Neuropeptides 4 : 175-182, 1984

SIMULTANEOUS EVALUATION OF THE CATECHOLAMINE PATHk!AY AND THREE OPIOID PEPTIDEpPRODUCINC SYSTEMS IN HUMAN PHEOCHROMOCYTOMAS

F. Cesselin, L. Pique: X. Bertagnat C. Benlct, J. Antreassian M.F. Prceschelt* F. Girard$+F. Zcgbi, J.C. Legrand, J.P. LutcAf J. DauchyXand J. Thibault:

Service de Bicchimie Medicale, C.H.U. Pitie-Salpe"triere ; +Labcratcire de la Clinique des Maladies Endccriniennes et Metabcliques, C.H.U. Ccchin ;++Service d'Explcraticns Fcncticnnelles, Hcpital Trousseau ;xLabcratcire de Bicchimie, Hbtel-Dieu ; 'Labcratcire de Biclcgie Mcleculaire, College de France ; Paris, France. (Reprint requests to F.C. 91, bd de l'hdpital, 75634 Paris Cedex 13)

ABSTRACT

Tyrcsine hydrcxylase (TH) and dcpamine B-hydrcxylase (DBH) activities, dopemine (DA), noradrenaline (NA), adrenaline (A), met 5-enkephalin (Met-Enk), leu -enkephalin (Leu-Enk), dyncrphin (Dyn) and B-endcrphin (B-end) were measu- red simultaneously in ten human phecchrcmccytcmas =

1 - TH activity was highly variable, from 22 to 2220 U/g tissue. 2- DBH activity, in contrast, was rather constant, from 96 to 582 U/g. 3 - Catechclamines (A and NA) concentrations showed only small variations. 4 - The four cpicid peptides were detected in all cases and exhibited a

wide range of tissue concentrations (Enk ) Dyn > B-end). 5 - Met-Enk and Leu-Enk concentrations were highly correlated ; no ccrre-

laticn was observed with the other cpicid peptides. 6 - A very strong correlation was observed between enkephalins concentrations

and both catechclamines concentrations and DBH activities.

These results are discussed in term of the significance of the cc-lccali- zaticn of these various biologically active substances, principally with regard to the possible regulation of catechclamine synthesis by cpicid peptides and conversely.

INTRODUCTION

Phecchrcmccytcmas are rare tumors which develop from the adrenal medulla. It is considered that catechclamine (CA) overproduction and secretion by the tumcral chrcmaffin cells are responsible for the clinical features through the peripheral effect of the bicgenic amines on their target cells (1).

Different cpicid peptides related to the enkephalins (Enk), Dynorphins (Dyn)

175

and B-endorphin (B-End) have been detected in adrenal medulla as well as in human pheochromocytomas (2-10). Three different precursor systems have been fully characterized which generate endogenous opioid peptides = pro-enkephalinA for Met-Enk, Leu-Enk, Met-Enk- A,rs6 - Phe7 and Met-Enk-Arg6-Gly7-Leu8 (11) ; pro-enkephalin B for 6-neo-endorphin and Dyn (12) and pro-opiomelanocortin (POMC) for B-End (13).

The present study was set up to allow the simultaneous evaluation of the CA pathway and that of the 3 opioid peptide precursor systems in 10 human pheochromocytomas. All correlations between the different systems were tested. The results are discussed in view of the significance of the co-localization of these various biologically active compounds in the same tumor cells

MATERIAL AND METHODS

Tumors were obtained at surgery in ten patients with unequivocal biological criteria of catecholamine hypersecretion. The pheochromocytomas were immedia- tely cut into small (10 X 10 mm) pieces which were frozen in liquid nitroaen and stored at -80°C. Each tumor was subsequently ground at -80°C in a Waring blendor to produce ahomogeneous powder ; when needed an aliquot of the powder, about lg, was homogenized in three volumes of 0.005 M phosphate buffer, pH 6.5, in a conical glass-glass homogeniser ; aliquots of the homogenate were used variably : for enzyme extraction the crude homogenate was diluted 10 fold with 20 ml of phosphate buffer ; for peptide and catecholamine determinations the crude extract was made one molar by the addition of concentrated HCl, kept on ice for one hour, centrifuged at 10,000 g for 20 min, and the super- natant was kept frozen.

The Tyrosine Hydroxylase (TH) assay was that of Lewitt et al (14) using 60 PM tyrosine and 720 VM 6-7 dimethyl 5,6,7,8 tetrahydropteridin. In our conditions 1 unit of TH corresponds to 1 nmole of L-Dopa formed per hour at 37°C. The assay was performed on crude tumor extract, lg in 40 ml 0.02 M phosphate buffer, pH 6.5.

For the Dopamine B-hydroxylase (DBH) assay we used the periodate cleavage of B-hydroxy-phenyl ethyl amines according to Wallace et al (15) with the follo- wing modifications : after the addition of sodium bisulfite, the parahydroxy- benzaldehyde formed during the reaction was extracted with 2 ml butanol and its concentration was measured by a spectrophotometer at 330 nm. One unit DBH corresponded to 1 nmole parahydroxybenzaldehyde formed per hour. DA, NA and A were assayed using the method of Jackman et al (16).

The Met5-enkephalin (Met-Enk) radioimmunoassay (PJA) was performed as previoutly described (17). The cross reactivity of Met-Enk-sulfoxide and Met- Enk-Arg -Phe7 were 360% and 19% on a molar basis, respectively. Leo-Enk, Tyr- Gly-Gly-Phe and Gly-Gly-Phe-Met tetrapeptides, and human B-endorphin (hBEnd) showed less than 0.4% cross reactivity.

For the Leu5-enkephalin (Leu-Enk) RIA, the anti-Leu-Enk antiserum (final dilution 1 = 6,000), the suitably diluted biological extracts, and 3 nCi of l251-Leu-Enk were incubated for 16-18h at 4°C in 0.0125 M Tris-HCl buffer, pH 7.6, containing 2.5 mg/ml lysoz me in a final volume of 0.3 ml. The assay was stopped by precipitating the IT5I-Leu-Enk-antibody complex, adding succes- sively 0.1 ml of diluted (1:,10) horse serum and 1 ml of propanol. Under these

176

conditions the detection limit of the PIA was 2.5 pg of Leu-Enk. The cross reactivities of peptides related to Leu-Enk were as follows : 3.1% for Bet-Enk. 0.8% for Leu-Enk-Arg and Leu-Enk-Lys, and 0.5% for hBEnd. Other related com- pounds cross reacted less than 0.01 X : Cly-Gly-Phe-Leu, Tyr-Gly-Gly-Phe, Tyr- Gly-Gly, Gly-Gly-Phe, Gly-Gly, Gly-Phe-Leu, Phe-Leu, Leu-Enk-NH2, Dvnorphins 1-8, 1-9, 1-13 and l-17, c1 neo-endorphin. Immunoreactive (IR)-Met-Ln;. and Leu-Enk were expressed as Met-Enk and Leu-Enk equivalents respectively in rig/g of tissue.

The hBEnd RIA was performed as previously described (18). Briefly it used antibody R 2489 directed toward the COOH-terminal end of the molecule. Purified hB-lipotropin(h81PlJ) cross-reacted 100X, purified h‘CILPH showed 0.5% cross reactivity, synthetic hLPH (37-58), hACTH, Leu-Enk. Met-Enk. Dynorphin (1-13) showed no significant cross reactivity.

The Dynorphin RIA used antiserum Lucia, kindly provided by Dr. Goldstein (Palo Alto, Ca, USA) as already described (19). Leu-Enk, Met-Enk, and hBEnd showed no significant cross reactivity.

RESULTS AND DISCUSSION

We report the tissue contents of CA, their key synthesizing enzymes and four opioid peptides in 10 pheochromocytomas obtained at surgery in man (Tables I and II). Although all these compounds have already been detected in human pheochromocytomas (2-7, 9, lo), to our Ikno~!le?le their simultaneous measurement has not yet been reported. With the recent demonstration that opioid peptides were present in adrenal medulla and that they might be involved in the regulation process of CA synthesis (213, 21), we decided to set up a strategy for a simultaneous evaluation of the two systems so that correlations could be adequately looked for = a single person collected all tumors, and tissue was ground into a homogeneous powder so that identical aliquots were used for the various extractions.

TABLE 1 Tumor levels of Tyrosine-hydroxylase (TH), Dopamine 6 hydroxylase (DBH), Dopamine (DA), Noradrenaline (NA) and Adrenaline (A)

TUMOR NUMBER TH x DBH x DA 5 NA 0 A 5

1 2

;

Z 7 8 9

10

391 22

Z 2220

;; 641 324 122

221

*Zi 262 508 278 406 365 582 451

0.0423 2.73 8.87 0.0317 2.20 0.04 0.0227 1.60 0.004 0.0339 1.23 0.24 0.0444 2.43 12.46 0.0363 1.19 5.31 0.0340 7.12 0.78

ND ND ND 0.0742 9.62 4.66

ND ND ND

x : unit/g 5 mg/g tissue . ND : not determined

177

Table II

Tumor levels of Met-Enkephalin (Met-Enk), Leu-Enkephalin (Leu-Enk), Dynorphin (Dyn), P Endorphin (B End)

TUMOR NUMBER Met-EnkX Leu-EnkX DynX

1 3440 1170 4.3

: :: 381 378 31.6 2.8 4 3077 531 27.2

2 26121 14470 2874 2953 1:*: 7 80587 2670 15:7 8 4520 432 2.2

1: 956700 13984 12045 2250 11.6 4.1

B EndX

6.4

0.5 0.3 3.8

1;:;

:*:

14:5 6.8

X rig/g tissue

The activities of the CA system was evaluated by measuring TH, DBH, DA, NA and A contents (Table I). Fowopioid peptides were measured (Table II) : Met-Enk and Leu-Enk which both derive from pro-Enkephalin A (ll), Dyn which derives from pro-Enkephalin B (12) and B-End which derives from pro-opiomela- nocortin (13), thus realizing an overall estimation of all known opioid systems. All the correlations which could be established are reported in table III.

Table III

Correlations between DBH activity, catecholamines and opioid peptides concen- trations in the pheochromocytomas (rank coefficient (r) of Spearman) (30)

VARIABLES r F<

1

Met-Enk Met-Enk Met-Enk Met-Enk Met-Enk

Leu-Enk DBE

2

Leu-Enk 0.96 0.001 DA+NAtA 0.81 0.05

NAtA 0.74 0.05 DBH 0.88 0.01

DBH 0.88 0.01

NAtA 0.88 0.01

-

178

TH is usually considered the key regulator enzyme for CA synthesis (22) ; others, however, have reported that the availability of the pteridine cofactor may be a limiting factor of its regulation (23). TH activities were highly variable, being 100 fold higher in tumor 2 compared to tumor 5, for example. DA concentrations, in contrast, were similar in all tumors ; although DA is a competitive inhibitor of the pteridine cofactor of TH,its possible interference with the TH assay seems unlikely since all tumors had almost identical DA concen trations. Thus, the low TH values of tumor 2, for example, most likely reflects a true TH deficiency.

In contrast with TH, DBH activity kept a more constant value between the tumors indicating that the 2 enzymes have different regulatory mechanisms in pheochromocytomas.

The IR-Enk concentrations were hiqhly variable (from 51.2 to 956,7CO and 378 to 12,D45 rig/g for IR-Met-Enk and IR-Leu-Enk. respectively). This has been reported by others who found values in the same order of magnitude as ours (2-5,7,9,10). Only tumor 9 had IR-Enk concentrations which largely exceeded (by more than 10 fold) the highest level measured in the others as well as those previously reported (2-5,7,9,1@). Assaying different areas of apparently healthy tissue of the same tumor, Wilson et al (5) reported a 4 fold variation in the opioid peptide concentrations. In our procedure, however, great care was given to the preparation of ahomogeneous powder so that the observed variations most likely reflected true differences between the various pheochromocytomas. Few data are available on Enk content of normal adrenal medulla in man. In an extract of one human adrenal, Clement-Jones et al (24) reported an IR-Met-Enk concentration of BOO rig/g.. We had the opportunity to measure IR-Met-Enk in a normal adrenal medulla and found a lower value of 35 rig/g (unpublished). Levels cf receptor-active opioid peptides ranging from 575 to 1,BGO rig/g were reported in post-mortem adrenal medulla from patients whose death followed prolonged disease (25). Thus, the mean IR-Met-Enk concentrations of the tumors reported here (16,256 + 8,534 ngjg, tumor 9 excluded) is much higher than those values. More interesting is the observation that tumor concentrations are extremely variable and in some cases (tumor 2 and 3) not higher than in normal.

A significant correlation was found between the levels of IR-Met and Leu-Enk. This was not unexpected since the two pentapeptides should have a common origin from pro-enkephalin A in the adrenal medulla (11). Although the IR-Net-Enk/IR-Leu-Enk ratio was most often close to the expected theoretical value, it was strikingly high in tumor 9 and low in tumors 2 and 3. In these two latter cases, the question arise wether excess IR-Leu-Enk (relative to IR-Met-Enk) could result from its production through another opioid system, namely pro-enkephalin B (12).

This hypothesis was tested by measuring Dyn which derives only from pro- Enkephalin B. Although we did find that Dyn had its highest level in tumor 2, it was at its lowest in tumor 3 ; furthermore no correlation could be found between IR-Dyn and IR-Leu-Enk concentrations in the 10 pheochromocytomas. Therefore another mechanism is probably responsible for the observed discre- pancies : altered processing of pro-enkephalin A resulting in the preferential release of one pentapeptide, or in the formation of an unusual peptide, remains an alternative explanaticn.

179

A fourth opioid peptide, hBEnd, was also measured in the tumor extracts. Although this assay cross reacts 100% with hB-LPH, which has no opioid activity, we had previously shown that the overall IR-hBEnd extracted from pheochromocytomas has gel filtration characteristics similar to if not identital with hBEnd (7) and that no hBLPH could be detected. hBEnd is known to derive from a third opioid peptide precursor system, namely pro-opio-melano- cortin (13). The fact that hBEnd concentrations strongly correlated with hXLPH concentrations, another pro-opiomelanocortin derived fragment (not shown) was good evidence that it actually reflected the local synthesis of pro-opiomela- nocortin which thus appears to be present in all pheochromocytomas. It is known that pheochromocytomas may be responsible for the ectopic ACTH-LPH syndrome (25). Our data would suggest that this phenomenon will probably occur whenever pro-opiomelanocortin is synthesized in large quantities and its proces- sing is directed toward the release of a biologically active ACTH fragment .

Since pheochromocytomas probably result from the growth of a single clone of tumoral cells (l), our data would provide the first evidence that the 3 known precursor systems for opioid peptides can be expressed simultaneously in the same cell. The fact that no correlation could be found between them confirms that they derive from 3 different structural genes with probably no common regulatory mechanism. Lastly, the high tissue concentrations of IR-Met and Leu-Enk compared to Dyn and hBEnd is indicative of a higher expression rate of the pro-enkephalin A gene in pheochromocytomas.

It has been postulated that feed back inhibition of TH might be lacking or deficient in pheochromocytomas (1) resulting in CA overproduction. Since CA (and CA agonists) have been reported to increase the opioid content of bovine adrenal chromaffin cells and also of dispersed cells from a human pheochromo- cytoma (27), this mechanism would explain why most pheochromocytomas contained high levels of IR-Enk and is supported by the observed correlation between CA and IR-Enk tumor concentrations (Table III). It was reported that, in bovine and hamster adrenal medulla, enkephalins are restricted to the adrenaline cells (28,29). It is noteworthy that the two tumors with the lowest A levels (cases no 2 and 3) also contain the smallest (i.e. "normal") amount of IR-Enk. However, no correlation was found between IR-EnK levels and A levels.

IR-Enk levels were strongly correlated with DBH activities, but not with TH activities. Morphine administration in animals provokes an induction of both DBH and TH when the adrenal qland is intact, but only of DBH when the gland is Idenervated (20,21). Assuming that endogenous opioid peptides could provoke a similar enzymatic induction as the opioid alcaloid, the observed correlation between IR-Enk and (only) DBH activity in the tumors could result from this mechanism since pheochromocytomas are non-innervated tumors (1). This pheno- menontiould induce a relative DBH excess compared to the other enzymatic activities ; it would explain why the enzyme substrate, DA, is low while the product, NA, is high ; it would also be consistent with the observed correla- tion between DBH and A t NP.!TaSle III)

Acknowledgements

.l';iis work was supported by grants frcm IWEWI (CRL 816005, 816010, 816029). We are grateful to Mrs Falcy (Pr. Fabiani lab.) who performed the catecholami- ne assays .

180

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

REFERENCES

Manger, W.H. and Gifford, R.W. Jr (1977) . Pheochromocytoma - Springer-Ver- lag, New York .

Sullivan, S.N., Bloom, S.R. and Polak, J.M. (1978) . Enkephalin in periphe- ral neuroendocrine turnouts . Lancet I : 986-987 .

Clement-Jones, V., Corder, R. and Lowry, P.J. (1980) . Isolation of human met-enkephalin and two groups of putative precursors (2K-pro-met-enkepha- lin) from an adrenal medullary tumour . Biochem. Biophys.Res.Commun. 95 : 665-673 .

Giraud, P., Eiden, L.E., Audigier, Y., Gillioz, P., Conte-Devolx, B., BOU- douresque, F., Eskay, R. and Oliver,C. (1981). Enkephalins, ACTH, MSH and B endorphin in human pheochromocytomas . Neuropeptides 1 : 237-252 .

Wilson, S.P., Cubeddu, L.X., Chang, K.J. and Viveros, O.H.(1981) . Met- enkephalin, leu-enkephalin and other opiate-like peptides in human pheo- chromo-cytoma tumors . Neuropeptides 1 : 273-281 .

Yoshimasa, T., Nakao, K., Oki, S., Tanaka, I., Nakai, Y. and Imura, H. (1981) . Presence of dynorphin-like immunoreactivity in pheochromocytomas. J. Clin. Endocrinol. Metab. 53 : 213-214 .

Bertagna, X., Pique, L., Ochoa, C., Luton, J.P., Bricaire, H., Serin, D., Girard, F., Plouin, P.F., Corvol, P., Cesselin, F. and Hamon, M. (1982) . Simultaneousmeasurementof B-endorphin, lipotrophins and met-enkephalin in pheochromocytomas . Acta Endocrinol. 101 : 72-77 .

Day, R., Denis, D., Barabe, I., Saint-Pierre, S. and Lemaire, S. (1982) . Dynorphin in bovine adrenal medulla . Int. J. Peptide Protein Res. 19 : IO- 17.

Giraud P., Castanas, E., Oliver, C., Conte-Devolx, B. Boudouresque, F., Taquet, H., Orlando, M., Jaquet, P., Eiden, L., Cesselin, F., Guns, G. and Trigano, M. (1982) . Enkephalins in human pheochromocytoma-Biosynthesis and secretion . Eur. Heart J. 3, supple : I?-22 .

Cubeddu, L.S., Wilson, S.P. and Viveros, O.H. (1983) . Opioid peptides and catecholamines in human pheochromocytoma . Life Sci 33 : 871-877 .

Comb, M., Seeburg, P.H., Adelman, J., Eiden, L. and Herbert, E. (1982) . Primary structure of the human met-and leu-enkephalin precursor and its m RNA. Nature 295 : 663-666 .

Kakidani, H., Furutani, Y., Takahaski, H., Noda, M., Morimoto, Y., Hirose, T., Asai, M., Inayama, S., Nakanishi, S. and Numa, S. (1982) . Cloning and sequence analysis of cDNA for porcine B-neo-endorphin/dynorphin precursor . Nature 298 : 245-249 .

Nakanishi, S., Inoue, A., Kita, T., Nakamura, M., Chang, A.C.Y., Cohen, S.N and Numa, S. (1979) . Nucleotide sequence of cloned cDNA for bovine corti- cotropin-B-lipotropin precursor . Nature 278 : 423-427.

Lewitt, M., Gibb, J.W., Daly, J.W., Lipton, M. and Udenfriend. S. (1966) . A new class of tyrosine-hydroxylase inhibitors and a simple assay of inhi- bition "in vivo" . Biochem. Pharmac. 16 : 1313-1321 .

Wallace, E.F., Krantz, M.J. and Lovenberg, W. (1973) . Dopamine B hydroxy- lase = a tetrameric glycoprotein . Proc. Natl. Acad. Sci USA . 78 : 2253- 2255 .

181

16. Jackman, G.P., Carson, V.J., Bobik, A. and Skews, H. (1980) . A simple and sensitive procedure for the assay of serotonin and catecholamines in brain by high performance liquid chromatography using fluorescence detection . J. Chromatogr. Biomed. Appl. 182 : 227-284 .

17. Cesselin, F., Soubrie, P., Bourgoin, S., Artaud, F., Reisine, T., Michelot R., Glowinski, J. and Hamon, M. (1981) . In vivo release of met-enkephalin in the cat brain . Neuroscience 6 : 301-313 .

18. Bertagna, X., Nicholson, W.E., Sorenson, G.D., Pettengill, O.S., Mount, C.D. and Orth, D.N. (1978) . Corticotropin, lipotropin, and B-endorphin produc- tion by a human non-pituitary tumor in culture : evidence for a common pre- cursor . Proc. Natl. Acad. Sci. USA 75 : 5160-5164 .

19. Ghazarossian, V.E., Chavkin, C. and Goldstein, A. (1980) . A specific radio- immunoassay for the novel opioid peptide dynorphin . Life Sci. 27 : 75-86 .

20. Patrick, R.L. and Kirshner, N. (1971) . Acetylcholine-induced stimulation of catecholamine recovery in denervated rat adrenals after reserpine . Mol. Pharmac. 7 : 389-396 .

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Anderson, T.R. and Scotkin, T.A. (1976) . The role of neural input in the effects of morphine on the rat adrenal medulla . Biochem. Pharmac. 25 : 1071-1074 .

Mandell, A.J. (1978) Redundant mechanisms regulating brain tyrosine and tryptophan hydroxylases . Ann. Rev. Pharmac. Toxicol. 18 : 461-493 .

Abou-Donia, M.M. and Viveros, H.O. (1981) . Tetrahydrobiopterin increases in adrenal medulla and cortex : a factor in the regulation of tyrosine hy- droxylase . Proc. Natl. Acad. Sci. USA. 78 : 2703-2707 .

Clement-Jones, V., Lowry, P.J., Rees, L.H. and Besser, G.M. (1980). Met- enkephalin circulates in human plasma. Nature 283 : 295-297 .

Saria, A., Wilson, S.P., Molnar, A., Viveros, O.H. and Lembeck, F. (1980) . Substance P and opiate-like peptides in human adrenal medulla . Neurosci. Lett. 20 : 195-200 .

Luton, J.P., Thieblot, P. and Bricaire, H. (1977) . Association syndrome de Cushing-pheochromocytome . Nouv. Presse Med. 43 : 4053-4057 .

Wilson, S.P., Chang, K.J. and Viveros, O.H. (1981) . Opioid peptide synthe- sis in bovine and human adrenal chromaffin cells. Peptides 2, Suppl. 1 : 83-88 . Livett, B.G., Day, R., Elde, R.P. and Howe, P.R.C. (1982) . Co-storage of enkephalins and adrenaline in the bovine adrenal medulla . Neuroscience 7 : 1323-1332 .

Pelto-Huikko, M., Salminen, T. and Hervonen, A. (1982) . Enkephalin-like immunoreactivity is restricted to the adrenaline cells in the hamster adre- nal medulla . Histochemistry 73 : 493-497

Gibbons, J.D. (1971) . Nonparametric statistical inference . MC Graw Hill, New York .

Received 16/12/83 Accepted 02/01/84

182