Kinetics and inhibition of glutamate carboxypeptidase II usinga microplate assay

Camilo Rojas, Scott T. Frazier, Juliet Flanary, and Barbara S. Slusher*

Guilford Pharmaceuticals Inc., 6611 Tributary Street, Baltimore, MD 21224, USA

Received 11 April 2002

Abstract

Glutamate carboxypeptidase II (GCPII or prostate-specific membrane antigen or NAALADase) is an enzyme that catalyzes the

hydrolysis of the neuropeptide N-acetylaspartylglutamate (NAAG) to N-acetylaspartate (NAA) and glutamate (G). Inhibitors of

GCPII provide neuroprotection in a variety of animal models of central nervous system disorders. Neuroprotection is probably the

result of increased NAAG concentrations and decreased levels of excess toxic glutamate. Consequently, GCPII inhibitors could be

useful therapeutic agents where increased glutamate levels are the result of increased GCPII activity. Current GCPII in vitro activity

assays are cumbersome or have limited sensitivity. In this report we describe a microplate assay to study GCPII inhibition that is

most sensitive, efficient, and generates little waste. GCPII turnover number (kcatÞ was 4s�1 and the binding constant (KmÞ for NAAG

and GCPII was 130 nM. The apparent association rate constant for GCPII and NAAG (kcat=KmÞ was 3� 107M�1 s�1. Inhibition

studies with the GCPII inhibitor 2-(phosphonomethyl)pentanedioic acid (2-PMPA) demonstrated competitive inhibition with a

Ki ¼ 0:2 nM.

� 2002 Elsevier Science (USA). All rights reserved.

Glutamate carboxypeptidase (GCPII)1 is a metallo-

peptidase that hydrolyzes the neuropeptide N-acetyl-

aspartylglutamate (NAAG) to glutamate (G) and

N-acetylaspartate (NAA) [1]. GCPII was initially

referred as N-acetylated-a-linked-acidic dipeptidase,

NAALA dipeptidase, or NAALADase [1,2]. Later on it

was shown that prostate-specific membrane antigen(PSMA), a prostate cancer marker that had been

reported earlier [3,4], exhibited the same substrate

and pharmacological characteristics as the brain dipep-

tidase [5]. Further, PSMA cDNA is 86% identical to

NAALADase rat brain cDNA, demonstrating that the

two enzymes are closely related molecular species [5].

The official name given to the enzyme is glutamate

carboxypeptidase II (GCPII, EC 3.4.17.21).

NAAG is thought to be a storage form of synaptic

glutamate. It activates mGluR3 and acts as a partial

NMDA receptor agonist, interactions associated with

neuroprotection [6]. On the other hand, excessive

glutamate levels are widely associated with neuronal

injury in many central nervous disorders [7]. Selective

NAALADase inhibition is neuroprotective in both invivo and in vitro models of ischemic brain injury [8]. The

mechanism of brain neuroprotection could be through

GCPII inhibition that in turn brings decreased gluta-

mate and increased NAAG levels. GCPII inhibitors

could be beneficial in the treatment of diseases where

excess glutamate levels are the result of increased GCPII

activity. Currently, GCPII is a pharmacological target

being pursued in the clinic for the regulation of gluta-mate and it is the focus of study of several academic and

industrial laboratories.

Traditionally GCPII activity in vitro is monitored

through the hydrolysis [3H]NAAG to NAA and

[3H]Glu. Even though this assay is most sensitive for

following glutamate production, it is cumbersome and

time consuming [1,9]. We report the development of a

microplate GCPII radioactivity-based assay that is

Analytical Biochemistry 310 (2002) 50–54

www.academicpress.com

ANALYTICAL

BIOCHEMISTRY

* Corresponding author. Fax: 1-410-631-6802.

E-mail address: slusherb@guilfordpharm.com (B.S. Slusher).1 Abbreviations used: GCPII, glutamate carboxypeptidase; NAAG,

N-acetylaspartylglutamate; G, glutamate; NAA, N-acetylaspartate;

PSMA, prostate-specific membrane antigen; 2-PMPA, 2-(phospho-

nomethyl)pentanedioic acid.

0003-2697/02/$ - see front matter � 2002 Elsevier Science (USA). All rights reserved.

PII: S0003 -2697 (02 )00286-5

sensitive, efficient, and generates little waste. The newprotocol was used to determine the steady-state pa-

rameters for the GCPII-catalyzed hydrolysis of NAAG

and to study the inhibition of the enzyme by 2-(phos-

phonomethyl)pentanedioic acid (2-PMPA).

Materials and methods

Reagents

NAAG and NAA [3H]G were obtained from Bachem

AG (Bubendorf, Switzerland) and Perkin–Elmer (Bos-

ton, MA), respectively. Human recombinant GCPII was

kindly provided by Dr. Jan Konvalinka from the Insti-

tute of Organic Chemistry and Biochemistry of the

Academy of Sciences of the Czech Republic. The detailsof the cloning and purification have been published [10].

2-(Phosphonomethyl)pentanedioic acid was synthesized

by the Medicinal Chemistry Department at Guilford

Pharmaceuticals [11]. The 96-well spin column was ob-

tained from Harvard Bioscience (Holliston, MA).

Enzyme activity assay

The radioactivity-based assay [1,9] served as a point of

departure for the new protocol. The reaction mixture

contained NAA[3H]G (30 nM, 4.3 nmol/lCi) and GCPII

(20 or 40 pM), in Tris–HCl (pH 7.4, 40mM) containing

CoCl2 (1mM), in a total volume of 50 ll. The reaction

was carried out at 37 �C for 15 min and stopped with ice-

cold sodium phosphate buffer (pH 7.5, 0.1M, and 50 ll).Blanks were obtained by incubating the reaction mixturein the presence of 2-PMPA (10 lM), a selective and po-

tent inhibitor of GCPII [11]. An aliquot of the reaction

mixture (90 ll) was transferred to a 96-well spin column

containing AG1X8 ion-exchange resin; the plate was

centrifuged at 900 rpm for 3–5min using a Beckman GS-

6R centrifuge equipped with a PTS-2000 rotor.

NAA[3H]G bound to the resin and [3H]G eluted in the

flowthrough. The columns were then washed twice withformate (1M, 90 ll) to ensure complete elution of [3H]G.

The flowthrough and the washes were collected in a deep

96-well block; an aliquot, 200 ll out of a total volume of

270 ll, was transferred to a solid scintillator-coated 96-

well plate (Packard) and dried to completion. The ra-

dioactivity corresponding to [3H]G was determined with

a scintillation counter (Topcount NXT, Packard,

counting efficiency 40%). Assay points for each experi-ment were mostly the average of eight determinations.

Data analysis

Km; Vmax; Km=Vmax and the error determinations of

these ratios were obtained by using Leonora [12], a

computer program that carries out a least-squares fit to

the Michaelis–Menten equation: v ¼ Vmax½S�=ðKm þ ½S�Þ.

Results

Assay protocol

The radioactivity-based assay [1,9] was miniaturized

to a 96-well format; the reaction volume was reduced 20-

fold from 1.0 ml to 50 ll. Small volumes were added

with an automatic ‘‘multidrop’’ pipettor for quick and

accurate volume dispensation and the hand-assembledPasteur pipette columns were replaced by a 96-well mi-

nicolumn assembly that eliminated radioactive glass

waste. The solid scintillator coated 96-well plates elimi-

nated liquid scintillation and related radioactive waste.

The counting time was also reduced because the top

count used 6 probes simultaneously in contrast to the

more traditional single probe scintillation counter.

Steady-state kinetics of GCPII

NAAG hydrolysis was dependent on GCPII con-

centration up to 160 pM (Fig. 1A) and on time of in-

cubation up to 50min (Fig. 1B). The reaction obeyed

Michaelis–Menten kinetics (Fig. 1C). The binding con-

stant (KmÞ for NAAG was 130 (40) nM, a value that

is in the range of what is reported in the literature:87–540 nM [1,13]. GCPII turnover (kcatÞ was 4 (2) s�1.

The apparent second-order rate constant for the reac-

tion of free GCPII with free NAAG, kcat=Km, was 3�107 ð1� 107ÞM�1 s�1. Values in parentheses are the

standard deviations for 5 independent determinations.

Inhibition of GCPII by 2-PMPA

GCPII inhibition by 2-PMPA exhibited competitive

inhibition (Fig. 2A); as the concentration of inhibitor

was increased from 0 to 10 nM, the apparent Michaelis

constant (Km appÞ increased. Even though the rates of

hydrolysis at 4 lM NAAG at different 2-PMPA con-

centrations was lower than Vmax in the absence of in-

hibitor, the trends of the curves suggested that at larger

NAAG concentrations the rates would be the same.Further, the reciprocal plot of the data (Fig. 2B) showed

the same intercept on the y-axis (1/VmaxÞ, thus confirm-

ing the same Vmax value under different 2-PMPA con-

centrations. The secondary plot of the slopes for each

line on the reciprocal plot (Km app=VmaxÞ vs inhibitor

concentration gave a straight line with an intercept on

the x-axis equal to �Ki. The value of Ki determined from

this secondary plot was 0.2 nM (Fig. 2C).

Discussion

The radioactivity-based microplate assay we report

here includes several improvements of the standard ra-

dioactivity-based assay making it much more expedi-

C. Rojas et al. / Analytical Biochemistry 310 (2002) 50–54 51

tious and efficient. The assay volume was reduced 20-

fold compared to the previous radioactivity-based assay

so that rate vs substrate concentration profiles ( in-

hibitor) are more practical to determine. Work-up steps

were developed to accommodate the new format; one

major modification was the use of solid scintillator

technology that significantly reduced radioactive waste.

It is now possible to monitor saturation of GCPII withNAAG at micromolar concentrations without high

levels of radioactivity. We used the new protocol to

determine the steady-state kinetic parameters of the

hydrolysis of NAAG catalyzed by GCPII and to de-

termine the equilibrium constant for the binding of

GCPII and 2-PMPA. The modified procedure was also

used successfully to determine GCPII activity in sciatic

nerve preparations (data not shown) and it could po-tentially be used to monitor GCPII activity in other

enzyme sources like prostate or brain membranes.

To our knowledge, this is the first time the turnover

number (kcatÞ and consequently the catalytic efficiency

(kcat=KmÞ values for GCPII are reported in the literature.

The turnover number for GCPII (4 s�1Þ is low when

compared to those for other enzymes like carbonic an-

hydrase (600,000 s�1Þ [14]. On the other hand, it is not

unusual; it is in the same order of magnitude as the

turnover number reported for DNA polymerase(15s�1Þ, tryptophan synthetase (2s�1Þ and lysozyme

(0:5s�1Þ [14]. NAAG exhibited tight binding to GCPII

(Km ¼ 130 nM) in accordance with previous reports

where Km is in the range 87–540 nM [1,13]. The Km ex-

hibits variability because it is in the range where the

reaction rate is linear with respect to substrate concen-

tration where small differences in substrate concentra-

tion translate into significant changes in rate thatcorrespondingly affect the Km determination. GCPII

exhibits a high catalytic efficiency (kcat=Km ¼ 3�

(A) (B)

(C)

Fig. 1. Dependence of rate of NAAG hydrolysis on (A) GCPII concentration, (B) time of incubation, and (C) NAAG concentration. Unless

otherwise specified, NAA [3H]G (30 pM, 4.3 nmol/lCi) was incubated with GCPII (40 pM) in Tris–HCl buffer (pH 7.4, 40 mM) containing CoCl2(1mM) in a final volume of 50ll. The reaction was carried out for 15min at 37 �C and terminated by the addition of ice-cold sodium phosphate (pH

7.4, 0.1M, and 50 ll). An aliquot (90ll) of the reaction mixture was added to an AG1X8 ion-exchange resin to capture NAA [3H]G; [3H]G product

was eluted with 1M formate and the radioactivity measured with a scintillation counter. The data are the average of 5 independent experiments.

Error bars correspond to the standard deviation for each determination. Signal to blank ratios at substrate concentrations around Km and Vmax values

were 40- and 3-fold, respectively. Specific counts per minute (total counts)blank counts) around Km and Vmax were about 1200 and 3000,

respectively.

52 C. Rojas et al. / Analytical Biochemistry 310 (2002) 50–54

107M�1 s�1Þ, similar to the catalytic efficiency of car-

bonic anhydrase [15]. In short, GCPII exhibits high

catalytic efficiency despite low turnover rate due to tight

binding of NAAG to GCPII.

Equilibrium constants for the binding of enzyme and

inhibitor (KiÞ are more rigorously determined from rate

vs substrate concentration profiles in the presence of

various fixed inhibitor concentrations [16]. The Ki for 2-PMPA we report here (0.2 nM) is similar to the Ki value

reported earlier [11]. The earlier value was inferred from

the IC50, the inhibitor concentration required to inhibit

50% of enzyme activity at a specific substrate concen-

tration. The Ki and the IC50 are similar for a competitive

inhibitor when the IC50 is determined at substrate con-

centrations well below the Km [17].

The Ki for 2-PMPA reported earlier was carried outwith a brain membrane preparation as GCPII source

[11]; the Ki we report here was obtained with the purified

extracellular portion of human recombinant GCPII

comprising amino acids 44–750 [10]. Even though re-

combinant GCPII does not contain the transmembrane

domain and it could exhibit differences in glycosylation

when compared to native GCPII, the Km for NAAG and

Ki for 2-PMPA are the same when using the two GCPII

sources, suggesting that the recombinant enzyme is a

reasonable approximation to the native enzyme.

The inhibition of GCPII by 2-PMPA (Fig. 2) exhib-

ited the hallmarks of competitive inhibition: no change

in Vmax and increase in Km app as the concentration of 2-

PMPA was increased. The rate constant of association

for GCPII and 2-PMPA (konÞ was reported recently as3� 107M�1 s�1 [18]. Interestingly, this value is the same

as kcat=Km for GCPII and NAAG determined from

steady-state kinetics experiments. This ratio is the ap-

parent second-order rate constant for the reaction rate

expression between substrate and enzyme and it is also a

measure of specificity for competing molecules [15]. If

we take kcat=Km as the rate of association for GCPII and

NAAG (k1Þ we can estimate the rate constant of disso-ciation for NAAG (k�1Þ from the equation for Km ob-

tained from the Michaelis–Menten kinetics mechanism:

Km ¼ ðk�1 þ kcatÞ=k1 [14]. Solving for k�1 and substi-

tuting Km; kcat and k1 for the corresponding values no-

ted above give k�1 0s�1. Even though this result is an

approximation, it clearly suggests that most of the

NAAG that binds GCPII goes on to form substrate.

Fig. 2. GCPII Inhibition by 2-PMPA. Various concentrations of NAA [3H]G (4.3 nmol/lCi) in the presence of different fixed concentrations of

2-PMPA were incubated with GCPII (20 pM). Reaction conditions were the same as those outlined in the legend for Fig. 1. The data illustrate the

results of a representative experiment. (A) Plot of rate of NAAG hydrolysis vs NAAG concentration in the presence of different fixed concentrations

of 2-PMPA. (B) Reciprocal plot of rate data to illustrate competitive inhibition. Vmax and Km were obtained by a least-squares fit to the Michaelis–

Menten equation (Materials and methods). (C) Secondary plot: slope (Km app=VmaxÞ of each reciprocal plot in (B) vs the corresponding 2-PMPA

concentration. The intercept on the x-axis is �Ki ¼ �0:2 nM.

C. Rojas et al. / Analytical Biochemistry 310 (2002) 50–54 53

Fig. 3 illustrates the similarities and differences between

NAAG and 2-PMPA when encountering GCPII as

suggested by the kinetic experiments: GCPII equally

recognizes both NAAG and 2- PMPA. However, in the

case of NAAG, the enzyme substrate complex goes on

to form products and free enzyme (kcat ¼ 4s�1Þ; very

little dissociates back to substrate. Recycled enzyme has

an equal probability to bind NAAG or 2-PMPA. In thecase of 2-PMPA, the rate constant of dissociation to free

enzyme and inhibitor (koffÞ is 0.01 s�1 [18], a rate ap-

proximately 400 times slower than the catalytic rate. As

a result, 2-PMPA effectively ‘‘locks’’ the enzyme in a

noncatalytic mode.

In summary, we used a microplate assay to determine

kcat; Km, and kcat=Km for the hydrolysis of NAAG cat-

alyzed by GCPII and to determine the Ki for the inhi-bition of GCPII by 2-PMPA. The results provide a

kinetic insight on the nature of catalysis by GCPII and

on the inhibition of the enzyme by 2-PMPA.

References

[1] M.B. Robinson, R.D. Blakely, R. Couto, J.T. Coyle, Hydrolysis

of the brain dipeptide N-acetyl-LL-aspartyl-LL-glutamate: identifica-

tion and characterization of a novel N-acetylated-linked acidic

dipeptidase activity from rat brain, J. Biol. Chem. 262 (1987)

14498–14506.

[2] B. Stauch Slusher, M.B. Robinson, G. Tsai, M.L. Simmons, S.S.

Richards, J.T. Coyle, Rat brain N-acetylated a-linked acidic

dipeptidase activity, J. Biol. Chem. 265 (1990) 21297–21301.

[3] J.S. Horoszewicz, E. Kawinski, G.P. Murphy, Monoclonal

antibodies to a new antigenic marker in epithelial prostatic cells

and serum of prostatic cancer patients, Anticancer Res. 7 (1987)

927–935.

[4] R.S. Israeli, C.T. Powell, W.R. Fair, W.D.W. Heston, Molecular

cloning of a complementary DNA encoding a prostate-specific

membrane antigen, Cancer Res. 53 (1993) 227–230.

[5] R.E. Carter, A.R. Feldman, J.T. Coyle, Prostate-specific mem-

brane antigen is a hydrolase with substrate and pharmacologic

characteristics of a neuropeptidase, Proc. Natl. Acad. Sci. USA 93

(1996) 749–753.

[6] J.H. Neale, T. Bzdega, B. Wroblewska, N-Acetylaspartylgluta-

mate: the most abundant peptide neurotransmitter in the

mammalian central nervous system, J. Neurochem. 75 (2000)

443–452.

[7] B.S. Meldrum, Protection against ischaemic neuronal damage by

drugs acting on excitatory neurotransmission, Cerebrovasc. Brain

Metab. Rev. 2 (1990) 27–57.

[8] B.S. Slusher, J.J. Vornov, A.G. Thomas, P.D. Hurn, I. Harukuni,

A. Bhardwaj, R.J. Traystman, M.B. Robinson, P. Britton, X.-

C.M. Lu, F.C. Tortella, K.M. Wozniak, M. Yudkoff, B.M. Potter,

P.F. Jackson, Selective inhibition of NAALADase, which con-

verts NAAG to glutamate, reduces ischemic brain injury, Nat.

Med. 5 (1999) 1396–1402.

[9] C. Tiffany, B.S. Slusher, Measurement of glutamate carboxypep-

tidase II (NAALADase) enzyme activity by the hydrolysis of

[3H]N-acetylaspartyl glutamate (NAAG), Curr. Protein Pharm.

3.10.1–3.10.12 (Suppl. 15) (2001).

[10] C. Barinka, M. Rinnov�aa, P. S�aacha, C. Rojas, P. Majer, B.S.

Slusher, J. Konvalinka, Substrate specificity, inhibition and

enzymological analysis of recombinant human glutamate car-

boxypeptidase II, J. Neurochem. 80 (2002) 477–487.

[11] P.F. Jackson, D.C. Cole, B.S. Slusher, S.L. Stetz, L.E. Ross, B.A.

Donzanti, D.A. Trainor, Design, synthesis, and biological activity

of a potent inhibitor of the neuropeptidase N-acetylated a-linkedacidic dipeptidase, J. Med. Chem. 39 (1996) 619–622.

[12] A. Cornish-Bowden, Analysis of Enzyme Kinetic Data, Oxford

University Press, Oxford, 1995.

[13] R. Luthi-Carter, A.K. Barczak, H. Speno, J.T. Coyle, Molecular

characterization of human brain, J. Pharmacol. Exp. Ther. 286

(1998) 1020–1025.

[14] L. Stryer, Biochemistry, third ed., W.H. Freeman and Company,

New York, 1988.

[15] A. Fersht, Structure and mechanism in protein science. A guide to

enzyme catalysis and protein folding, W.H. Freeman and Com-

pany, New York, 1999.

[16] I.H. Segel, Enzyme Kinetics. Behavior and Analysis of Rapid

Equilibrium and Steady-State Enzyme Systems, Wiley, New York,

1975.

[17] Y.-C. Cheng, W.H. Prusoff, Relationship between the inhibition

constant (KiÞ and the concentration of inhibitor which causes 50%

inhibition (I50Þ of an enzymatic reaction, Biochem. Pharm. 22

(1973) 3099–3108.

[18] C.W. Tiffany, N.S. Cai, C. Rojas, B.S. Slusher, Binding of

the glutamate carboxypeptidase II (NAALADase) inhibitor

2-PMPA to rat brain membranes, Eur. J. Pharm. 427 (2001)

91–96.

Fig. 3. Comparison of rate constants for NAAG and 2-PMPA in their reaction with GCPII.

54 C. Rojas et al. / Analytical Biochemistry 310 (2002) 50–54