Adaptation of an enzymatic fluorescence assay for l -glutamic acid decarboxylase

ANALYTICAL

BIOCHEMISTRY

192,

78-81

(1991)

Adaptation of an Enzymatic Fluorescence Assay for L-Glutamic Acid Decarboxylase Rainer

Wolf1

and Harald

Max-Planck-Znstitute

Received

May

Klemisch

for Psychiatry,

Munich 40, Federal Republic of Germany

l&l990

The activity of L-glutamic acid decarboxylase (GAD) is commonly estimated by several radiometric methods, whereas a fluorimetric assay based on an enzymatic formation of NADPH as described by Y. Okada and C. Shimada [(1975) Brain Res. 98, 20%2061 has been given little attention in biochemical and pharmacological investigations. A simple modification of this assay is presented to permit rapid and sensitive GAD measurements in unpurified tissue homogenates. This method, employing a linear NADPH standard curve, is demonstrated to be a valid assay system for a pharmacolological approach using 3-mercaptopropionic acid. o lt~l Academic

Press,

Inc.

L-Glutamic acid decarboxylase (GAD,’ L-glutamate 1-carboxyl-lyase, EC 4.1.1.15) catalyzes a-decarboxylation of L-glutamate to form y-aminobutyric acid (GABA) and CO,. GAD is a highly specific enzyme that requires pyridoxal Y-phosphate (PLP) as a cosubstrate and is believed to be the rate-limiting step that normally determines the steady-state levels of the inhibitory neurotransmitter GABA in the nervous system of invertebrates and vertebrates (1). GAD activity has commonly been assayed by various radiometric procedures using 14C- or 3H-labeled glutamate as substrate (2). However, measuring ‘*CO, as a reaction product may overestimate GAD activity up to

1 To whom correspondence and reprint requests should be addressed at Max-Planck-Institute for Psychiatry, Kraepelinstrasse 2, D-8000 Munich 40, FRG. ’ Abbreviations used: GAD, L-glutamate l-carboxyl-lyase (glutamate decarboxylase); PLP, pyridoxal 5’-phosphate; GABA-T, 4aminobutyrate:2-oxoglutarate aminotransferase (GABA transaminase); SSA-DH, succinate-semialdehyde:NAD(P) oxidoreductase (succinate dehydrogenase); GABase, an enzyme mixture from Pseudomonas fluoresceas consisting of GABA-T and SSA-DH; AET, 2aminoethylisothironium bromide; MPA, 3-mercaptopropionic acid.

10% because of unspecific 14C0, production by alternate pathways of other decarboxylases (3); measuring [14C]or [3H]GABA as product requires separation of GABA and glutamate. The fluorescence-based GAD assay method described by Okada and Shimada (4) needs neither radioactively labeled material nor a separation step. However, the oil well technique used in this method to enable working with volumes in the range of 0.01~1 combined with the cycling step of Lowry and Passonneau (5) may be a potential source of failure. Moreover, because an enzyme reaction is employed as an indicator reaction, this procedure needs clear-cut validation for the actual pharmacological approach. Therefore, the aim of our study was to establish basic kinetic data with a NADPH-coupled GAD assay method, utilizing a simple assay procedure with a sensitivity sufficiently high to render the assay applicable to tests of the effects of drugs, e.g., the convulsant drug 3-mercaptopropionic acid (MPA), in crude tissue homogenates.

MATERIALS

AND

METHODS

Source of materials. 2-Aminoethylisothironium bromide (AET), Triton X-100, pyridoxal 5’-phosphate, Lglutamate, and Tris buffer were purchased from Sigma (Deissenhofen, FRG). a-Ketoglutarate, NADP (oxidized form), and GABase, an enzyme preparation from Pseudomonas fluorescens consisting of GABA-T (EC 2.6.1.19) and SSA-DH (EC 1.2.1.16) with a specific activity of about 4 U/mg protein (25°C GABA as substrate), were from Boehringer (Mannheim, FRG). All other substances were obtained from Merck (Darmstadt, FRG). Homogenate preparation. A 10% (w/v) homogenate of whole-brain tissue was prepared in ice-cold 0.1 M sodium phosphate buffer, containing 1 mM AET, 0.1% Triton X-100,20 PM PLP, adjusted to pH 7.0. The tissue samples were homogenized with an ultrasonic cell

78 All

Copyright 0 1991 rights of reproduction

0003-2697/91$3.00 by Academic Press, Inc. in any form reserved.

ENZYMATIC

ASSAY

FOR

L-GLUTAMIC

disruptor (Branson, Heusenstamm, FRG) and centrifuged at 5000g for 30 min at 4°C. Enzyme reaction. Using a conial1.5-ml test tube, lo~1 tissue homogenates in an appropriate dilution of 0.2, 1, 2, 2.5, 3, 4, or 5% (w/v) were mixed with 10 ~1 GADsubstrate solution. The composition of this reagent was 50 mM glutamate, 250 PM PLP, and 0.4% 2-mercaptoethanol in 0.1 M sodium phosphate buffer, adjusted to pH 7.0. These GAD reaction mixtures were allowed to stand at 38°C for 15, 30, 45, 60, 75, or 90 min, respectively. Then 10 ~1 0.25 N HCl was added to stop the reaction and the mixture was heated for 5 min at 100°C to destroy endogenous NADPH in the tissue. The tube was then cooled on ice. Indicator reaction. An aliquot of the terminated GAD reaction mixture was added to a solution of 50 ~1 of the GABA assay reagent, which consisted of 6.0 mM a-ketoglutarate, 0.1 mM NADP, 6.0 mM 2-mercaptoethanol, and 10 U GABase in 0.3 M Tris buffer, adjusted to pH 8.4. Because the end point of the indicator reaction was reached after 8 min, the samples were allowed to stand for 15 min at 38°C. Then ice-cold distilled water was added to give a total volume of 1 ml and the fluorescence was measured in a filter fluorimeter (LS-BB, Perkin-Elmer, Munich, FRG) with a wavelength of 340 nm for excitation and 460 nm for emission. Calculations. The activity of GAD is expressed as 1 pmol GABA formed by GAD per hour (pmol/h) under standard assay conditions. Maximum velocity (urnaX) is given in micromoles of GABA per hour and gram of brain tissue (pmollh . g). Since endogenous GABA is present in each homogenate sample, a blank was performed by adding the substrate, i.e., 50 mM glutamate, to the reaction mixture after the termination step of the GAD reaction. Thus, for each sample an individual blank value was subtracted. In the case of 1% (w/v) homogenates the blank value was about 300-400 pmol endogenous GABA/tube (0.1 mg tissue). RESULTS

AND

DISCUSSION

1. Sensitivity, Linearity, Indicator Reaction

and Reproducibility

GABA

GABA-T,PLP_ L-glutamate

+ ol-ketoglutarate

+ succinate Succinate

semialdehyde

+ NADP+

of the

semialdehyde s

+ NADPH

[la]

succinate + H+

[lb]

According to Eqs. [la] and [lb] the fluorescent reaction product NADPH is formed in an equimolar ratio with GABA; however, for use as an indicator reaction in

ACID

Activity

0

79

DECARBOXYLASE

Inmol GABA/hj

0.1

0.2

0.3

Brain tissue

0.4

0.5

0.6

lmgl

FIG. 1. The effect of the amount of GAD in brain tissue on the production of GABA. Incubation time was 30 min at 38°C. The assay mixture contained 50 mM glutamate per tube.

kinetic or pharmacological studies, it was necessary to demonstrate that the GABA/NADPH standard curve was affected neither by the reagents of the enzyme reaction (Eq. [2]) nor by the drug used in the pharmacological tests. The standard curve was performed in triplicate using 10,100,500,1000,5000, or 10,000 pmol GABA/test tube. The correlation coefficient was 0.998 and the interassay variance was less than 3.2%. In the presence of the GAD reagents there was no significant change in these assay parameters. MPA, when added in a concentration range from 10 pM to 10 mM, did not disturb the assay system. Thus, Eqs. [la] and [lb] can be considered a valid indicator system for the GAD activity assay. 2. Sensitivity, Linearity, Enzyme Reaction L-Glutamate

=

and Reproducibility

GABA + CO,

of the

PI

For accurate estimation of the activity of GAD it is important to measure the initial velocity of the enzyme reaction (Eq. [2]), i.e., to use that range of assay conditions when substrate degradation per time or product formation per time is constant. Thus, there are two criteria for the assay conditions, which were not examined for the previous NADPH-coupled GAD method: first, the linear relationship between the velocity of substrate turnover (e.g., increase in fluorescent NADPH) and the amount of enzyme (e.g., mg tissue) while the assay time is held constant; and second, the linear relationship between velocity and reaction time while the amount of enzyme is held constant. Figure 1 shows a linear increase in the amount of GABA formed per hour within a range of 0.02 to 0.5 mg brain tissue per tube. The relative coefficient of variation ranged from 10.4 to 4.4%. The assay time was 30

80

WOLF

Activity 1001

AND

KLEMISCH TABLE

Inmol GABA/hl Kinetic Method: Reference: Source:

80. 80.

K,, (mM)” K (/.db

0

20

40

Time FIG. 2. ,The effect of incubation GABA. The assay mixture contained glutamate per tube.

60

100

80

time at 38°C 0.25 mg brain

3. Effect of 3-Mercaptopropionic

on production of tissue and 50 mM

Acid

MPA has been reported to cause severe convulsions in rats after intraperitoneal administration (7). This effect has been related to the inhibition of neuronal L-glutamic acid decarboxylase by MPA, which has been observed in brain homogenates in both in vitro and in vivo tests (8,9). Figure 3 depicts the Lineweaver-Burk plot (10) of the kinetic data for the inhibition of GAD by MPA, studied l/V

l/(~mol/h-g) /d

YC

O.l0.08

-

0.06 0.04

-

0.02 0

0.05

l/lGlutamateI

of GAD

‘%O, (11) Mouse

“co, (7) Rat

‘TO, (9) Mouse

3.0-7.9 -

3 8.1’

0.7 1.8

NADPH This study Rat 24.2 76.8

’ For glutamate. ’ For MPA with regard to glutamate. ’ Calculated by the present authors.

[minl

min. Glutamate concentration was 50 mM; a concentration range between 5 and 200 mM glutamate has been tested and found to give reproducible results. Figure 2 depicts a linear increase in the amount of GABA formed per hour within a range of 15 to 90 min. The relative coefficient of variation ranged from 8.3 to 4.2%. The amount of brain tissue was 0.25 mg; glutamate concentration was 50 mM. The interassay variation for the enzyme reaction was about 18%, when an assay time of 30 min and 0.25 mg tissue were used.

0.121

Parameters

1

0.1 0.15 l/(mM)

0.2

FIG. 3. The effect of 0 (a, circles), 100 (b, triangles), 200 (c, stars), and 300 (d, squares) PM MPA with regard to glutamate on the production of GABA. Incubation time was 30 min at 38°C. The assay mixture contained 0.25 mg brain tissue and 5, 10, 50, 100, or 200 mM glutamate per tube, respectively.

at inhibitor concentrations [I] of 100, 200, and 300 PM MPA/tube, with glutamate concentrations varying between 5 and 200 mM glutamate/tube. The data suggest a competitive inhibition mode for MPA with regard to L-glutamate. The Michaelis-Menten constant, K,, for L-glutamate in the absence of MPA was 24.2 mM; maximum velocity, v,,, , was 78.2 pmol/h . g. The mean value of the inhibitor constant for MPA, Ki, was calculated to be 76.8 PM, using the appropriate intersection (-l/K,) of the abscissa on the Lineweaver-Burk plot: Kj = [I]/ (KJK,,, - l), where K, is the apparent K,,, in the presence of MPA at concentration [I]. Previous studies of enzyme preparations derived from mammalian brain homogenates using the carbonyl trapping method (7,9,11) have also revealed a competitive mode of inhibition for MPA. However, the values for K, and Ki are lower than our data obtained by the NADPH method (Table 1). These differences may be caused in part by the different assay methods used, but the main factor seemsto be the different procedures for homogenate/enzyme preparation from mammalian brain. A similar situation has been reported for bacterial GAD, purified from Escherichia coli, in which K,,, values of 2.6,4.5, and 27.0 mM have been estimated. The variance has been suggested to arise from different isolation and purification procedures (12). Employing mouse brain tissue, Susz et al. (11) have reported K,,, values for the crude enzyme (3 mM) at pH 6.4 that were lower than those observed after preparation with calcium phosphate elution (3.6-4.0 mM) or with Sephadex chromatography (7.9 mM), measured at pH 7.2. This is surprising, and the authors suggest that the purification steps employed may have produced conformational changes in the enzyme protein which resulted in a change in pH optimum and in a decreased affinity of the enzyme for the substrate. In contrast, since amino acids such as aspartate, malate, and glutarate and inorganic anions such as Cl- have been described as competitive inhibitors of GAD (3), then one should expect that without purification of the enzyme, the effective concentrations of such inhibitors should be higher, and therefore the value of K,,, for a crude tissue homogenate should be higher than that of the Km of an purified enzyme prepa-

ENZYMATIC

ASSAY

FOR

L-GLUTAMIC

ration, This latter view is supported by the present data when compared with previous reports (Table 1). Thus, further kinetic studies using one of the above-mentioned purification procedures (7,9,11) followed by the present NADPH-coupled GAD method may elucidate these variations in K,,, values. In all reports mentioned, the source of the mammalian brain enzyme was the “crude mitochondrial” fraction refined by different purification steps (e.g., centrifugation, chromatography). Therefore, the estimated value of K,,, represents only that for mitochondrial GAD, whereas the cytosolic or the membrane-bound GAD, which may be involved predominantly in neuronal activity and which may possess different kinetic characteristics (13-15), did not contribute to these reported Kmvalues (Table 1). Thus, the existence of multiple forms of GAD in brain (3,14,15) may have to be taken into consideration for such kinetic studies. In conclusion, this paper offers a modification of the GAD assay method of Okada and Shimada (4) which is useful for pharmacological tests in unpurified tissue. Two aspects, namely the GAD purification grade and the existence of multiple forms, may be of importance for estimation and interpretation of drug effects on the GAD-dependent GABAergic neurotransmission in mammalian brain. ACKNOWLEDGMENTS The authors and Physiological

thank Professor Biochemistry,

E. Holler, Regensburg,

Institute for Biophysics for helpful discussions.

ACID

81

DECARBOXYLASE

This work was supported und Technologie (BMFT),

by the Bundesministerium Bonn, FRG.

fiir

Forschung

REFERENCES 1. Kravitz, E. A. (1967) in The Neurosciences (Quarton, G. C., Melnechnuk, T., and Schmitt, F. O., Eds.), pp. 433-444, Rockefeller Univ. Press, New York. 2. Wu, J.-Y., Su, Y. Y. T., Lam, D. M. K., Schousboe, A., Chude, 0. (1981) in Research Methods in Neurochemistry (Marks, N., and Rodnight, R., Eds.), Vol. 5, pp. 129-177, Plenum, New York. 3. Martin, D. L. (1986) in Neuromethods (Boulton, A. A., Baker, G. B., and Yu, P. H., Eds.), Vol. 1, pp. 361-388, Humana, Clifton, NJ. 4. Okada, Y., and Shimada, C. (1975) Brain Res. 98,202-206. 5. Lowry, 0. H., and Passonneau, J. V. (1972) A Flexible System of Enzymatic Analysis, Academic Press, New York. 6. Storm-Mathisen, J., and Fonnum, F. (1971) J. Neurochem. 18, 1105-1111. 7. Lamar, C. (1970) J. Neurochem. 17, 165-170. A., Fonnum, F., Malthie-Sorenssen, a. Karlsson, Mathisen, J. (1974) Biochem. Pharmacol. 23, 9. Wu, J.-Y., 10. Lineweaver, 658-666.

D., and 3053-3061.

and Roberts, E. (1974) J. Neurochem. H., and Burk, D. (1934) J. Amer.

11. Susz, J. P., Haber, 2870-2876.

B., and

Roberts,

E. (1966)

Storm-

23, 759-767. Chem. Sot. 56, Biochemistry

5,

12. Shukuya, R., and Schwert, G. W. (1960) J. Biol. Chem. 235, 1649-1652. 13. Spink, D. C., Porter, T. G., Wu, S. J., and Martin, D. L. (1985) Biochem. J. 23 1,695-703. L. A., and Wu, J.-Y. (1985) J. Neurochem. 44,957-965. 14. Denner, 15. Spink, D. C., Wu, 40,1113-1119.

S. J., and Martin,

D. L. (1983)

J. Neurochem.