ACTH and Prostaglandin Receptors in Human Adrenocortical Tumors
APPARENT MODIFICATION OF A SPECIFIC COMPONENT OF THE ACTH-BINDING SITE
JOSÉ M. SAEZ, ALICE DAZORD, and DOMINIQUE GALLET From Unité de Recherches Endocriniennes et Métaboliques chez l’Enfant. I.N.S.E.R.M., Hôpital Debrousse, 69322 Lyon Cedex 1, France
ABSTRACT The failure of certain adrenal tumors to respond to ACTH was investigated in vivo by ad- ministration of corticotropin-(1-24)-tetracosapeptide (ACTH1-2) and dexamethasone and in vitro by study- ing the binding properties of ACTH1-2 and prosta- glandin EI (PGE1) and their effect on adenylate cyclase activity of the tumors’ crude membranes ; in addition, in five cases the stimulation of cortisol production in iso- lated adrenal cells by both hormones and dibutyryl cy- clic adenosine 3’,5’-monophosphate (cAMP) was also studied. The results obtained in 13 hormone-producing tumors of the human adrenal cortex, i.e. 10 carcinomas and 3 adenomas, were compared with those found in normal human adrenal glands.
According to the adenylate cyclase responses to ACTH1-24 and PGE1, the tumors fall into different cate- gories. In the first group are six tumors in which the adenylate cyclase was stimulated by both ACTH1-24 and PGE1; in addition specific binding could be demon- strated for the two hormones in all six. The binding affinity for 13I-ACTH1-2 was found to be about 10 times higher than that for 12I-ACTH11-24. In the one tumor in which the experiment was performed, bound 13%I- ACTH1-2 was displaced by ACTH1-10. These results are similar to the ones found in normal human adrenal preparations. For two tumors of the group in which ACTH did not increase steroidogenesis in vivo, the biochemical abnormality might be located beyond cAMP formation.
A preliminary report of this work was presented at the 55th Meeting of the Endocrine Society, June 1973.
Received for publication 10 May 1974 and in revised form 22 May 1975.
A second group encompasses six tumors in which the steroidogenesis in vivo and the adenylate cyclase ac- tivity were insensitive to ACTH1-2 but in which the en- zyme was stimulated by PGE1 and NaF. However, these preparations bound 13I-ACTH1-24 and 125I-ACTH11-24, the binding affinity being similar for both peptides but 10 times lower than the one found in normal adrenal cor- tex for 125I-ACTH1-2. In the only case of this group where it was tested, ACTH1-10 did not displace bound 125I-ACTH1-2. This result strongly suggests the possi- bility of a modification or a loss of the receptor site that binds the N-terminal sequence (1-10) of ACTH, the biologically active part of the molecule.
In the last tumor, both PGE and ACTH were unable to stimulate adenylate cyclase activity and steroid pro- duction in a preparation of isolated adrenal cells, al- though steroidogenesis was stimulated by dibutyryl cAMP. No specific binding for PGE1 could be demon- strated. However, 125I-ACTH1-2 and 12I-ACTH11-24 were found to be bound to the tumor with the same affinity.
INTRODUCTION
It is well known that in most of the human adrenocor- tical tumors ACTH does not stimulate steroidogenesis (1). The biochemical anomaly responsible for this fail- ure of stimulation of steroidogenesis by ACTH is still unknown. The same ACTH insensitivity has also been shown in vivo and in vitro for several types of adreno- cortical tumors of the rat. In the tumor first described by Snell and Stewart (2), the main biochemical ab- normality seemed to be located beyond the formation of cyclic AMP (3) since neither ACTH nor cyclic AMP could increase steroidogenesis (4 6), although adenylate
| Patients | Sex | Age | Etiology | Tumor wt | Basal | Dexamethasone* | ACTHỊ | |||
|---|---|---|---|---|---|---|---|---|---|---|
| 17-KS | 17-OCHS | 17-KS | 17-OCHS | 17-KS | 17-OCHS | |||||
| yr | g | mg/24 h | mg/24 h | mg/24 h | ||||||
| 1 | F | 115 | R. carcinoma§ | 165 | 24-28 | 1.1-1.7 | 26 | 1.3 | - | - |
| 2 | M | 6 | L. carcinoma | 70 | 18 | 1.4 | 20 | 1.4 | - | - |
| (recurrence) | ||||||||||
| 3 | F | 7 | R. carcinoma | 320 | 50 | 25 | ||||
| 4 | F | 18 | L. carcinoma | 170 | 400-600 | 9-12 | 399 | 0.5 | 540 | 5 |
| 5 | F | 36 | R. carcinoma | 210 | 157 | 8 | 185 | 30 | 180 | - |
| 6 | F | 38 | L. carcinoma | 1,350 | 94-159 | 47-91 | 84 | 45 | 114 | 59 |
| 7 | F | 39 | L. carcinoma | 750 | 200 | 43 | 134 | 19 | 210 | 47 |
| 8 | F | 47 | L. carcinoma | 400 | 216-270 | 24-28 | 206 | 24 | 215 | 30 |
| 9 | F | 51 | R. carcinoma | 510 | 39 | 4.8 | 31 | 2.5 | 25 | 3.9 |
| 10 | M | 54 | R. carcinoma | 1,815 | 150 | 19 | 127 | - | 202 | 33 |
| 11 | F | 7 | L. adenoma | 72 | 39-72 | 10-16 | 41 | 14 | 50 | 15 |
| 12 | F | 25 | R. adenoma | 80 | 65-75 | 5-10 | 69 | 3 | 72 | 22 |
| 13 | M | 53 | L. adenoma | 120 | 5-8 | 15-20 | 6 | 21 | 10 | 60 |
* After dexamethasone, 8 mg/day for 3 days.
# After Synacthen-Depot®, 1 mg/day for 2 days.
§ R., right; L., left.
cyclase activity in subcellular fractions was stimulated by ACTH (3). In other mutant adrenal cell lines stud- ied by Shimmer (7, 8) the anomaly was localized in the cell membrane since adenylate cyclase in adrenal sub- cellular preparations was not stimulated by ACTH, but cyclic AMP was still able to stimulate steroidogenesis.
The purpose of this work was to establish whether in human adrenocortical tumors abnormalities of the ACTH receptors on the plasma membrane could be demonstrated. Therefore, we have studied the interaction of ACTH and also of prostaglandin E1 (PGE1)1 with subcellular fractions prepared from 13 tumors, and the results have been compared to those obtained with the same type of preparations from normal human adrenals.
METHODS
Materials. 13 human adrenocortical tumors were used for this study (10 carcinomas and 3 adenomas). The type and weight of the tumors as well as the output of 17- ketosteroids (17-KS) and 17-hydroxycorticosteroids (17- OHCS), measured by the method described by Few (9)
1 Abbreviations and trivial names used in this paper: ACTH1-10, corticotropin-(1-10)-decapeptide; ACTH1-2, cor- ticotropin-(1-24)-tetracosapeptide; ACTH11-2, corticotropin- (11-24)-tetradecapeptide; androstenedione, androst-4-ene-3, 17-dione; cortisol, 118,17a,21-trihydroxypregn-4-ene-3,20-di- one; DcAMP, dibutyryl cyclic adenosine 3’,5’-monophos- phate; dehydroepiandrosterone (DHA), 36-hydroxyandrost- 5-en-17-one; DHAS, DHA sulfate; EGTA, ethylene glycol bis (B-aminoethyl ether) N,N’-tetraacetic acid; 17-KS, 17- ketosteroids; 17-OHCS, 17-hydroxycorticosteroids; PGE1, PGE,, prostaglandins E, and Ea, respectively.
and Glenn and Nelson (10), respectively, are shown in Table I. In all the patients, there were signs of virilization except in patient 10, who presented signs of marked femini- zation, and in patient 13, who did not present any clinical signs of excessive hormone secretion. Normal adrenal glands (confirmed histologically) were obtained from women during surgery for breast cancer. None of them had been given any previous antimitotic, hormonal, or X-ray therapy. A pool of 10 normal adrenal glands was used for all the binding and adenylate cyclase studies. The adrenal tumors were ob- tained at surgery.
Corticotropin-(1-24)-tetracosapeptide (ACTH1-2), corti- cotropin-(11-24)-tetradecapeptide (ACTH11-x), and corticot- ropin-(1-10)-decapeptide (ACTH1-10) were generously pro- vided by Drs. Rittel and Desaulles (Ciba-Geigy AG., Basel, Switzerland). Labeling of both ACTH’s with 125I was per- formed by the method described previously (11). PGE, was a gift from Dr. J. E. Pike, Upjohn Company, Kalamazoo, Mich. [H]PGE, (sp act 68.5 Ci/mol) was purchased from New England Nuclear, Boston, Mass. Microfine silica, Quso G-32, was obtained from Philadelphia Quartz Co., Philadel- phia, Pa. Other chemicals were of reagent grade.
Membrane preparation. Before homogenization, the me- dulla of normal adrenals and the necrotic portions of the tumor were removed. Normal adrenal or tumor tissue was homogenized in 20 mM Tris-HC1 buffer, pH 7.4, containing 0.25 M sucrose with a Potter-Elvejehm glass homogenizer (10 strokes) followed by filtration through two layers of gauze and 5 strokes of the loose pestle in a Dounce homoge- nizer. The homogenate was centrifuged twice at 800 g for 10 min. The sediment was collected, and the supernate was centrifuged at 20,000 g for 30 min. The pellet obtained at 20,000 g was washed with 20 mM Tris-HCl buffer, pH 7.4, centrifuged, and kept in small aliquots (crude membranes). In certain tumors, highly purified preparations of plasma membranes were made (and confirmed by electron micros- copy) by the method of Finn, Widnell, and Hofmann (12). This method was inapplicable to normal human adrenals be-
| Patients | Basal | Dexamethasone* | ACTH# | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DHAS | DHA | Andros- tene- dione | Cortisol | DHAS | DHA | Andros- tene- dione | Cortisol | DHAS | DHA | Ahdros- tene- dione | Cortisol | |
| ug/100 ml | pg/100 ml | μg/ 100 ml | ||||||||||
| 3 | - | 0.3 | - | 22 | 0.05 | 16 | ||||||
| 6 | 1,520 | 2.9 | - | 29 | 1,340 | 2.8 | 6.1 | 31 | 1,500 | 2.5 | 6 | 26 |
| 7 | 1,551 | 3.4 | 5.9 | 16 | 1,384 | 3.2 | 4.7 | 10 | ||||
| 10 | 960 | 1.9 | 0.9 | 28 | 910 | 1.3 | 0.8 | 22 | ||||
* After dexamethansone, 8 mg/day for 3 days.
# After Synacthen-Depot ®, 1 mg/day for 2 days.
cause of too low a yield (less than 0.1 mg membrane protein per gram of adrenal tissue). All the preparations were kept in liquid nitrogen and thawed shortly before use. In these conditions the binding capacity and the adenylate cyclase activity remained unaltered at least for 4 mo.
Binding studies. Measurement of the binding for [3H]- PGE1, 18%I-ACTH1-24, and 125I-ACTH11-24 to normal adrenal particulate fractions has been described in detail elsewhere (13, 14). Briefly, 0.25 ml of the membrane protein prepara- tion in 20 mM Tris-HC1 buffer pH 7.4 containing 1% albu- min and the labeled hormone ([3H]PGE, or 13I-ACTH) was incubated at 4℃. When equilibrium was reached (30 min for 12I-ACTH and 90 min for [&H]PGE1), the sample was layered over 1 ml of 20 mM Tris-HC1 buffer, pH 7.4, 0.25 M sucrose containing 2% albumin and centrifuged im- mediately at 50,000 g for 10 min at 0℃. The supernate was removed, and the pellet was retained. The pellet containing [3H]PGE, was dissolved in 0.1 ml of Soluene® (Packard Instrument Co., Inc., Downers Grove, Il1.) for 1 h at 60℃ and transferred with 0.4 ml of methanol in 10 ml Bray’s scintillation fluid. The pellet containing 125I-ACTH was counted in a well-type scintillation counter.
All binding determinations were performed in six repli- cates, three containing only the radioactive hormone and three containing the radioactive plus the unlabeled hormones (10 µg of PGE1 or 200 µg of ACTH1-24). The mean of the three latter figures, which is an estimate of nonspecific bind- ing, was subtracted in each instance from the average of the first three. In all the experiments the nonspecific binding represented 1-6% of the total binding for 125I-ACTH1-24 and 125I-ACTH11-24 and 8-12% of the total binding of [H]PGE1.
Assay for ACTH degradation. The degradation of 12%I- ACTH1-x was measured according to a method described previously (11). Briefly, after incubation of 13I-ACTH1-24 with the crude membrane preparation for 30 min at 4°C, the sample was layered over 2.5 ml of 20 mM Tris-HCI (pH 7.4) containing 0.25 M sucrose and 2% albumin and centrifuged at 50,000 g for 20 min at 0℃. Only the 1st ml of the supernate was aspirated and kept (unbound fraction). The pellet was washed once with 2 ml of the same buffer and centrifuged at 50,000 g for 20 min. The supernate was discarded and the pellet was resuspended in 0.5-1.0 ml of 1% acetic acid, shaken for 30 min at 22℃, and centrifuged at 50,000 g for 10 min. The supernate (bound fraction) was removed and neutralized with 1 N NaOH. About 70-80% of the radioactivity present in the pellet was recovered in the acetic acid fraction. The degradation of unbound and bound ACTH was measured by absorption of the labeled material
on microfine silica (Quso G-32) and from its ability to bind to fresh adrenal crude membranes. The results were ex- pressed as the percentage of the hormone that remains intact in relation to a control specimen that was incubated under the same conditions but without membranes.
Preparation of isolated cells. Isolated adrenal cells from normal adrenals and tumors were prepared according to a modification (15) of the method described by Sayers, Swal- low, and Giordano (16). An aliquot of the cell suspension stained with trypan blue showed more than 85% of viable cells. Steroid production by the isolated adrenal cells was measured after 2 h at 37℃ in 0.8 ml of Krebs-Ringer bi- carbonate containing 0.2% glucose and 4% albumin. ACTH or other substances were added in a volume of 0.1 ml of vehicle (0.9% sodium chloride, pH 3, for ACTH and di- butyryl cyclic AMP [DcAMP] and 0.9% sodium chloride containing 0.1% ethanol for PGE1). The mixture was gassed with 95% oxygen and 5% CO2 and incubated in a Dubnoff metabolic shaker.
Assay of steroids. Cortisol was measured by a competi- tive protein-binding method (17). Testosterone, androstene- dione, and dehydroepiandrosterone (DHA) after separation and purification on Celite columns (Johns-Manville Products Corporation, New York) were measured by a sensitive radio- immunoassay (18). The first steps of purification of DHA sulfate (DHAS) were performed as described (19). After solvolysis, the compound was measured by radioimmunoassay as DHA.
Enzymatic activities. 5’-Nucleotidase was determined ac- cording to Heppel and Hilmor (20), adenosine 3’,5’-mono- phosphate phosphodiesterase was assayed by the method of Rutten, Schoot, and de Pont (21), and adenylate cyclase was measured as described elsewhere (13). Protein content was estimated by the method of Lowry, Rosebrough, Farr, and Randall (22) with bovine serum albumin as standard.
RESULTS
In vivo effects of ACTH and dexamethasone adminis- tration. The urinary excretion of steroids after adminis- tration of ACTH or dexamethasone is shown in Table I and the plasma content of steroids under the same conditions in patients where the test was available is shown in Table II. In addition, in patients 11 and 13 a perfusion of Synacthen® (Ciba Pharmaceutical Com- pany, Summit, N. J.), 250 µg in 500 ml of saline, was
| Subjects | |||||
|---|---|---|---|---|---|
| 3 | 6 | 7 | 9 | 10 | |
| ng/2 h per 106 cells | |||||
| Basal | 26±3* | 21 ±1.2 | 22±3 | 3.6±0.6 | 7.5±1.2 |
| ACTH, 30 nM | 247±15# | 19±2 | 24±3 | 2.9±0.4 | 9.8±2.1 |
| ACTH, 3 MM | 258±18# | 21±2 | 23±4 | 2.6±0.3 | 8.9±2.3 |
| PGE1, 60 nM | 200±15# | 65±14+ | 63±84 | ||
| PGE2, 2 MM | 63±8 | 9.6±1.3 | 8.6±1.8 | ||
| DcAMP, 1 mM | 286±28+ | 128±14# | 181±12本 | 37.4±2.4+ | 26.6±3.1本 |
* Mean ±SD (three observations).
# P < 0.05 compared to the first value in its own column.
administered for 6 h. Plasma cortisol was measured at 0, 2, 4, and 6 h. The following results were obtained- patient 11 : 27, 20, 25, and 20 µg/100 ml, respectively; patient 13: 17, 51, 84, and 65 µg/100 ml, respectively. These results suggest that patients 4-8, 10, and 11 did not respond to either of the two tests. Patient 13 re- sponded to ACTH but not to dexamethasone. The re- sults of patient 12 are difficult to explain.2
2 The interpretation of the tests using ACTH and dexa- methasone administration in patients presenting an adrenal tumor is difficult for two reasons : (a) The output of urinary steroids may vary from one day to another in the same patient within limits as high as 100%. (b) Some adrenal
Cortisol production by isolated adrenal cells (Table III). Since the interpretation of the ACTH stimulation test in vivo was difficult, we decided, in the last five pa- tients of our study (nos. 3, 6, 7, 9, and 10), to investi- gate the response of isolated adrenal cells to ACTH, PGE1, and DcAMP. Stimulation of cortisol production with ACTH was obtained in only one case (no. 3).
tumors of the virilizing type present a low secretion of cor- tisol, probably not submitted to the feed-back control mecha- nism which regulates the function of the normal adrenal. This could explain the responses to the tests in the patient 12, who, besides the adenoma, had an adrenal of normal macroscopic and microscopic aspect.
| Patients | Cyclic AMP* | |||||
|---|---|---|---|---|---|---|
| Basal | ACTH (10 μM) | PGE1 (10 μM) | ACTH (10 μM) + PGE1 (10 M) | EGTA (0.1 mM) | NaF (6 mM) | |
| pmol/10 min per mg protein | ||||||
| 1 | 258±40 | 380±25+ | 348±511 | 425±20§ | 530±52+ | 3,820±140本 |
| 2 | 103±15 | 185±21本 | 145±16+ | 212±238 | 147±14+ | 1,581 ±102 |
| 3 | 122±16 | 360±32本 | 302±38+ | 460±37 | 181±15+ | 3,580±180+ |
| 4 | 236±32 | 520±481 | 850±68+ | 1,110±90§ | 395±42+ | 9,300±244 |
| 5 | 659±50 | 793±50本 | 1,029±70 | 1,120±74§ | ㅡ | 6,580±180本 |
| 6 | 198±21 | 204±28 | 309±321 | 304±28 | 274±22: | 2,950±130本 |
| 7 | 288±15 | 277±28 | 373±22+ | 384±23 | 360±181 | 4,852±210本 |
| 8 | 91±8 | 88±9 | 138±11+ | 135±11 | 188±15+ | 2,789±160本 |
| 9 | 463±40 | 349±29 | 1,120±80本 | 635±50 | 617±52+ | 8,540±280本 |
| 10 | 135±13 | 141±15 | 140±12 | 137±9 | ㅡ | 3,980±160 |
| 11 | 135±13 | 132±14 | 202±19+ | ㅡ | 一 | 2,548±140本 |
| 12 | 220±28 | 222±25 | 294±30本 | 296±42 | 292±38+ | 1,475±70本 |
| 13 | 882±86 | 2,586±145+ | 1,496±155↑ | 3,172±134§ | 1,202±103本 | 7,089±310+ |
| Normal human adrenal | 358±20 | 848±40# | 910±36+ | 1,271 ±80§ | 690±50+ | 3,720±142# |
* Mean±SD (12 observations).
# P <0.05 compared to the first value in its own line.
§ P <0.05 compared to the values of ACTH or PGE1.
1000
c AMP pmol / mg protein /20 min
750
500
250
10
10
10
PGE, (M)
This preparation also responded to DcAMP and PGE1. Namely, the stimulation obtained with 3 × 10-8 M ACTH was similar to that induced by 10 M DcAMP and higher than that due to 2 × 10- M PGE1. In the other four cases ACTH was unable to stimulate the cortisol production, but a stimulatory effect could be obtained by using DcAMP and PGE1 in three of the tumors (nos. 6, 7, and 9), while in tumor 10 only DcAMP in- creased slightly the cortisol production.
Adenylate cyclase activity. The enzymatic activity of crude membranes under basal conditions varied from one tumor to another (Table IV). With the exceptions of tumors 5, 9, and 13, it was generally lower than the activity found in the same preparation from normal hu- man adrenals. Otherwise, the adenylate cyclase of the tumor preparations had similar characteristics to that from the normal adrenals. In the presence of a fixed concentration of ATP (0.53 mM), increasing amounts of Mg2+ stimulated the enzyme and reached a plateau towards 4 mM. On the other hand, in the presence of a fixed concentration of Mgª (7.5 mM) the maximal enzymatic activity was obtained with 0.8-1 mM ATP. Further increase of the ATP concentration inhibited the adenylate cyclase activity. This inhibition was virtually complete at 3 mM ATP. Ca2+ inhibited the enzyme and the inhibition was maximal at 10 mM.
Ethylene glycol bis (8-aminoethyl ether) N,N’-tetra- acetic acid (EGTA) and NaF stimulate the adenylate cyclase activity in crude membrane preparations of adrenal gland from several species (23). In the normal
human adrenal cortex the stimulation was maximal with concentrations of 0.1 mM EGTA and 6 mM NaF. At these concentrations, the stimulation was about 80 and 1,000% of the basal activity, respectively. In the tumor preparations the percent stimulation induced by EGTA was similar to that observed in normal adrenals, but the one induced by NaF was generally higher (Table IV).
PGE1 stimulates the adenylate cyclase activity in par- ticulate preparations of human adrenals (13). Maximal stimulation (100% of basal activity) is obtained with 5 × 10- M of the hormone. All tumors except no. 10 were sensitive to PGE1 (Table IV and Fig. 1). How- ever, the percentage of stimulation varied from one case to another. The stimulation was normal in cases 3-5, 9, and 13, low but significant in cases 1, 2, 6-8, 11, and 12, and undetectable in case 10.
In crude membranes of normal human adrenals ACTH is able to induce a 100% stimulation of basal adenylate cyclase activity when used at a 10-5 M concentration; higher values lead to a reversal of the stimulation (Fig. 2). Half-maximal stimulation is obtained with 6 × 10-7 M of ACTH.
In the tumors of our present investigation (Table IV) the stimulation of adenylate cyclase activity induced by 10-5 M ACTH was unobtainable for seven cases (nos. 6-12), low but significant in three cases (nos. 1, 2, and 5) and normal in the last three cases (nos. 3, 4, and 13). The half-maximal stimulation was obtained with about 6 × 10-7 M ACTH. This value is calculated from ex- periments performed in tumors 3-5, where dose-re- sponse curves to increasing ACTH concentrations could be established (Fig. 2).
c AMP pmol /mg protein/20 min
750
500
250
107
10
10°
10
ACTH1.24 (M)
| Patients | Cyclic AMP degraded |
|---|---|
| nmol/15 min per mg protein | |
| 1 | 14 |
| 2 | 26 |
| 3 | 38 |
| 5 | 28 |
| 6 | 34 |
| 7 | 42 |
| 8 | 44 |
| 9 | 17 |
| 10 | 21 |
| 12 | 30 |
| 13 | 33 |
| Normal human adrenal | 87 |
Crude membranes (100 µg of protein) were incubated in 0.2 ml of 66 mM Tris-HCI, pH 7.4, 1 mM MgCl2, and 1 mM CyclicAMP at 37℃ for 15 min. The values are mean of three replications.
An additive effect of ACTH and PGE1 on adenylate cyclase stimulation has been shown for normal ovine and human adrenals (Table IV and reference 3). The same additive effect can be described for the tumors respon- sive to both hormones (nos. 1-5 and 13). Concentrations (10-5 M) of ACTH and PGE1 responsible for maximal stimulation of adenylate cyclase activity when added separately induce a further stimulatory effect when they are added simultaneously. In the tumors insensitive to ACTH, the PGE stimulation is not changed by the
presence of ACTH, except in tumor 9, in which ACTH inhibits in part PGE1 stimulation.
Other hormones have been shown to be effective in stimulating the adenylate cyclase activity of the particu- late fraction from the tumor 494 described by Snell and Stewart (2). The tumors tested in the present study (nos. 2 and 8-10) did not respond to concentrations as high as 10-5 M of epinephrine, luteinizing hormone, glu- cagon, and insulin (data not shown).
The inability of ACTH1-2 to stimulate the accumula- tion of cyclicAMP in tumors 6-12 could be attributed to one or several of the following abnormalities :
(a) Increase of the phosphodiesterase activity of the tumor crude membranes. This hypothesis is very un- likely since the stimulation of adenylate cyclase induced by other stimuli was normal or higher than normal and, when measured, the phosphodiesterase activity was lower than in normal adrenals (Table V). These re- sults are similar to those reported by Sharma (24), who showed a lower phosphodiesterase activity in adrenal cortical tumors than in the normal adrenal in rats.
(b) Abnormal distribution of adenylate cyclase ac- tivity in subcellular fractions of adrenal tumors. The distributions of 5’-nucleotidase and adenylate cyclase in the different subcellular fractions of tumors (nos. 5 and 12) are given in Table VI and compared to results ob- tained in normal human adrenals. These two enzymatic activities were present in all of the particulate frac- tions in both tumors and the normal adrenal. The dis- tribution of adenylate cyclase activity differed from the one observed in the normal adrenal (Table VI). How- ever, this table clearly shows that in none of the subcel- lular fractions obtained from tumor 12 was the adenylate cyclase sensitive to ACTH.
| Normal adrenal | Tumor 12 | Tumor 5 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 5'-Nucleotidase | Adenylate cyclase* | Adenylate cyclase* | Adenylate cyclase* | ||||||||
| Basal | ACTH (10 µM) | NaF (6 mM) | 5'-Nucleotidase | Basal | ACTH (10 µM) | NaF (6 mM) | Basal | ACTH (10 4M) | NaF (6 mM) | ||
| mol Pi liber- ated/h per mg protein | pmol/20 min per mg protein | mol Pi liber- ated/h per mg protein | pmol/20 min per mg protein | pmol/20 min per mg protein | |||||||
| Homogenate | 1.2 | 109 | 359 | 1,980 | 2.8 | 244 | 250 | 2,260 | 271 | 241 | 1,645 |
| 800 g | 2.2 | 92 | 370 | 2,480 | 4.2 | 590 | 610 | 7,651 | 303 | 363 | 2,277 |
| 20,000 g | 4 | 330 | 829 | 3,690 | 1.7 | 215 | 218 | 1,520 | 640 | 784 | 6,580 |
| 105,000 g | 4.8 | 103 | 117 | 1,400 | 4.1 | 589 | 579 | 5,921 | 353 | 357 | 1,785 |
| Purified | - | - | 15.3 | 1,608 | 1,641 | 20,280 | - | - | - | ||
| membranes | |||||||||||
* Mean value of three replications.
# Prepared by the method of Finn et al. (12).
| Patients | Percentage of hormone remaining intact | |||
|---|---|---|---|---|
| Unbound | Bound | |||
| Adsorption to Quso | Binding to fresh membranes | Adsorption to Quso | Binding to fresh membranes | |
| 1 | 11 | 3 | 97 | 107 |
| 2 | 73 | 54 | 99 | 114 |
| 3 | 19 | 9 | 96 | 107 |
| 4 | 22 | 14 | 98 | 110 |
| 5 | 14 | 5 | 95 | 104 |
| 6 | 36 | 24 | 98 | 111 |
| 7 | 28 | 17 | 98 | 109 |
| 8 | 52 | 33 | 99 | 114 |
| 9 | 20 | 8 | 97 | 107 |
| 10 | 47 | 33 | 99 | 110 |
| 11 | 39 | 28 | 97 | 108 |
| 12 | 18 | 6 | 97 | 106 |
| 13 | 17 | 10 | 96 | 104 |
| Normal human adrenal | 45±2.8* | 31±3 | 99±6 | 110±5 |
Crude membranes (1.4 mg of protein/ml) were incubated 30 min at 4℃ with 125I-ACTH1-24 (22 nM). Bound and unbound 125I-ACTH1-24 were separated by cen- trifugation and bound 125I-ACTH1-24 extracted by 1% acetic acid (see Methods). Degradation of 125I-ACTH1-24 in both fractions was measured by absorption to Quso and ability to bind to fresh adrenal membranes as described under Methods. * Mean value of six replications.
(c) Increase degradation of ACTH1-2 by tumor crude membranes. It has been recently demonstrated that crude membrane preparations of normal adrenals of several species can degrade ACTH (11). This degra- dation is independent from the binding, and it only af- fects the ACTH that is not bound to its receptor. Prep- arations of crude membranes prepared from adrenal tu- mors also degraded unbound 125I-ACTH1-24 but did not degrade bound 13I-ACTH1-24 (Table VII). The degra- dation of unbound ACTH estimated by binding to fresh membranes was, in all the tumors (as observed in nor- mal adrenals), greater than that estimated by absorp- tion to Quso.3 The degree of degradation of ACTH varied from one tumor to another, but these quantitative variations could not explain why in certain tumors adenylate cyclase was not stimulated by ACTH. The degradation system was less potent in tumors 6, 8, 10,
and 11 although in all of them ACTH did not stimulate adenylate cyclase. In the other tumors (except for no. 2) the degradation was greater than normal even in cases where adenylate cyclase responded to ACTH.
(d) Modification of the ACTH binding receptor in tumor crude membranes. Specific binding of 125I- ACTH1-4 and 15I-ACTH11-24 to crude membranes of tumoral origin was shown by the same criteria as those used for normal adrenal preparations (14) : (i) Bound radioactivity was displaced only by ACTH1-24 and some of its analogues (see below). Insulin, glu- cagon, and PGE1 at high concentrations (10-5 M) were without effects. (ii) Heating the membranes at 60℃ for 30 min or pretreatment with trypsin (250 ug/mg of membrane protein) completely inhibited the binding.
According to these criteria all the membranes pre- pared from the tumors specifically bound both labeled ACTH’s. However, the apparent binding affinity of ACTH1-2 for the tumors in which the adenylate cyclase was not stimulated by ACTH was about 10 times lower than that for normal adrenals and for tumors in which the adenylate cyclase was stimulated by ACTH. On the other hand, the binding affinity of ACTH11-24 was simi- lar for both normal adrenals and all the tumors and about 10 times lower than the one of ACTH1-2 in a
3 In two cases, normal adrenals and tumor 10, the degra- dation of bound and unbound 125I-ACTH1-24 was also mea- sured by the ability of bound and unbound hormone to stim- ulate the cortisol secretion by the isolated ovine adrenal cells. The percentage of hormone remaining intact in the unbound fraction was 28 and 30 in normal adrenal and tumor 10, respectively, and in the bound fraction the percentage was 112 and 110, respectively. These results are similar to those found when the degradation was measured by the ability to bind to fresh adrenal membranes (Table VII).
542 J. M. Saez, A. Dazord, and D. Gallet
| Subjects | 12BI-ACTH1-24 | 125 I-ACTH11-24 |
|---|---|---|
| μ.Μ | ||
| 1 | 0.34 | |
| 2 | 0.55 | |
| 3 | 0.41 | 2.5 |
| 4 | 0.42 | 3.8 |
| 5 | 0.62 | |
| 6 | 2.1 | 4.1 |
| 7 | 3.9 | 2.5 |
| 8 | 3.9 | |
| 9 | 4.1 | 5.4 |
| 10 | 6.2 | 7.1 |
| 11 | 4 | 5.2 |
| 12 | 2.7 | |
| 13 | 0.24 | |
| Normal human adrenal | 0.32 | 4.1 |
The table gives the concentrations of ACTH1-24 and ACTH11-24 required to displace 50% of the bound labeled hormone.
normal gland (Table VIII). Figs. 3 and 4 are repre- sentative experiments of the displacement of 13I- ACTH1-24 and 13I-ACTH11-24 by ACTH1-24 and ACTH11-24, respectively, in normal adrenals, one tumor that re- sponds to ACTH (no. 3), and four ACTH nonrespond- ing tumors (nos. 6-9).
In normal adrenal preparations ACTH1-10 displaced bound 1ªI-ACTH1-24 but not bound 13I-ACTH11-24 (14). Owing to insufficient quantities of unlabeled ACTH1-10, we could study the effect of this peptide in only two cases. The results in Table IX show that ACTH1-10 dis- placed bound 13I-ACTH1-24 in tumor 3 (ACTH re- sponder) but did not in no. 6 (ACTH nonresponder) and suggest that the binding of the sequence 1-10 of ACTH may be specifically impaired in tumors 6-12.
| Subjects | Control | +ACTH1-10(0.1 mM) |
|---|---|---|
| 3 | 100±3* | 71±4₫ |
| 6 | 100±4* | 116±61 |
| Normal human adrenal | 100±3* | 65±31 |
Membranes were incubated 30 min at 4℃ in 0.25 ml of 20 mM Tris-HCI (pH 7.4) containing 1% albumin and 1 nM 125 I-ACTH1-24 with or without ACTH1-10 (0.1 mM). Results for specific binding obtained under control conditions are given an arbitrary value of 100.
* Mean±SD (six observations).
# Mean±SD (three observations).
| Subjects | [H ]PGE1 bound |
|---|---|
| pmol/mg protein* | |
| 1 | 0.18±0.05 |
| 2 | 0.16±0.04 |
| 3 | 0.85±0.10 |
| 4 | 1.6 ±0.12 |
| 5 | 0.94±0.09 |
| 6 | 0.64±0.07 |
| 7 | 0.49±0.06 |
| 8 | 0.34±0.02 |
| 9 | 1.10±0.11 |
| 10 | 0± |
| 11 | 0.52±0.03 |
| 12 | 0.45±0.07 |
| 13 | 1.10±0.12 |
| Normal human adrenal | 1.20±0.09 |
Membranes (about 100 µg of protein) were incubated at 4℃ for 90 min in 0.25 ml of 20 mM Tris-HCI, pH 7.4, containing 1% albumin and 12 nM [3H]PGE1. Specific binding of the hormone was measured as described under Methods.
* Mean±SD (nine observations).
# See text.
However, these studies do not exclude an associated abnormality of the coupling system in this group of tumors.
Specific binding of [‘H]PGE, to crude membranes obtained from adrenal tumors. Since the adenylate cy- clase of tumor 10 was not stimulated by PGE1 (Table IV), we decided to investigate whether any abnormality of the binding receptor of the hormone could be dem- onstrated in this tumor. As in normal human adrenal preparations (13) a specific binding of [H]PGE was observed in all the tumors, except in no. 10 (see below). The specificity of the binding was shown by the fact that in the presence of 10 µg of PGE1, a 88-92% de- crease in total radioactivity bound was noted. ACTH, insulin, and glucagon at high concentrations (10-5 M) had no effect (data not shown). On the contrary, 200 ug of crude membranes from tumor 10 in the presence of 1.2 × 10-8 M [‘H]PGE, bound only 1.1±0.3% of the total radioactivity, and after addition of 10 µg of un- labeled PGE1 the radioactivity bound was the same (1±0.4%). In the other tumors the specific binding varied from one tumor to another and in general was much lower than that of normal human adrenals (Ta- ble X).
DISCUSSION
In the normal human adrenal ACTH is essential for the continuous production of glucocorticoids and an-
ACTH and PGE: Receptors in Human Adrenal Tumors 543
drogens (25). The first detectable step in the mecha- nism of action of ACTH is the binding of this hormone to specific adrenocortical receptors (26, 27). This in- teraction is followed by a stimulation of the adenylate cyclase enzyme which is responsible for an increase in intracellular cyclic AMP level (28). This cyclic nu- cleotide, by mechanisms still not understood, increases steroidogenesis (29). The failure of certain adrenal tu- mors to respond to ACTH in vivo despite the existence of a high rate of steroidogenesis may suggest either the presence of one or several abnormalities in the bio- chemical sequence initiated by the hormone or that steroidogenesis in these tumors is already accelerated to a maximal rate by endogenous ACTH or other factors.
The study in vitro of human adrenal tumors has shown a great heterogeneity in all the parameters that have been studied. However, according to the adenylate cy- clase response to ACTH and PGE the tumors investi- gated here can be divided into three categories : (a) tu- mors in which the adenylate cyclase was stimulated by both ACTH and PGE1 (nos. 1-5 and 13) ; (b) those in which adenylate cyclase was stimulated by PGE1 but not by ACTH (nos. 6-9, 11 and 12); (c) the one in which adenylate cyclase was stimulated neither by PGE1 nor by ACTH (no. 10).
In the first group of tumors adenylate cyclase was stimulated by the same factors as in the normal adrenal. The quantitative variations observed might be accounted for by the heterogenous composition of the crude mem- brane preparations used in both normal adrenals and adrenal tumors and/or by an abnormality of the adenylate cyclase system. In this group, only in two cases (nos. 4 and 5) was an insensitivity to ACTH demonstrated in vivo. The results in these two patients were similar to those observed in the 494 adrenocortical tumor of the rat where ACTH was unable to stimulate steroidogenesis in vivo and in vitro (3-6) whereas adenylate cyclase in particulate preparations was stimu- lated by the hormone (3). Since in this tumor cyclic AMP did not stimulate steroidogenesis (4, 5), it has been suggested (3) that the main anomaly responsible for the insensitivity to ACTH would occur after the formation of cyclic AMP. As our results suggest but do not confirm (since we could not test the effect of DcAMP on steroid production), the same anomaly could be present in some human adrenal cortical tu- mors. However, the similarity between human and rat tumors is far from complete. We have not found re- ceptors for epinephrine and luteinizing hormone in the human tumors as were found in the rat tumor (30). It also remains to be established if the multiple anomalies in the mechanism of action of ACTH on the rat tumor, described by Sharma (6, 31, 32), are present in this group of human tumors.
The failure of ACTH to stimulate the cyclic AMP accumulation in our second group of tumors cannot be explained by an increased activity of the nucleotide phosphodiesterase (Table V). Moreover, it is probably not due to an anomaly of the catalytic sites of adenylate cyclase, since the enzymatic activity was stimulated by PGE1, EGTA, and NaF. Therefore, the nonresponse to ACTH is probably related to a membrane anomaly. Adenylate cyclase insensitivities to specific hormones have already been reported for ACTH by Shimmer (8) in a mutant adrenal cell line and for thyroid-stimulating hormone by Macchia, Meldolesi, and Shiariello (33) in a rat thyroid tumor. In these cases, as in human adrenal tumors, the adenylate cyclase was stimulated by NaF.
The membrane anomaly in our second group of tu- mors could be located in the hormone binding receptor and/or at the level of the system coupling discriminator and catalytic sites. Although the binding of 12I-ACTH1-24 is specific in those tumors, the first hypothesis cannot be eliminated.
Recent studies concerning the relationship between the structure and the function of ACTH (12, 27, 34-37) have shown that the peptide sequences necessary for the binding and for the biological action of this hor- mone are localized in two distinct parts of the molecule. The C-terminal (ACTH11-24) is important for binding (12) but has no biological action (37) whereas the N-terminal sequence (ACTH1-10) has a very low binding affinity but is essential for biological action. In addition, it has been shown (11) that in normal adrenals the binding affinity of 13I-ACTH1-24 is about 10 times higher than that of 13I-ACTH11-24 and that ACTH1-10 displaces bound 1ªI-ACTH1-24 but not ACTH11-24. This could be explained by the existence in the normal adrenal mem- branes of two distinct binding sites for the ACTH molecule, one related to the 1-10 sequence and the other one to the 11-24 sequence.
Data (Figs. 3 and 4 and Tables VIII and IX) from the binding studies in tumors of our second group (where adenylate cyclase does not respond to ACTH) clearly show that the binding affinity for ACTH1-2 is much lower than normal. It is similar to that for ACTH11-24 and not 10 times higher as in the normal gland or in the tumors sensitive to ACTH. This loss could not possibly be related to a modification of the site which binds the 11-24 sequence of the hormone since the binding affinity of the ACTHn-2 remained normal. Alternatively, the failure of ACTH1-10 to dis- place bound 125I-ACTH1-24 in the tumor 6 strongly sug- gested that, at least in this example, the defect is prob- ably due to a modification or a loss of the membrane component which normally binds the 1-10 sequence of ACTH. However, the possibility of an associated anomaly of the coupling system of the ACTH receptor cannot be ruled out by our study; nevertheless, this
INHIBITION OF “I-ACTH, BOUND ( % of maximum )
100-
50-
10
10
10
ACTH , 24 (M)
anomaly would not be extended to the coupling system of the PGE1 receptor since the hormone stimulated the adenylate cyclase activity of those tumors.
All the results presented so far for both tumors and normal adrenals can be reconciled in a diagram repre- senting the receptors for ACTH and PGE, such as the one proposed in Fig. 5. Although no biological action has yet been described for ACTH11-24, a possible role for the 11-24 sequence could be to increase the binding affinity of ACTH and therefore to allow the sequence 1-10 to fit in the second site. Only when this second site is occupied would the adenylate cyclase be stimu- lated.
INHIBI TION OF I-ACTH,, __ BOUND ( % of maximum )
100
50
10
10°
10°
ACTH,,_„ (M)
CELL MEMBRANE
ACTH
receptor
11-24
?
ATP
1-10
Coupling
A.C
CAMP
…
ATP
PGE
receptor
Coupling
A.C.
CAMP
CELL MEMBRANE
This model would also explain the case of the tumor (no. 10) of the last category which, in addition to the anomalies of the ACTH receptor, as for the second group, presents also an associated defect of the PGE1 receptor. The PGE1 did not stimulate the adenylate cy- clase of this preparation, and no significant binding could be observed with the tritiated hormone; however, it could not be demonstrated whether the defect con- sisted in a modification or a loss of the PGE1-binding site, which could or could not be associated with an anomaly of its coupling system.
It can be concluded from our study that in human adrenocortical tumors, a loss of sensitivity to ACTH can be related to a variety of biochemical abnormali- ties. They can affect different components of the ACTH receptor-adenylate cyclase complex or the message ini- tiated by the formation of cyclicAMP. Further work in progress in this laboratory indicates that their detailed observation and analysis, despite an apparent hetero- geneity, will give valuable information concerning the physiological role of the different peptide sequences of ACTH in the normal adrenal cell.
ACKNOWLEDGMENTS
The authors are most grateful to Dr. G. P. E. Tell for helpful suggestions in preparing the manuscript. We wish to thank the clinicians who allowed us to study these tumors : Profs. M. David, Drosdowsky, R. Francois, M. Jeune, P. Guinet, R. Mornex, P. C. Sizonenko, and J. Tourniaire. We are indebted to Drs. W. Rittel and P. A. Desaulles for the generous gift of various ACTH analogues and to Dr. Y. Guichard for the electron microscopy studies. We express our thanks to Drs. M. G. Forest and E. de Peretti for their kind supply of antiserum anti-testosterone and anti-DHA. We also thank Prof. J. Bertrand for his continuous en- couragement and interest in this work. The secretarial as- sistance of Miss J. Bois is also appreciated.
This work was supported in part by Centre National de la Recherche Scientifique grant no. 429904.
REFERENCES
1. Lipsett, M. B., R. Hertz, and G. T. Ross. 1963. Clinical and pathophysiological aspects of adrenocortical car- cinoma. Am. J. Med. 35: 374-383.
2. Snell, K. C., and H. L. Stewart. 1959. Variations in histological pattern and functional effects of a trans- plantable adrenal cortical carcinoma in intact, hypophy- sectomized and newborn rats. J. Natl. Cancer. Inst. 22: 1119-1155.
3. Schorr, I., and R. L. Ney. 1971. Abnormal hormone responses of an adrenocortical cancer adenyl cyclase. J. Clin. Invest. 50: 1295-1300.
4. Ney, R. L., N. J. Hochella, D. G. Grahame-Smith, R. N. Dexter, and R. W. Butcher. 1969. Abnormal regu- lation of adenosine 3’,5’-monophosphate and corticoste- rone formation in an adrenocortical carcinoma. J. Clin. Invest. 48: 1733-1739.
5. Sharma, R. S., and K. Hashimoto. 1972. Ultrastruc- tural studies and metabolic regulation of isolated adrenocortical carcinoma cells of rat. Cancer Res. 32: 666-674.
6. Sharma, R. K. 1973. Metabolic regulation of steroido- genesis in adrenocortical carcinoma cells of rat. Eur. J. Biochem. 32: 506-512.
7. Schimmer, B. P. 1969. Phenotypically variant adrenal tumor cell cultures with biochemical lesions in the ACTH-stimulated steroidogenic pathway. J. Cell. Phys- iol. 74: 115-122.
8. Schimmer, B. P. 1972. Adenylate cyclase activity in adrenocorticotropic hormone-sensitive and mutant adren- ocortical tumor cell lines. J. Biol. Chem. 247: 3134- 3138.
9. Few, J. D. 1961. A method for the analysis of urinary 17-hydroxycorticosteroids. J. Endocrinol. 22: 31-64.
10. Glenn, E. M., and D. H. Nelson. 1953. Chemical method for the determination of 17-hydroxycorticosteroids and 17-ketosteroids in urine following hydrolysis with ₿ glycuronidase. J. Clin. Endocrinol. Metab. 13: 911-921.
11. Saez, J. M., A. Dazord, A. M. Morera, and P. Bataille. 1975. Interactions of ACTH with its adrenal receptors. Degradation of ACTH1-24 and ACTH11-24. J. Biol. Chem. 250: 1683-1689.
12. Finn, F. M., C. C. Widnell, and K. Hofmann. 1972. Localization of an adrenocorticotropic hormone recep- tor on bovine adrenal cortical membranes. J. Biol. Chem. 247: 5695-5702.
13. Dazord, A., A. M. Morera, J. Bertrand, and J. M. Saez. 1974. Prostaglandin receptors in human and ovine adrenal glands: binding and stimulation of adenyl cy- clase in subcellular preparations. Endocrinology. 95: 352-359.
14. Saez, J. M., A. M. Morera, A. Dazord, and P. Bataille. 1974. Interactions of ACTH with its adrenal receptors. Specific binding of ACTH1-21, its o-nitrophenyl sulfenyl derivative and ACTH11-21. J. Steroid Biochem. 5: 925- 933.
15. Kitabchi, A. E., and R. K. Sharma. 1971. Corticos- teroidogenesis in isolated cells of rats. Effect of cor- ticotropins and 3’,5’-cyclic nucleotides on corticosterone production. Endocrinology. 88: 1109-1116.
16. Sayers, G., R. L. Swallow, and N. O. Giordano. 1971. An improved technique for the preparation of isolated rat adrenal cells: a sensitive, accurate and specific
method for the assay of ACTH. Endocrinology. 88: 1063-1068.
17. Murphy, B. E. P. 1967. Some studies of the protein- binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J. Clin. Endocrinol. Metab. 27: 973-990.
18. Forest, M. G., P. C. Sizonenko, A. M. Cathiard, and J. Bertrand. 1974. Hypophysogonadal function in hu- mans during the first year of life. I. Evidence for tes- ticular activity in early infancy. J. Clin. Invest. 53: 819- 828.
19. Saez, J. M., and J. Bertrand. 1968. Studies on testicular function in children: plasma concentrations of testoste- rone, dihydroepiandrosterone and its sulfate before and after stimulation with human chorionic gonadotropin. Steroids. 12: 749-761.
20. Heppel, L. A., and R. J. Hilmoe. 1951. Purification and properties of 5-nucleotidase. J. Biol. Chem. 188: 665- 676.
21. Rutten, W. J., B. M. Schoot, and J. J. H. de Pont. 1973. Adenosine 3’,5’-monophosphate phosphodiesterase assay in tissue homogenates. Biochim. Biophys. Acta. 315: 378-383.
22. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with Folin phenol reagent. J. Biol. Chem. 193: 265-275.
23. Dazord, A., D. Gallet, J. Bertrand, and J. M. Saez. 1973. Action du calcium, de l’EGTA et de l’EDTA sur l’activité adényl-cyclasique de préparations subcellulaires de corticosurrénales. C. R. Acad. Sci. (Paris). 277: 1917-1920.
24. Sharma, R. K. 1972. Studies on adrenocortical carci- noma of rat cyclic nucleotide phosphodiesterase activi- ties. Cancer Res. 32: 1734-1736.
25. Rosenfeld, R. S., L. Hellman, H. Roffwarg, E. D. Weitzman. D. K. Fukushima, and T. F. Gallagher. 1971. Dehydroisoandrosterone is secreted episodically and synchronously with cortisol by normal man. J. Clin. Endocrinol. Metab. 33: 87-92.
26. Lefkowitz, R. J., J. Roth, W. Pricer, and I. Pastan. 1970. ACTH receptors in the adrenal: specific binding of ACTH-125I and its relation to adenyl cyclase. Proc. Natl. Acad. Sci. U. S. A. 65: 745-752.
27. Hofmann, K., W. Wingender, and F. M. Finn. 1970. Correlation of adrenocorticotropic activity of ACTH analogs with degree of binding to an adrenal cortical particular preparation. Proc. Natl. Acad. Sci. U. S. A. 67: 829-836.
28. Grahame-Smith, D. G., R. W. Butcher, R. L. Ney, and E. W. Sutherland. 1967. Adenosine 3’,5’-monophosphate as the intracellular mediator of the action of adreno- corticotropic hormone on the adrenal cortex. J. Biol. Chem. 242: 5535-5541.
29. Gill, G. N. 1972. Mechanisms of ACTH action. Metab. Clin. Exp. 21: 571-588.
30. Schorr, I., P. Rathnam, B. B. Saxena, and R. L. Ney. 1971. Multiple specific hormone receptors in the adeny- late cyclase of an adrenocortical carcinoma. J. Biol. Chem. 246: 5806-5811.
31. Sharma, R. K., and J. S. Brush. 1974. Metabolic reg- ulation of steroidogenesis in adrenocortical carcinoma cells of rat. Effect of adrenocorticotropin and adenosine cyclic 3’,5’-monophosphate on the incorporation of (20 S)-20-hydroxy-(7a-3H) cholesterol into deoxycorticoste- rone and corticosterone. Biochim. Biophys. Res. Com- mun. 56: 256-263.
546 J. M. Saez, A. Dazord, and D. Gallet
32. Sharma, R. K. 1974. Metabolic regulation of steroido- genesis in isolated adrenal and adrenocortical carcin- oma cells of rat. Effect of adrenocorticotropic hormone and adenosine cyclic 3’-5’-monophosphate on the plasma membranes. FEBS (Fed. Eur. Biochem. Soc.) Lett. 38: 197-201.
33. Macchia, V., M. F. Meldolesi, and M. Shiariello. 1972. Adenyl-cyclase in a transplantable thyroid tumor : loss of ability to respond to TSH. Endocrinology. 90: 1483- 1491.
34. Ramachandran, J., and C. H. Lee. 1967. Structure- activity relationships of the adrenocorticotropins and melanotropins. The synthetic approach. Adv. Enzymol. 29: 391-477.
35. Tesser, G. I., R. Maier, L. Schenkel-Hulliger, P. L. Barthe, B. Kamber, and W. Rittel. 1973. Biological activity of corticotropin peptides with homoarginine, lysine, or ornithine substituted for arginine in position 8. Acta Endocrinol. 74: 56-66.
36. Geiger, R., and H. G. Schröder. 1973. Synthetische analoga des Corticotropins. Hoppe-Seyler’s Z. Physiol. Chem. 354: 156-162.
37. Sayers, G., S. Seelig, S. Kumar, G. Karlaganis, R. Schwyzer, and M. Fujino. 1974. Isolated adrenal cor- tex cells : ACTH,2 (NH2), ACTH&-2, ACTH6-2, and ACTH7-23 (NH2) ; cyclic AMP and corticosterone pro- duction. Proc. Soc. Exp. Biol. Med. 145: 176-181.