Characterization of Adrenal Autonomy in Cushing’s Syndrome: A Comparison between in Vivo and in Vitro Responsiveness of the Adrenal Gland
STEVEN W. J. LAMBERTS, JOKE ZUIDERWIJK, PIET UITTERLINDEN, JOLANDA J. BLIJD, HAYO A. BRUINING, AND FRANK H. DE JONG Departments of Medicine and Surgery, Erasmus University, Rotterdam, The Netherlands
ABSTRACT. We measured cortisol and precursor steroid pro- duction in response to ACTH, cholera toxin, and forskolin by the dispersed adrenocortical cells prepared from the adrenal glands of 10 patients with different forms of Cushing’s syndrome.
The cells prepared from the hyperplastic adrenal glands from 4 patients with Cushing’s disease responded in a dose-dependent manner to ACTH, cholera toxin, and forskolin. The adrenal cells prepared from 4 encapsulated adrenal adenomas showed no (n = 2), a lowered (n = 1), or a clear (n = 1) response of cortisol release to ACTH. The cells prepared from the adrenal glands of 1 patient with dysplastic micronodular adrenal glands showed a limited response to ACTH, while the cells from an adrenocortical carcinoma, which secreted very little cortisol per cell, were unresponsive to ACTH, cholera toxin, and forskolin. The reaction of the dispersed adrenal cells from these 10 patients to ACTH, cholera toxin, and forskolin showed a close correlation (P < 0.001 in all instances). This suggests that the defect in autonomous glands is not located at the level of the ACTH receptor, but, rather, involves the adenylate cyclase complex as a whole or its coupling to cAMP-dependent protein kinase. The release into the medium of the cortisol precursors deox-
ycortisol, 17-hydroxyprogesterone, and progesterone showed that the four autonomous nodules were characterized by a sig- nificantly higher deoxycortisol/cortisol ratio in the medium (P < 0.01), suggesting a relative blockade of 118-hydroxylase in these adrenal adenomas. This was further substantiated in cells from several adrenals by a significant increase in the release of these precursors in response to ACTH in the absence of a cortisol response.
We conclude the following. 1) Adrenal adenoma formation in patients with Cushing’s syndrome is accompanied by a parallel decrease in the stimulation of the release of steroid hormones in response to ACTH, cholera toxin, and forskolin. This points to a defect in the adenoma cells beyond the ACTH receptor. 2) Adrenal adenoma formation in patients with Cushing’s syn- drome is accompanied by a relative blockade of 118-hydroxylase activity. 3) By comparing the preoperative dynamic tests of the pituitary-adrenal axis, the plasma ACTH concentration, the morphology of the adrenal glands, and their in vitro responsive- ness, a gradual transition from pituitary to (partial) adrenal autonomy could be recognized in several patients. (J Clin En- docrinol Metab 70: 192, 1990)
P REVIOUS studies with adrenal glands obtained from patients with Cushing’s syndrome have shown that evidence of autonomy of the adrenal gland in the dynamic tests in vivo is in most cases reflected by a lowered responsiveness of the cells to ACTH in vitro (1- 5). However, both from in vivo tests with ACTH admin- istration and from in vitro investigations it has been shown that adrenal adenomas often retain a sensitivity to the stimulatory effect of ACTH (6-8). In recent years a number of patients with supposedly pituitary-depend- ent Cushing’s disease have been described in whom a single adrenal adenoma was also detected (9-14). These patients might represent different stages of the transition from pituitary to adrenal-dependent Cushing’s syn- drome.
Received May 11, 1989.
Address all correspondence and requests for reprints to: Steven W. J. Lamberts, Department of Medicine, University Hospital Dijkzigt, 40 Dr. Molewaterplein, 3015 GD Rotterdam, The Netherlands.
In the present study we investigated the in vitro re- sponsiveness of the dispersed adrenal cells obtained from 10 patients with different forms of Cushing’s syndrome to ACTH, cholera toxin, and forskolin. Apart from the amount of cortisol released into the medium, we also measured the concentrations of the precursors deoxycor- tisol, 17-hydroxyprogesterone, and progesterone. By comparing the preoperative dynamic tests of the pitui- tary-adrenal axis, the plasma ACTH concentration, the morphological aspect of the adrenal glands, and their in vitro responsiveness, several aspects of the transition from pituitary to (partial) adrenal autonomy were inves- tigated in several of these patients.
Subjects and Methods
Patients
Ten patients with Cushing’s syndrome were investigated (Table 1). All had hypercortisolism, as measured by the cortisol secretion rate, the absence of a diurnal rhythm of plasma
cortisol, and insufficient overnight suppression of plasma cor- tisol in response to 1 mg dexamethasone at 2300 h. Subse- quently, the diagnosis of pituitary-dependent Cushing’s disease was made in patients 1-4 by demonstrating an increase of plasma cortisol by more than 200 nmol/L in response to 1 ug/ kg CRH, iv, and/or an increase in plasma deoxycortisol levels to more than 420 nmol/L in response to six doses of 750 mg metyrapone and a decrease in plasma cortisol by more than 200 nmol/L at the end of a 5-h infusion of 5 mg dexamethasone (1 mg/h) (15). In four patients transsphenoidal operation re- sulted in incomplete removal of a microscopically proven ACTH-secreting pituitary adenoma, while plasma ACTH levels remained virtually unchanged thereafter. Because of persisting hypercortisolism all four patients underwent bilateral adrenal- ectomy thereafter. The adrenal glands of patients 1-3 showed diffuse adrenocortical hyperplasia. The adrenal glands of pa- tient 4 showed microscopically diffuse adrenocortical hyperpla- sia, but in the left adrenal a single well encapsuled adrenal adenoma with a diameter of 4 cm was found, which was used separately for in vitro studies. Microscopically this was an adenoma. Limited studies were also performed with the sur- rounding hyperplastic adrenocortical tissue of this adrenal gland (patient 4). Patients 5-8 all showed the biochemical picture of an autonomous cortisol-secreting adrenal tumor, which did not react to metyrapone and dexamethasone in the presence of undetectable ACTH levels, except in the case of patient 8, who had a normal plasma ACTH concentration. Patients 5, 6, and 7 had encapsulated adenomas with diameters of 4 (left adrenal), 2 (right adrenal), and 3 (right adrenal) cm,
respectively, while the remaining part of the adrenal cortex was microscopically atrophic. In patient 8 the left adrenal gland contained a 4-cm encapsulated adenoma, while the remaining adrenal gland was microscopically clearly hyperplastic. Patient 9 also showed a biochemical picture compatible with that of an adrenal tumor. However, at computed tomographic scanning two seemingly normal adrenal glands were seen. Because of the severity of the clinical symptoms it was decided to carry out bilateral adrenalectomy in this patient. Macroscopically the glands looked normal, but microscopically, dysplastic micro- nodular adrenocortical hyperplasia with atrophic zones in be- tween were seen. There was no pigmentation within the adrenal gland. Finally, patient 10 presented with a clinically very ag- gressive Cushing’s syndrome. At computed tomographic scan- ning an adrenal tumor with a diameter of 15 cm was seen on the right side. No further biochemical testing was carried out, and she was operated upon immediately. An adrenal carcinoma infiltrating the liver and blood vessels was found.
Adrenocortical cells
The adrenal glands were transported in culture medium and reached the laboratory within 15 min after removal. Special care was given to handle the tissue consistently, and the dis- persion was always carried out by the same person (J.Z.).
Between 1.5-6.8 g adrenocortical tissue were used in the preparation of dispersed cells (11, 16). The tissue was chopped into 1 × 1-mm pieces, and cells were isolated after digestion with 3 mg/ml collagenase (type I; Sigma Chemical Co., St. Louis, MO) and mechanical dispersion using a plastic pipet.
| Patient no. | Sex | Age (yr) | Cortisol secretion rate (pmol/ 24 h) | Adrenal histology | 1 mg Dex suppression plasma cortisol (nmol/L) | Plasma ACTH (pmol/L) | CRH (1 µg/ kg), cortisol (nmol/L) | Deoxycortisol After 4.5 g metyrapone (nmol/L) | iv Dex (5 mg/5 h), Cortisol (nmol/ L) | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Basal | Max | Basal | 3 h | 5 h | ||||||||
| 1 | F | 37 | 211 | Bilateral diffuse adrenocortical hyperplasia | 630 | 27.5 | 650 | 960 | 1250 | 630 | 420 | 220 |
| 2 | M | 26 | 489 | Bilateral diffuse adrenocortical hyperplasia | 830 | 17.8 | 960 | 1200 | 850 | 1050 | 780 | 650 |
| 3 | F | 48 | 213 | Bilateral diffuse adrenocortical hyperplasia | 710 | 13.2 | 680 | 640 | 440 | 320 | ||
| 4 | F | 43 | 536 | Bilateral diffuse adrenocortical hyperplasia + 1 single ade- noma (4 cm) | 640 | 6.2 | 730 | 1480 | 720 | 690 | 270 | 240 |
| 5 | F | 62 | 225 | Adenoma (left) of 4 cm | 410 | 0 | 710 | 740 | 70 | 460 | 450 | 430 |
| 6 | M | 48 | 120 | Adenoma (right) of 2 cm | 490 | 0 | 80 | 590 | 580 | 620 | ||
| 7 | M | 42 | 139 | Adenoma (right) of 3 cm | 510 | 0 | 180 | 520 | 540 | 580 | ||
| 8 | M | 49 | 195 | Adenoma (left) of 3 cm + dif- fuse hyperplasia surrounding cortex | 410 | 8.8 | 190 | 450 | 460 | 460 | ||
| 9 | F | 33 | 184 | Bilateral micronodular adreno- cortical hyperplasia | 580 | 0 | 800 | 820 | 80 | 840 | 880 | 860 |
| 10 | F | 31 | 468 | Adrenal carcinoma (right) of 15 cm | 820 | 0 | ||||||
| Normal | values | 140 | 4.4-14.2 | >420 | Decrease by | |||||||
| 26-80 | >200 | >200 | ||||||||||
Dex, Dexamethasone.
The viability of the cells, as measured with the trypan blue exclusion technique, amounted to 85% or more in all cases. Cells (105) were incubated in a final volume of 1.0 mL Krebs- Ringer bicarbonate buffer containing 0.2% glucose, 0.5% BSA, and Ca2+ (7.65 mmol/L; pH 7.4). Synthetic ACTH(1-24) (te- tracosactin, Organon, Oss, The Netherlands) was added in four concentrations of 13.6 pmol/L (62 pg/mL), 33.9 pmol/L (154 pg/mL), 135.5 pmol/L (615 pg/mL), and 1355 pmol/L (6154 pg/mL). In addition, experiments were carried out with cholera toxin (Sigma) and forskolin (Calbiochem, La Jolla, CA). Max- imally active concentrations of cholera toxin and forskolin were used in these studies. In earlier (unpublished) studies with dispersed adrenocortical cells prepared from the hyperplastic adrenal glands of two other patients with Cushing’s disease we showed that 10 ug/mL cholera toxin and 10 uM forskolin were the maximally active cortisol release-stimulating concentra- tions.
All experiments were started by adding the test compounds. There were four vials per group. The cell suspensions were incubated at 37 C in a Dubnoff shaker in an atmosphere of 95% O2-5% Co2 for 120 min. The media then were removed, and different steroids were measured.
Steroid determinations
Cortisol was estimated using the y-coat RIA kit supplied by Clinical Assays (Cambridge, MA). 11-Deoxycortisol, 17-hy- droxyprogesterone, and progesterone were estimated by RIA after extraction. The antisera used were raised in the Depart- ment of Biochemistry, Division of Chemical Endocrinology, Erasmus University (Rotterdam, The Netherlands). The main cross-reacting steroids in the 11-deoxycortisol assay were 17- hydroxyprogesterone (11.4%), cortisol (1.4%), and progesterone (0.8%); those in the 17-hydroxyprogesterone assay were pro- gesterone (5.5%), 3,17-dihydroxy-5a-pregnan-20-one (1.6%), and 5a-pregnane-3,20-dione (1.0%). The specificity of the anti- sera used for the estimation of progesterone was previously described (16). Interassay variations for the estimations of cortisol, 11-deoxycortisol, progesterone, and 17-hydroxyproges- terone, expressed as coefficients of variations, were 9.5%, 10%, 12.1%, and 10%, respectively.
Statistics
The results are expressed as the mean ± SEM. Statistical evaluation was performed by analysis of variance.
Results
Basal cortisol release by the dispersed adrenocortical cells from patients 1-9 varied between 124 ± 10 and 281 ± 17 pmol/105 cells. 2 h (Table 2). There was no signifi- cant difference in cortisol release per cell between cells derived from hyperplastic adrenal glands and from (au- tonomous) adenomas. However, basal cortisol release from the cells prepared from the adrenal carcinoma of patient 10 amounted to less than 8% of that observed in the other cells (Table 2).
The responsiveness of the adrenocortical cells to four
concentrations of ACTH is shown in Table 2. Cortisol release from the cells obtained from diffusely hyperplas- tic adrenocortical cells (patients 1, 2, 3, and 4b) re- sponded to ACTH in a dose-dependent manner. The sensitivity to ACTH of the cells prepared from the in vivo autonomous adrenal adenomas varied considerably; no stimulation of cortisol release by ACTH was observed in the cells from patients 6 and 7, while the cells from patient 5 reacted only to the highest concentration of ACTH. In contrast the cells prepared from the adenoma from patient 8 reacted to the lowest concentration of ACTH with a significant increase in cortisol release. The single adenoma removed from the otherwise hyperplastic gland of patient 4 was not responsive to ACTH in vitro (Table 1, patient 4a). The cells prepared from the two micronodular adrenal glands obtained from patient 9 reacted to only a limited extent to ACTH, while the adrenocortical carcinoma cells from patient 10 were un- responsive to ACTH.
In the media obtained after the 2-h incubation of the dispersed adrenocortical cells from seven patients we also measured deoxycortisol, 17-hydroxyprogesterone, and progesterone (Table 2). The stimulatory effect of ACTH on the release of these precursor steroids tended to be higher than that on the release of cortisol (calcu- lated as a percentage of baseline control release). A statistically significant increase in deoxycortisol (and to a minor extent also in 17-hydroxyprogesterone and/or progesterone) was observed in reaction to lower concen- trations of ACTH in several autonomous cells than those needed to stimulate cortisol release (cf. patients 4a, 7, and 8).
Apart from their responsiveness to ACTH, the stimul- ability of hormone release of the dispersed adrenocortical cells was also investigated in response to maximally active concentrations of cholera toxin and forskolin. There was a parallel reaction of the cells to the maximal doses of ACTH, cholera toxin, and forskolin, (Fig. 1; P < 0.001 in all three instances). Again (as in their response to ACTH), the stimulatory effects of cholera toxin and on the release of deoxycortisol, 17-hydroxyprogesterone, and progesterone were more powerful (expressed as a percentage of control) than those on cortisol release. In patient 6, in whom no cortisol response to cholera toxin was found, and in patient 7, in whom no response of cortisol to either cholera toxin or forskolin had been observed, the cells responded to these secretagogues with a significant increase in cortisol precursors (Table 2).
Finally, it was asked whether measurement of the release of four different steroids from these dispersed adrenal cells might provide information with regard to the presence of possible enyzmatic blocks in the adrenal glands obtained from this group of patients with a variety of causes of Cushing’s syndrome. In Table 3 we expressed
| Patient no. | Hormone release (pmol/105 cells . 2 h) | |||
|---|---|---|---|---|
| Cortisol | Deoxycortisol | 17-Hydroxyprogesterone | Progesterone | |
| 1 | ||||
| Control | 128 ± 7 | 8.4 ± 0.2 | 4.2 ±0.2 | 7.8 ±0.6 |
| ACTH | ||||
| 13.6 pmol/L (I) | 358 ± 7ª | 24.2 ± 0.4ª | 17.2 ± 0.2ª | 25.0 ± 0.3ª |
| 33.9 pmol/L (II) | 450 ± 86 | 32.4 ± 0.66 | 21.0 ± 0.6b | 38.1 ± 0.8b |
| 135.5 pmol/L (III) | 517 ± 14° | 35.5 ± 0.96 | 26.5 ± 0.56 | 51.6 ± 1.2€ |
| 1355 pmol/L (IV) | 537 ± 12º | 41.7 ± 0.9€ | 32.0 ± 0.9cd | 65.3 ± 0.8c,d |
| Cholera toxin (10 µg/ml) | 382 ± 11ª | 24.4 ± 1.5ª | 22.3 ± 0.7ª | 41.4 ± 1.8ª |
| Forskolin (10 M) | 609 ± 18ª | 37.4 ± 2.3ª | 36.0 ± 0.7ª | 78.0 ± 2.3ª |
| 2 | ||||
| Control | 159 ± 5 | 10.6 ± 0.8 | 3.8 ± 0.1 | 2.8 ± 0.2 |
| ACTH | ||||
| I | 339 ± 19ª | 19.4 ± 1.4ª | 8.0 ± 0.6ª | 4.2 ± 0.4ª |
| II | 459 ± 96 | 29.4 ± 1.66 | 12.1 ± 0.76 | 6.1 ± 0.46 |
| III | 517 ± 13€ | 29.8 ± 0.66 | 13.9 ± 0.4ª | 6.1 ± 0.46 |
| IV | 545 ± 20€ | 35.2 ± 0.8€ | 14.3 ± 0.1€ | 6.5 ± 0.36 |
| Cholera toxin | 305 ± 21ª | 18.6 ± 1.4ª | 10.1 ± 0.4ª | 6.5 ± 0.6ª |
| Forskolin | 453 ± 13ª | 34.1 ± 3.0ª | 14.5 ± 0.7ª | 10.9 ± 0.4ª |
| 3 | ||||
| Control | 124 ± 8 | 10.1 ± 0.5 | 3.1 ± 0.3 | 2.4 ± 0.4 |
| ACTH | ||||
| I | 166 ± 9ª | 22.5 ± 0.7ª | 5.9 ± 0.2ª | 3.8 ± 0.2ª |
| II | 186 ± 7ª | 29.7 ± 2.86 | 8.1 ± 0.36 | 5.6 ± 0.36 |
| III | 214 ± 7b | 34.5 ± 1.7b | 10.6 ± 0.5b | 6.9 ± 0.4 |
| IV | 229 ± 56 | 35.8 ± 2.1b | 11.4 ± 0.76 | 8.2 ± 0.4€ |
| Cholera toxin | 180 ± 3° | 29.1 ± 1.2ª | 9.3 ± 0.2ª | 5.9 ± 0.3ª |
| Forskolin | 264 ± 4ª | 52.9 ± 4.1ª | 17.2 ± 0.8ª | 10.5 ± 0.3ª |
| 4a | ||||
| Control | 171 ± 6 | 149.4 ± 7.6 | 40.6 ± 0.3 | 17.5 ± 0.8 |
| ACTH | ||||
| I | 167 ± 13 | 148.2 ± 15.2 | 38.8 ± 0.6 | 15.5 ± 0.6 |
| II | 178 ± 11 | 142.6 ± 9.2 | 40.0 ± 1.5 | 14.1 ± 0.9 |
| III | 174 ± 8 | 164.4 ± 8.6 | 44.1 ± 0.6 | 17.5 ± 0.6 |
| IV | 179 ± 2 | 211.9 ± 8.9ª,b | 51.1 ± 0.3ª,b | 20.9 ± 0.9b,c |
| Cholera toxin | 199 ± 6ª | 186.0 ± 6.4ª | 76.3 ± 0.4ª | 30.7 ± 1.0ª |
| Forskolin | 214 ± 7ª | 217.0 ± 7.6ª | 107.0 ± 6.3ª | 49.0 ± 1.0ª |
| 6 | ||||
| Control | 166 ±8 | 63.0 ± 2.7 | 16.9 ± 0.8 | 21.2 ± 0.8 |
| ACTH | ||||
| I | 162 ± 16 | 76.3 ± 6.1 | 20.8 ± 0.7 | 23.8 ± 0.3 |
| II | 160 ± 9 | 60.8 ± 10.4 | 19.3 ± 0.6 | 23.5 ± 1.1 |
| III | 165 ± 12 | 67.8 ± 1.4 | 18.7 ± 0.5 | 22.3 ± 0.8 |
| IV | 171 ± 8 | 70.3 ± 4.7 | 20.5 ± 1.1 | 22.9 ± 1.7 |
| Cholera toxin | 190 ± 7 | 99.3 ± 3.0ª | 35.4 ± 2.3ª | 46.7 ± 1.2ª |
| Forskolin | 212 ± 9ª | 123.0 ± 7.0ª | 41.1 ± 1.3ª | 51.6 ± 2.5ª |
| 7 | ||||
| Control | 250 ± 5 | 116.8 ± 1.6 | 23.2 ± 0.5 | 13.0 ± 0.8 |
| ACTH | ||||
| I | 244 ± 17 | 126.0 ± 2.4 | 24.6 ± 0.5 | 13.9 ± 0.2 |
| II | 225 ±7 | 118.2 ± 4.3 | 22.7 ± 0.4 | 12.7 ± 0.7 |
| III | 238 ± 8 | 121.8 ± 4.7 | 25.0 ± 0.4 | 13.8 ±0.6 |
| IV | 257 ± 6 | 170.0 ± 8.3ª | 31.1 ± 0.2ª,b,c | 20.1 ± 0.7a, b, c |
| Cholera toxin | 263 ± 12 | 170.0 ± 6.9ª | 38.1 ± 0.9ª | 23.3 ± 0.7ª |
| Forskolin | 261 ± 6 | 191.8 ± 10.1ª | 55.1 ± 1.4ª | 31.0 ± 0.7ª |
Values are the mean ± SEM (n = 4 vials/group).
ª P < 0.01 vs. control.
· P < 0.01 vs. ACTH (13.6 pmol/L).
· P < 0.01 vs. ACTH (33.9 pmol/L).
d P < 0.01 vs. ACTH (135.4 pmol/L).
| Patient no. | Hormone release (pmol/105 cells . 2 h) | |||
|---|---|---|---|---|
| Cortisol | Deoxycortisol | 17-Hydroxyprogesterone | Progesterone | |
| 8 | ||||
| Control | 212 ± 4 | 49.2 ± 1.9 | 9.3 ± 0.6 | 9.5 ± 0.8 |
| ACTH | ||||
| I | 285 ± 5ª | 54.5 ± 4.3 | 12.3 ± 1.4 | 9.9 ± 0.4 |
| II | 266 ± 7ª | 61.8 ± 5.4ª | 10.0 ± 0.1 | 9.5 ± 0.9 |
| III | 294 ± 5ª | 64.5 ± 2.4ª | 12.3 ± 0.7 | 11.4 ± 1.3 |
| IV | 361 ± 5ª,6 | 87.0 ± 5.1%,b | 19.1 ± 1.8a,b, c | 21.0 ± 3.4a, b, c |
| Cholera toxin | 314 ± 8° | 82.3 ± 4.1ª | 17.6 ± 0.9ª | 17.3 ± 0.9ª |
| Forskolin | 337 ± 8ª | 101.5 ± 7.5ª | 26.3 ± 0.8ª | 28.8 ± 1.9ª |
| 9 | ||||
| Control | 281 ± 7 | |||
| ACTH | ||||
| I | 313 ± 9 | |||
| II | 353 ± 10° | |||
| III | 360 ± 9° | |||
| IV | 358 ± 6ª | |||
| Cholera toxin | 324 ± 5° | |||
| Forskolin | 414 ± 11ª | |||
| 10 | ||||
| Control | 9.5 ± 0.6 | 0.8 ± 0.1 | ||
| ACTH | ||||
| I | 8.6 ± 0.6 | 1.7 ± 0.2 | ||
| II | 8.7 ± 0.6 | 1.8 ± 0.1 | ||
| III | 8.8 ± 0.7 | 2.4 ± 0.1 | Not detectable | |
| IV | 11.6 ± 0.9 | 2.7 ± 0.1 | ||
| Cholera toxin | 8.6 ± 0.6 | 3.0 ± 0.1 | ||
| Forskolin | 10.5 ± 0.6 | 2.7 ± 1.1 | ||
600-
600-
maximal reaction to ACTH as % of basal
500-
500-
400
400-
300
300
200
200
100-
P=0.96
100
r=0.97
p<0.001
p<0.001
0
0
0
100
200
300
400
500
100
200
300
400
500
600
maximal reaction to cholera toxin as % of basal
maximal reaction to forskolin as % of basal
basal hormone release by the adrenal cell suspensions as a ratio of the different precursor steroids over cortisol. These data indeed show a higher relative amount of deoxycortisol secreted by the cells prepared from four autonomous adrenal adenomas than from hyperplastic adrenal glands (P < 0.01; Table 3).
Discussion
In the present study we investigated steroid hormone release by the dispersed cells prepared from the hyper- plastic or tumorous adrenal glands obtained from 10
patients with different types of Cushing’s syndrome.
If expressed per cell the amounts of cortisol released by cells from adenomatous or hyperplastic glands were similar. No difference in basal cortisol secretion was observed between different groups of Cushing’s patients. Only the cells removed from one patient with an adrenal carcinoma secreted less than 10% of the amount of cortisol released from the cells from the other nine pa- tients. It was previously recognized that adrenal carci- nomas are in general much larger than adenomas at the time of diagnosis. This might be attributed to a decreased
| Patient no. | Deoxycortisol/cortisol | 17-Hydroxyprogesterone/cortisol | Progesterone/cortisol |
|---|---|---|---|
| 1. Hyperplastic cells | 0.07 | 0.03 | 0.06 |
| 2. Hyperplastic cells | 0.07 | 0.02 | 0.02 |
| 3. Hyperplastic cells | 0.08 | 0.03 | 0.02 |
| 4a. Autonomous cells | 0.87 | 0.24 | 0.10 |
| 6. Autonomous cells | 0.38 | 0.10 | 0.13 |
| 7. Autonomous cells | 0.47 | 0.20 | 0.05 |
| 8. Autonomous cells | 0.23 | 0.04 | 0.04 |
The deoxycortisol/cortisol ratio between 1, 2, 3, and 4a, 6, 7, and 8 is significantly different (P < 0.01).
efficacy of steroidogenesis in carcinomatous tissue (2, 6).
The responsiveness of hormone release from the dis- persed adrenal cells to different concentrations of ACTH and to cholera toxin and forskolin yielded new data on the regulation of adrenal steroid production in different forms of Cushing’s syndrome. The concentrations of ACTH used and the steroid responses measured were similar to those previously observed in normal adrenal glands and in diced adrenal tissue from breast cancer patients. Honn and Chavin (17) found that a maximally stimulatory concentration of 3 nmol/L ACTH stimulated cortisol output to levels 4-5 times higher than control values. Most previous studies with human adrenal glands obtained from patients with Cushing’s syndrome showed that (partial) autonomy of the adrenal gland in vivo is generally also reflected by the functions of cells or tissues incubated in vitro, in that they become less responsive to ACTH than normal tissue (1-5). However, in one previous study a tumor causing Cushing’s syndrome, which was apparently not suppressible by dexametha- sone or stimulated by ACTH in vivo, responded to ACTH in vitro (7). This was later confirmed by O’Hare et al (8), who showed that most of 20 benign adrenal adenomas were responsive to ACTH in culture, whereas 4 of 5 malignant tumors showed abnormal or nonexistent re- sponses to ACTH.
Our study shows that the cells prepared from the hyperplastic adrenal glands from three patients with classical Cushing’s disease responded to ACTH in a dose- dependent manner. The cells prepared from the four patients with Cushing’s syndrome caused by an autono- mous cortisol-secreting adrenal adenoma did not respond to ACTH in two cases, showed a low sensitivity to ACTH in one, and clearly responded to ACTH in a dose-de- pendent fashion in another case. In fact, the ACTH sensitivity of the adrenal cells from one of the patients with Cushing’s disease (no. 3) was similar to that ob- served in this adenoma patient (no. 8). The partial re- sponsiveness of these adenoma cells to ACTH was not unexpected. Bertagna and Orth (6) studied a number of patients with adrenocortical tumors in vivo and also
found that adenomas which were uniformly independent of endogenous ACTH stimulation frequently responded to exogenous ACTH. Nishikanwa et al. (18) described in vitro studies with ACTH-unresponsive and -responsive adenomas. Basal steroid production and cholesterol ester content were high, and cAMP was low in the ACTH- unresponsive cells. We extended our study with regard to the dynamics of hormone production by the adreno- cortical cells by measuring their reaction to maximally active concentrations of the secretagogues cholera toxin and forskolin, two potent stimulators of the Ns protein and the catalytic subunit of adenylate cyclase, respec- tively (19). A close correlation was found between the responsiveness of the dispersed cells to ACTH, on the one hand, and to cholera toxin or forskolin, on the other. The parallel attenuation in the reaction to these three compounds in autonomous cells suggests that the defect in these cells is not located at the level of the ACTH receptor, but, rather, involves the adenylate cyclase com- plex as a whole or its coupling to cAMP-dependent protein kinase or even a stage beyond this point. In this respect the transplantable adrenocortical carcinoma in Osborne-Mendel rats (tumor 494) seems similar, in that the tissue is not responsive to stimulation by ACTH or cyclic nucleotides (20, 21), while a defect in cAMP- binding protein kinase activity was found (22). The low- ered responsiveness of autonomous adrenal adenomas to ACTH, cholera toxin, and forskolin differs considerably from the desensitizing effect of long term exposure of the normal adrenal gland to the ACTH. In rats bearing transplantable ACTH-secreting pituitary tumors, which have adrenocortical hyperplasia, we showed that the excessive activation of corticosterone release by the adre- nocortical cells in vitro was accompanied by a loss of sensitivity to ACTH (23). However, the sensitivity of these cells to cholera toxin and forskolin remained un- changed, suggesting that the mechanism involved in this phenomenon includes down-regulation of the ACTH re- ceptors or uncoupling of the ACTH-receptor complex from the Ns subunit, leaving the following train of intra- cellular events intact (23).
One potential problem with the data presented in this study is that the adrenal tissues obtained from these 10 patients were all processed on different days. Thus, if there was any variation in responsiveness due to subtle changes in the dispersant techniques between patients, this might influence the interpretation of the results. To investigate this possibility further, we prepared dispersed rat adrenal cells on 5 different days and measured the responsiveness of corticosterone release to ACTH. We found that the maximal response to ACTH varied by only 11%.
Earlier studies in which direct comparisons were made between hyperplastic and adenomatous adrenal tissue showed little or no difference between steroid secretion patterns (24-26). Only D’Agata et al. (27), studying the in vitro activities of several adrenocortical enzymes (not including 116-hydroxylase) in six adrenal adenomas, found increased 21-hydroxylase activity in adrenal ade- nomas compared with that in normal adrenal tissue. We measured the amounts of cortisol and its three major precursors deoxycortisol, 17-hydroxyprogesterone, and progesterone, as they were released over a period of 2 h into the medium. We observed significantly higher deox- ycortisol/cortisol ratios in the medium from adrenal ad- enoma cells than in that from hyperplastic adrenal cells. This suggests that a relative blockade of 118-hydroxylase activity might be a characteristic of adenoma formation in the human adrenal gland. We cannot exclude, how- ever, that the relative blockade of 118-hydroxylase might also be caused by differing degrees of ACTH stimulation of the tissues before adrenalectomy. In accordance with this suggestion is the observation that the adrenal cells of patient 8, which presumably had been exposed to ACTH in vivo, seem to show ratios intermediate to those of the hyperplasia and adenoma groups.
Our studies shed some further light on the different forms of autonomy at the adrenal level and the possible transition from pituitary to adrenal autonomy, as they have been hypothesized in a number of patients with different forms of Cushing’s syndrome (9-14). In this respect the in vivo and in vitro data of patient 4 are helpful. She showed the classical picture of pituitary- dependent Cushing’s disease with normal circulating ACTH levels, clear stimulability and suppressibility of the pituitary-adrenal axis in the dynamic tests, and a high sensitivity of cortisol release by the cells prepared from the two diffuse hyperplastic adrenal glands to ACTH. The diagnosis had been previously further sub- stantiated by the incomplete removal of an ACTH-se- creting pituitary adenoma. Surprisingly, a single encap- sulated adenoma with a diameter of 4 cm was detected in the otherwise hyperplastic left adrenal gland. Cortisol release by the cells prepared from this adenoma was unresponsive to ACTH, but responded to a minor extent
to cholera toxin and forskolin. In addition, deoxycortisol, 17-hydroxyprogesterone, and progesterone release by these cells was stimulated by high dose ACTH treatment. This patient differed from the previously described pa- tients with Cushing’s disease with a coexisting single macronodule of the adrenal cortex (9-14, 28, 29) in that these nodules always consisted of hyperplastic tissue and were not surrounded by a capsule, while in vitro studies (as carried out in three cases) showed a normal or a hyperresponse of cortisol to ACTH compared with that of the surrounding diffuse hyperplastic adrenal tissue. A dual control with various degrees of autonomy of the adrenal in some cases and the predominance of pituitary control in others has been hypothesized (12, 14, 29, 30). During the course of the disease such an adrenal nodule may become autonomous, which eventually will result in suppression of pituitary ACTH release and of the unin- volved adrenal cortex (28, 29). Thus, the solitary ma- cronodules in the patients previously described (9, 11, 12, 14, 28) may have represented an intermediary stage in the transition between diffuse adrenal hyperplasia and the appearance of an autonomous adrenocortical ade- noma, while the single encapsulated adenoma of patient 4, which was unresponsive in vitro to ACTH, might even represent another further step in the process of this transition, just before autonomous cortisol release by this adenoma (completely) suppresses pituitary ACTH release and eventually also suppresses the activity of the surrounding and contralateral adrenal gland. The final and last step of this transition might be represented by patient 8. In this patient a 4-cm encapsulated adenoma was found in the right adrenal gland, which was sur- rounded by diffuse adrenocortical hyperplasia. In vivo cortisol release by this adenoma was autonomous, in that there was an insufficient stimulation of the pituitary- adrenal axis in response to metyrapone, while dexameth- asone did not suppress circulating cortisol levels. Plasma ACTH levels, however, were normal. In vitro, the ade- noma cells responded to ACTH, cholera toxin, and for- skolin in a manner similar to, for example, the response of the cells prepared from the diffuse hyperplastic glands from patient 3 with classical Cushing’s disease. A differ- ence, however, was the very high deoxycortisol/cortisol ratio of the steroid output by these adenoma cells.
In conclusion, we have presented evidence that adrenal adenoma formation in patients with Cushing’s syndrome is characterized by a parallel decrease in the stimulation of the release of steroid hormones by the dispersed ad- renal adenoma cells to ACTH, cholera toxin, and forsko- lin. This means that the defect in these adenoma cells is located beyond the ACTH receptor. Adrenal adenoma formation is further characterized by a relative blockade of 11-hydroxylase. Careful comparison of in vivo and in vitro data in several of our patients with Cushing’s syn-
drome indeed pointed to the presence of a gradual tran- sition from pituitary to (partial) adrenal autonomy.
References
1. Whitehouse BJ, Vinson GP. Steroid hormone production in normal and abnormal human adrenocortical tissue. In: Fotherby K, Pal SB, eds. Hormones in normal and abnormal human tissues. Berlin: de Greuter; 1981;9:215-55.
2. Lipsett MB, Hertz R, Ross GT. Clinical and pathophysiological aspects of adrenocortical carcinoma. Am J Med. 1963;35:374-83.
3. Neville AM. The human adrenal cortex: correlation of structure and function. In: Griffiths K, Cameron EHD, eds. The human adrenal cortex and its relation to breast cancer. Cardiff: Alpha Omega Alpha; 1969;28-35.
4. Saez JM, Dazord A, Gallet D. ACTH and prostaglandin receptors in human adrenocortical tumours. Apparent modification of a specific component of the ACTH binding site. J Clin Invest. 1975;56:536-47.
5. Cowan JS, Kinson GA, Poznanski WJ. Effect of corticotrophin on cortisol and testosterone secretion in suspensions of normal and abnormal human adrenal cells. J Endocrinol. 1977;72:247-8.
6. Bertagna C, Orth DN. Clinical and laboratory findings and results of therapy in 58 patients with adrenocortical tumors admitted to a single medical center (1951-1978). Am J Med. 1981;71:855-75.
7. Voigt KH, Fehm HL, Pfeiffer EF. Dissociation of in vivo and in vitro “autonomy” in a human adrenocortical tumour. Acta Endo- crinol (Copenh). 1975;78:302-5.
8. O’Hare MJ, Ellison ML, Neville AM. Tissue culture in endocrine research; perspectives, pitfalls and potentials. In: James VHT, Martini L, eds. Current topics in experimental endocrinology. London: Academic Press; 1978;3:1-56.
9. Steingart DE, Tsao HS. Coexistence of pituitary adrenocorticotro- pin-dependent Cushing’s syndrome with a solitary adrenal ade- noma. J Clin Endocrinol Metab. 1980;50:961-6.
10. Aron DC, Findling WJ, Fitzgerald PA et al. Pituitary ACTH dependency of nodular adrenal hyperplasia in Cushing’s syndrome: report of two cases and review of the literature. Am J Med. 1981;71:302-6.
11. Lamberts SWJ, Bons EG, Bruining JA. Different sensitivity to adrenocorticotropin of dispersed adrenocortical cells from patients with Cushing’s disease with macronodular and diffuse adrenal hyperplasia. J Clin Endocrinol Metab. 1984;58:1106-10.
12. Leiba S, Shindel B, Weinberger I, et al. Cushing’s disease coexisting with a single macronodule simulating adenoma of the adrenal cortex. Acta Endocrinol (Copenh). 1986;112:323-8.
13. Watanoba H, Kawagishi T, Hirai Y, et al. Cushing’s syndrome presenting the coexistence of a pituitary corticotropic cell hyper- plasia and a unilateral functional adrenal adenoma. Acta Endocri- nol (Copenh). 1985;110:302-7.
14. Hermus AR, Gerlach F, Pieters F, et al. Transition from pituitary- dependent to adrenal-dependent Cushing’s syndrome. N Engl J
Med. 1988;318:966-70.
15. Lamberts SWJ, Klijn JGM, de Jong FH. The definition of true recurrence of pituitary-dependent Cushing’s syndrome after trans- sphenoidal operation. Clin Endocrinol (Oxf). 1987;26:707-12.
16. Lamberts SWJ, Bons EG, Bruining HA, de Jong FH. Differential effects of the imidazole derivatives etomidate, ketoconazole and miconazole and of metyrapone on the secretion of cortisol and its precursors by human adrenocortical cells. J Pharmacol Exp Ther. 1986;240:259-64.
17. Honn KV, Chavin W. In vitro temporal cAMP and cortisol re- sponses to ACTH by the normal human adrenal gland. Acta Endocrinol (Copenh). 1977;85:823-31.
18. Nishikawa M, Mikami K, Tamura Y, Yamamoto M, Kumagai A. Comparative study of cyclic AMP-generating system and lipid metabolism in vitro in ACTH responsive and unresponsive adrenal tumours. Endocrinol Jpn. 1979;26:9-17.
19. Gilman AG. G Proteins and dual control of adenylate cyclase. Cell. 1984;36:577-9.
20. Ney RL, Hochella NJ, Grahame-Smith DB, Dexter RN, Butcher RW. Abnormal regulation of adenosine-3’5’-monophosphate and corticosterone formation in an adrenocortical carcinoma. J Clin Invest. 1969;48:1733-9.
21. Sharma RK, Hashimoto K. Ultrastructural studies and metabolic regulation of isolated adrenocortical carcinoma cells of rat. Cancer Res. 1972;32:666-74.
22. Sharma RK, Shankar G, Ahrens H, Ahmed NK. Partial purifica- tion and characterization of the defective cyclic adenosine 3’5’ monophosphate binding protein kinase from adrenocortical carci- noma. Cancer Res. 1977;37:3297-300.
23. Lamberts SWJ, Bons EG, Zuiderwijk-van der Roest JM. Studies on the mechanism of corticotrophin-mediated desensitization of corticosterone secretion by rat adrenocortical cells. Mol Cell En- cocrinol. 1987;52:243-9.
24. Charreau EH, Dufau ML, Villee DB, Villee CA. Synthesis of 5a- pregnane-3,20-dione by fetal and adult human adrenals. J Clin Endocrinol Metab. 1968:28:629-32.
25. Cameron EHD, Beynon MA, Griffiths K. The role of progesterone in the biosynthesis of cortisol in human adrenal tissue. J Endocri- nol. 1968;41:319-26.
26. Ward PJ, Grant JK. Metabolism of progesterone-4-C14 by adre- nocortical tissue from patients with Cushing’s syndrome. J Endo- crinol. 1963;26:139-47.
27. D’Agata R, Malozowski S, Barkan A, Cassorla F, Loriaux D. Steroid biosynthesis in human adrenal tumors. Horm Metab Res. 1987;19:386-8.
28. Josse RG, Bear R, Kovacs K, Higgins HP. Cushing’s syndrome due to unilateral nodular adrenal hyperplasia: a new pathophysio- logical entity? Acta Endocrinol (Copenh). 1980;93:495-504.
29. Smals AGH, Pieters GFFM, Van Haelst UJG, Kloppenborg PWC. Macronodular adrenocortical hyperplasia in long standing Cush- ing’s disease. J Clin Endocrinol Metab. 1984;58:25-31.
30. Bricaire H, Luton JP, Ghozland M, Forest M. La maladie microad- enomatose de la cortico-surrenale dans le syndrome de Cushing. Ann Med Intern (Paris). 1970;121:755-77.