal saturation kinetics at high 2-oxoglutarate concentration (Peraino and Pitot, 1963: Daune et al., 1988). For rat brain 2-oxoglutarate concentrations of 0 . 1 5 9 + 0 . 0 6 # m o l / g wet wt have been reported (Chapman et al., 1977). Assuming that concentrations in mouse brain are of similar magnitude, it is evident that in vivo the 2-oxoglutarate concentration is below the in vitro observed optimum. Inhibition of 5-FMOrn-resistant OAT by

2-oxoglutarate, and prevention of the competitive inhibition of the enzyme by 5-FMOrn is, therelbre. unlikely. 5-FMOrn-resistant OAT was observed m brain with highest activity in the brainstem, and was also tbund in liver and kidney, but seems present only at low activity (or absent) in most other tissues (Table 3). From the fact that repeated administration of 5-FMOrn did not enhance the residual OAT activity in any tissue, one may conclude that the 5-FMOrn-resistant OAT was not induced. It has been previously reported (Daune et al., 1988) that brain Orn concentrations of 5-FMOrn-treated mice are decreasing before a significant recovery of the 5-FMOrn-sensitive enzyme could be observed. It was proposed that Orn concentrations are enhanced in Hvo above the concentration that would be observed due to the inactivation of OAT, because 5-FMOrn is inhibiting the residual enzyme activity. This assumption is supported by the present demonstration that 5-FMOrn is a competitive inhibitor of this portion of OAT. The disappearance of free 5-FMOrn from brain before 24 h (Daune et al., 1988) allows the transamination of Orn by the residual enzyme, demonstrating that 5-FMOrn-resistant OAT plays an important role in the metabolism of Orn in t it~o.

One of the major functions of OAT is presumably the regulation of tissue Orn concentrations. The rate of Orn degradation in brain is considerable, in spite of a low steady-state level of Orn (Daune and SeileL 1988; Daune et al., 1988). It has been known tor many years that glutamate and GABA are formed from labelled Orn in brain (Seiler and Knodgen, 1971), and that brain contains the appropriate transport system for the active uptake of Orn into nerve endings (Seiler and Deckardt, 1976; Shank and Campbell, 1983) and the enzymes for its transformation within the nerve endings into glutamate and GABA (Murrin, t980; Wong et al., 1981; Yoneda et al., 1982; Shank and Campbell, 1983). Furthermore, evidence was recently presented for the in vivo feedback inhibition of OAT by GABA within the nerve endings (Seiler et al., 1987; Daune and Seiler, 1988). Wong et al. (1981) reported a high portion of brain OAT in nerve ending fractions. These facts together with our present observations, i.e. the highest proportion of 5-FMOrn-resistant OAT in brain, with a significant activity of this enzyme in nerve endings, is rather strong support in favour of an intraneuronal role of 5-FMOrn-resistant OAT in the formation or regulation of glutamate and or GABA. The concept of OAT as an enzyme involved in the generation of

Interrelationships between ornithine, glutamate and GABA--III

391

2-substituted propargylamines as potential catalytic irreversible enzyme inhibitors. Tetrahedron Lett. 3337-3340. Mueckler M. M. and Pitot H. C. (1985) Sequence of the precursor to rat omithine aminotransferase deduced from a cDNA clone. J. biol. Chem. 260, 12993-12997. Murphy B. J. and Brosnan M. E. (1976) Subcellular localREFERENCES ization of ornithine decarboxylase in fiver of control and Bey P. (1978) Substrate-induced irreversible inhibition of growth-hormone-treated rats. Biochem. J. 157, 33-39. ~-aminoacid decarboxylases. Application to glutamate, Murrin L. C. (1980) Ornithine as a precursor for aromatic-L-~-aminoacid and ornithine decarboxylases. 7-aminobutyric acid in mammalian brain. J. Neurochem. In: Enzyme-Activated Irreversible Inhibitors (Seller N., 34, 1779-1781. Jung M. J. and Koch-Weser J., eds), pp. 27-41. Peraino C. and Pitot H. C. (1963) Ornithine-6-transaminase Elsevier/North-Holland Biomedical Press, Amsterdam. in the rat. I. Assay and some general properties. Biochim. Boernke W. E., Stevens F. J. and Peraino C. (1981) Effects biophys. Acta 73, 222-231. of self-association of omithine aminotransferase on its Peraino C., Bunville L. G. and Tahmisian T. N. (1969) physicochemical characteristics. Biochemistry 20, Chemical, physical and morphological properties of or115-121. nithine aminotransferase from rat liver. J. biol. Chem. Chapman A. G., Meldrum B. S. and Siesjo B. K. (1977) 244, 2241-2249. Cerebral metabolic changes during prolonged epileptic Roberts E. (1981) Strategies for identifying sources and sites seizures in rats. J. Neurochem. 28, 1025-1035. of formation of GABA-precursor transmitter glutamate Danne G. and Seller N. (1988) Interrelationships between in brain. In: Glutamate as Neurotransmitter (DiChiara G. ornithine, glutamate and GABA--II. Consequences of and Gessa G. L., eds), pp. 91-102. Raven Press, New inhibition of GABA-T and ornithine aminotransferase in York. brain. Neurochem. Res. 13, 69-75. Rosenthal G. A. (1978) The biological and biochemical Daune G., Gerhart F. and Seller N. (1988) properties of L-canaline, a naturally occurring structural 5-Fhioromethylornithine, an irreversible and specific inanalogue of L-ornithine. Life Sci. 23, 93-98. hibitor of L-ornithine: 2-oxoacid aminotransferase. BioSeller N. (1980) On the role of GABA in vertebrate polychem. J. 253, 481-488. amine metabolism. Physiol. Chem. Phys. 12, 411-429. Seller N. and Deckardt K. (1976) Association of putrescine, Deshmuk D. R. and Srivastava S. K. (1984) Purification and spermidine, spermine and GABA with structural elements properties of oruithine aminotransferase from rat brain. of brain cells. Neurochem. Res. I, 469--499. Experientia 40, 357-359. Dodd P. R., Hardy J. A., Oakley A. E., Edwardson J. A., Seller N. and Eichentopf B. (1975) 4-Aminobutyrate in Perry E. K. and Delaunoy J.-P. (1981) A rapid method for mammalian putrescine catabolism. Biochem. J. 152, preparing synaptosomes: comparison with alternative 201-210. Seller N. and Knodgen B. (1971) Die Umwandlung yon procedures. Brain Res. 226, 107-118. Glutaminsaure, Putrescin und Ornithin in die Drejer J. and Schousboe A. (1984) Ornithine-f7-Aminobuttersaure im Gehirn. Hoppe-Seyler 's Z. Physaminotransferase exhibits different kinetic properties in iol. Chem. 352, 97-105. astrocytes, cerebral cortex interneurons and cerebellar granule ceils in primary culture. J. Neurochem. 42, Seller N., Spraggs H. and Danne G. (1987) Interrelations between ornithine, glutamate and GABA--I. Feed-back 1194-I 197. inhibition of ornithine aminotransferase by elevated brain Francois J.-P. and Gittos M. W. (1979) A preparative synthesis of D,L-gabaculine. Synth. Commun. 9, 741-750. GABA levels. Neurochem. Int. 10, 391-397. Shank R. P. and Campbell G. L. (1983) Ornithine as a Grillo M. A. and Colombatto S. (1986) Is rat liver ornithine precursor of glutamate and GABA. Uptake and metabodecarboxylase localized in both cytoplasm and nucleus? lism by neuronal and glial enriched cellular material. In: Biomedical Studies o f Natural Polyamines (Caldarera C. M., Clo C. and Guarnieri C., eds), pp. 15-18. CLUEB, J. Neurasci. 9, 47-57. Shuil J. D. and Pitot H. C. (1986) Analysis of gene structure Bologna. Jung M. J. and Seller N. (1978) Enzyme-activated irresuggests the existence of a single gene for ornithine aminotransferase in the rat. Fed. Proc. 45, 1620. versible inhibitors of L-ornithine:2-oxoacid aminoSimmaco M., John R. A., Barra D. and Bossa F. (1986) The transferase. J. biol. Chem. 2.$3, 7431-7439. primary structure of ornithine aminotransferase. Kato T. and Kurokawa M. (1967) Isolation of cell nuclei from the mammalian cerebral cortex and their assortment Identification of active-site sequence and site of posttranslational proteolysis. FEBS Lett. 199, 39-42. on a morphological basis. J. Cell Biol. 32, 649-~52. Kitz R. and Wilson B. (1962) Esters of methanesulfonic acid Vogel H. J. and Davis B. D. (1952) Ghitamic semialdehyde and A~-pyrroline-5-carboxylic acid intermediates in as irreversible inhibitors of acetylcholinesteras¢. J. biol. the biosynthesis of proline. J. Am. chem. Soc. 74, 109Chem. 237, 3245-3249. 112. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol Wong P. T. H., McGeer E. G. and McGeer P. L. (1981) A sensitive radiometric assay for ornithine aminoreagent. J. biol. Chem. 193, 265-275. transferase: regional and subcellular distribution in rat Matsuzawa T., Sugimoto N., Sobue M. and Ishignro I. brain. J. Neurochem. 36, 501-505. (1982) Ornithine oxoacid aminotransferase found in AH Yoneda Y., Roberts E. and Dietz G. W. (1982) A new 130 ascites hepatoma ceils. Biochim. biophys. Acta 714, synaptosomal biosynthetic pathway of glutamate and 356-360. GABA from ornithine and its negative feedback inMetcalf B. W. and Casara P. (1975) Reglospecific hibition by GABA. J. Neurochem. 20, 1686-1694. 1,4-addition of a propargylic anion. A general synthon for neurotransmitter amino acids (Seller, 1980; Roberts, 1981) is currently under investigation in several laboratories.