Neurochemical Research, Vol. 17, No. 2, 1992, pp, 193-i99

Accumulation of Labeled Gamma-Aminobutyric Acid into Rat Brain and Brain Synaptosomes After I.P. Injection L. Vignolo 1,2, A. Cupello 1,3, P. Mainardi 2, M. V. RapallinC, A. Patrone 2, and C. Loeb m (Accepted July 23, 1991)

The accumulation of labeled GABA into brain and brain nerve endings was studied in the adult rat after i.p. injection of large doses of neurotransmitter (740 mg/Kg). In the first 5-30 minutes after the injection the exogenous neurotransmitter reaches a stable plasma level of around 5 raM. The accumulation of radioactive GABA into the brain presents a latency of a few minutes from the time of the injection. Thereafter, the accumulation of the neurotransmitter is almost linear with time. Once in the brain tissue labeled GABA is in part broken down. The exogenous neurotransmitter is taken up in GABA-ergic nerve endings with a steep increase between 20 and 30 minutes after the injection. From a quantitative point of view, the data show that the brain accumulation of labeled GABA at 30 minutes post injection is minimal in the respect of the steady state average concentration of the endogenous neurotransmitter (0.014%). However, the amount of radioactive GABA which accumulates in the nerve endings, at the same post injection time, is around 7% of the endogenous neurotransmitter in that compartment. The data thus show a selective enrichment of exogenous systemic GABA in a physiologically important compartment of the brain. KEY WORDS: Gamma-aminobutyric acid; blood brain barrier; rat brain; nerve endings; subsynaptosomal fractions.

INTRODUCTION

positive reports about the prevention of seizures in mice (8) and even improvements in epileptic patients (9,10) after systemic administration of GABA. More recently, intravenous injections in mice of a large dose of the neurotransmitter have been reported to provide protection against 3-mercaptopropionic acid induced convulsions (11). A lower dose has been shown to reduce the activity of epileptic loci in the cat, provided the blood brain barrier was locally altered by irradiation (12,13). However, the attempts to directly administrate plain GABA as an anticonvulsant have become very few in favour of experimental trials with molecules derived from it but with a more lipophilic character (14-16). Other molecules derived from GABA and with a marked lipophilic character so to pass readily the BBB have been recently synthetised and shown active in neurophysiological assay (17-19).

The issue whether systemically administered gammaamino-butyric acid (GABA) can penetrate the blood brain barrier and be taken up into mammalian brain tissue in adult animals is a rather controversial one. Early experiments using tracer amounts of the neurotransmitter appeared to exclude its passage from the blood stream into the brain (1--4). However, other experiments have shown that large doses, grams/Kg, of the neurotransmitter result in an elevation of its cerebral content (5-7). From a therapeutic point of view, there were early 1 Centro di Studio per la Neurofisiologia Cerebrale, C.N.R. Genova. 2 Clinica Neurologica dell'Universit~ di Genova, Italy. 3 To whom to address reprint requests: Dr. Aroldo Cupello, Centro di Neurofisiologia Cerebrale, C.N.R., Via De Toni 5, 16132 Genova Italy. tel. 010 509058 telefax 010 354180

193 0364-3190/92/0200-0193506.50/09 1992PlenumPublishingCorporation

194

Vignolo, Cupello, Mainardi, Rapallino, Patrone, and Loeb

Referring to the direct use of G A B A as an anticonvulsant, protection against penicillin and isoniazid induced epilepsy in rats has been demonstrated in our lab by administration i.p. of large doses of G A B A (740 rag/ Kg) sonicated with phosphatidylserine (PS). The same amount of plain G A B A was without an effect in the same models of experimental epilepsy (20,21). The present experiments were undertaken in order to understand the modalities of the passage of the exogenous neurotransmitter into the blood and to the brain and nerve endings compartment after the i.p. injection, at the same dose (740 mg/Kg) used in our previous pharmacological experiments. The final goal being that of understanding where and how phosphatidylserine makes a difference leading to the antiepileptic effect of GABAPS in comparison with plain GABA. In the same time these experiments were an attempt to better define the more general issue of the fate of blood borne exogenous GABA and its possibility of being taken up in the brain, and then accumulated in a physiologically important compartment such as the nerve endings.

brain vascular space. We have evaluated this space to be 3.1% in the adult rat, by intravenous [14C]inulininjections (data not presented here). Synaptosomes were prepared by a proc.edure derived from Cotman and Matthews (23), involving a discontinuous Ficoll gradient (24). The synaptosomal fraction underwent then osmotic shock with distilled water (3 ml/synaptosomes obtained from a, rat brain). After the osmotic shock, aliquots were taken for the evaluation of the protein in the synaptosomal fraction. The suspension was then centrifugated at 105,000g for 1 hr at 4~ The pellet containing synaptic vesicles, intraterminal mitochondria and synaptic membraneswas resuspended and counted for 14C-radioactivity.From the supernatant, aliquots were taken for the determination of soluble 14-C radioactivity. The rest of the supematant was analysed by the electrophoreticmethod for determining the proportion of intact [14C]GABA(22). Determination of the Octanol/WaterPartition Coefficientfor GABA and Serine. The n-octanol/waterpartition coefficientswere determined for (14-C)-GABA and (14-C)-Serine by the technique of Yunger and Cramer (25). A differencewas that as the water phase we used distilled water instead of 10 mM phosphate buffer, pH 7.0. In these analyses we used the same batch of [14C]GABA used throughout the present experiments. However, [~4C]GABAwas further purified by preparative TLC on Silicagel-plates (solvent system, butanol:acetic acid:water 4:1:1), before the partition experiments. L-[U-~4C]Serine, s.a. 141 mCi/mmol, freshly arrived from Amersham International (U.K.), was used for the determination of serine partition coefficient.

EXPERIMENTAL PROCEDURE

RESULTS

Animals and [14C]GABAAdministration Procedure. In this work, we used Sprague Dawley rats weighing 220-260 grams, maintained on standard laboratory diet. On the day of the experiment, the animals were injected i.p. with a mixture of "cold" GABA (740 mg/Kg) associated with 35 ~Ci of [U-14C]GABA,228 mCi/mmol, Amersham, U.K. The neurotransmitter was dissolved in 3 ml of distilled water, with a final concentration of 0.55 M. Groups of seven animals were sacrificed respectively at 5, 10, 20, and 30 minutes after the injection of labeled GABA. These procedures were performed always between 8:00 - 9:30 in the morning. The sacrifice was by decapitation of the animals, after which the brains were rapidly sampled, blood samples were also collected in the process, in heparinized tubes. Analysis of the Blood Samples. Plasma was obtained from the blood samples by centrifugation. The plasma samples were kept frozen at - 80~ until the biochemical assays. For the assays, 100 ~1 aliquots were diluted 1:75 with distilled water, added with 1/10 of their volume of buffer E, made 10% in sucrose and run by an electrophoreticmethod for the determination of the proportion of intact [14C]GABA (22). Evaluation of the radioactivity in the samples before the electrophoresis provided the information about the level of radioactivity in plasma. Evaluation of 1~C-Radioactivity and % of Unchanged [14C]GABA in Brain Tissue and Synaptosomes. The brain tissue was washed three times with buffer A (0.32 M sucrose; Tris-HC1, 10 raM; pH 7.4). The tissue was then homogenizedwith 10 volumes of buffer A by 7 strokes in a teflon-glass homogenizer. Aliquots were taken from the homogenates for determining the radioactive content and for the evaluation, by the electrophoretic procedure, of the proportion of [14C]GABA within tissue 14C-radioactivity.Radioactivitycontent of the brain tissue was determined after subtraction of the contribution by blood in the

We report in Figure 1A the level of plasma (14-C)radioactivity in the first 30 minutes of the post-injection period. The data show that in this period the blood borne radioactivity remains practically constant at the level it reaches at 5 minutes post-injection. Most of the 14-C radioactivity is intact [14C]GABA, as it can be seen in the same figure from the data representing the proportion of labeled G A B A within plasma radioactivity. Such a proportion, in the post injection period considered, is n e v e r b e l o w 8 4 % . T h e l e v e l of u n b r o k e n d o w n [14C]GABA in the plasma is reported in Figure 1B, that level is stable at 4-5 raM. The accumulation of radioactivity in the brain is reported in Figure 2A together with the proportion of [14C]GABA within brain total radioactivity. Taking in account that p4C]GABA in the blood is largely protected against catabolism (Figure 1A), one can assume that brain radioactivity accumulating within 30 minutes aetuaIly represents labeled G A B A which originally entered the brain. The tissue radioactivity, subtracted of the contribution by the blood in the brain inulinic space, increases linearly in the time period considered. The slope of the best fit line is 1.07 x 103 dpm/min/gr, which in terms of rate for G A B A entering the brain is 3.9 x 10 .4 ~mol/ sec/gr.

GABA Accumulation

@

195

|

@

o ?

O

30

2 M

$

-o

o o

0

2

IO0

2O

a 2. B 84

g U Cl

r

">

m L

100

G

10

50

/

Z

g (A)

O 10

3o

20

(A)

rnin.

(0

=

a o < m

q)

@ .-

5

=

4

2oo

~5

@

O Z

@

E

loo

2: g

u

/

G/

c

o o

3

go
,

< (0


o "0

I

0

10

lo

30

=o

(A) |

rain.

|

E "0

4.

0

lp~ o.242; n s p5= 0.078; ns Pl0= 0.014 p~= 0.064: n s m o

I

(c)

10

mitt.

(e)

[

20

30

min

cles + intraterminal mitochondria). In both cases there appears to be a major jump in the label accumulation between 20 and 30 minutes. In fact, for the soluble radioactivity the increase between the 20 and the 30 minutes value is by 104% (p = 0.002), whereas that between 10 and 20 minutes is only 41% (p --- 0.019). For the radioactivity in the particulate fraction, the increase between the 20 and the 30 minutes value is by 150% (p < 5 X 10 -7) and that between the 10 and the 20 minutes one is only 60% (p = 0.005). Considering the soluble radioactivity which is actually associated with GABA (Figure 3B), the increase in the accumulation of the labeled neurotransmitter between 20 min and 30 min is still prominent (+ 56%; p = 0.016). The increase between 20 and 10 min is lower (+42%; p --- 0.028). In Figure 3C we report the ratio between soluble

Fig. 3. A: accumulation of (14C)radioactivity into brain synaptosomes. The lower curve (. . . . . . ) represents the radioactivity in the synaptosomal particulate fraction. The intermediate curve ( . . . . . . . ) represents the supematant and the continuous line is the sum of the previous two. The data have been normalized by the amount (nag) of synaptosomal protein. B: accumulation of [x4C]GABA (nM/mg synaptosomal protein) in the synaptosomal soluble component (continuous line). The broken line represents the proportion of the [I'C]GABA within soluble (14C)-radioactivity. C: ratios of soluble to particulate ~4C-radioactivity within synaptosomes at the various times post injection. The Pt values refer to the statistical comparison of the relevant ratio with that at time t. The vertical bars and the numbers in parentheses have the usual meaning.

and particulate radioactivity within synaptosomes as a function of time. The results show that this ratio tends to decrease with time. This indicates a gradual redistribution of synaptosomes associated radioactive material between the synaptic sap and intraterminal vesicles and mitochondria. Finally, we report in Table I an evaluation of the 30 min accumulation of exogenous intact GABA in comparison with the published steady state levels of the neurotransmitter in brain tissue and brain nerve endings fraction. The data show that in the tissue the accumulated radioactive neurotransmitter represents only 0.014% of the endogenous counterpart. However, in synaptosomes that proportion is much higher (6.8%). Thus, the data show a preferential accumulation of exogenous GABA by GABA-ergic nerve endings. The enrichment factor in the nerve endings being 6.8 : 0.014, i.e. 486.

GABA Accumulation

197

Table I. Accumulationof ExogenousGABAin Brain and Brain Synaptosomesin ComparisonWith the EndogenousLevel

Fraction Whole tissue"

Synaptosomesb

Endogenous GABALevel (>5)d 1.88 _ 0.07~ (from Bohlen et al. 1979;ref.29) (12) 11.8 _-,2-0.7 (fromWoodet al. 1988;ref.30)

Accumulation of Exogenous GABAWithin 30 Minutes

one representing passive diffusion across the blood brain barrier. In other words, the cerebrovascular permeability-surface area product (PA) for a certain neutral aminoacid was assumed to be:

Rc

(6) 0.00027 -+ 0.00004

0.00014

(5) 0.80 - 0.22

0.068

" mmols/1 b nmol/mgprotein c ratio betweencolumns3 and 2 a withinparenthesesthe numberof experiments

PA =

V,~,, + KD Cp,. . . . )

(Km +

if that aminoacid was the sole to engage the carrier (as it was the case in Smith et al. perfusion system). However, when the particular amino acid is in the presence of other neutral amino acids, substrates for the same carrier, Km must be substituted by: Km(~pp~i

= Kin(i)[1+

SD

DISCUSSION

The data we report here appear to confirm that gamma-amino-butyric acid when given systemically in large doses can accumulate into the adult mammalian brain (5-7). However, in our experiments the actual concentration reached in the brain by the exogenous neurotransmitter is below 1 ~M and appears to be negligible in the respect of the average brain concentration of GABA (see Table I). The fact that other authors found higher accumulations in both mice (5-7) and rats (6) may be explained taking into account that these authors have followed experimental protocols involving higher doses and/or more prolonged periods of GABA administration. In any case, the dose (740 mg/Kg; i.p.) which we have utilized ensures a costant plasma level of exogenous neurotransmitter for at least 30 minutes after the injection. The protection of labeled GABA from enzymatic catabolism is assured by its relatively high plasma level, 4 - 5 mM. This high and constant plasma concentration of the exogenous neurotransmitter is the basis for its continued, linear accumulation into the brain (Figure 2A). In connection with this, we would like to discuss in some detail the possible mechanisms underlying GABA influx into the brain. In a recent paper Smith et al. (26) have studied the kinetics of the transport of neutral aminoacids (NAA) across the blood brain barrier in the adult rat. They assumed that the passage was due to two components, one representing saturable facilitated transport and another

(1)

~ (~/1 j-1 \ Kin(j) ] J j~i

(2)

where i represents the aminoacid of interest and j one of the other n-1 neutral amino acids able to engage the carrier; Cpl(j)'s represent their plasma concentrations. The same authors demonstrated that for 14 NAA's there is a linear relationship between log 1/Kin and the octanol/water partition coefficient. Since the authors quoted did not include GABA in their study, we evaluated its octanol/water partition coefficient in order to have an indication of what could be its Km for the facilitated transport carrier. We found a value of (9) 0.96 + 0.40 x 10-3, which is very similar to that for serine (3) 0.97 _ 0.20 x 10 -3. Since the Krn value for serine is 6.6 mM (26), the same value should apply to GABA. Then, taking in account the other NAA's present in plasma and their Km's and concentrations reported by Smith et al., from equation (2) one comes up with a Knq(app) for GABA of 153 mM. In our experiments the plasma level of GABA is between 4 and 5 mM. Thus, the neurotransmitter is not engaging the facilitated transport system at a significant extent. Actually, in this plasma concentration range, [GABA]plasm a < < KIII(app)GABA, SO that for the unidirectional influx of GABA into brain: Vbrain influxGABA = Vm~

Krn(app)GABA

(3)

X [GABA]p I....

+ KDa~A x [GABA]plasma

-(

Vm~'~ + KD~,~A)• [GABA]pl..... \ KITI(app)GABA

198

Vignolo, Cupello, Mainardi, Rapallino, Patrone, and Loeb

Phenomenologically, the situation is analogous to a plain diffusion process with V = K x C. This influx law and the constancy of [GABA]plasm a in the first 30 minutes in the post injection period explains the constant rate of the labeled neurotransmitter accumulation in brain tissue (Figure 2A). Moreover, the negligible contribution to GABA passage by carrier mediated transport indicates that amino acid exchange between plasma and brain (27) can not play a significant role under the present conditions. The data in Figure 3A, B show that a certain amount of radioactivity (Figure 3A) and intact [14C]GABA (Figure 3B) accumulate in brain synaptosomes. The data show in addition that with time there is a tendency to a redistribution of radioactivity in the nerve endings. In particular, the relative amount of 14C-radioactivity in the particulate compartments (synaptic membranes, synaptic vesicles and intraterminal mitochondria) tends to increase with time. Although the radioactivity we are speaking about does not necessarily represent [14C]GABA, it is tempting to suggest that there is a continued accumulation of the exogenous neurotransmitter in such compartments. A further detail which seems worth noting is the speeding up in the accumulation of 14C-radioactivity (Figure 3A) and [14C]GABA (Figure 3B) into synaptosomes between 20 and 30 minutes. We interpret this result assuming that labeled exogenous GABA is reaching in this period an extracellular concentration such to allow interaction with presynaptic GABA uptake carriers. We found in previous experiments (unpublished data) a Kin for GABA uptake by adult rat brain synaptosomes of (2) 10.1 ___ 3.8 IxM, quite in agreement with the 13 I~M value reported by Levi and Raiteri (28). If one looks at the average brain concentration reached by exogenous GABA in the 20-30 minutes period one finds levels of 0.2-0.3 IxM. However, it seems reasonable to assume a higher concentration in the extracellular space. Especially, taking in account that this material is newly arriving from the blood stream. We compared the amount of exogenous GABA accumulated into brain and brain synaptosomes with the level of the endogenous neurotransmitter, using data from the literature (29,30). This comparison shows (Table I) that the proportion of exogenous GABA accumulated in the nerve endings at 30 minutes after the injection (6.8%) is much higher than its counterpart for the brain tissue (0.014%). This indicates a factor 486 for a more efficient accumulation in the nerve endings. The reason behind this phenomenon is most likely the presence of high affinity transport systems almost exclusively in presynaptic GABA-ergic boutons.

These data and interpretation together with the indication that the exogenous GABA may accumulate with time in the synaptic vesicles pool (see above) go along with the experimental evidence that in GABA-ergic neurons in culture exogenous [3H]GABA is preferentially taken up in the neurotransmitter pool (31). In any case, this situation appears of potential utility in further attempts to administer GABA as an antiepileptic treatment. Referring to our starting point, it is possible that the simultaneous administration of GABA and phosphatidylserine (PS) can improve the synaptic accumulation of the exogenous neurotransmitter (32) by two mechanisms. Firstly, PS may facilitate the first passage of GABA from the i.p. site of injection to the blood stream, and indeed we have such an indication from previous experiments (33). This event would result in an increased [GABA]plasma, at least for a certain time span. Since brain influx of the neurotransmitter follows a law of the type V = K • Cplasma (see above), this would automatically result in an increased brain concentration of exogenous GABA. This would apply in particular to the extracellular space with a consequent great increase in GABA synaptic accumulation. Secondly, exogenous PS may further enhance in vivo GABA synaptic uptake (34,35). These events may drive the accumulation of the exogenous inhibitory neurotransmitter into the relevant nerve endings beyond the critical threshold for the occurrence of an antiepileptic effect (36,37).

ACKNOWLEDGMENTS This work has been supported by C.N.R. Target Project "Biotechnology and Bioinstrumentation." L.V. is the recipient of a research grant by Fidia Laboratories, Abano Terme.

REFERENCES 1. Van Gelder, N. M., and Elliot, K. A. C. 1958. Disposition of gamma-amino-butyric acid administered to mammals. J. Neurochem. 3:139-143. 2. Hespe, W., Roberts, E., and Prins, H. 1969. Autoradiographic investigation of the distribution of (14C)-GABA in tissues of normal and aminooxyacetic acid-treated mice. Brain Res. 14:663671. 3. Kuriyama, K., and Sze, P. Y. 1971. Blood-brain barrier to 3Hgamma-amino-butyric acid in normal and amino oxyacetic acidtreated animals. Neuropharmacology, 10:103-108. 4. Oldendorf, W. H. 1971. Brain uptake of radiolabeled amino acids, amines, and hexoses after arterial injection. Am. J. of Physiol. 221:1629-1639. 5. Levi, G., Amaldi P., and Morisi, G. 1972. Gamma-aminobutyric acid (GABA) uptake by the developing mouse brain in vivo. Brain Res. 41:435-451. 6. Toth, E., and Lajtha, A. 1981. Elevation of cerebral levels of

GABA Accumulation

7. 8.

9.

10.

11. 12. 13. 14. 15.

16.

17. 18. 19.

20.

21. 22.

nonessential amino acids in vivo by administration of large doses. Neurochem. Res. 6:1309-1317. Loscher, W. 1981. Effect of inhibitors of GABA aminotransferase on the metabolism of GABA in brain tissue and synaptosomal fractions. J. of Neuroehem. 36:1521-1527. Hawkins, J. E., and Sarett, L. H. 1957. On the efficacy of asparagine, glutamine, gamma-aminobutyric acid and 2-pirrolidinone in preventing chemically induced seizures in mice. Clinica Chimica Acta 2:481--484. Tower, D. B. 1976. GABA and seizures: clinical correlates in man. Pages 461--478, in E. Roberts, T. N. Chase and D. B. Tower (eds), GABA in Nervous System Function, Raven Press, New York. Floris, V., Morocutti, C., Oaggino, G., and Napoleone-Capra. 1962. L'azione degli acidi gamma-amino-butirricoe gamma-aminobeta-idrossi-butirrico sul traceiato elettroencefalografico dei soggetti normali e sui grafoelementi epilettici. Boll. Soc. Ital. Biol. Sper. 38:538--541. Toth, E., Lajtha, A., Sarhan, S. and Seller, N. 1983. Anticonvulsant effects of some inhibitory neurotransmitter amino acids. Neurochem. Res. 8:291-302. Remler, M. P., and Marcussen, W. H. 1981. Radiation controlled focal pharmacology in the therapy of experimental epilepsy. Epilepsia. 22:153-159. Remler, M. P., and Marcussen, W. H. 1983. A GABA-EEG test of the blood brain barrier near epileptic foci. Appl. Neurophysiol. 46:276-285. Frey, H.-H., and Loscher, W. i980. Cetyl OABA: effect on convulsant thresholds in mice and acute toxicity. Neuropharmacology 19:217--220. Bianchi, M., Deana, R., Quadro, G., Mourier, G., and Galzigna, L. 1983.4-amino butyric acid methyl ester hydrochloride, a precursor of 4-aminobutyric acid, Biochem. Pharmacol. 32:10931096. Galzigna, L., Bianchi, M., Bertazzon, A., Barthez, A., Ouadro, G., and Coletti-Previero, M. A. 1984. An N-protected gammaamino-butyric acid dipeptide with anticonvulsant action. J. of Neurochem. 42:1762-1766. Hesse, O. W., Shashoua, V. E., and Jacob, J. N. I985. Inhibitory effect of cholesteryl gamma-amino butyrate on evoked activity in rat hippocampal slices. Neuropharmacology 24:139-146. Hesse, G. W., Jacob, J. N., and Shashoua, V. E. 1988. Uptake in brain and neurophysiological activity of two lipid esters of gamma-amino-butyric acid. Neuropharmacology 27:63%640. Jacob, J. N., Hesse, G. W., and Shashoua, V. E. 1990. Synthesis, brain uptake and pharmacological properties of a glyceryl lipid containing GABA and the GABA-T inhibitor gamma-vinyl-GABA. J. Med. Chem. 33:733-736. Loeb, C., Benassi, E., Besio, G., Martini, M., and Tanganelli, P. 1982. Liposome entrapped GABA modifies behavioral and electrographic changes of penicillin-induced epileptic "activity. Neurology 32:1234-1238. Loeb, C., Besio, G., Mainardi, P., Scotto, P. E., Benassi, E., and Bo, 0. P. 1986. Liposome entrapped GABA inhibits isoniazid-induced epileptogenic activity in rats. Epilepsia 27:98-102. Cupello, A., Rapallino, M. V., and Besio, G., and Mainardi, P.

199

23. 24. 25. 26. 27. 28. 29. 30.

31.

32.

33.

34. 35. 36.

37.

1987. An electrophoretic method for the determination of the proportion of gamma-amino-butyric acid in a mixture of labeled neurotransmitter and its catabolites. Analyt. Biochem. 160:14-16. Cotman, C. W. and Matthews, D. A. t971. Synaptic plasma membranes from rat brain synaptosomes: isolation and partial characterization. Biochem. Biophys. Acta. 249:380-394. Cupello, A., and Hyden, H. 1975. Fractionalion of RNA from brain synaptosomes and cytoplasmic subcellular fractions. J. Neurochem. 25:399--406. Yunger, L. M., and Cramer, R. D. 1981. Measurement and correlation of partition coefficients of polar amino acids. Mol. Pharmacol. 20:602-608. Smith, O. R., Momma, S., Aoyagi, M. and Rapoport, S. I. 1987. Kinetics of neutral amino acid transport across the blood brain barrier. J. Neurochem. 49:1651-1658. Toth, J., and Lajtha, A. 1977. Rates of exchanges of free amino acids between plasma and brain in mice. Neurochem. Res. 2:149160. Levi, G., and Raiteri, M. 1971. Detectability of high and low affinity uptake system for GABA and glutamate in rat brain slices and synaptosomes. Life Sciences 12:81-88. Bohlen, P., Huot, S., and Palfreyman, M. G. 1979. The relationship between GABA concentrations in brain and cerebrospinal fluid. Brain Res. 167:297-305. Wood, J. D., Kurylo, E., and Lane, R. 1988. Gamma-aminobutyric acid release from synaptosomes prepared from rats treated with isonicotic acid hydrazide and gabaculine. J. Neurochem. 50:I839-t843. Gram, L., Larsson O. M., Johnsen, A. H. and Schousboe A. 1988. Effects of valproate, vigabatrin and amiuooxyacetic acid on release of endogenous and exogenous GABA from cultured neurons. Epilepsy Res. 2:87-95. Loeb, C., Marinari, U. M., Benassi, E., Besio, G., Cottalasso, D., Cupello, A., Maffini, M., Mainardi, P., Pronzato, M. A., and Scotto, P. A. 1988. Phosphatidilserine increases in vivo the synaptosomal uptake of exogenous GABA in rats. Exp. Neurol. 99:440-446. Loeb, C., Besio, G., Maffini, M., Berti, C., and Benassi, E. 1982. Distxibuzione tissutale del GABA e det GABA veicolato dai liposomi. Bollettino Lega Italiana contro l'Epilessia 37/38:1517. De Medio, G. E., Trovarelli, G., Hamberger, A., and Porcellati, G. 1980. Synaptosomal phospholipid pool in rabbit brain and its effect on GABA uptake. Neurochem. Res. 5:171-179. Chweh, A. Y., and Leslie, S. W. 1982. Phosphatidylserine enhancement of (3H)-gamma-amino-butyricacid uptake by rat whole brain synaptosomes. J. Neurochem. 38:691-695. Wood, J. D., Russel, M. P., and Kurylo, E. 1980. The gammaamino-butyrate content of nerve endings (synaptosomes) in mice after the intramuscular injection of gamma-amino-butyrate elevating agents: a possible role in anticonvulsant activity. J. Neurochem. 35:125-130. Wood, J. D., Kurylo, E., and Tsui, S. K. 1981. Interactions of di-n-propylacetate, gabaculine and aminooxyacetic acid: anticonvulsant activity and the gamma-amino-butyrate system. J. Neurochem. 37:1440-1447.