294

Bratn Re.sear~/t. 5t)b; (1992) 294 3()1 ~'~ 1992 Elsevier Science Publishers B.V. All rights reserved (}O06-S993/q2/$O5.t)(p

BRES 18356

Characterization of the release of Met-enkephalin from isolated nerve terminals: release kinetics and cation-dependence Matthijs Verhage a, Wim E.J.M. Ghijsen b and Victor M. Wiegant " " Rudolf Magnus Institute, Department of Medical Pharmacology, Unicersity of Utrecht, Utrecht (The Netherlands') and b Department of Experimental Zoology, Unicersity of Amsterdam, Amsterdam (The Netherlands) (Accepted 21 July 1992)

Key words: Met-enkephalin; Neuropeptide release; Synaptosome; Exocytosis; Ca 2. Dependence; Rat

The release of the neuropeptide Met-enkephalin (Met-ENK) from isolated nerve terminals (synaptosomes) of the rat forebrain was characterized with respect to the subcellular distribution, the release upon addition of various stimulatory agents, the release kinetics, the cation-dependence of release and the relationship between Met-ENK release and elevations of the intraterminal free Ca2÷-concentration ([Ca]i). A highly specific radioimmunoassay was developed for determination of Met-ENK (H-Tyr-Gly-Gty-Phe-Met-OH). Truncated and elongated forms of Met-ENK, Leu-enkephalin,/3-endorphin and dynorphin displayed negligible cross-reactivity. Met-ENK-like immunoreactivity (Met-ENK-LI) is enriched in the purified synaptosomal fraction of rat forebrain homogenates and is released in a strictly Ca2+-dependent manner upon chemical depolarization or stimulation with the Ca 2+ ionophore, ionomycin. A correlation exists between the release of Met-ENK-LI and the elevations of [Ca] i. Barium ions are able to replace Ca 2+ in triggering Met-ENK-LI release. The release of Met-ENK-LI is initiated within 20 s after depolarization and is terminated after 3-5 min, although depolarization and [Ca] i elevation are maintained. At this time, > 90% of the initial Met-ENK-LI is still present inside the synaptosomes. Repolarization and renewed stimulation again evokes Ca2+-dependent release of this retained Met-ENK-LI. It is concluded that Met-ENK release from isolated nerve terminals is exocytotic, and that exocytosis is terminated by a regulatory mechanism in synaptosomes after 3-5 min of depolarization, a process which can be reversed by repolarization. The characteristics of Met-ENK release are compared to those of other neuropeptides, of catecholamines and of amino acid transmitters, in similar preparations.

INTRODUCTION

Met-enkephalin (Met-ENK) is an endogenous pentapeptide with an affinity for opiate receptors that exerts opiate-like effects. Nerve cells, fibers and terminals containing M e t - E N K are widely distributed in the brain. M e t - E N K is synthesized as part of the high molecular weight precursor, proenkephalin A. Following post-translational enzymatic liberation from this precursor, M e t - E N K is stored in synaptic vesicles and can be released to exert its effects on brain functions as a neuropeptide transmitter (see for a review refs. 10, 23). Neuropeptides appear to be stored in large, densecored vesicles, distinct from the small synaptic vesicles, in which only classical transmitters are believed to be stored. There is ample evidence that neuropeptides (e.g. Met-ENK) can be co-localized with classical trans-

mitters in the same neurons and nerve terminals 3"tl't3. In systems where both types of transmitter can be studied together, the stimulation conditions effective for the release of neuroactive peptides are distinct from those effective for classical transmitters (see for a review ref. 3), yielding indications for multiple signalling at the (same) synapse. In this paper we characterize the in vitro release of M e t - E N K in the rat brain with the use of a highly specific radioimmunoassay for the pentapeptide, in order to discriminate between this active form of enkephalin and other, truncated or elongated forms that may occur in nerve terminals. The release characteristics are compared to those of classical transmitters and other neuropeptides. To study the release of different types of transmitters, purified nerve terminals or synaptosomes were used. Iso-osmotically purified synaptosomes are met-

Correspondence: M. Verhage, Institute for Molecular Biology, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands. Fax: (31) (30) 581 208.

295

abolically active, maintain a high A T P / A D P ratio, a negative membrane potential, and a low intra-terminai free calcium concentration ([Ca] i, see refs. 20, 29, 31). Synaptosomes release amino acid transmitters in an energy- and Ca2+-dependent manner from a non-cytosolic compartment (see for a review ref. 20). In this study, we present data on the release of Met-ENK-LI from Percoll-purified synaptosomes upon different levels of K + depolarization or the Ca 2+ ionophore, ionomycin. Comparisons are made with the release of amino acids 29'-~j, catecholamine transmitters 33 and of another abundant neuropeptide, cholecystokinin-83°, in the same preparation and with the regulation by [Ca] 29. MATERIALS

AND METHODS

Materials Percoll was obtained from Pharmacia (Uppsala, Sweden); bacitracin from Sigma (St Louis, MO, USA.); ionomycin from Calbiochem (La Jolla, CA, USA); and Fura-2-AM (fura-2-acetoxymethyl ester) from Molecular Probes (Eugene OR, USA). All synthetic peptides were obtained from Organon (Oss, The Netherlands). All other chemicals, of the purest grade available, were obtained from Merck (Darmstadt, FRG) and Jansen Pharmaceutica (Beerse, Belgium).

Synaptosomal preparation Synaptosomes were prepared from the whole forebrain of male Wistar rats (180-240 g). Preparations were made with a modification of the methods of Nagy and Delgado-Escueta t9 and Dunkley et al. 7 as described before 29, with 7 . 5 / 1 0 / 2 3 % discontinuous Percoll/0.32 M sucrose gradients. Synaptosomes were collected from the 10-23% interface and diluted in artificial cerebrospinal fluid (ACSF) of the following composition (in raM): NaCI 132; KCI 3; MgSO 4 2; NaH2PO 4 1.2; D-glucose 10; HEPES 10; CaCI 2 0.02. The pH was adjusted at 7.4 with Tris. Synaptosomes were pelleted and resuspended in ACSF with 1.5 mM CaCI2, kept on ice and used for experiments within 5 h. Protein content was measured according to Bradford 4 with bovine serum albumin as a standard.

Enkephalin analysis Met-ENK was extracted from tissue homogenates, synaptosomes, and from supernatants and perfusion fluid of release experiments (see below) by incubating the samples for 15 min at 90°C in 1 M acetic acid. In the case of tissue homogenates and synaptosomes the samples were sonicated for two periods of 30 s and spun in a BHG table centrifuge (12,000× g) for 1 rain, and the supernatants were taken for analysis. Met-ENK was assayed by radioimmunoassay using a rabbit antiserum (M8) raised against bovine thyroglobulin-conjugated synthetic Met-ENK. Synthetic Met-ENK was used as a standard and, in radio-iodinated form, as a tracer. Samples were evaporated to dryness and reconstituted in 250 ~,1 RIA buffer: 125 mM phosphatebuffered saline containing 0.2% (w/v) sodium azide and 0.1% (w/v) Triton X-100 (pH 7.5). Standard solutions of Met-ENK were regularly made with this RIA buffer. In several control experiments standards were made in ACSF, evaporated and treated further as other samples. Samples and standards (100 #1) were preincubated for 24 h at 4°C with 50/,tl antiserum (final dilution x2,000 in RIA buffer). Samples were assayed in three dilution's. Tracer Met-ENK was added (approximately 10,000 cpm in 50/~1 R1A buffer) and the mixture was incubated for another 20 h at 4°C. After addition of 50 p,M horse serum as the carrier protein, the incubation was terminated by precipitation of the bound fraction with 1 ml 20% polyethyleneglycol in R1A buffer containing 0.5% (v/v) of the detergent,

Tween-20. The precipitate was collected by centrifugation at 3,000 rpm for 15 rain and was counted for radioactivity. Data were analyzed with a splime curve fitting program (Cobra, Packard, CT, USA). The detection limit of the assay was established at 10% displacement of tracer, which corresponded to 1 pg Met-ENK per sample. Cross-reactivities of synthetic peptides were determined at 50% displacement of the tracer and expressed as a percentage on a mass basis (Table 1). Serial sample dilutions yielded displacement curves that paralleled the Met-ENK standard curve.

Release experiments (batch assay) Synaptosomal atiquots (750/~g protein) were spun down before release experiments in a BHG table centrifuge (12,000x g, 1 rain) and resuspended in 1.5 ml ACSF with 0.2 m g / m l bacitracin, a non-specific peptidase inhibitor 16, and either 1.5 mM CaCI 2 or 50 ~ M EGTA. Synaptosomes were preincubated for 5 min in a shaking water bath at 36°C. After preincubation, 10 ~1 of one of the different stimulants was added: KCI, BaCl 2 or ionomycin. After addition, the samples were gently mixed. For the estimation of initial release (20 s), incubations were terminated by the addition of an equal volume of ice-cold 10 mM EGTA (in water, pH 8.2 with NaOH). Synaptosomes were immediately pipetted onto 200 ~1 50:50% (v/v) silicone oil (Dow Corning 550) and dinonylphtalate, and spun for 90 s in a BHG table centrifuge (12,000× g). For longer incubations synaptosomes were incubated again at 36°C and were spun through oil after 1, 3 or 5 rain. Samples (500 ~1) were taken from the supernatant, pipetted into an equal volume of 2 M acetic acid, and Met-ENK-content was assayed as described above. Pellets were resuspended and enkephalins were extracted as described above. Ca 2+ dependency of release was studied by parallel incubation of two samples (see also ref. 29): one with normal ACSF containing 1.5 mM CaCI 2 and one with ACSF without added Ca 2+ and with 50 p,M EGTA to buffer endogenous Ca 2+.

Release experiments (superfusion assay) In order to stimulate synaptosomes repetitively, they were immobilized in a rapid perfusion chamber -~2. Before transferring synaptosomes to the chamber, an aliquot of 100 p,I of the suspension (100 p,g protein) was lysed with 0.2% (w/v) SDS to determine the total (initial) Met-ENK content. This sample was diluted 10-fold with ACSF to reduce the SDS concentration of the sample before MetENK was assayed as described above. Purified synaptosomes (200400 # g protein) were immobilized in Sephadex GI0 (coarse; Pharmacia, Uppsala, Sweden) in an isobar laminar flow perfusion chamber of 400 /.tl provided with horizontal flow frits (Chrompack, Darmstadt, Germany) and a media separator. Immobilized synaptosomes were perfused at 1 m l / m i n with ACSF. Synaptosomes were perfused for 15 rain, at the end of which 3 samples representing basal effiux were taken. Synaptosomes were stimulated by equimolarly replacing Na + with K +. The Ca2+-dependent release was estimated by subtracting Ca2+-independent release from the total release obtained in two parallel experiments. Synaptosomes were stimulated for 3 rain with 30 mM K + with samples taken over periods of 30 s, were then allowed to depolarize in normal ACSF (3 mM K + ) for 3 rain and finally stimulated again for 3 rain with 30 mM K +. Immediately after the last stimulation period the synaptosomes were perfused with 0.2% (w/v) SDS to determine the retained Met-ENK content. Perfusates were immediately stored at - 2 0 ° C in the presence of 0.1% trichloracetic acid and 1 # M citrullin as an internal standard. To investigate the exact time profile of the K + diffusion and -elevation, KCI was replaced by Phenol red in both basal and depolarizing medium with a 10-fold concentration difference between the two media. The Phenol red concentration in perfusates was estimated photometrically at 595 nm. The data obtained with Phenol red are identical to direct estimations of the K + concentrations with flame ionization. Protein content of the perfusates was < 0.2 ~ g / m i n (the detection limit of the assay), which was < 0.1% of the amount of synaptosomal protein placed onto the gel. Ca 2+ dependency of release was studied as in batch experiments with parallel perfusion of two samples +_Ca 2+ (see ref. 31).

296 dilution with ACSF, incubations were continued lot It) rain. Emis sion of Fura-2-AM was measured at 510 nm (slit 20 nm) and automatically averaged over 3.8-s intervals with excitation alternatin,g between 336 and 380 nm (slit 2.5 nm) every 3.8 s (program provided by Perkin Elmer).

700 . . . . . . . . . . . . .

A

600 500-

~ - -

+Ca2+

-------C~

.C;a~+ ~

~ / _ T. ~ 1

~ K+

/

4OO 300

Cah'ulations and statistics Data were stored on Apple-Macintosh computers and evaluated. averaged and visualized with Microsoft Excel, Cricket Graph and MacDraw 11. Data are expressed as the m e a n s + S.E.M. of n experiments. Statistical significance was calculated using the Student's two tailed t-test.

200 UJ

100~

~t3OmMK +

-5

0

1

2

3

4

5

6

RESULTS

time (rain) 400

a

.z_

I

**

I

®

2

ca o)

3oo

E

o

E 200 ro i1) ~ v Z m,

100

~

Net O.

i

1

Ca2+-dependentcomponent --i

2

p

i

i

3

4

5

time (min) Fig. 1. Characteristics of Met-ENK-LI release from isolated nerve terminals. A: kinetics of Met-ENK-LI release after K + depolarization in the presence of 1.5 mM Ca 2+ during (pre)incubation (filled symbols) or with no added Ca 2+ and 50/zM EGTA during (pre)incubations (open symbols). 30 mM K + was added at timepoint zero (arrow) in the indicated traces. In the other traces (basal), no K ÷ was added. Data points represent means + S.E.M. of 5 independent preparations (except timepoint 5 min: n = 3). B: the net Ca2+-depen dent component of Met-ENK-LI release. This component was calculated by subtraction within each experiment of K+-evoked MetENK-LI release in the absence of Ca e+ from release in its presence. This net Ca2+-dependent component was averaged and found to be significantly higher than zero at * P < 0.002; * * P < 0.001.

[Ca] i measurement Estimation of [Ca] i was carried out as described before 2~. Synaptosomes (3-4 mg p r o t e i n / m l ) were incubated with 4-5 /zM fura-2acetoxymethyl ester (Fura-2-AM) for 30 rain at 30°C. After a 10-fold

Detection and recot~'ery of Met-enkephalin To evaluate the specificity of the radioimmunoassay used for analysis of Met-ENK, the cross-reactivity of a number of endogenous peptides that are structurally related to Met-ENK was determined over a wide concentration range. Substitution of Leu for Met 5 (in Leu-ENK and dynorphin), N-terminal truncation (in des-TYR1-Met-ENK) and C-terminal elongation (MetENK-Arg6-Phe 7, Met-ENK:Arg6-GlyT-Leu 8, /3-endophin]_3j and 3,-endorphin) completely abolished the immunoreactivity (Table 1). This illustrates the high specificity of the radioimmunoassay for Met-ENK and suggests that the immunoreactivity detected with this assay in biological samples predominantly represents authentic, free Met-ENK. When analyzed in this assay, sonicated homogenates of the rat forebrain yielded 1.1 +0.2 pmol of immunoreactive material per mg protein (n = 3). After purification of synaptosomes using Percoll density gradients 2.4 _+ 0.4 pmol/mg protein was found (n = 5). Thus, a 2.2-fold enrichment of Met-ENK-LI was obtained in purified nerve terminals with respect to whole forebrain extracts. Upon thorough washing of synaptosomes (3 × spinning-down and resuspension), 0.5 pmol Met-ENK/mg protein was retained in the supernatant (n--4), suggesting that some of the Met-ENK-LI is present in or associated with 'leaky' synaptosomes or

TABLE I

Cross-reactit'ities of synthetic peptides in the radioimmunoassay for Met-enkephalin For details on the radioimmunoassay conditions see Materials and Methods.

Peptides

Amino acid sequence

Cross-reacticity (% on mass basis)

Met-ENK ( =/3-endorphin a_5) Met-ENK-Arg6-Phe 7 Met-ENK-Arg6-Gly 7-Leu s Leu-ENK des-TYR t-Met-ENK ( = ,8-endorphin ~_ 5) ,8-Endorphin 2- 7 /3-Endorphin i- 31 /3-Endorphin]-]7 ( = ,/-endorphin) Dynorphin 1- 13

Tyr-Gly-Gly-Phe-Met Tyr-Gly-Gly-Phe-Met-Arg-Phe Tyr-Gly-Gly-Phe-Met-Arg-Gly-Leu Tyr-Gly-Gly-Phe-Leu Gly-Gly-Phe-Met G ly-G ly-Phe-Met- Thr- Ser Tyr-Gly-Gly-Phe-Met-(AA ) e6 Tyr-Gly-Gly-Phe-Met- (AA )12 Tyr-Gly-Gly-Phe-Leu-(AA)s

100 < 0.001 < 0.001 < 0.001 0.006 < 0.001 < 0.001 < 0.001 < 0.001

297

B

A

EXOGENOUS Met-ENK

12

12 A¢.-

Q.

~ 8 O~

E

E ~ 1° ~ , ~ _ . . A .

o

E t3.

1...4k-.L ~z~'-'-

PELLET SUPERNATANT

4

Z

Z

ILl

9

0 initial content

released retained

81

.

0

1

.

. 2

.

.

3 4 time (rain)

5

6

Fig. 2. Release, retention and retrieval of Met-ENK in isolated nerve terminals. A: the released and retained fractions of the initial (endogenous) Met-ENK-LI content of synaptosomes. Synaptosomes were depolarized with 30 mM K + in the presence of Ca 2+ and incubated for 3 min at 36°C before synaptosomes (retained fraction) were separated from the medium (released fraction) by spinning through oil (see Materials and Methods). Bars represent means + S.E.M. of 5 independent preparations. The initial content of synaptosomes was taken as 100% (absolute value: 2.4 pmol/mg protein, see text). B: extracellular breakdown and synaptosomal uptake of synthetic (exogenous) Met-ENK. At timepoint zero 10 pmol/mg protein synthetic Met-ENK was mixed with synaptosomes and incubated for 3 min at 36°C. Data points represent means of two independent preparations.

contaminating membranes or may have escaped from the terminals during spinning.

marginal increase of Met-ENK release above basal efflux. The presence or absence of Ca 2+ did not influence the basal efflux (with 3 mM K ÷) of Met-ENK-LI. Fig. 1B shows the net K÷-evoked CaE+-dependent component of Met-ENK release, calculated from the data in Fig. 1A. Within 20 s a significant Ca2+-dependent release was detected. Between 3 and 5 min the K+-evoked release of Met-ENK-LI levelled off, even when the depolarization was maintained (Fig. 1A). At the time that the release of Met-ENK-LI ceased, a large amount of Met-ENK-LI was retained in the synaptosomes (Fig. 2A). Compared to the initial content, approximately 90% of the peptide was retained inside the synaptosomes after prolonged K + depolarization, and approximately 5% was released from the

The kinetics of Met-enkephalin release Fig. 1 shows the kinetics of Met-ENK-LI release from synaptosomes upon depolarization with 30 mM K +. In the presence of 1.5 mM Ca 2+ in the medium (Fig. 1A, e), the release increased significantly above basal levels within 20 s of depolarization (P < 0.05). The amount of Met-ENK-LI released increased in time until approximately 3 min after depolarization. Between 3 and 5 min K ÷ after depolarization, no further significant release of Met-ENK was observed. In the absence of added Ca 2+, but with 50/zM EGTA present (Fig. 1A, o), depolarization only induced a

350 ...................... ~-30 ~ 300. Z.~_ UJ

'3S°l1B

~.- 30

"~ v A 300

g O:

"3. =E== 2s° 1

'-E

8

eo

?

®

V g o

150

.........

v,= 3

,I (..)

150"

"........

~:-3 i

time (rain)

time (rain)

Fig. 3. The kinetics of Ca2+-dependent Met-ENK release from synaptosomes immobilized in a rapid perfusion chamber upon a single depolarization (A) and repetitive depolarization (B) with 30 mM K +. The rise of the K + concentration inside the chamber is depicted in the dotted line (see Verhage et al.~2). Fractions were collected over periods of 30 s, while synaptosomes were perfused with 1 ml/min. The net Ca 2+-dependent component was calculated after parallel peffusion of two samples, one in the presence of 1.5 mM Ca 2+, the other with no added Ca 2+ and in the presence of 50/.tM EGTA. Data show a representative experiment. The results were reproduced in two further independent preparations.

298 synaptosomes (Fig. 2A). The retained amounts were independent of the treatment of samples (depolarized or non-depolarized, presence or absence of Ca 2 +). The total recovery of Met-ENK in the retained and the released fractions added up to approximately 95% of the initial content. The final 5% was not retrieved in either fraction and might reflect a limited breakdown of the peptide during incubations. Therefore the extracellular breakdown of Met-ENK during the incubations was tested by adding exogenous, synthetic MetENK to a synaptosome suspension (Fig. 2B). Only marginal breakdown of exogenous Met-ENK was found in synaptosome suspensions (in the presence of the non-specific peptidase inhibitor, b a c i t r a c i n ' ) and no binding to, or uptake into synaptosomes of, exogenous Met-ENK was observed (Fig. 2B). The final 5% that was not retrieved in either fraction may reflect experimental loss of material during handling of the samples, rather than biological breakdown. The large percentage of Met-ENK-LI retained in the synaptosomes during prolonged depolarization may be present in a population of nerve terminals with defective release mechanisms. In order to test this possibility and to further investigate the release of Met-ENK-LI, synaptosomes were immobilized in a rapid perfusion set-up 32 to allow for repetitive stimulation of the terminals. Fig. 3 shows the net Ca2+-depen dent release of Met-ENK from immobilized synaptosomes, calculated from parallel determinations in two samples of the total release (in the presence of 1.5 mM Ca 2+) and of the Ca2+-independent release (in the absence of added Ca 2+ but with 50 g M EGTA). The amounts of Met-ENK-LI released during a period of 3

=

M e t - E N K release

---0--

6oo A

300

'"

0

, 0

, 10

, 20

[K ÷ ] (raM)

, 30

min permanent depolarization with 3 0 mM KCI were comparable to the results obtained with a batch assay (Fig. 1B), although there may be a difference in basal efflux, possibly due to perfusion. Furthermore, in the rapid perfusion set-up, the kinetics of Met-ENK-LI release were slightly different in the initial seconds after depolarization as compared to the batch assay. This difference is primarily due to the fact that, in a perfusion system, the stimulus can not be applied in a step-wise manner, but only more gradually when changing basal media for stimulating media. The dotted lines in Fig. 3 indicate the rise of the stimulus itself (30 mM K +) in the perfusion chamber. Fig. 3A shows that immobilized synaptosomes exhibited termination of Met-ENK-LI release, in agreement with the observations in batch experiments. When synaptosomes were repolarized after stimulation by renewed perfusion with basal media (3 mM KCI) and then stimulated again with 30 mM K +, the synaptosomes again released a significant amount of Met-ENK-LI in a Ca2+-depen dent manner (Fig. 3B). Both with respect to the release kinetics and the amounts released, the release during the second period of stimulation resembled the characteristics of release during the first period. After the second stimulation the immobilized synaptosomes were lysed with 1% SDS (see Materials and Methods). The retrieved amounts of Met-ENK-LI after one (Fig. 3A) and two (Fig. 3B) periods of stimulation revealed that the total Met-ENK-LI content of the synaptosomes did not significantly change during periods of stimulation, indicating that production or breakdown of Met-ENK inside the terminals did not bias the above results (data not shown).

Intracellular free C a 2+ ~>51~M

r

500

/

300

_,

L

2oo

z~,,,

~ ~ I

:~

,oo

300

o 0 0

, 1

2

[ionomycin] (~M)

Fig. 4. The release of M e t - E N K and corresponding elevation of [Ca] i induced by different mechanisms of Ca 2+ entry. A: release of Met-ENK-LI (filled symbols) and elevation of [Ca] i (open symbols) as a function of [K ÷ L. B: release of MeI-ENK-LI and elevation of [Ca] i as a function of [ionomycin]. Release was estimated 3 min after K + or Ionomycin addition. Synaptosomes were preincubated and incubated in the presence of 1,5 m M Ca 2+ (circles). The squares represent stimulation with 30 m M K + (A) or 1 /xM ionomycin (B) in the presence of 5 0 / x M E G T A and no added Ca z+. Data points represent means+_S.E.M, of 4 - 5 independent preparations.

299

Dicalent cations and Met-enkephalin release Fig. 4 shows the release of Met-ENK and the corresponding intracellular free Ca 2÷ concentrations ([Ca] i) obtained by different ways of evoking Ca 2÷ entry. Fig. 4A shows the relationship between depolarizationevoked Ca2+-entry and the Ca2+-dependent release of Met-ENK (estimated after 3 min). The level of [Ca] i (open symbols) and the amount of Met-ENK-LI released CaZ+-dependently (filled symbols) rose in parallel with increasing depolarization. The levels of [Ca] i can also be elevated by the use of ionophores, such as ionomycin. In this way, larger increases in [Ca]~ were obtained than with K ÷ depolarization, which only induced [Ca] i elevations up to 400-500 nM (Fig. 4A). A low dose of ionomycin (1 /~M) also evoked a parallel increase in [Ca] i and Met-ENK release (Fig. 4B). However, a higher dose of ionomycin (2 ~zM), evoked a very large increase in [Ca] i , notably beyond the detection limit of the Ca 2+ indicator (Fura-2: [Ca]~ > 5/xM), but without a further increase in Met-ENK release. It has been reported before that Ba z+ can substitute for Ca 2÷ in Ca2÷-dependent secretion of catecholamines and amino acid transmitters 9"21'24'34. Fig. 5 shows that Ba 2÷ was also a potent substitute for Ca 2+ in the case of secretion of a neuropeptide from isolated nerve terminals. Half maximal stimulation of MetENK-LI release was found at [Ba 2÷] < 0.5 mM. The maximal amounts of Met-ENK-LI released by Ba 2÷ were comparable to those released by Ca 2+ entry upon depolarization, or addition of ionomycin. Addition of K ÷, together with Ba 2÷, in order to open voltage-gated Ca 2÷ channels, did not potentiate the effect of Ba 2+ alone (data not shown). DISCUSSION In this study, the secretion of Met-enkephalin from isolated nerve terminals has been characterized. The c

600

E o

400

v z

w,

200 0

1

use of a highly specific radioimmunoassay for Met-ENK allowed us to study specifically its presence in and secretion from isolated nerve terminals in a quantitative manner.

Analysis of released Met-enkephalin The negligible cross-reactivity of the antibodies used in this study allowed discrimination between intact Met-ENK and truncated and elongated forms of MetENK and structurally related peptides from distinct precursors, particularly proenkephalin B (proendorphin) and pro-opiomelanocortin. In this way it was possible to specifically study the release of authentic, free Met-ENK. Membrane-bound and soluble peptidases have been shown to specifically inactivate Met-ENK. The most important of these endogenous inactivators of MetENK activity are a broad-spectrum aminopeptidase, which cleaves the Tyr~-Gly 2 bond, and more specific zinc metallopeptidases (endopeptidase 24.11 or 'enkephalinase', and endopeptidase 24.15), which cleave the Gly3-Phe 4 bond 1'6J4'15. The endopeptidases responsible for this breakdown are preferentially found in the synaptosomal fraction of brain homogenates ~'5"12, although in less purified synaptosomal fractions than that used in the present study. Nevertheless, exogenous Met-ENK added to synaptosomes was not significantly hydrolysed during the incubations (Fig. 2B). This indicates that the addition of the non-specific peptidase inhibitor, bacitracin 16, the relatively short incubations (in total, 10 min or less) and the high dilution of the nerve terminals with (protease-free) medium are sufficient to minimize hydrolysis of Met-ENK. Furthermore, the present data suggest that isolated nerve terminals do not contain mechanisms to bind or incorporate Met-ENK to any appreciable extent (Fig. 2B), confirming the idea that nerve terminals do not have mechanisms to re-capture released neuropeptides. We conclude that this lack of re-uptake and the minimal extracellular breakdown of the peptide in these experiments justifies quantitative analysis of Met-ENK secretion from these terminals, and confirms that the pentapeptide, Met-ENK, is present in, and secreted from, nerve terminals. In addition, nerve terminals may secrete structurally related peptides derived from distinct precursor peptides (pro-dynorphin-derived peptides25), which are probably not detected with the present methodology.

2

[barium](mM) Fig. 5. Ba z+-evoked release of Met-ENK. Release was estimated 3 min after addition of the indicated a m o u n t s of Ba 2+, added as BaCI2. Data points represent m e a n s of 5 independent preparations _+S.E.M.

Subcellular distribution of enkephalins Endogenous Met-ENK is 2.2-fold enriched in purified nerve terminals as compared to whole forebrain homogenates. These fractionation data point towards a

300 specific occurrence of Met-ENK-LI in nerve terminals. The analysis of retained contents of synaptosomes reveals that their Met-ENK content is independent of the treatment of the samples. Thus, presynaptic activity (depolarization or elevation of [Ca] i) induces no appreciable intracellular metabolism of Met-ENK, and precursor peptides (proenkephalin A) are not processed to yield free Met-ENK in nerve terminals. Together, these data suggests that processing of the precursor peptide occurs predominantly before peptide-containing vesicles arrive at the nerve terminal. Furthermore, the present data confirm that the pentapeptide, Met-ENK, is a major (active) form in enkephalinergic neurotransmission. Release characteristics Upon depolarization and Ca 2+ entry, the release of Met-ENK from synaptosomes increases rapidly. This evoked release is entirely Ca 2+ dependent. This observation is in line with earlier studies in endocrine cells s. Here, we show that Ba 2÷ can effectively replace Ca 2÷ in the induction of Met-ENK release. This finding indicates that release mechanism of this neuropeptide shares similar affinities for the two divalent cations as the mechanisms responsible for secretion of classical transmitters 9,21,24,34. Little is known about the release kinetics of neuropeptides from synapses. The release of another neuropeptide, CCK-8, was found to develop slowly, with no significant release during the first 20 s 3°. This was interpreted as typical for a neuropeptide, in analogy with the relatively slow effects that neuropeptides exhibit on post-synaptic cells. The release kinetics of Met-ENK and CCK-8 clearly show that considerable differences may exist between the release mechanisms of the different neuropeptides. It is generally believed that neuropeptides are present only in large dense cored vesicles. However, the regulation of intracellular traffic and release may still vary among Met-ENK and CCK-8 containing vesicles. On the other hand, MetENK may be co-localized within the same vesicles with classical transmitters. Evidence from the Electric ray nerve terminals indicates that, in certain cases, MetENK is present in a morphologically uniform population of small electron-lucent vesicles with the classical transmitter acetylcholine ~s'28. Indeed, the release kinetics of Met-ENK resemble those of classical transmitters, like glutamate 17'31, GABA 26'32, acetylcholine 2 and catecholamines 33, rather than those of the peptide CCK-8 3°. Only approximately 5% of the total amount of MetENK-LI present in purified synaptosomes is released from the nerve terminals during prolonged stimulation.

The release of Met-ENK-LI is terminated after 3-5 rain permanent depolarization. This observation can not be explained by breakdown or re-uptake of released Met-ENK during prolonged incubations, since these processes were negligible under the present conditions, as discussed above. The Met-ENK-LI-containing vesicles may be subject to some regulatory mechanism limiting the release. The data from rapid perfusion and repetitive stimulation of immobilized synaptosomes indicate that this termination of release is indeed an actively regulated process associated with the function of the terminals, which can be reversed by repolarization. The termination of release may be induced by stimulation of presynaptic receptors by some agonist that is released from the terminals. It has been suggested that released Met-ENK evokes a feed-back inhibition through presynaptic, predominantly K-opioid receptors in the brainstem 27. However, two important arguments argue against such a mechanism in our studies using forebrain. First, the termination of release is also prominent when synaptosomes are perfused with high flow (! ml/min), and released agonists are thus rapidly separated from the terminals. The termination of release in the rapid perfusion set-up closely resembles that in a batch assay. Second, both in the batch assay and in perfusion the synaptosomes are well diluted and the medium concentrations of dominant neurotransmitters in this preparations are very low. For instance under similar conditions amino acid release experiments yield extracellular concentrations always below 2 /zM 29'31, for catecholamine release below 100 nM 33, and for release of the neuropeptide CCK-8 below 1 nM 3°. Therefore, we conclude that the termination of release is guided by intracellular mechanisms. The termination of release after prolonged, permanent depolarization and [Ca] i elevation is a property that appears to be shared among different transmitters. Some evidence suggests that a population of vesicles with amino acid transmitters is withheld from fusing with the membrane during a single K + stimulation 3~, probably by phosphorylation-dependent interactions with the cytoskeleton 2°. Further insight into the cellular factors interacting with neuropeptide-containing vesicles near their release site, and the homogeneity of these vesicles, is necessary to fully understand this apparent limitation of neuropeptide release. ABBREVIATIONS ACSF [Ca]i CCK-8

artificial cerebrospinal fluid intra-terminal free Ca2+ concentration cholecystokinin-8

301

Met-ENK Met-ENK-LI Leu-ENK SDS

methionine-enkephalin (H-Tyr-Gly-Gly-Phe-MetOH) methionine enkephalin-like immunoreactivity leucine-enkephalin (H-Tyr-Gly-Gly-Phe-Leu-OH) sodium dodecyl sulphate

Acknowledgements. The authors thank Ank Frankhuijzen and Elly Besselsen for their skillful technical assistance, and Drs. M. van der Velde for his contributions to experiments with the rapid perfusion set-up.

REFERENCES 1 Acker, G.R., Molineaux, C. and Orlowski, M., Synaptosomal membrane-bound form of endopeptidase-24.15 generates leu-enkephalin from dynorphin, a- and fl-neoendorphin, and Met-enkephalin from Met-enkephalin-Arg6-Gly7-Leus, J. Neurochem., 48 (1987) 284-292. 2 Adam-Vizi, V. and Ashley, R.H., Relation of acetylcholine release to Ca 2÷ uptake and intra-terminal Ca 2+ concentration in guinea-pig cortex synaptosomes, J. Neurochem., 49, (1987) 10131021. 3 Bartfai, T., lverfeldt, K., Fisone, G. and Serf6z6, P., Regulation of the release of coexisting neurotransmitters, Annu. Rec. PharmacoL ToxicoL, 28 (1988) 285-310. 4 Bradford, M.M., A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye binding, AnaL Biochem., 72 (1976) 248-252 5 Demmer, W. and Brand, K., A dipeptidyl carboxypeptidase in brain synaptic membranes active in the metabolism of enkephalin containing peptides, Biochem. Biophys. Res. Commun., 114 (1983) 804-812. 6 Dua, A.K., Pinsky, C. and LaBella, F.S., Peptidases that terminate the action of enkephalins: consideration of physiological importance for amino-, carboxy-, endo-, and pseudoenkephalinase, Life Sci., 37 (1985) 985-992. 7 Dunkley, P.R., Jarvie, P.E., Heath, J.W., Kidd, G.J. and Rostas, J.A.P., A rapid method for isolation of synaptosomes on Percoll gradients, Brain Res., 372 (1986) 115-129. 8 Eiden, L.E., Giraud, P., Dave, J.R., Hotchkiss, A.J. and Affolter, H.U., Nicotinic receptor stimulation activates enkephalin release and biosynthesis in adrenal chromaffin cells, Nature, 312 (1984) 661-663. 9 Heldman, E., Levine, M., Raveh, L. and Pollard, H.B., Barium ions enter chromaffin cells via voltage-dependent calcium channels and induce secretion by a mechanism independent of calcium, J. Biol. Chem., 264 (1989) 7914-7920. 10 Khatchaturian, H., Lewis, M.E., Sch~ifer, M.K.H. and Watson, S.J., Anatomy of the CNS opioid systems, Trends Neurosci., 8 (1985) 111-119. 11 Klein, R.L., Wilson, S.P., Dzielak, D., Yang, W.H. and Viveros, O.H., Opiod peptides and noradrenaline coexist in large dense cored vesicles from sympathetic nerve, Neuroscience, 7 (1982) 2255-2261. 12 Lane, A.C., Rance, M.J. and Walter, D.S., Subcellular localization of leucine-enkephalin hydrolysing activity in rat brain, Nature, 269 (1977) 75-76. 13 Lundberg, J.M. and H6kfelt, T., Coexistence of peptides and classical neurotransmitters, Trends Neurosci., 6 (1983) 325-333. 14 Malfroy, B., Swerts, J.P., Guyon, A., Roques, B.P. and Schwartz, J.C., High-affinity enkephalin-degrading peptidase in brain is increased after morphine, Nature, 276 (1978) 523-525. 15 Malfroy, B., Kuang, W.J., Seeburg, P.H., Mason, A.J. and Schofield, P.R., Molecular cloning and amino acid sequence of human enkephalinase (neutral endopeptidase), FEBS Lett., 229 (1988) 206-210.

16 McKelvy, J.M., Leblanc, P., Loudes, C., Perrie, S., GrimmJorgensen, Y. and Kordon, C., The use of bacitracin as an inhibitor of the degradation of thyrotropin-releasing hormone and luteinizing hormone-releasing hormone, Biochem. Biophys. Res. Commun., 73 (1976) 507-515. 17 McMahon, H.T. and Nicholls, D.G., Transmitter glutamate release from isolated nerve terminals: evidence for biphasic release and triggering by localized Ca 2÷, J. Neurochem., 56 (1991) 86-93. 18 Michaelson, D.M. and Wien, D., Enkephalin uptake into cholinergic synaptic vesicles and nerve terminals, Ann. N.Y. Acad. Sci., 493 (1987) 234-249. 19 Nagy, A. and Delgado-Escueta, A.V., Rapid preparation of synaptosomes from mammalian brain using nontoxic isoosomotic gradient material (Percoll), J. Neurochem., 43 (1984) 1114-1123. 20 Nicholls, D.G., Release of glutamate, aspartate and y-aminobutyric acid from isolated nerve terminals, J. Neurochem., 52 (1989) 331-341. 21 Nichols, R.A. and Sihra, T.S., Concentration dependence of the effect of barium on endogenous glutamate release from rat brain synaptosomes, Soc. Neurosci. Abstr., 15 (1989) 474. 22 Nichols, R.A., Sihra, T.S., Czernik, A.J., Nairn, A.C. and Greengard, P., Calcium/calmoduline-dependent kinase-ll increases glutamate and noradrenaline release from synaptosomes, Nature, 343, (1990) 647-650. 23 Numa, Y., Opioid peptide precursors and their genes. In Udenfrind and Meienhofer (Eds.), The Peptides, Academic Press, USA, 1984. 24 Tagliatela, M, Canzoniero, L.M.T., Amoroso, S., Fatalis, A., Di Renzo, G.F. and Annunziato, L., Cobalt-sensitive and dihydropyridine-insensitive stimulation of dopamine release from tuberoinfundibular neurons by high extracellular concentrations of barium ions, Brain Res., 488 (1989) 114-120. 25 Terrian, D.M., Johnston, D., Claiborne, B.J., AnsahYiadom, R., Strittmatter, W.J. and Rea, M.A., Glutamate and dynorphin release from a subcellular fraction enriched in hippocampal mossy fiber synaptosomes, Brain Res. Bull., 21 (1988) 343-343. 26 Turner, T.J. and Goldin, S.M., Multiple components of synaptosomal [3H]GABA release resolved by a rapid superfusion system, Biochemistry, 28 (1989) 586-593. 27 Ueda, H., Fukushima, N., Ge, M., Takagi, H. and Satoh, M., Presynaptic opioid K-receptor and regulation of the release of met-enkephalin in the rat brainstem, Neurosci. Lett., 81 (1987) 309-313. 28 Unsworth. C.D. and Johnson Jr., R.G., Opioid peptides of the electric ray Narcine braziliensis, Ann. N.Y. Acad. Sci., 493 (1987) 406-408. 29 Verhage, M., Besselsen, E., Lopes da Silva, F.H. and Ghijsen, W.E.J.M., Ca2+-dependent regulation of presynaptic stimulus secretion coupling, J. Neurochem., 53 (1989) 1188-1194. 30 Verhage, M., Ghijsen, W.E.J.M., Nicholls, D.G. and Wiegant, V.M., Characterization of the release of cholecystokinin-8 from isolated nerve terminals and comparison with exocytosis of classical transmitters, J. Neurochem., 56 (1991) 1394-1400. 31 Verhage, M., Lopes da Silva, F.H. and Ghijsen, W.E.J.M., Activity dependent recruitment of endogenous glutamate for exocytosis, Neuroscience, 43 (1991 ) 59-66. 32 Verhage, M., Sandman, H., Mosselveld, F., Van de Velde, M., Hengst, P., Lopes da Silva, F.H. and Ghijsen, W.E.J.M., Immobilized isolated nerve terminals as a model for studies on the regulation of transmitter release: release of different endogenous transmitters, repeated stimulation and high time resolution, J. Neurochem., 58 (1992) 1313-1320. 33 Verhage, M., Ghijsen, W.E.J.M., Boomsma, F. and Lopes da Silva, F.H., Endogenous noradrenaline and dopamine in nerve terminals of the hippocampus: differences in levels and release kinetics, J. Neurochem., 59 (1992) 881-887. 34 Weiss, C., Sela, D. and Atlas, D., Divalent cations effectively replace Ca 2÷ and support bradykinin-induced noradrenaline release, Neurosci. Lett., 119 (1990) 241-244.