Androgen Synthesis in a Patient with Virilizing Adrenocortical Carcinoma1
DOROTHY B. VILLEE,2 HOWARD ROTNER,3 BERNARD KLIMAN,3 CAESAR BRIEFER, JR.,3 AND DANIEL D. FEDERMAN3 Departments of Pediatrics and Medicine of the Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts
ABSTRACT. A 16-yr-old girl with a virilizing adrenocortical carcinoma was studied clinically before and after surgery, and portions of the tumor were incubated in vitro with a variety of labeled steroids. The virilization of the patient was explained by a higher than normal concen- tration of testosterone in her plasma and by in vitro evidence of extensive androgen synthesis. Compared to adrenals from normal adults, the tumor showed a relative lack of 118-hydroxyla- tion, but very active 16a-hydroxylation in vitro.
When the tumor was exposed to 10-5M proges- terone in organ culture for 24 hr, and subse- quently incubated with progesterone-4-14C, more 21-, 17-, and 16a-hydroxylated products of progesterone were isolated than from compara- ble incubations of explants previously cultured without progesterone. In vivo, the tumor re- sponded slightly to ACTH stimulation but not to dexamethasone suppression. (J Clin Endocr 27: 1112, 1967)
F UNCTIONING adrenocortical carci- nomas may produce a variety of meta- bolic disturbances including Cushing’s syn- drome, feminizing and virilizing syndromes, hypokalemic alkalosis, and hypoglycemia (1-5). We have recently cared for a 16- year-old girl with a virilizing adrenocortical
Received September 30, 1966; accepted April 18, 1967.
Supported by USPHS Grants AM 08026, AM 6731, AM 4501, AM 5205 and American Cancer Society Grant IN-42-D.
1 Trivial names used refer to the following: preg- nenolone =38-hydroxy-45-pregnen-20-one; 17-hy- droxypregnenolone = 36,17-dihydroxy-45-pregnen- 20-one; 16a-hydroxypregnenolone =36,16a-dihy- droxy-45-pregnen-20-one; dehydroepiandrosterone =36-hydroxy-45-androsten-17-one; 16@-hydroxy- dehydroepiandrosterone =36,16a-dihydroxy-45-an- drosten-17-one; 17-hydroxyprogesterone =17-hy- droxy-44-pregnene-3,20-dione; 16a-hydroxyproges- terone =16a-hydroxy-44-pregnene-3,20-dione; an- drostenedione = 44-androstene-3,17-dione; 1. hydroxyandrostenedione = 118-hydroxy-44- andro- stene-3,17-dione; 11-deoxycortisol =17,21-dihydroxy- 44. pregnene-3,20-dione; 16@-hydroxydehydroepi- androsterone acetate =3a,16a-diacetoxy-45-andro- sten-17-one; 16a-hydroxypregnenolone acetate =3a, 16a-diacetoxy-45-pregnen-20-one; dexamethasone =9a-fluoro-16a-methyl-118,17,21-trihydroxy-41,4- pregnadiene-3,20-dione.
2 Biochemical studies.
3 Clinical studies.
carcinoma which eventually produced overt Cushing’s syndrome. The concentration of testosterone in her plasma exceeded that of the normal adult female. Incubation of homogenized tumor with labeled steroid substrates showed that it was capable of synthesizing relatively large amounts of dehydroepiandrosterone, androstenedione, and testosterone. The pattern of hormone production of the tumor in vitro, as well as its sensitivity to hormonal influences, both in vivo and in vitro, was studied. Evidence is presented for a relative decrease of 118-hydroxylation and a high degree of 16a-hydroxylation in the tumor in vitro, compared to normal adrenal tissue.
Case Report
C. B. (MGH 130-90-47), a 16-yr-old Cauca- sian girl, entered the Massachusetts General Hospital in October 1964 because of hirsutism of approximately 1 yr’s duration. She had been well all her life and had had the onset of regular menses at age 11. The patient had noted exces- sive growth of hair, deepening of her voice, and a more oily skin. The remainder of the past his- tory and family history was unremarkable.
The patient, an attractive, alert young girl, was 172 cm in height and weighed 63 kg. The
| Urine 17-OHCS (mg/24 hr) | Urine 17-KS (mg/24 hr) | Plasma cortisol (ug/100 ml) | Plasma testosterone (ug/100 ml) | |
|---|---|---|---|---|
| Normal | ||||
| Base line | 3-8 | 6-15 | 5-25 | F 0.00-0.08 |
| After ACTH | 3- to 5-fold increase | 2- to 3-fold increase | 30-55 | M 0.30-0.80 |
| After dexamethasone | <50% control | <50% control | 4 | |
| 0.5 mg/6 hr for 2 days | ||||
| Patient C.B. preoperative | ||||
| Control | 6.8 | 76.6 | 13 | 0.23 |
| Control | 5.5 | 83.5 | ||
| After ACTH | 10.1 | 107.1 | 18 | 0.23 |
| After dexamethasone 2 | 9.0 | 102.2 | 12 | 0.14 |
| mg/6 hr for 2 days | ||||
| Patient C.B. postoperative | ||||
| Control | 6.2 | 1.4 | 16 | |
| Control | 4.2 | 1.2 | ||
| After ACTH | 17.3 | 8.2 | 38 | |
| After dexamethasone 0.5 | 3.2 | 1.3 | 0.05 | |
| mg/6 hr for 2 days | ||||
| Patient C.B. after relapse | 12 | 450-600 | 22 | 0.31 |
Abbreviations: 17-OHCS =17-hydroxycorticosteroids; 17-KS =17-ketosteroids.
pulse was 76 and the blood pressure 125/85. There was hypertrichosis of the face, breasts, abdomen, and extremities, with a male escutch- eon but no recession of the hairline. Although the skin was slightly oily, no acne was present. In addition, there were no striae, ecchymoses, or abnormal fat deposits and the skin was of normal texture. Breast development was nor- mal for her age with no lactorrhea. Abdominal examination revealed a hard, immobile, non- tender mass deep in the left upper quadrant near the midline. The clitoris was enlarged, measuring 2 by 0.5 cm.
Vaginal smears showed no malignant cells and 100% basal cells. Buccal smears revealed 29% chromatin bodies. Roentgenograms of the chest and skull were normal. Intravenous py- elogram revealed the upper pole of the left kid- ney to be displaced toward the midline. The urinary 17-ketosteroids were 76 mg/24 hr (nor- mal range 6-15 mg), and the 17-hydroxyster- oids were 6.8 mg/24 hr (normal range 3-8 mg). A diagnosis of virilizing adrenal tumor, prob- ably carcinoma, was made and a laparotomy was performed.
At surgery, a large, irregular, hard tumor was found to have totally replaced the left adrenal gland. The tumor was removed and examined by Dr. Richard B. Cohen, Pathologist. Adreno- cortical carcinoma with invasion of surrounding lymphatic tissue was found. The tumor mea- sured 16 ×12 ×6.8 cm and was firm, non-encap- sulated, and yellow in color, with extensive hemorrhage throughout. Microscopic examina- tion showed nests and cords of tightly packed
fairly uniform cells. They contained scant, oc- casionally vacuolated, acidophilic cytoplasm and small round or oval nuclei which were uni- form and only occasionally hyperchromatic. The cells were, in general, smaller than normal adrenal cortical cells. Nowhere in the tumor was there noted to be any attempt at duplica- tion of normal adrenal cortical architecture. Mitotic figures were infrequent. Fat stains re- vealed areas in which the tumor cells contained considerable cytoplasmic lipid. Many of the cells suggested, in their appearance and lipid content, an adrenocortical origin.
Postoperatively, the patient’s rate of hair growth decreased. The urinary 17-ketosteroids were still slightly elevated, 20 mg/24 hr, where- as the 17-hydroxycorticosteroids were in the low normal range, 3-5 mg/24 hr. The patient was admitted to the hospital for the second time in June 1965 with complaints of right rib and shoulder pain and a swelling in the right epigastrium of 4 days’ duration. She had missed her last menstrual period. She now appeared acutely ill. Recurrent hirsutism and, for the first time, extensive acne were present. The cli- toris was further enlarged. The urinary 17-keto- steroids were markedly elevated to 450-580 mg /24 hr, and, for the first time, the 17-hydroxy- corticosteroids in the urine were also elevated to values between 7 and 12 mg/24 hr. Chest x-ray examination showed pulmonary metastases in both lung fields. A vaginal smear showed 72% basal cells, 23% intermediate cells, and 5% superficial cells. Urinary gonadotropins were less than 5 rat U/24 hr. The patient pursued a
| Estrone (ug/24 hr) | Estriol (ug/24 hr) | Estradiol (ug/24 hr) | |
|---|---|---|---|
| Normal | |||
| Follicular phase | 1-15 | 1-15 | 0.1-5 |
| Luteal phase | 10-23 | 10-27 | 2-8 |
| After ACTH | |||
| Follicular phase | 2- to 3-fold increase | ||
| Luteal phase | same as control values | ||
| Patient C.B. before surgery | |||
| Control (luteal) | 13.2 | 24.9 | 4.2 |
| After ACTH (luteal) | 16.5 | 14.0 | 3.2 |
| Patient C.B. after relapse | |||
| Control | 1430 | 1360 | 221 |
downhill course in the hospital and died a few weeks later of massive gastro-intestinal bleed- ing.
Clinical Studies
Methods. Urinary 17-hydroxycorticosteroids were determined by the method of Porter and Silber (6). Urinary 17-ketosteroids were mea- sured using a modified Zimmermann reagent (7). Plasma cortisol and 11-deoxycortisol were determined by the double isotope dilution method of Kliman and Peterson (8). Plasma testoterone was assayed using double isotope dilution with the separation of plasma testos- terone by paper, thin layer and gas chromatog- raphy as reported by Kliman and Briefer (9). The ACTH test was performed by infusing 50 IU ACTH in a liter of normal saline intrave- nously at a constant rate from 8:00 AM to 4:00
PM. Blood samples for plasma cortisol were drawn just before and 4 hr after the start of the infusion. Pituitary suppression was attempted by oral administration of dexamethasone, either 0.5 or 2.0 mg every 6 hr for 2 days. Urine was collected for measurement of 17-hydroxycorti- costeroids and 17-ketosteroids, and blood was drawn for determination of cortisol, 11-deoxy- cortisol, and testosterone, before and after the administration of dexamethasone. Urinary es- trogens were determined by a combined fluoro- metric and isotope method of high specificity (10).
Results. Preoperatively, the urinary 17-keto- steroids were markedly elevated, from 76-83 mg/24 hr (Table 1). They increased slightly to 107 mg/24 hr during ACTH infusion, but showed no decrease during dexamethasone ad-
| Vessel | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| Substrate | Preg-7a-3H (7 uc) | 17-OH Preg-7a-3H (13 uc) | And-4-14C (6 uc) | Prog-4-14C (4 uc) |
| SA | 10 mc/umole | 10 mc/umole | 30 uc/umole | 30 uc/umole |
| % substrate not metabolized % of radioactive products: | 5 | 8.4 | 69.2 | 10 |
| Dehydroepiandrosterone | 1.5 | 24.2 | ||
| Androstenedione | 7.6 | 36.9 | 69.2 | 0 |
| Testosterone | 1.7 | 7.3 | 12.1 | 0 |
| 118-OH-androstenedione | + | + | 3.6 | + |
| Corticosterone | 2.2 | + | ||
| Cortisol | 0.38 | 5.3 | 3.6 | |
| 16a-OH-cortisol | 10.2 | 2.8 | ||
| Radioactivity in 16a-OH DHA region | 27.0 | 7.5 | ||
| 16a-OH-progesterone | 11.1 | |||
| 17-OH-progesterone | 9.1 | |||
| Deoxycorticosterone | 8.2 | |||
| 17-OH-pregnenolone | 6.3 | |||
| Radioactivity in region of 11-deoxycortisol | 15.0 | |||
| Radioactivity in column washing | 15 | 3.5 | 1.1 | 0.2 |
In each vessel 635 mg homogenized tissue was incubated with the radioactive substrate in the presence of 1 ml phosphate buffer and 0.3 ml each of DPN (10-2M), TPN (10-2M) and glucose-6-phosphate (0.1M) for a period of 1 hr at 37 C with constant shaking. + =small amount of radioactivity associated with steroid. Abbreviations: Preg-7a-3H = pregnenolone-7a-3H, 17-OH-preg-7a-3H =17-hydroxypregnenolone- 7a-3H, And-4-14C = androstenedione-4-14C, prog-4-14C = progesterone-4-14C, 16a-OH-DHA =16a-hydroxy- dehydroepiandrosterone.
CPM
OD205
or
☒
☐ OD
240
80,000
ANDROSTENEDIONE
DEHYDROEPIANDROSTERONE
TESTOSTERONE + 17-OH PROGESTERONE
11 8-OH ANDROSTENEDIONE
16 a- OH PROGESTERONE
CPM
0.4
OD
☒ ☐
16 a-OH DHA AC.
30,000
0.3
S AC.
B AC.
60,000
0.3
20,000- 0.2
10,000- 0.1
40,000
0.2
0
0
11-DEOXYCORTISOL
TUBE NO.
0
100
200
20,000
0.1
11-DEOXYCORTISOL + CORTICOSTERONE
0
0
+
+
3:1
2:1
PROGESTERONE
60,000
DEHYDROEPIANDRO-
17-OH PROGESTERONE
0.3
ANDROSTENEDIONE
STERONE
DEOXYCORTICOSTERONE
TESTOSTERONE
11 A-OH ANDROSTENEDIONE
17-OH PREGNENOLONE
40,000
0.2
CORTISOL
20,000
0.1
0
0
;
TUBE NO.
0
100
200
300
400
500
600
700
800
900
+
+
3:1
2:1
ministration. The urinary 17-hydroxycortico- steroids and the plasma cortisol value were nor- mal preoperatively and showed no response to ACTH or dexamethasone. The plasma con- tained 0.23 ug testosterone/100 ml, well above the range for normal females, and was not ap- preciably changed by ACTH or dexametha- sone. Total urinary estrogens (Table 2) were within normal limits for the luteal phase of the cycle and, appropriately, did not increase when ACTH was administered (11). After surgery the 17-ketosteroids decreased to subnormal lev- els but increased 6-fold in response to ACTH. The resting urinary 17-hydroxycorticosteroids and plasma cortisol were also normal, and now showed normal responses to stimulation and
suppression. The concentration of testosterone in the plasma was now within the normal female range. At the time of widespread recurrence, shortly before the patient’s death, a 24-hr urine sample contained 500 mg of 17-ketosteroids, 12 mg of 17-hydroxycorticosteroids, and over 3 mg of estrogens. The concentration of testos- terone in the plasma was again high.
Biochemical Studies
Materials and methods
Incubation. 2.54 g Tumor tissue obtained at operation was homogenized immediately in phosphate buffer (12) and aliquots of the ho- mogenized tissue were pipetted into each of 4 incubation flasks containing radioactive sub-
| Radioactive product | Column or CCD | Crystallization | Derivative | Cryst. or TLC of derivative | Total cpm X10-3 | % of Total radio- activity | |||
|---|---|---|---|---|---|---|---|---|---|
| And-3H | 384 | CI (A) | 209 | ||||||
| (1) | CII (B) | 240 | monoxime | 232 | 417 | 7.6 | |||
| MLII | 213 | dioxime | 232 | ||||||
| And-3H | 2460 | CI (B) | 2990 | dioxime | 2500 | 3525 | 36.9 | ||
| (2) | MLI | 2550 | |||||||
| DHA-3H | CI (A) | 63 | |||||||
| (1) | CII (B) | 54 | acetate | CI | (B) | 55 | 83 | 1.5 | |
| MLII | . 63 | MLI | 50 | ||||||
| DHA-3H | 1310 | CI (A) | 1310 | ||||||
| (2) | CII (B) | 1131 | acetate | CI | (B) | 1082 | 2307 | 24.2 | |
| MLII | 1183 | MLI | 1056 | ||||||
| Test-3H | 281 | CI (A) | 166 | ||||||
| (1) | CII (B) | 150 | |||||||
| CIII (B) | 136 | acetate | CI | (A) | 140 | 95 | 1.7 | ||
| MLIII | 151 | MLI | 131 | ||||||
| Test-3H | acetate | CI | (A) | 340 | |||||
| (2) | MLI | 350 | |||||||
| CII | (B) | 332 | 704 | 7.3 | |||||
| MLII | 335 | ||||||||
| Test-14C | CI (A) | 180 | acetate | CI | (A) | 195 | 430 | 12.1 | |
| (3) | MLI | 170 | MLI | 175 | |||||
Numbers represent cpm/ug of steroid. Numbers in parentheses refer to the number of the incubation vessel. Each radioactive product was isolated from material subjected to column chromatography and/or countercurrent distribution (CCD). Additional carrier steroid was added to this material, and the mixture was crystallized. Subsequently, a derivative of the crystallized material was formed and the derivative was purified by thin layer chromatography (TLC) or crystallization (Cryst.). Total cpm of the product were calculated on the bases of the final specific activity, corrected for dilution at the time of crystalliza- tion. (A) =methanol and water, (B) =acetone and water, And-3H =androstenedione-3H, DHA-3H =de- hydroepiandrosterone-3H, Test-3H = testosterone-3H, Test-14C =testosterone-14C.
strates (Table 3). Glucose-6-phosphate, DPN, and TPN were added in excess. The vessels were placed in a Dubnoff incubator and kept at 37 C with constant shaking for 1 hr. The con- tents of the vessels were then frozen until the time of analysis.
Analysis. Steroid carriers were added to the contents of the incubation flasks and the mix- ture was then extracted exhaustively with redis- tilled ethyl acetate. The extract was dried in vacuo and the total cpm of the residue was deter- mined. The residue was chromatographed on a 20 g Hyflo Super-Cel partition column using 2,2,4-trimethylpentane as the moving phase and 90% methanol in water as the stationary phase. The column elution technique was simi- lar to that reported previously (12); however, experience has shown that in order to separate certain pairs of steroids (in particular, testos- terone and 17-hydroxyprogesterone) it is nec- essary to use a stepwise gradient consisting ini- tially of a mixture of trimethylpentane (3
parts) and 1,2-dichloroethane (1 part). This mixture was placed in an upper reservoir con- nected to a lower reservoir containing 400 ml of trimethylpentane. The lower reservoir was equipped with a stirring bar to ensure complete mixing of the gradient with the contents of the lower reservoir. The effluent of the lower reser- voir ran through the column containing the sample, and the column effluent was collected in 1 ml fractions. The fractions were analyzed for radioactivity and steroid content as de- scribed before (12). The recovery of added ste- roids, calculated from the weights of the car- riers added, ranged from 70 to 90%. A 2:1 gra- dient (2 parts trimethylpentane plus 1 part di- chloroethane) was introduced at a later stage, and still later a 1:1 gradient (Fig. 1). Pure di- chloroethane was placed in the upper reservoir for the last fractions and the column was finally washed with redistilled methanol.
Purification of radioactive products. A variety of methods including countercurrent distribu-
CPM
OD 280
or
OD 240
80,000
0.4
PROGESTERONE
ANDROSTENEDIONE
DEOXYCORTICOSTERONE
17-OH PROGESTERONE + TESTOSTERONE
11-BOH ANDROSTENEDIONE
11-DEOXYCORTISOL + CORTICOSTERONE
TESTOSTERONE
11-BOH ANDROSTENEDIONE
60,000
0.3
40,000
>ESTRADIOL-178
0.2
A
ESTRONE
20,000
0.1
0
0
1
TUBE NO. 0
100
200
300
400
500
0
100
200
300
3:1
2:1
DCE
tion, thin layer chromatography, crystallization and derivative formation were used to assess the radiochemical homogeneity of the products (Tables 4-7). Steroid acetates were formed by dissolving the steroids in redistilled pyridine and reacting them with redistilled acetic anhy- dride overnight at 37 C. Excess reagents were re- moved under a stream of nitrogen. Steroid ox- imes were prepared by the method of Brooks et al. (13).
Solvents and reagents. All organic solvents ex- cept methanol (reagent grade, J. T. Baker Chemical Co.) were redistilled. The radioactive steroids were obtained from New England Nu- clear Corp. and were tested for impurities by column chromatography. DPN, TPN, glucose- 6-phosphate and the steroids used as carriers were obtained from Sigma Chemical Co. The 16a-hydroxyprogesterone and 16x-hydroxycor- tisol were obtained through the generosity of Dr. Samuel Solomon, Montreal, Canada.
Organ culture of tumor tissue. Explants (each approximately 5 mg) of tumor tissue were pre- pared and cultured in the presence and absence of 10-5M progesterone (14). The pattern of ac- tivity of enzymes involved in the metabolism of progesterone was assessed after 24 hr in organ culture. Explants were homogenized in phos- phate buffer and incubated in the presence of
progesterone-4-14C (1 µc), DPN (2.7 umoles), TPN (2.7 umoles), and glucose-6-phosphate (24 umoles) for 1 hr at 37 C with constant shak- ing. Steroid carriers were added after incuba- tion and the mixture was extracted with ethyl acetate. Radioactive products were purified as described for the incubation of the fresh tumor tissue.
Results
Incubations of fresh tissue in vitro
Androgen synthesis. Androstenedione-3H, dehydroepiandrosterone-3H, and testoster- one-3H were formed from both pregneno- lone-7a-3H and 17-hydroxypregnenolone- 7a-3H (Fig. 1). Sixteen times as much dehydroepiandrosterone-3H was isolated from vessel 2 (Table 3) as from vessel 1, suggesting that the 17-hydroxylation of pregnenolone may be a rate-limiting step in the synthesis of dehydroepiandrosterone. There was a 4- to 5-fold difference in the amount of androstenedione-3H and testos- terone-3H isolated from the two vessels. No testosterone-14C or androstenedione-14C
| Radioactive product | Column or CCD | Crystallization | Derivative | Cryst. or TLC of derivative | Total cpm X10-3 | ||
|---|---|---|---|---|---|---|---|
| Cortisol-3H | 36 | CI (B) | 28 | acetate | 23 | 20.7 | |
| (1) | MLI | 39 | |||||
| Cortisol-3H | 425 | CI (B) | 430 | acetate | 440 | 506.0 | |
| (2) | MLI | 450 | |||||
| Cortisol-14C | 150 | CI (A) | 130 | ||||
| (4) | CII (B) | 135 | |||||
| MLII | 150 | ||||||
| CIII(C) | 140 | acetate | CI MLI (A) | 128 | 172.2 | ||
| MLIII | 150 | 132 | |||||
| 16a-Hydroxycortisol-3H (1) | 790 | acetate | CCD | 870 | |||
| TLC | 800 | ||||||
| CI (A) | 750 | 562.0 | |||||
| MLI | 820 | ||||||
| 16a-Hydroxycortisol-3H (2) | 258 | acetate | TLC | 260 | |||
| CI (A) | 255 | 266.0 | |||||
| MLI | 260 | ||||||
| 118-Hydroxyandrostene- | 232 | CI (A) | 183 | ||||
| dione-14C | CII (B) | 174 | dioxime | CI (A) | 158 | ||
| (3) | MLII | 208 | CII (B) | 156 | 128.0 | ||
| MLII | 160 | ||||||
| Corticosterone-3H | 220 | CI (A) | 112 | ||||
| (1) | CII (B) | 96 | monoxime | 112 | 122.0 | ||
| MLII | 100 | dioxime | 95 | ||||
Numbers represent cpm/ug of steroid. Numbers in parentheses refer to the number of the incubation vessel. In each case the product isolated from either column chromatography or countercurrent distribu- tion (CCD) was crystallized and/or a derivative of the material was formed. The derivative was subse- quently purified by crystallization, countercurrent distribution, or thin layer chromatography (TLC). The total cpm are calculated from the final specific activity, corrected for the addition of carrier at the time of crystallization. (A) =methanol and water, (B) =acetone and water, (C) =cyclohexane and ethyl acetate.
could be detected in vessel 4 (Fig. 2). This finding suggests that, in this tissue in vitro, androgens are formed from pregnenolone via 17-hydroxypregnenolone and dehydro- epiandrosterone and not from progesterone. With androstenedione-4-14C as the sub- strate, a 12% conversion to testosterone was found (Fig. 2). It is interesting to note that the ratio of androstenedione/testoster- one was approximately five in vessels 1, 2 and 3, suggesting a state of equilibrium be- tween these two androgens.
11ß-Hydroxysteroid synthesis. In all four incubations the extent of 118-hydroxyla- tion was less than that found in comparable experiments with either human fetal adre- nals (15) or hyperplastic adrenals (16). A small amount of radioactivity was associ- ated with 118-hydroxyandrostenedione in
vessels 1, 2 and 4; however, radiochemical homogeneity was not achieved. In vessel 3, with androstenedione-4-14C as substrate, 3.6% of the total radioactivity could be accounted for as 118-hydroxyandrostene- dione-14C. The synthesis of both cortico- sterone and cortisol in vitro was relatively low (Table 3).
16a-Hydroxysteroid synthesis. The tumor tissue showed a very great capacity for hydroxylating steroid molecules in the 16a-position in vitro. 16a-Hydroxycortisol- 3H was isolated from both vessels 1 and 2. Unfortunately, this product was not looked for in vessel 4. Radioactivity was also found in the region of the column chro- matograms from vessels 1 and 2, where both 16a-hydroxydehydroepiandrosterone and 16a-hydroxypregnenolone would appear
August 1967
| Radioactive products | Column or CCD | Crystallization | Derivative | Cryst. or TLC of derivative | Total cpm X10-3 | |||
|---|---|---|---|---|---|---|---|---|
| 16a-Hydroxyprogester- | 615 | CI (A) | 618 | |||||
| one-14C | CII (B) | 630 | acetate | CI | (A) | 600 | 534 | |
| (4) | MLII | 618 | MLI | 636 | ||||
| 17-Hydroxyprogester- | 327 | CI (A) | 296 | |||||
| one-14C | CII(B) | 266 | dioxime | 260 | 447 | |||
| (4) | MLII | 266 | ||||||
| Deoxycorticosterone-14C | 650 | CI (A) | 552 | |||||
| (4) | CII(B) | 555 | acetate | CI (A) | 544 | 394 | ||
| MLII | 540 | MLI | 562 | |||||
| 17-Hydroxypregnenol- one-3H | 1340 | CI (A) | 513 | |||||
| CII(B) | 385 | acetate | CI (A) | 378 | ||||
| (1) | MLII | 400 | CII (B) | 360 | ||||
| CIII(C) | 356 | 348 | ||||||
| MLIII | 362 | |||||||
Numbers represent cpm/ug of steroid. Numbers in parentheses refer to the number of the incubation vessel. In each case, the product isolated from column chromatography or countercurrent distribution (CCD) was crystallized and a derivative was formed from the final crystals. The derivative was subse- quently purified by either crystallization or thin layer chromatography (TLC). The total cpm were cal- culated on the basis of the final specific activity, corrected for the addition of carrier at the time of crystal- lization. (A) =methanol and water, (B) =acetone and water, (C) =cyclohexane and ethyl acetate.
(tubes 550-650). When material from this region of the column chromatogram of vessel 1 (representing 30% of the total radioactivity) was acetylated and re- chromatographed on Hyflo Super-Cel, 640,000 cpm (39%) moved with carrier 16a-hydroxydehydroepiandrosterone ace- tate (Fig. 1). The oxime of 16a-hydroxy- dehydroepiandrosterone acetate was formed and subjected to thin layer chroma- tography. The specific activity of the oxime was identical after chromatography in two systems (90% chloroform, 10% ethanol, and 50% cyclohexane, 50% ethyl acetate). Calculations based on this specific activity revealed that 12.4% of the total radioactivity of vessel 1 could be accounted for as 16x-hydroxydehydroepiandrosterone. The rest of the radioactivity which was originally associated with 16x-hydroxy- dehydroepiandrosterone may represent, in part at least, 16x-hydroxypregnenolone-3H.
In vessel 4 (substrate: progesterone-4- 14C), 11% of the total radioactivity could be accounted for as 16x-hydroxyproges- terone-14C. There was little or no radio- activity in the region where 16a-hydroxy-
cortisol-14C would appear. Though conju- gated steroids were not looked for specifi- cally, the total radioactivity in the strip- ping of the column (more polar than 16x- hydroxycortisol) was measured and com- pared in the four incubations. There was far more “very polar radioactivity” in vessel 1 than in the others. However, in all vessels this material represented a rela- tively small amount of the total radioac- tivity (Table 3).
Incubations of tumor explants. A sum- mary of the purification procedures for the radioactive products from the incubations of the tumor explants with progesterone-4- 14C is shown in Table 7. Previous exposure of the tumor explants to progesterone (10-5M) for 24 hours resulted in increased conversion of progesterone-4-14C to radio- active products when compared to explants cultured for 24 hours with no progesterone added to the medium. Increased hydroxyla- tion at positions 16a, 21 and 17 were noted in the explants previously exposed to progesterone (Table 7). Explants not ex- posed to progesterone, and subsequently incubated with progesterone-4-14C, were
| Steroid | Purification | Specific activity | Total cpm ×10-3 | ||
|---|---|---|---|---|---|
| 0 | + | 0 | + | ||
| Prog-4-14C | Column chromatography | 2400 | 2142 | 2496 | 2356 |
| (84) | (68) | ||||
| DOC-14C | Column chromatography | 36 | 134 | ||
| CI (A) | 31 | 130 | |||
| CII (B) | 29 | 128 | |||
| MLII | 24 | 125 | |||
| Acetylation | |||||
| CI-acetate (B) | 26 | 125 | 36 | 132 | |
| MLI | 31 | 129 | (1.2) | (3.8) | |
| 17-OH-prog-14C | Column chromatography | 240 | 380 | ||
| CI (A) | 198 | 386 | |||
| CII_(B) | 205 | 396 | |||
| MLII | 197 | 388 | |||
| Formation of dioxime and TLC of dioxime | 212 | 363 | 274 (9.3) | 419 (12.2) | |
| 16a-OH-prog-14C | Column chromatography | 226 | 368 | ||
| CCD | 223 | 363 | |||
| CI (A) | 215 | 365 | |||
| MLI | 260 | 370 | |||
| Acetylation, TLC of acetate | 220 | 360 | 230 | 520 | |
| (5.3) | (15.1) | ||||
Specific activity is expressed in cpm/ug of steroid. The total cpm were calculated on the basis of the final specific activity. 0 =no progesterone added to the medium during the culture period; + =10-5M progesterone added to the culture medium. (A) =methanol and water, (B) =acetone and water. Prog-4- 14C =progesterone-4-14C, DOC-14C =deoxycorticosterone-14C, 17-OH-prog-14C =17-hydroxyprogesterone- 14C, 16a-OH-prog-14C =16a-hydroxyprogesterone-14C.
Numbers in parentheses refer to the per cent of the total radioactivity for which each steroid accounts.
unable to convert the substrate to labeled 11-deoxycortisol, cortisol, or corticosterone. However, explants previously exposed to progesterone were able to form small amounts of radioactivity coincident with these three carrier steroids. Less than 1% of the total radioactivity in both incuba- tions was more polar than cortisol.
Discussion
This study represents an attempt to assess the biochemical potential of a virilizing adrenal carcinoma in vitro and to correlate these findings with clinical studies of the patient before and after operation.
The tumor tissue in vitro possessed a re- markable ability to synthesize androgens -and in particular testosterone. Both dehydroepiandrosterone (17) and testoster- one (18) have been extracted from virilizing adrenocortical carcinomas; hence, some androgens were probably present in the tissue at the beginning of the incubation. The finding of Kowel et al. (19) that dehy- droepiandrosterone inhibits the rate of oxidation of C21 45 steroids to 44-3 ketoste- roids may explain the relatively small amounts of progesterone and 17-hydroxy- progesterone formed by the tumor from their respective 36-hydroxy precursors.
The ratio of androstenedione to testoster- one was remarkably constant in the three incubations from which these steroids were isolated. This strongly suggests an equi- librium between the two androgens. Such an equilibrium is completely compatible with previous studies of the interconversion of androstenedione and testosterone. The potential for androgen synthesis demon- strated in vitro and the increased concen- tration of testosterone found in the plasma may explain the virilization of the patient.
In contrast to the propensity for andro- gen synthesis, this tissue showed relatively little activity of the 118-hydroxylase en- zymes in vitro. Under similar experimental conditions human fetal adrenals (15) and hyperplastic adrenals (16) synthesized 4- to 10-fold more cortisol-14C from progester- one-4-14C. This difference in 116-hydrox- ylation is particularly striking when the incubations of tumor and hyperplastic adrenal with androstenedione-4-14C are compared. Recent observations of ours (unpublished) showed that about 24% of the androstenedione-4-14C was converted to 118-hydroxyandrostenedione-14C in non- virilizing hyperplastic adrenals and very little radioactivity was associated with carrier testosterone. The converse was true with this virilizing adrenocortical carci- noma. It is possible that 116-hydroxylation was inhibited by dehydroepiandrosterone, androstenedione, or testosterone (20); how- ever, in the incubation with labeled proges- terone in which no radioactive androgens could be found as products, the extent of 118-hydroxylation was still limited.
Though the tissue had relatively poor 118-hydroxylase activity in vitro, the amount of enzyme present was obviously sufficient to maintain a normal concentra- tion of cortisol in the circulating plasma. It is important to remember that in vitro data can give some idea of the enzyme potential of a tissue, but cannot reveal the extent of enzyme activity that would prevail under in vivo conditions.
Lipsett and Wilson (21) described a
patient whose tumor produced mainly 11-deoxycortisol. In our patient, however, direct measurement of plasma 11-deoxycor- tisol gave a value of 0.4 ug/100 ml when the cortisol value was 11.7 ug/100 ml. In the later course of the patient, when she was losing weight and showing the signs of Cushing’s syndrome, her urinary excretion of 17-hydroxycorticosteroids was 12 mg/ day. The concentration of cortisol in the plasma was now 22 ug/100 ml and that of 11-deoxycortisol 2.0 µg/100 ml. This final concentration of 11-deoxycortisol in the plasma was elevated compared to the ini- tial value and to the concentration in the plasma of 50 normal subjects (0.0-0.5 µg/ 100 ml) (22).
Estrogen production by the tumor early in the course of the disease could not be demonstrated unequivocally, but, in view of her normal menstrual periods and the normal concentrations of estrogens in the urine, we assume that the tumor did not produce estrogens in significant amounts. After the tumor had recurred, however, the estrogens in the urine were measured in milligram amounts and there is little doubt that the tumor was responsible for this very high excretion. Interestingly, except per- haps in the vaginal cytology, even these high concentrations of estrogens were in- sufficient to counter the effects of the overproduction of androgen, and the find- ings clinically and at autopsy were those of virilization.
The extent of 16a-hydroxylation of var- ious steroids was large in the tumor tissue in vitro. Similar results were found in hu- man fetal adrenals (15) and in hyperplastic adrenals (16).
In vivo the tumor tissue showed some response to ACTH but was not suppressed by dexamethasone. In vitro, the tumor responded to exposure to 10-5M progester- one by increasing its ability to convert progesterone-4-14C to more polar steroids. These results are similar to the effects pre- viously reported for human fetal adrenals and mouse adrenals (14). The mechanism
of this effect is not known. It may represent simply substrate stabilization of pre-exist- ing enzyme molecules or induction of synthesis of new enzyme molecules. A sim- ilar phenomenon with dehydroepiandros- terone, which is formed in large amounts in vitro and which has been isolated from tumor tissue (17), might explain the high activity of the C19 36-hydroxysteroid de- hydrogenase-isomerase system.
Acknowledgments
The authors are grateful to Dr. Joseph Barlow of the Boston Lying-In Hospital for performing the urinary estrogen determinations. In addition, they would like to thank Miss Barbara Apple and Miss Sally Greenstein for their expert technical assis- tance, and the nursing staffs of the Mallinckrodt Research Unit and the Vincent Memorial Hospital who cared for the patient during her illness.
References
1. Lipsett, M. D., R. Hertz, and G. T. Ross, Amer J Med 35: 374, 1963.
2. Gabrilove, J. L., D. C. Sharma, H. H. Wotiz, and R. I. Dorfman, Medicine (Balt) 44: 37, 1965.
3. Crane, M. G., J. J. Harris, and R. Herber, Ann Intern Med 63: 494, 1965.
4. Eymontt, M. J., G. Gwinup, F. A. Kruger,
D. E. Maynard, and G. J. Hamwi, J Clin En- docr 25: 46, 1965.
5. Williams, R., L. E. Kellie, A. P. Wade, E. D. Williams, and T. M. Chalmers, Quart J Med 30: 269, 1961.
6. Silber, R. H., and C. C. Porter, J Biol Chem 210: 923, 1954.
7. Epstein, E., Clin Chim Acta 7: 735, 1962.
8. Kliman, B., Endocr. Soc. Meeting, New York, June 1965, p. 101 (Abstract).
9. Kliman, B., and C. Briefer, Jr., Endocr. Soc. Meeting, Chicago, June 1964, p. 53 (Abstract).
10. Barlow, J. J., Anal Biochem 6: 435, 1963.
11. - , J Clin Endocr 24: 586, 1964.
12. Villee, D. B., L. L. Engel, J. M. Loring, and C. A. Villee, Endocrinology 69: 354, 1961.
13. Brooks, S. G., R. M. Evans, G. F. H. Green, J. S. Hunt, A. G. Long, B. Mooney, and L. J. Wyman, J Chem Soc, 1958, pp. 4614-28.
14. Villee, D. B., Advances Enzym Reg 4: 269, 1966.
15. Villee, D. B., and S. G. Driscoll, Endocrinology 77: 602, 1965.
16. Villee, D. B., J Clin Endocr 24: 442, 1964.
17. Plantin, L. O., E. Diczfalusy, and G. Birke, Nature (London) 179: 421, 1957.
18. Anliker, R., O. Rohr, and M. Marti, Helv Chim Acta 39: 1100, 1956.
19. Kowal, J., E. Forchielli, and R. I. Dorfman, Steroids 4: 77, 1964.
20. Sharma, D. C., E. Forchielli, and R. I. Dorf- man, J Biol Chem 238: 572, 1963.
21. Lipsett, M. B., and H. Wilson, J Clin Endocr 22: 906, 1962.
22. Kliman, B., Excerpta Med, Second Intern. Con- gress on Hormonal Steroids, Milan, 1966, p. 132 (Abstract).