Life Sciences, Vol. 32, pp. 287-294 Pergamon Press Printed in the U.S.A.

DISTINCT DISTRIBUTION OF IMMUNOREACTIVE DYNORPHIN AND LEUCINE ENKEPHALIN IN VARIOUS POPULATIONS OF ISOLATED

ADRENAL CROMAFFIN CELLS

Michel Dumont, Robert Day and Simon Lemaire*

Department of Physiology and Pharmacology Centre Hospitalier Universitaire

Sherbrooke, Quebec, Canada JIH 5N4

(Received in final form October I, 1982)

Summary

The distribution of inmmnoreactive-dynorphin (ir-Dyn) in isolated subpopulations of bovine adrenal chromaffin cells was examined and compared with that of adrenaline (A), noradrenaline (NA) and ir- Leucine-Enkephalin (ir-Leu-Enk). Using a stepwise bovine serum albumin (BSA) gradient, various populations of catecholamine-storing cells were separated and designated as cell layers I, II and III. Cell layer I contained more NA than A; cell layer II contained slightly more A than NA whereas cell layer Ill was highly enriched in A. The original cell preparation contained 2.9 times more ir- Leu-Enk than ir-Dyn (4.7 and 1.6 pmoles per 106 cells, respectively). After separation of the cells on BSA gradient, ir-Dyn was mainly detected in cell layer I (4.0 pmoles/106 cells) whereas ir-Leu-Enk was concentrated in cell layer III (8.3 pmoles/106 cells). Both peptides were secreted in response to acetylcholine (S x I0 -s M), but the amount secreted was in accordance with the cell content in each peptide. After subcellular fractionation of the adrenal medulla, the neuropeptides were found in close association with catecholamines in the secretory granules. These results indicate that bovine adrenal chromaffin cells can be isolated according to their specific content in A, NA and opioid peptides and are consistent with the hypothesis of distinct biosynthetic pathways for Dyn and the Enk.

Dyn, the most potent opioid peptide in the guinea pig ileum assay, was first discovered by Goldstein et al. in the porcine pituitary (I). Since then, immunocytochemical and chemical studies have identified Dyn in the brain, the neurointermediate pituitary, the adenohypophysis (2-6), the gut (7) and the ad- renal medulla (8-9). However, the first endogenous opioid compounds identified in the adrenal medulla were the Enk. Both Leu- and Met- Enk-immunoreactive ma- terials were found in nerve terminals and in gland cells of the adrenal medulla (I0-II). Opiate like-peptides were also demonstrated to be co-secreted with catecholamines from perfused adrenal glands (12). Starting from these data, several laboratories have undertaken the isolation and chemical identification of Enk-containing peptides at this level (13-20). Some of these peptides were demonstrated to contain more than one copy of Leu- and Met-Enk (18). The dis= covery of these precursors represented a large advance in the understanding of the biosynthetic pathway of the Enk. It was then postulated that both Leu- and Met-Enk are generated from the same precursor molecule after the proper cleav- age by trypsin-like enzymes. Finally, the complete structure of the eDNA for bovine adrenal preproenkephalin was established and demonstrated to contain the sequence for both Leu- and Met-Enk but not for Dyn (21,22). These data were

0024-3205/83/030287-08803.00/0 Copyright (c) 1983 Pergamon Press Ltd.

288 Dynorphin in Adrenomedullary Cells Vol. 32, No. 3, 1983

further supported by the discovery of the nucleotide sequence of a complete cDNA copy of enkephalin precursor mRNA from human phaeochromocytoma (23). Con- versely, we have recently demonstrated the presence (8) and the secretion (9) of Dyn in the bovine adrenal medulla and freshly isolated adrenal chromaffin cells. The present study was aimed at determining if adrenal ir-Dyn and ir-Leu- Enk are localized in the same or in distinct types of chromaffin cells and if they are particularly associated with A, NA or both catecholamines.

Materials and Methods

Collagenase (CLS type I) was obtained from Worthington. BSA (Path-O-Cyte 4) was purchased from Miles Lab. Acetylcholine was a product of Sigma, St-Louis, Missouri. [3H]-Dyn-(l-13) (50 Ci/mmol) and [1251]-Leu-Enk (I000 ~Ci/ ~g) were obtained from New England Nuclear. The antiserum to Dyn was produced in our laboratory as described by Day et al. (9). The antiserum to Leu-Enk was a generous gift from Dr. K.J. Chang, Wellcome Res. Lab., N.C. Leu-Enk and Dyn- (1-13) were synthesized in our laboratory as described by Lemaire et al. (37).

Separation of adrenal chromaffin cell subpopulations. Fresh bovine adre- nal glands were obtained at a local slaughterhouse and brought to the labora- tory on ice. The chromaffin cells were isolated as described by Lemaire et al. (24). Briefly, the medullary tissue was extracted free of cortical tissue and submitted to 5 successive digestions with collagenase (0.1S% in modified Krebs buffer, pH 7.3; buffer I, ref. 24). Chromaffin cells were collected from each digestion and purified by various washes and differential sedimentations at unit gravity. They were then stored overnight in a capped plastic tube at room temperature and used the next day. This preparation was designated as the original cell preparation. For fractionation, the cells were layered on top of a stepwise BSA gradient (from top to bottom, 3 ml each of 10%, 15%, 19% and 23% BSA in buffer I) and spun at 16,200 x g for 30 min at 4°C. After centrifuga- tion, layers of cells were observed at the interfaces of the BSA gradient; that is, cell layer I between 10% and 15% BSA, cell layer II between 15% and 19% and cell layer III between 19% and 23%. Each cell layer was collected, washed with two 15 ml portions of a modified Krebs buffer, pH 7.3 (buffer If, ref. 24) and counted with a hemocytometer. In a typical experiment, starting with 50 x 106 cells of the original cell preparation, 6.6 x 106 , 14.2 x i0 ~ and 17 x 106 cells were obtained in layers I, II and III, respectively. Cell viability (trypan blue exclusion) was estimated in each layer as > 95%.

Secretion and cell content in ir-Dyn, ir-Leu-Enk and catecholamines. The secretion assay was started by the addition of 1.0 x 106 cells to a prewarned (37oc) solution (0.8 ml) of Krebs buffer (24) in the presence or absence of acetylcholine (50 ~M). The secretion was stopped after 5 min by transferring the tubes to an ice-water bath (0°C) followed by centrifugation at II0 x g for 10 min. The supernatants were collected and lyophilized. The dry powder ob- tained after lyophilization was dissolved with 450 ~I of H20 and its content in Dyn and Leu-Enk was measured by radioimmunoassay with i00 ~i aliquots ac- cording to the technique of Day et al. (9). The lower limit of detection for both radioimmunoassays was approximately 15 fmoles. The Dyn-antiserum did not have any cross-reactivity with Leu-Enk. Conversely, the Leu-Enk antiserum cross-reactivity with Dyn was lower than .0001%. Each data represents the mean ± SEM of four sets of duplicates. The secretion represents the acetyl- choline-evoked release of neuropeptides after the subtraction of the basal release at 37 ° in the absence of drug.

The cellular content in ir-Dyn and ir-Leu-Enk was also determined by ra- dioimmunoassay after acid extraction. The extracts were obtained by adding

Vol. 32, No. 3, 1983 Dynorphin in Adrenomedullary Cells 289

8 ml of a hot (95 ° ) solution of 0.i M HCI - 0.01% thiodiethanol to cell pel- lets (i x 106 cells) obtained by centrifugation. The tubes were incubated at 95oc for 15 min and cooled down to 0oc in an ice-water bath. The mixtures were homogenized with a glass-teflon homogenizer and the pH was increased to 7.4 with 0.5 M Tris. The homogenates were centrifuged for 30 min at 27,000 x g and the supernatants were collected and lyophilized. The dry powders were dissolved with 450 ~i H20 and their contents in ir-Dyn and ir-Leu-Enk were measured as described above.

The cellular contents in catecholamines (A, NA or total) were measured by spectrofluorometry as described (24). Briefly, the cells (0.i x I0 ~) were centrifuged at ii0 x g for i0 min and 0.7 ml of a cold solution of 5% trifluo- roacetic acid was added to the pellets. The cells were then frozen and thawed 3 times and the protein precipitate was discarded by centrifugation at 2,000 x g for 20 min. Catecholamines were measured in the supernatant (200 ~l ali- quots) according to the method of Miura et al. (25). Fluorescence was moni- tored on a spectrofluorometer (Carl Zeiss, model AFMz) at excitation/emission wavelengths of 410/510 and 455/510 and the amounts of A and NA were determined as described (25). The data represent the mean ± SEM of six separate sets of duplicates.

The subcellular fractions of bovine adrenal medulla (cytosol, secretory granules, mitochondria and microsomes) were prepared as described by Winkler et al. (26). Their contents in ir-Dyn and ir~Leu-Enk were measured by radio- immunoassay after extraction with the HCI solution as described for the cellu- lar extracts. Their contents in catecholamines were also estimated as de- scribed above. Proteins were measured by the method of Lowrey et al. (27). The data represent the mean ± SEM of two duplicate. All the results were eval- uated using the student T-test and P < 0.05 was considered significant.

Results

Fig. 1 and Table I show the relative contents of the various isolated sub- populations of adrenal chromaffin cells in A, NA, ir-Dyn and ir-Leu-Enk. The original cell preparation (before separation on BSA gradient) contained 1.6 times more A than NA (20.1 and 12.2 nmoles/106 cells, respectively). However, the cell content in both amines varied greatly between each isolated cell pop- ulation, the A/NA ratios being 0.6, 1.3 and 3.3 in cell layers I, II and III, respectively. The high NA content (24.4 nmoles/106 cells, Fig. i) in cell layer I corresponded to a high level of ir-Dyn (4.0 pmoles/106 cells) whereas the cell layer enriched in A (cell layer III, 29.4 nmoles/106 cells) had a high content in ir-Leu-Enk (8.3 pmoles/106 cells). The original cell prepara- tion contained 2.9 times more ir-Leu-Enk than ir-Dyn. However, the ir-Leu-Enk/ ir-Dyn ratio varied between 0.73 in cell layer I and 7.5 in cell layer IIl.

The original chromaffin cell preparation has already been shown to be highly functional by its secretion of catecholamines in response to choliner- gic and depolarizing agents (24). To verify if the intracellularneuropeptides were also secreted, cells were incubated in the presence of acetylcholine (5 x I0 -s M) for 5 min and the amount of opioid peptides released into the incuba- tion medium was measured. Starting with the original cell preparation, a re- lease of 14% and 19% of total content in ir-Dyn and ir-Leu-Enk, respectively, was observed (Table I). After separation of the cells on BSA gradient, ir-Dyn was mostly released from cell layer I whereas ir-Leu-Enk was rather secreted from cell layer III. The relative percentage of the neuropeptides released from the isolated cell populations corresponded more likely to their respec- tive content in each peptide.

290 Dynorphln in Adrenomedullary Cells Vol. 32, No° 3, 1983

. ° , , t "..::..:.i A 30 i!i!i! 2sl i iili!i

ii!iiii , iiiiiii ',iiiil

,o i!iiiii i!iili iiiiiii ililil g s

0 O.C.P. I I I I I I

Fig. 1 Content in adrenaline (A) and noradrenaline (NA) in the original cell preparation (O.C.P.), cell layer I, cell layer II and cell layer III. Results are the mean ± SEM of 6 experiments performed in duplicate. A and NA were measured by spectrofluorometry as described under "Materials and Methods".

Table II shows the distribution of catecholamines (A plus NA), ir-Dyn and ir-Leu-Enk in subcellular fractions obtained from the bovine adrenal medulla. The concentration of catecholamines in the granular fraction was at least 3.6 times higher than that found in the other fractions (1097 nmoles/mg protein as compared with 215-296.8 nmoles in the other fractions). Similarly, ir-Dyn and ir-Leu-Enk were found highly concentrated in the secretory granules (3.24 and 261 pmoles/mg protein, respectively, as compared with 0.i to 0.6 and 30 to I00 pmoles in the other fractions). The high ratio (from 66 to 480) found between the content of the subcellular fractions in ir-Leu-Enk and ir-Dyn could possi- bly be explained by some more important loss of ir-Dyn during the preparation of the subcellular fractions than that of the chromaffin cells. This assumption is supported by the ratio found between the catecholamine and opioid peptide contents in the granular and the cellular extracts. The intracellular levels of catecholamines in the original cell preparation was 32.3 nmoles per i0 ~ cells (Fig. i). Based on this value, the catecholamine/ir-Dyn and catecholamine/ir- Leu-Enk ratios in the isolated cells are 20,187 and 6872, respectively (Table I). Conversely, these ratios as calculated from the granular extraction (Table II) become 338,796 and 4,199 for ir-Dyn and ir-Leu-Enk, respectively. One can s~e that the ratio remains approximately the same for Leu-Enk whereas it in- creases by a factor of more than 16-fold for Dyn in the granular extract. Such important increase in this ratio is relevant to a large loss of ir-Dyn. We may

TABLE I

Cell Content and Release of ir-Dynorphin

and ir-Leucine-Enkephalin

in

Subpopulations

of Isolated Adrenal Chromaffin Cells

o LO

bO

Cell

A/NA*

Content

Secretion

population

ratio

ir-Dyn

ir-Leu-Enk

Leu-Enk/Dyn

ir-Dyn

ir-Leu-Enk

Leu-Enk/Dyn

ir-Dyn

(pmoles/106

cells)

ratio

(pmoles/106

cells)

ratio

Percent released

ir-Leu-Enk

~o

OO

Lo

Original cell

preparation

1.6

1.6 ±

0.i

4.7 ±

0.9

2.9

0.23 ±

0.02

0.9 +

0.03

3.9

14

BSA gradient

Layer I

0.6

4.0 ±

0.8

2.9 ±

0.4

0.73

0.37 +

0.04

0.ii±0.01

0.3

9 Layer II

1.3

1.3 ±

0.3

5.3 +

0.8

4.1

0.08 ±

0.04

0.41±0.001

5.1

6 Layer III

3.3

i.i

± 0.3

8.3 ±

1.0

7.5

0.03 ±

0.01

1.03±0.03

34.0

3

19 4 8

12

o

*According to Fig.

i.

Experiments were performed as

described under "Materials

and Methods".

TABLE II

Subcellular Distribution of Catecholamines,

ir-Dynorphin

and ir-Leucine-Enkephalin

in bovine adrenal medulla

fD

O

Extracts

Catecholamines

ir-Dyn

ir-Leu-Enk

Leu-Enk/Dyn

(nmoles/mg prot)

% of Total

(pmoles/mg prot)

% of Total

(pmoles/mg prot)

% of Total

ratio

O~

b~

Granules

1097.7± 36.8

58.6

3.24 ±

0.14

72.6

261.4±23.7

58.7

81

Mitochondria

296.8±

6.0

15.8

0.65 ±

0.05

14.6

I01.i ±

4.5

22.7

155

Microsomes

263.2 ±

2.6

14.1

0.11 ±

0.01

2.5

52.8 ±

2.0

11.8

480

Cytosol

215.0 ±

2.3

11.5

0.46 ±

0.01

10.3

30.3 ±

1.3

6.8

66

CD

Experiments were performed as

described under "Materials

and Methods".

292 Dynorphin in Adrenomedullary Cells Vol. 32, No. 3, 1983

assume that when the cells are fractionated into their subcellular compartments more Dyn is exposed to proteolytic cleavage whereas Leu-Enk remains particular- ly stable.

Discussion

The present study was undertaken to explore the possibility of separating various types of catecholamine-storing cells and analysing their specific con- tent in A, NA, ir-Dyn and ir-Leu-Enk. After digestion with collagenase, various populations of catecholamine-storing cells were isolated on BSA gradients ac- cording to their specific content in A and NA. Such technique allowed the sep- aration of three cell populations: cell layer I with a high NA content, cell layer II with a high content in both amines and cell layer III with a high A content. Interestingly, ir-Dyn and ir-Leu-Enk were particularly concentrated into chromaffin cells enriched in NA and A, respectively. The finding of a distinct localization for ir-Dyn and ir-Leu-Enk is in accordance with the dif- ferent distribution patterns of these two peptides in the brain (6). Further- more, recent immunohistochemical studies in the bovine adrenal medulla have shown that Leu- and Met-Enk are stored in the A-cells (28). Such association of Dyn and Leu-Enk with two distinct types of catecholamine-storing cells strongly suggests that these two opioid peptides originate from distinct pre- cursor molecules although it does not exclude the possibility that some of the Leu-Enk can still be generated from Dyn or prodynorphin which, for instance, could be cleaved less effectively in cell layer I. In this regard, cDNA se- quence studies (21-23) have recently demonstrated that bovine adrenal and human phaeochromocytoma proenkephalins do not contain the Dyn sequence, thus indica- ting that Dyn cannot be produced by the proteolytic cleavage of the enkephalin precursor.

Using either perfused bovine adrenal glands (12) or isolated adrenal chro- maffin cells (29,30), various authors have already demonstrated that the Enk are coreleased with catecholamines in the presence of nicotinic drugs. The present studies not only support these data but further indicate that Dyn can also be released from adrenal chromaffin cells in the presence of acetylcholine and the secretions of Dyn and Enk are preferentially obtained from cells en- riched in NA and A, respectively (Table I). The percentages of ir-Dyn and ir- Leu-Enk released (from 3% to 19% of total cell content) are also in the same range than those observed for the release of catecholamines (7% to 13%, ref. 31). However, these percentages varied according to the content of each peptide in each cell layer (Table I). Such phenomenon was reflected by a large difference between the ratios of Leu-Enk/Dyn for cell secretion in layer III and Layer I (factor of 113) as compared with that found for cell con- tent in opioid peptides (factor of I0) and in catecholamines (factor of 6 for the A/NA ratios) in these populations. Our data do not provide any explanation for such difference but we may postulate that the particular low levels of se- creted ir-Leu-Enk (4%) and ir-Dyn (3%) from cell layers I and III, respectively, result from a higher speed of enzymic breakdown. Finally, the concept of co- storage and co-secretion of opioid peptides with catecholamines is supported by the finding of high concentrations of ir-Leu-Enk and ir-Dyn along with catecho~ amines in the secretory granules (Table II). Thus, the levels of opioid pep- tides and of catecholamines are shown to be from 3 to 30 times higher in the secretory granules than in the other subcellular fractions. These findings strongly support the possibility of an exocytotic release of these opioid pep- tides with other granular constituants.

In our original cell preparation, there was almost 2 times as much A as NA (Fig. i). The particular association of Dyn with NA-storing cells, a smaller component of the adrenomedullary cell population, may indicate that Dyn has a modulatory role at this level and controls the release of some more abundant neu- rohormones or neurotransmitters. Sterospecific high affinity opiate binding

Vol. 32, No. 3, 1983 Dynorphin in Adrenomedullary Cells 293

sites have been identified in membrane preparations of bovine adrenal medulla (32-34) and Dyn was shown to be the most potent opioid peptide in inhibiting the nicotine-evoked release of catecholamines from isolated adrenal chromaffin cells (34). However, no correlation could be made between this inhibitory ef- fect and the binding characteristics of the stereospecific high affinity opiate receptor (34,35). Therefore, the inhibition obtained with Dyn was not blocked by naloxone and it was mimicked by dextrorphan, an inactive o~iate enantiomer. On the other hand, Dyn has been found to specifically stimule K opiate receptors (36). The specific stimulation of some putative < opiate receptors at this lev- el may regulate other functions of adrenomedullary cells such as the synthesis, storage or the release of catecholamines or of some other neurotransmitters or peptides. Whether the adrenal opiate receptor is Dyn-specific and whether Dyn at this level has a modulatory or hormonal function are questions which remain to be answered and which constitute the subject of our present investigations.

Acknowledgements

We thank Mrs. Paulette Mercier for technical assistance and Mrs. Danielle Lau- rendeau for typing the manuscript. This work was supported by Grant PG-20 to SL from the Medical Research Council of Canada, the Canadian Heart Foundation and "La Fondation des Maladies du Rein". M.D. and R.D. are recipients of F.C. A.C. and I.R.S.S.T. studentships.

References

1.

2

3

4

5

6

7 8.

9

I0

ii

12

13

14

15

16

17.

A. GOLDSTEIN, S. TACHIBANA, L.I. LOWNEY, M. HUNKAPILLER and L. HOOD, Proc. Natl. Acad. Sci. 76, 6666-6670 (1979). B.R. SEIZINGER, V. HOLLT and A. HERZ, Biochem. Biophys. Res. Commun. 103, 256-263 (1981). V. HOLLT, J. HAARMAN, K. BOVERMANN, M. JERLICZ and A. HERZ, Neurosci. Lett. 1__88, 149-153 (1980). A. GOLDSTEIN and V.E. GHARZAROSSIAN, Proc. Natl. Acad. Sci. 77, 6207-6210 (1980). M.J. MILLAN, Y.F. TSANG, R. PREWLOCKI, V. HOLLT and A. HERZ, Neurosc i . Lett. 24, 75-79 (1981). S.J. WATSON, H. AKIL, V.E. GHAZAROSSIAN and A. GOLDSTEIN, Proc. Natl. Acad. Sci. 78, 1260-1253 (1981). S. TACHIBANA, K. ARAKI, S. OHYA and S. YOSHIDA, Nature 295, 339-340 (1982). D. DENIS, R. DAY and S. LEMAIRE, Int. J. Peptide Prot. Res. 19, 18-25 ( 1 9 8 2 ) . R. DAY, D. DENIS, J . BARABE, S. ST-PIERRE and S. LEMAIRE, I n t . J . P e p t i d e P r o t . Res . 19 , 10-17 ( 1 9 8 2 ) . M. SCHULTZBERG, T. HOKFELT, M.J . LUNDBERG, L. TERENIUS, L.G. ELBIN and R. ELDE, A c t a P h y s i o l . Scad . 103, 4 7 5 - 4 7 ? ( 1 9 7 8 ) . M. SCHULTZBERG, J .M. LUNDBERG, T. HOKFELT, L. TERENIUS, R.P. BRANDT and R. ELDE, N e u r o s c i e n c e ~ , 1169-1186 ( 1 9 7 8 ) . O.H. VlVEROS, E . J . DILEBERTO, E. HAZUM and K . - J . CHANG, Mol. Pha rmac . 1 6 , 1101-1108 ( 1 9 7 9 ) . A.S. STERN, R.V. LEWIS, S. KIMURA, J . ROSSIER, L.D. GERBER, L. BRINK, S. STEIN and S. UDENFRIEND, P r o c . N a t l . Acad. S c i . 7 6 , 6 6 8 0 - 6 6 8 3 ( 1 9 7 9 ) . R.V. LEWIS, A.S . STERN, J . ROSSIER, S. STEIN and S. UDENFRIEND, Biochem. B i o p h y s . Res . Commun. 8 9 , 822-829 ( 1 9 7 9 ) . A.S . STERN, R.V. LEWIS, S. KIMURA, J . ROSSIER, S. STEIN and S. UDENFRIEND, A r c h . Biochem. B i o p h y s . 205, 606 - 613 ( 1 9 8 0 ) . S. KIMURA, R.V. LEWIS, A.S . STERN, J . ROSSIER, S. STEIN and S. UDENFRIEND, P r o c . N a t l . Acad. S c i . 77_?, 1681-1685 ( 1 9 8 0 ) . B.N. JONES, A.S. STERN, R.V. LEWIS, S. KIMURA, S. STEIN, S. UDENFRIEND and J . E . SHIVELY, A r c h . Biochem. B i o p h y s . 204 , 592-595 ( 1 9 8 0 ) .

294 Dynorphin in Adrenomedullary Cells Vol. 32, No. 3, 1983

18. D.L. KILPATRICK, T. TANIGUCHI, B.N. JONES, A.S. STERN, J.E. SHIVELY, J. HULLIHAN, S. KIMURA, S. STEIN and S. UDENFRIEND, Proc. Natl. Acad. Sci. 78, 3265-3268 (1981).

19. K. MIZUNO, N. MINAMINO, K. KANGAWA and H. MATSUO, Biochem. Biophys. Res. Commun. 95, 1482-1488 (1980).

20. K. MIZUNO, N. MINAMINO, K. KANGAWA and H. MATSUO, Biochem. Biophys. Res. Commun. 97, 1283-1290 (1980).

21. M. NODA, Y. FURUTANI, H. TAKAHASHI, M. TOYOSATO, T. HIROSE, S. INAYAMA, S. NAKANIOHI and S. NUMA, Nature 295, 202-206 (1982).

22. V. GUBLER, P. SEEBURG, B. HOFFMAN, L.P. GAGE and S. UDENFRIEND, Nature 295, 206-208 (1982).

23. M. COMB, P.H. SEEBURG, J. ADELMAN, L. EIDEN and E. HERBERT, Nature 295, 663-666 (1982).

24. S. LEMAIRE, G. DEROME, R. TSENG, P. MERCIER and I. LEMAIRE, Metabolism 30, 462-468 (1981).

25. Y. MIURA, V. CAMPESE, V. DEQUATTRO and D. MEIJER, J. Lab. Clin. Med. 89, 421-427 (1977).

26. H. WINKLER, H. HORTNAGL and A.D. SMITH, Biochem. J. 118, 303-310 (1970). 27. O.H. LOWREY, N.J. ROSEBROUGH, A.L. FARR and R.J. RANDALL, J. Biol. Chem.

193, 265-270 (1951). 28. B. LIVETT, R. DAY, R. ELDE and P.C. HOWE, Neuroscience ~, 1323-1332 (1982). 29. S.M. STINE, H.Y.T. YANG and E. COSTA, Neuropharmacol. 19, 683-685 (1980). 30. B.G. LIVETT, D.M. DEAN, L.G. WHELAN, S. UDENFRIEND and J. ROSSIER, Nature,

289, 317-319 (1981). 31. G. DEROME, R. TSENG, P. MERCIER, I. LEMAIRE and S. LEMAIRE, Biochem.

Pharmac. 30, 855-860 (1981). 32. C. CHAVKIN, B.M. COX and A. GOLDSTEIN, Mol. Pharmacol. IS, 751-753 (1979). 33. K. KUMAKURA, F. KAROUM, A. GUIDOTTI and E. COSTA, Nature 283, 489-492. 34. S. LEMAIRE, B. LIVETT, R. TSENG, P. MERCIER and I. LEMAIRE, J. Neurochem.

36, 886-892 (1981). 35. D.M. DEAN, S. LEMAIRE and B. LIVETT, J. Neurochem. 38, 606-614 (1982). 36. C. CHAVKIN, I.F. JA~,~S and A. GOLDSTEIN, Science 215, 413-415 (1982). 37. S. LEMAIRE, A. BERUBE, G. DEROME, I. LEMAIRE, J. MAGNAN, D. REGOLI and

S. ST-PIERRE, J. Med. Chem. 21, 1232-1235 (1978).