Enkephalin convertase in the rat spinal cord

Neuropeptides8: 367-376, 1986

ENXEPHALIN CONVERTASE

IN THE RAT SPINAL CORD

Jerzy Silberring, Wzadysraw

Lason, Barbara Przewlocka and

Ryszard Przewzocki Polish Academy of Sciences, Institute of Pharmacology, 12

Smetna Street, 31-343

Krakow, Poland

(Reprint requests

to JS) ABSTRACT 'H-Guanidinoethylmercaptosuccinic

acid

('H-GEMSA) a very

selective inhibitor of enkephalin convertase, crude rat spinal cord homogenates

binds to the

saturably, reversibly

and

with high affinity. Scatchard analysis revealed two classes of binding sites with KD: 4.5 nM and 215 nM. A plot of dissociation experiment was nonlinear with the T min, respectively.

l/2:

2 min and 6

'H-GEMSA binding sites are evenly distrib-

uted throughout the rat spinal cord and their high density might suggest a physiological

significance

of enkephalin con-

vertase in that tissue. INTRODUCTION Enkephalin convertase

(EC 3.4.17.10), a carboxypeptidase

B-like enzyme, was er and Snyder

recently purified and characterized by Frick2+ (1). That Co -stimulated enzyme was localized

mainly in anterior pituitary, brain

gra-

(1,3) and, in smaller amounts, in different peripheral

nules

tissues

(2). As suggested by Lynch et al. (4), enkephalin con-

vertase is probably responsible

for the enkephalin processing

from a larger precursor. 367

NEURO.

(2), adrenal chromaffin

D

Despite a detailed characterization

of enkephalin converta-

se in the brain, little is known about this enzyme in the spinal cord. This is of particular

interest, as that tissue contains

large amounts of enkephalins,

and the spinal enkephalin

system

seems to play an essential role in the pain transmission. In our studies we characterized

enkephalin convertase

the rat spinal cord using the tritiated enkephalin Phe-Ala-Arg

OH-Bz-FAR),

in

analog 3H-Bz-

and the tritiated, highly specific in-

hibitor 'H-quanidinoethylmercaptosuccinic

acid

MATERIALS AND METHODS All rats were 300-350 g Wistar males. 3H-GEMSA and 'H-Bz-Phe-Ala-Arq

(30 Ci/mmol) were purchased from the New

England Nuclear. GEMSA was bought from Calbiochem, Crude homogenates homogenization

(89 Ci/mmol) La Jolla, USA.

of the spinal cord tissue were prepared by

with Polytron, Brinkman

(20 s, setting 3) in 20

volumes of ice-cold 100 mM NaAc, pH 5.7. Since we determined the total convertase,

thus the homogenate was not further proces-

sed. The protein content was determined according to Lowry et al. (5). The radiometric

carboxypeptidase

assay was performed as

described by Stack et al. (6). A 3H-GEMSA binding assay was performed in duplicates

at 4'C as follows: 4.7 nM of tritiated

ligand were incubated together with 200 ~1 of the tissue homogenate

(diluted 1 : 20 w/v) and with increasing concentrations

of the unlabeled GEMSA

(from 0 to 5 x 10-7M) . The total in-

cubation volume was 300 ~1 and all the substances were dissolved in 100 mM NaAc, pH 5.7. After 40 min of incubation at 4OC the samples were filtered through polyethyleneimine GF/B filters

100 mM ?JaAc, pH 5.7. The filters were equilibrated scintillation

- presoaked

(Millipore), followed by three 2-ml washings with in a Bray

cocktail for a minimum of 24 h and counted in a

368

Beckman liquid scintillation

counter. Each sample was measured

for at least 5 min, with the counting error smaller than 2%. A specific binding was calculated after subtraction of a nonspecific binding

("infinite" concentration

of the unlabeled GEMSA).

The evaluation of binding parameters was performed with the aid of a "LIGAND" program microcomputer

(7) modified for the Apple II plus

by M.H.Teicher

(BCTIC/MED-58).

RESULTS A displacement

curve for the 'H-GEMSA binding to the rat

spinal cord homogenates

is shown in Fig. 1.

%Bound

i

e

7

6 -loglGEMSA.MI

Figure 1. Displacement

curve of 'H-GEMSA binding to crude homogenates of lumbar part of the rat spinal cord. Each point represents the mean value + SEM (n=13).

This binding is a saturable process with IC50 of 13.0 nM. The binding was always smaller than 8% of the specific

nonspecific

binding. The Scatchard analysis

(Fig. 2) revealed two classes of

binding sites, with KD equal to 4.5 + 0.11 nM and 215 + 51 nM, respectively. That four-parameter model was significantly better than the other models, as tested by the F-test and the residual 369

B/F

1

2

3 Bound

lnM1

Figure 2. Scatchard plot, obtained from the data, presented in Fig. 1. Each point represents the mean value

variance, either. Appropriate

(n=13).

values of Bmax were: 530 + 20

fmol/mg protein and 1327 + 154 fmol/mg protein, respectively. The inhibition constants Ki were calculated according to the Chang-Prusoff

formula

(8) and these parameters equalled: 6.4

nM and 12.7 nM. Kinetics of the 'H-GEMSA binding reached equilibrium 30-40 min at 4'C, as was shown in Fig. 3A. Association

after

rate

constants were calculated by an initial velocities method and -1 -1 -1 equalled to 0.0093 nM min and 0.0037 nM-'min . The data from Fig. 3A were replotted, assuming a pseudo-first action

order re-

(9), as shown in Fig. 3B. The obtained plot exibits non-

linearity, suggesting again two classes of binding sites. The calculated k,, values, compared with those obtained from Scatchard analysis were in close agreement, plied model

(Tab. 1).

370

independent of the ap-

Figure 3(A): Association

of 3H-GEMSA to spinal cord homogenates.

(B): Pseudo-first order kinetics of "H-GEMSA binding. Data replotted from Fig. 3A.

Table I - Comparison of physico-chemical

parameters,

obtained

by different methods

K' D [nM]

K*[nM] D

k*

+I

-1 -1 [nM min ]

site 1

site 2

k-l

site 1

Scatchard

4.46

215.0

0.025

0.002

0.13

Pseudo-order

3.67

116.7

0.030

0.003

-

Dissociation

-

1 min

1

site 2

0.11

- was calculated, assuming: 4.7 nM 'H-GEMSA and B *k +1 max estimated from Scatchard plot.

371

-1

0.64

0.35

A dissociation

2

experiment

Figure 4. Dissociation homogenates.

1L

10

6

(Fig. 4) presented second-order

18

22

26 ml"

of 'H-GEMSA from the spinal cord

The ligand was displaced after the indicated time

intervals with an "infinite" concentration

of the unlabeled

GEMSA. The nonspecific binding was subtracted kinetics with half-lives:

from each point.

2 min, and 6 min for each component.

The comparison od physico-chemical

parameters,

obtained in

different ways, is presented in Table 1. In order to find out whether an upward concavity of the Scatchard plot was caused by a negative cooperativity

effect, we also performed dissocia-

tion experiments with dilution alone, as well as dilution plus the unlabeled GEMSA

(data not shown), according to DeMeyts

Both the obtained curves were curvilinear,

(IO).

exibiting similar

shapes. The distribution

of the 3H-GEMSA binding activity- in the

rat spinal cord is presented in Table 2. During that experiment we found a nearly equal distribution

of enkephalin convertase

along the.spinal cord. We also compared the 3H-GEMSA binding

372

Table II - Distribution

of 'H-GEMSA binding sites in the rat

spinal cord*

Ventral

Dorsal

Cervical

141.6

110.0

Thoracic

159.1

170.9

Lumbar

194.3

163.8

*Values expressed in fmoles 3H-GEMSA per mg of protein. to the lumbar part of the spinal cord of the rats subjected to chronic pain

(local inflammation of the hind limb) and we found a 30% increase in K1 and 44% increase in K 2 in arthritic rats, D D while the remaining parameters were unchanged. On the other hand, the enzymatic activity measured with 3H-Bz-FAR as a substrate was unchanged

in either group.

DISCUSSION Physico-chemical crude homogenates

parameters of the 'H-GEMSA binding in

of the rat spinal cord, estimated by different

methods, are in a close agreement. The dissociation

constant

KA for the higher affinity site, as well as the reaction rate constant k,, and IC50 are similar to those obtained by Strittmatter et al. (2) for the soluble fraction of the rat brain membranes.

The above parameters obtained in our study suggest

that 'H-GEMSA was bound to enkephalin convertase, whose presence in the spinal cord was previously determined by autoradiography (4). The Scatchard analysis reveals two subpopulations ing sites. This effect may be explained

373

of bind-

in several ways:

(i).

the presence of two classes of binding sites; of negative cooperativity

(ii) the existence

among binding sites;

between soluble and membrane-bound

fractions:

(iii) differences (iiii) the pres-

ence of two distinct enzymes, binding specifically

'H-GEMSA.

The results obtained after application of an unlabeled tor for preparing the displacement first possibility,

inhibi-

curve are in favour of the

thus we were able to extend the concentra-

tion range up to the micromolar

level, which permitter a lower

affinity site to be revealed. In addition, both the association and dissociation

experiments

exhibited a nonlinear relation-

ships indicating that the 3H-GEMSA binding was a more complex' reaction. The results obtained from the dissociation formed according to DeMeyts negative cooperativity

experiment, per-

(IO), exclude the existence of a

phenomenon.

It is also quite unlikely

that the two components may account for soluble and membranebound fractions, as the second Kg is too high in relation to that observed by other investigators

in brain homogenates

(2).

It might be interesting to assume that we deal with two different enzymes which can be readily inhibited by GEMSA. Since GEMSA is a highly specific inhibitor of enkephalin convertase, its K D for other carboxypeptidase of a micromolar

B-like enzymes are

range; thus, the influence of the second enzyme

is less likely, according to the present knowledge. Autoradiographic distribution

studies showed that, in general, the

of 'H-GEMSA binding corresponds

immunocytochemical

closely to an

localization of enkephalin-containing

neurons

(4), though it was not observed in the pituitary. As the GEMSA binding sites evaluated in our study are rather evenly distributed in the rat spinal cord, this observation

suggest

the lack of a close correlation between enkephalins

and the

enzyme content in the tissue. Moreover, the chronic pain, which enhances the enkephalin

level in the rat spinal cord (II), did

374

not evoke any changes in the enzyme activity, although 30% a raise in K' D was actually observed. Those observations

confirm

the hypothesis that enkephalin convertase may be also involved in the processing

of other peptide( REFERENCES

1. Fricker, L.D., Snyder, S.H.V. Purification

(1982). Enkephalin convertase:

and characterization

synthesizing carboxypeptidase

of a specific enkephalin-

localized to adrenal chromaf-

fin granules. Proc. Natl. Acad. Sci., USA 79: 3886-3890. 2. Strittmatter,

S.M., Lynch, D.R., Snyder, S.H. (1984). 3H-

Guanidinoethylmercaptosuccinic homogenates.

acid and binding to tissue

J. Biol. Chem. 259: 11812-11817.

3. Strittmatter,

S.M., Lynch, D.R., DeSouza, E.B., Snyder, S.H.

(1985). Enkephalin convertase demonstrated

in the pituitary

and adrenal gland by ["H]Guanidinoethylmecaptosuccinic auroradiography:

dehydration decreases neurohypophyseal

levels. Endocrinology 4.

117:

Lynch, D.R., Strittmatter, lin convertase

acid

1667-1674.

S.M., Snyder, S.M. (1984) Enkepha

localization by ('Hj-guanidinoethylmecapto-

succinic acid autoradiography: enkephalin-containing

selective association with

neurons. Neurobiology

81:

6543-6547.

5. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, Ri-J. (1951).

Protein measurement

Chem. 193:

with the folin reagent. J. Biol.

265-275.

Stack, G., Fricker, L.D., Snyder, S.H. (1984). A sensitive radiometric

assay for enkephalin convertase ana other carbo-

xypeptidase

B-like enzymes. Life Sci. 34: 113-121.

Munson,P.,

Rodbard, D. (1980) LIGAND: A versatile comput-

erized approch for characterization tems. Anal. Biochem. 107: 220-239.

375

of ligand-binding

sys-

a. Cheng, Y., Prusoff, W.H. inhibition constant

(1973). Relationship

(Ki) and the concentration

which causes 50% inhibition

between the of inhibitor

(I~~) of an enzymatic reaction.

Biochem. Pharmac. 22: 3099-3108. 9. Pollet, R.J., Standaert, M.L., Haase, B.A.

(1977). Insulin

binding to the human lymphocyte receptor. J. Biol. Chem. 252: 5828-5834. 10. De Meyts, P. (1976). Insulin and growth hormone receptors in human cultured lymphocytes and peripheral blood monocytes.

In:

Blecher, M. (ed.) Methods in receptor research. Marcel Dekker, New York., p.301-383. 11. Cesselin, F., Bourgoin, S., Artaud, F., Hamon, M. (1984). Basic and regulatory mechanisms

of in vitro release of met-

enkephalin

from the dorsal zone of the rat spinal cord. J.

Neurochem.

43: 763-773.

Received26/g/86 Accepted24/g/86

376