Androgen Release and Synthesis in Vitro by Human Adult Adrenal Glands

ANDROGEN RELEASE AND SYNTHESIS IN VITRO BY HUMAN ADULT ADRENAL GLANDS

George L. Cohn, Patrick J. Mulrow

J Clin Invest. 1963;42(1):64-78. https://doi.org/10.1172/JCI104697.

Research Article

Find the latest version: https://jci.me/104697/pdf Journal of Clinical Investigation Vol. 42, No. 1, 1963 ANDROGEN RELEASE AND SYNTHESIS IN VITRO BY HUMAN ADULT ADRENAL GLANDS * By GEORGE L. COHN AND PATRICK J. MULROW (Fromn the Department of Internal Medicine, Yale University School of Medicine, New Haven, Conn., and the Medical Service, Veterans Administration Hospital, West Haven, Conn.) (Submitted for publication June 14, 1962; accepted September 20, 1962) It is well established that surviving human synthesis in adult human tissues are also not well adrenal slices 1) release corticosteroids into the defined. Alternative pathways, two of which by- media; 2) increase their output in response to pass progesterone have been postulated (20, 21) ACTH, and 3) convert radioactive precursors to and are shown in Figure 1. Reactions 2 and 4 re- biologically active corticosteroids (1-11). In con- quiring a 3,j-ol dehydrogenase and double-bond trast, in vitro human adult androgen (C-19) pro- shift and reactions 1 and 3 requiring a 17a-hy- duction has not been well studied. Lombardo and droxylase are well recognized intermediate steps Hudson (7) and Cooper and associates (3) iso- for mammalian and human corticosteroid biosyn- lated 11,/3-hydroxyandrostenedionel from the thesis. In 1957, Bloch, Dorfman, and Pincus re- media after incubation of adrenal tissue from pa- ported the conversion of acetate-C14 to C14-labeled tients with metastatic breast carcinoma, prostatic dehydroepiandrosterone, androstenedione, and carcinoma, and hypertension. Other C-19 steroids 11,8-hydroxyandrostenedione by adrenal slices such as androstenedione or dehydroepiandros- from a 28-year-old female with androgenital syn- terone were not identified by these workers," al- drome (22). Recently, Goldstein, Gut, and Dorf- though these steroids and 11,8-hydroxyandro- man (23) and Lipsett and H6kfelt (24) dem- stenedione have been demonstrated in human onstrated dehydroepiandrosterone synthesis with adrenal vein blood (14-17). The adrenal origin adult human adrenal homogenates and slices re- of the aforementioned androgens is well docu- spectively via reactions 1 and 5 (pregnenolone mented by 17-ketosteroid secretory rate studies 17a-hydroxypregnenolone -> dehydroepiandros- measured by isotope dilution (18, 19). terone). Other alternative pathways of androgen Furthermore, pathways of adrenal androgen bio- biosynthesis were not described in these studies. However, Solomon, Lanman, Lind, and Lieber- * Supported by grants no. 254 (C-8 and C-9), C-3998, man human fetal adrenal homoge- and A-4783 from the U. S. Public Health Service. (25) incubated 1 Compounds referred to are: Pregnenolone: 3,j-hy- nates with progesterone-4-C14 and observed 17- droxy-5-pregnene-20-one; 17a-hydroxypregnenolone: 3,b,- hydroxylation and side-chain cleavage (reactions 17a-dihydroxy-5-pregnene-20-one; progesterone: 4-preg- 3 and 7, Figure 1 ) with the formation of an- nene-3,20-dione; 17a-hydroxyprogesterone: 17a-hydroxy- drostenedione. The biosynthetic capacity of fetal 4-pregnene-3,20-dione; 21-deoxycortisol: 11p,17a-dihy- incubated with sodium-C14 acetate droxy-4-pregnene-3,20-dione; 1l-deoxycortisol, (Com- adrenal slices pound S): 17a,21-dihydroxy-4-pregnene-3,20-dione; corti- was verified further by Bloch and Benirschke (26) sol: 1 1,,17a,21-trihydroxy-4-pregnene-3,20-dione; dehy- who isolated C-21 intermediates pregnenolone and droepiandrosterone (DHEA): 3,8-hydroxy-5-androstene- 17a-hydroxyprogesterone and C-19 steroids-de- 17-one; androstenedione (4-andro): 4-androstene-3,17- hydroepiandrosterone, androstenedione, and 1 1,8- dione; 1 lp-hydroxyandrostenedione (11,6-OH-4-andro): hydroxyandrostenedione. ll-hydroxy-4-androstene-3,17-dione; testosterone: 17#- hydroxy-4-androstene-3-one; andrenosterone: 4-andro- Since androgen production and biosynthetic stene-3,11,17-trione; 2,4-DNPH; 2,4-dinitrophenylhydra- pathways were not completely resolved in adult zone derivative; ACTH: adrenocorticotropin. adrenal tissue, an investigation was undertaken to 2 Plantin, Diczfalusy, and Birke identified dehydroepi- study these aspects of C-19 metabolism with "nor- androsterone after extraction (without incubation) of a virilizing adrenal tumor removed from a 62-year-old mal," atrophic, "hypertensive," hyperplastic, ade- woman (12). In addition, Anliker, Rohr, and Marti iso- nomatous adrenal gland slices. Adrenal carci- lated testosterone from a virilizing adrenal tumor (13). nomatous tissues from two children were also 64 HUMAN ADULT ADRENAL ANDROGENS 65

Pregnenolone 17a-hydroxypregnenolone {2 j5 Progesterone 4 Dehydroepiandrosterone 13 16

1 7a-hydroxyprogesterone Androstenedione 8 9 i11 10 2 1-deoxycortisol 1 0B-hydroxyandrostenedione >t 1-deoxycortisol 13

Cortisol

FIG. 1. POSSIBLE PATHWAYS OF ADRENAL ANDROGEN BIOSYNTHESIS. investigated. This paper is a report of 1) isola- MATERIALS AND METHODS measurement, and identification of adrenal tion, Subjects. The clinical subjects included eight patients androgens released into the incubation media with with Cushing's syndrome, two with adrenal virilism, one slices alone and with the addition of radioactive patient with malignant hypertension (E.W.), one with a steroid precursors and 2) the in vitro effect of suspected pheochromocytoma (J.K.), and one with' car- ACTH on androgen release and substrate con- cinoma of the prostate (J.C.) (Table I). Normal adreno- version. Studies of corticosteroid production and cortical function was observed in Patients E.W. and J.K. but not studied in Patient J.C. biosynthesis by some of these tissue slices were The patients with Cushing's syndrome demonstrated reported previously ( 11, 27). the typical clinical manifestations associated with the

TABLE I Clinical and pathological findings of patients with normal and hyperadrenal function

Laboratory Symptoms and signs 17- 17- Keto- Keto- Masculiniza- Menstrual ir- ster- genic Patient Sex Age Clinical diagnosis Adrenal pathology tion Hirsutism regularities oids steroids mg/24 hrs Normal adrenal function E.W.* F 47 Malignant hypertension Bilateral normal Absent Absent Menopausal 14.0 8.6 J.K.* M 41 Pheochromocytoma Right normal Normal Normal 8.4 7.4 J.C. M 63 Carcinoma of prostate Bilateral normal Normal Normal Adrenal hyperfunction L.H.* F 30 Cushing's syndrome Bilateral hyperplasia Absent Present Amenorrhea 31.5 48.5 Right adenoma L.J.* F 45 Cushing's syndrome Left adenoma Absent Present Oligomenorrhea 19.1 20.6 Left atrophy M.C. F 28 Cushing's syndrome Left adenoma Present Present Amenorrhea 22.6 9.2 Right atrophy G.H. F 35 Cushing's syndrome Bilateral hyperplasia Absent Present Amenorrhea 16.7 33.1 A.D. F 39 Cushing's syndrome Bilateral hyperplasia Absent Absent Amenorrhea 19.8 28.0 B.H. M 24 Cushing's syndrome Bilateral normal Normal male Normal 10.1 48.0 T.N.* M 12 Cushing's syndrome Bilateral hyperplasia Normal male Present 16.7 30.5 P.P. M 12 Cushing's syndrome Bilateral hyperplasia Normal male Absent 9.4 19.7 C.S.t F 3 Adrenal virilism Right carcinoma Present Present 6.8 2.9 Left atrophy E.S.t F 1I Adrenal virilism Right carcinoma Present Present 2.8 3.0

* The clinical features of these patients were summarized in a previous communication (11). t Genetic studies indicated the patients were female. No clinical stigmata of Cushing's syndrome were evident. 66 GEORGE L. COHN AND PATRICK J. MULROW disease. Laboratory studies revealed elevated 8:30 a.m. crograin quantities of carrier steroid were also added plasma cortisol levels (except L.J., whose basal level was prior to incubation. The solvents were evaporated under within the normal range) and exaggerated cortisol re- nitrogen or air (at 370 C) and propylene glycol was sponses to 5-hour infusions of ACTH (28). The 24- added to facilitate solubilization of the substrates. The hour urinary 17-ketogenic steroids (29) usually mirrored C00-labeled progesterone wdvas purified by sequential pa- the findings in the plasma. Elevated total 17-ketosteroids per chromatography in the systems of Bush C (33), pe- were observed in six (L.H., M.C., L.J., A.D., G.H., and troleum ether, 75 per cent methanol, and the cyclohexane: T.N.) of the eight patients. Gradient elution column dioxane: methanol: water (100: 100: 50: 25) system of chromatography of the urinary 17-ketosteroids (30) of Kliman and Peterson (34). Tritium-labeled progesterone patients M.C. and L.H. revealed elevation of the 11-oxy- was similarly processed after the addition of unlabeled genated 17-ketosteroids and etiocholanolone of the 11- progesterone.5 The 17a-hydroxypregnenolone-7a H3 was deoxy-17-ketosteroid fraction similar to the observations subjected to three separate solvent partitionings in n-hex- of Jailer, Vande Wiele, Christy, and Lieberman (31). ane: ethanol: water (100: 2: 98 vol/vol) at 2° C, n-hex- The clinical impression of adrenal virilism (patients ane: ethanol: water (100: 20: 80: vol/vol) at 250 C, and E.S. and C.S.) was corroborated by normal endogenous n-hexane: carbon tetrachloride: ethanol: water (30: 70: cortisol secretion (32) and studies of corticosteroid me- 47.5: 52.5 vol/vol) at 250 C as described by Carstensen, tabolism (27, 29). Urinary 17-ketosteroids were ele- Oertel, and Eik-Nes (35). The compound was stored in vated, predominantly of the 11-deoxy-17-ketosteroid frac- a frozen state in benzene: methanol (9: 1) until used. tions, dehydroepiandrosterone and etiocholanolone. Adre- Radiochemical homogeneity was observed after the addi- nal cortical suppression tests were not performed.3 tion of stable 17a-hydroxypregnenolone and paper chro- The patients (except L.J., whose small right adrenal matography in cyclohexane: benzene: methanol: water gland was not removed) underwent total bilateral adrenal- (100: 50: 100: 25) and toluene: propylene glycol systems ectomies at a one-stage operation (28); they received followed by reverse phase paper chromatography in the cortisone acetate or cortisol hemisuccinate the evening benzene: petroleum ether: methanol: water (20: 10: 40: prior to and the morning of surgery. Blood and cortisol 10) system. A second minor peak of radioactivity esti- hemisuccinate were administered during the operative mated to contain less than 5 per cent of the radioactivity procedure. The pathological diagnoses of these glands of the 17a-hydroxypregnenolone peak was noted in one are listed in Table I. system. The preparation and purification of dehydroepi- Incubation technique. The adrenal glands removed at androsterone-7a-H3 from the acetate derivative were de- operation were immediately iced and kept cold through- scribed previously (36). Dehydroepiandrosterone-7a-H out pathological examinations and experimental proce- was chromatographed in the heptane: 96 per cent metha- dures. The glands were weighed, sectioned, freed of fat, nol system (37) within 72 hours of use. reweighed, and sliced with a Stadie-Riggs microtome. Steroid separation and measurement. After incubation, The medullary tissues were not separated. Slices from the media were carefully decanted, and the tissues and multiple areas of each adrenal were taken for each ex- flasks were washed three times with cold isotonic saline. perimental flask. Incubation of weighed slices was per- The appropriate media and washings were pooled. In formed in a Dubnoff metabolic incubator for 3 hours at order to correct for chemical losses incurred during 37° C in an atmosphere of 95 per cent oxygen, 5 per cent manipulations, tracers of tritium-labeled dehydroepiandros- carbon dioxide in Krebs-Ringer bicarbonate buffer with terone, SA 1.25 mc per smole; androstenedione-4-C14, SA 200 mg per 100 ml glucose. The incubations were started 2.13 juc per Amole; and 1l-hydroxyandrostenedione-4-CI4,6 within 3 hours after excision of the adrenals. Ten units SA 1.4 /Ac per jumole, were added to some media prior to of ACTH 4 per gram of tissue was added to selected extraction (with the exception of the radioactive con- flasks. version studies). Constant specific activities of the free Addition of radioactive substrates. Prior to the addi- C-19 tracers were observed after reverse isotope dilution tion of buffer and slices, certain flasks received tracer with paper chromatography. quantities of either progesterone-16-H3 (specific activity, The pooled washings and media were extracted twice SA, 0.506 mc per /Amole), dehydroepiandrosterone-7a-HW with 3 volumes of dichloromethane which had been puri- (SA 1.25 mc per Amole), or 1 7a-hydroxypregnenolone-7a- fied by passage through a silica-gel column. The di- H3 (SA 0.863 mc per ,.mole) dissolved in absolute etha- chloromethane extract was washed once with 1/10 vol- nol: methanol (1: 1) or benzene: methanol (9: 1). In ume of 0.05 N sodium hydroxide and twice with 1/10 one study, approximately equimolar amounts of 17a- volume of distilled water. The water was removed with hydroxypregnenolone-H3 and progesterone-4-C1' (SA 16.2 anhydrous sodium sulfate and the extract evaporated in per were added. In several experiments mi- jxc /mole) 5 The tritium-labeled steroids were purchased from the 3 The exception was C.S., a 3-year-old girl with a New England Nuclear Corp., Boston, Mass. Proges- clinical diagnosis of "adrenogenital syndrome." Twenty terone-4-C1" was generously donated by the Endocrine mg of prednisone was administered per day for 30 days Study Section, National Institute of Health, Bethesda, without effecting a change of the total urinary 17-keto- Md. steroid excretion. 6 Dr. H. Leon Bradlow, Sloan-Kettering Institute, New 4Armour beef adrenocorticotropin. York, N. Y. graciously made this isotope available. HUMAN' ADULT ADRENAL ANDROGENS 67 vOacto at 370 C. The residues were applied to washed Model C110 B, Nuclear-Chicago Corp., Des Plaines, M\hatmian no. 1 filter paper and developed for 72 hours Ill.). The efficiency for tritium wlvas 18 to 20 per cent. in a modification of the toluene: propylene glycol system The final value was determined, taking into account the (38).7 The effluents were collected and chromatographed absorbance of the blank, chemical amount of the added for 20 hours in the cyclohexane: dioxane: methanol: wa- carbon-14 tracer (0.20 /Ag or less), dilution factor, and ter system (100: 100: 50: 25) which separated corticos- the isotope recovery (40). Since the chemical amount teroids (corticosterone and 11-deoxycortisol) (34) from of the added dehydroepiandrosterone tracer was less than androgens. Dehydroepiandrosterone and androstenedione 0.01 lug, the steroid was calculated as the free compound ran off the paper and were collected. On occasion the without chemical correction. The calculation of the an- more polar C-19 steroids were located at the bottom of drogens isolated from media of the radioactive conversion the paper. The effluents and paper eluates 8 were dried experiments was uncorrected for isotope recovery, al- in vacuo and transferred quantitatively with chloroform though other results were so corrected. to washed WN hatman no. 1 filter paper (28 x 1 inch strips Identtification. The isolated androgens from the in- with a common origin) in the heptane: 96 per cent cubations without added radioactive substrates were pooled methanol (1: 1) system of Bush and Willoughby. Fifteen separately. The identification of A'-3-keto C-19 steroids to twenty Ag each of reference compounds, dehydroepi- was substantiated by characteristic mobilities and con- androsterone, androstenedione, 118-hydroxyandrostenedi- stant specific activities in two additional paper chromato- one, and adrenosterone were applied together in a single graphic systems, heptane: propylene glycol (41) and spot on each side of the "extract area." The paper was ligroin: benzene: methanol: water (667: 333: 800: 200) equilibrated for 4 hours and the mobile phase allowed to (42). Areas of ultraviolet absorption and Zimmermann develop for 14 to 16 hours,9 after which the paper was positive staining coincided with the radioactive peaks lo- air dried. The C-19 steroids with a A-3-ketone configu- cated by a gas-flow paper chromatogram scanner (Model ration were located on the chromatograms by the ultra- C-100 B, Actigraph II, Nuclear-Chicago Corp., Des violet scanning technique and the A5-3,8-ol steroid (dehy- Plaines, Ill.). Androstenedione-C" and 11,8-hydroxyan- droepiandrosterone standard) by Zimmermann reagent drostenedione-C4 bis-2,4-dinitrophenylhydrazoines were stain (alkaline inz-dinitrobenzene). Appropriate areas formed (prior acetylation was unsuccessful) and con- corresponding to the standards and paper blanks were firmed by characteristic absorption maxima of 378 to 380 cut out and eluted with 4.0 ml of absolute ethanol as m/A (22) in the Beckman DU spectrophotometer. The previously described (36). specific activities of the hydrazones were determined by Measurements of the isolated androgens with a 5-3-ke- absorbance at 378 mu and calculated for the free com- tone configuration were performed on portions of the elu- pound. Radioactivity was measured by counting at in- ates by the alkaline fluorescent method of Abelson and finite thinness on aluminum planchets to a standard er- Bondy (39). Authentic androstenedione was used in con- ror of + 2 per cent in the Nuclear-Chicago, Model C-110 struction of the standard curve. The C-19 steroids were B micro-window gas-flow counter. The efficiency for determined also by absorbance (0.8 ml microcuvets with a C1' was 24 to 28 per cent by this technique. 1.0 cm light path) at 240 m~u (ethanol) in a Beckman Dehydroepiandrosterone was verified as the mono-2, model DU spectrophotometer. The extinction coefficients 4-dinitrophenylhydrazone derivative by the following cri- at 240 mA were used for the chemical determination. teria: 1) absorption maximum of 368 mu (chloroform); Dehydroepiandrosterone was converted to the 2,4-dini- 2) characteristic position of only one radioactive peak on trophenylhydrazone derivative and measured spectropho- elution from micro-alumina columns (36); and 3) mno- tometrically at 368 mA (36). bility of the mono-2,4-dinitrophenylhydrazone (1 cm per A 0.5 ml portion of the ethanol eluate was counted hour) developed for 12 hours in the heptane: propylene to a standard error of + 2 per cent in a liquid phosphor glycol system (22). The appropriate yellow radioactive scintillation counter (Model LP-2, Technical Measure- area and paper blanks were cut into thirds, eluted with ments Corp., BNew Haven, Conn.), which had a counting chloroform, and specific activities determined by spectro- efficiency of 45 to 48 per cent for carbon-14. The tritium- photometry and windowless planchet counting with the labeled dehydroepiandrosterone 2,4-dinitrophenylhydrazone Nuclear-Chicago Model C-110 B gas-flow counter. The was plated at infinite thinness on aluminum planchets specific activities were calculated on the basis of free and counted to a ± 2 per cent standard error in a window- dehydroepiandrosterone. less gas-flow counter (Automatic Sample Changer, Radioactive androgens from the precursor experiments 7 The stationary phase was propylene glycol diluted with were identified after paper chromatographic separation methanol (1: 1) before application to the paper. The and spectrophotometric and isotope measurements. One methanol was then allowed to evaporate. hundred micrograms of carrier androstenedione, ll3-hy- 8 Radioactive C-19 tracers were added at this stage of droxyandrostenedione, and dehydroepiandrosterone were the steroid isolation if omitted previously, in order to cor- added to pooled paper eluates from experiments with a rect for losses in purification. common radioactive substrate. The separate eluates were 9 Paper chromatography of the 17-ketosteroids was car- dried in vacuto at 370 C and the dried residues of each ried out in a room maintained at a constant temperature A'-3-ketone androgen product subj ected to a 40-tube of 200 C and relative humidity of less than 50 per cent. countercurrent distribution (E-C Apparatus Company, 68 GEORGE L. COHN AND PATRICK J. MULROW

TABLE II Androgen production by "normal," atrophic, and "hypertensive" adrenal glands

Steroids* Wt of - - incubated Andro- 1 1t-OH-4 Patient Pathology tissue DHEA stenedione Andro g jsg/g/3 hrst J.K. "Normal" 1.661 2.6 2.9 1.5 J.C.- "Normal" 2.524 0.4 1.8 1.5 L.J. Atrophic 0.984 0 1.1 2.0 Atrophic + ACTH t 0.474 4.4 3.8 17.7 M.C. Atrophic 0.700 14.3 0 0 Atrophic + ACTH t 0.158 5.3 2.7 34.2 E.W. Left hypertension 0.573 0 3.5 5.4 Right hypertension 0.570 0 3.7 4.9 * DHEA, dehydroepiandrosterone; 11,B-OH-4 andro, 11,3-hydroxyandrostenedione. t The total mean 4±SD per cent recoveries of C14- and tritium-labeled tracers added to media before extraction were 49 ± 8.2. The total per cent isotopic recoveries for tracers added to effluents and eluates from cyclohexane :dioxane: methanol:water system were 82 ± 10.1. t The amount of ACTH added was 10 units per gram of tissue.

Swarthmore, Pa.) at room temperature. Theoretical in ethanol, and the peak tube and the two adjoining tubes countercurrent distribution curves were calculated ac- on both sides of the peak pooled for further identification. cording to the procedure of Way and Bennett (43). The Mobilities and specific activities of the free steroids and solvent system employed for androstenedione (K = 0.397) 2,4-dinitrophenylhydrazone derivatives were calculated and ll -hydroxyandrostenedione (K = 0.178) was ethyl after paper (Bush and Zaffaroni systems) and micro- acetate: hexane: methanol: water (120: 180: 180: 120). A alumina column chromatographic separation, respec- 20-tube transfer was carried out for dehydroepiandros- tively.10 In addition, a portion of the pooled transfers terone in the ethyl acetate: cyclohexane: ethanol: water from the countercurrent distribution of 118-hydroxyan- (30: 70: 55: 45) system, K = 1.02, of Savard and col- drostenedione from one series of incubations with 17a- laborators (44). At the end of the distribution, the total hydroxypregnenolone-H3 was oxidized with chromic acid contents of each tube (20 ml) were transferred to test tubes and dried in air at 370 C. After specific activities 10 Only dehydroepiandrosterone 2,4-dinitropheniylhydra- were determined as described, the residues were dissolved zone was chromatographed on alumina columns.

TABLE III Androgen production by hyperplastic, "normal," adenomatous, and carcinomatous tissue from patients with adrenal hyperfunction

Steroids* Steroids (with ACTHt) Wt of - Wt of incubated-- incubated 1 l1-OH-4 tissue 1 1B-OH-4 Patient Pathology tissue DHEA 4-Andro Andro DHEA 4-Andro Andro g yg/g/3 hrs g lig/g/3 hrs Cushing's syndrome L.H. R. hyperplasia 1.005 0 3.1 11.3 1.010 0 4.1 25.0 L. hyperplasia 1.025 0 3.6 8.7 .500 0 12.6 47.4 R. adenoma 1.845 2.3 0.9 3.4 1.200 2.5 11.7 15.0 T.N. R. hyperplasia 2.056 0 0 4.3 2.008 0 1.0 6.3 L. hyperplasia 2.482 0 0 3.8 2.532 0 0 5.2 P.P. R. hyperplasia 0.984 1.3 0 4.3 1.326 0.7 0 2.6 L. hyperplasia 1.027 0 1.2 1.2 .986 0.8 1.6 2.1 A.D. L. hyperplasia 1.019 0 1.3 0.9 1.008 0 0.7 1.3 B.H. R. "normal" 1.033 1.010 0 1.2 4.6 L. "normal" 0.977 0 4.5 0 .992 1.6 0.6 4.2 L.J. Adenoma 1.554 0 1.3 3.0 1.544 1.5 0 6.6 M.C. Adenoma 2.380 0 3.3 7.4 2.406 0 1.8 6.3 Adrenal virilism C.S. Carcinoma 1.504 2.7 0 0 1.508 4.5 2.7 0 E.S. Carcinoma 1.007 0 0 8.0 .940 3.5 8.9 12.9

* DHEA. dehydroepiandrosterone; 4-andro, androstenedione; 11 -OH-4 andro, 1 1,l-hydroxyandrostenedione. t 10 units of ACTH per gram of tissue. HUMAN ADULT ADRENAL ANDROGENS 69 according to the method of Lieberman and collaborators and androstenedione at a rate comparable with (45). The identity of the radioactive product, adrenos- that of adrenal tissue from some patients with terone, was verified by its characteristic mobility on paper Cushing's syndrome. in the heptane: propylene glycol and 96 per cent methanol: heptane systems. The identical ultraviolet light ab- The androgen production by adrenal slices from sorbing and radioactive zone from each chromatogram patients with Cushing's syndrome and adrenal was eluted and specific activities determined by planchet virilism is listed in Table III. Small amounts of counting and absorbance at 240 as described. mgu endogenous dehydroepiandrosterone were detected in incubations of adenomatous (L.H.), hyper- RESULTS plastic (P.P.), and carcinomatous tissues (C.S.). A. Adrenal androgen production The addition of ACTH to the system did not stim- Three adrenal androgens, dehydroepiandroste- ulate dehydroepiandrosterone production consist- rone, 1 1,8-hydroxyandrostenedione, and androstene- ently. However, ACTH stimulated the carci- dione were isolated from the media after 3-hour in- nomatous tissues (C.S. and E.S.) to produce cubation. Adrenosterone was not detected. In amounts of dehydroepiandrosterone in the range Table II is listed the steroid production (ug per of the atrophic tissues of L.J. and AI.C. (Table gram of slice) by tissue slices from patients with II). normal and with atrophic adrenals. The results of On the other hand, incubation of "Cushing Patients J.K. and J.C. represent the pattern of pro- adrenal tissue" (Table III) in the presence of duction by "normal" tissues. The functional ca- ACTH resulted in a one and a half to sixfold in- pacity of the atrophic glands (L.J., ipsilateral, and crement of 11,8-hydroxyandrostenedione produc- M.C., contralateral) is indicated by the endoge- tion. Nevertheless, the most exaggerated 11fl- nous and ACTH-stimulated androgen release, es- hydroxyandrostenedione response to ACTH is pecially the striking increase in 11,8-hydroxyan- found in the atrophic tissues (Table II), associ- drostenedione. The "hypertensive" glands of Pa- ated clinically with adenomata. tient E.W. produced 11,/3-hydroxyandrostenedione The effect of ACTH on androstenedione release

TABLE IV Total androgen production per gland

Steroidst Steroids (with ACTH) Total weight of 1l0-OH- 11-0H- Patient Pathology tissue* DHEA 4-Andro 4-Andro Total DHEA 4-Andro 4-Andro Total g pg/gland/3 hrs ug/gland/3 hrs J.K. Normal 8.3 21.6 24.1 12.5 58.2 J.C. Normal 4.7 1.9 8.5 7.1 17.5 E.W. L. hypertension 16.0 0 56.0 86.4 142.4 R. hypertension 12.5 0 46.3 61.3 107.4 L.H. L. hyperplasia 13.5 0 48.6 117.5 166.1 0 170.1 639.9 810 R. hyperplasia 12.5 0 38.8 141.3 180.1 0 51.3 312.5 363.8 R. adenoma 32.0 73.6 28.8 108.8 211.2 80 374.4 480.0 934.4 L.J. L. adenoma 22.02 0 28.6 66 94.6 33 0 145.2 178.2 R. atrophic 3.0 0 3.3 6 9.3 13.2 11.8 53.1 77.7 M.C. L. adenoma 51.9 0 171.3 384.1 555.4 0 93.4 327 420.4 R. atrophic 0.7 10.0 0 0 10.0 3.7 1.9 23.9 29.5 A.D. L. hyperplasia 18.0 0 23.4 16.2 39.6 0 12.6 23.4 36 B.H. L. normal 4.0 0 18.0 0 18.0 6.4 2.4 16.8 25.6 R. normal 5.2 0 6.2 23.9 30.1 T.N. L. hyperplasia 11.5 0 0 43.7 43.7 0 0 59.8 59.8 R. hyperplasia 14.5 0 0 62.4 62.4 0 14.5 91.4 105.9 P.P. L. hyperplasia 7.7 0 9.2 9.2 18.4 6.2 12.3 16.2 34.7 R. hyperplasia 6.7 8.7 0 28.8 37.5 4.7 0 17.4 22.1 C.S. R. carcinoma 9.0 24.3 0 0 24.3 40.5 24.3 0 64.8 E.S. R. carcinoma 7.2 0 0 57.6 57.6 25.2 64.1 92.9 182.2

* These values represent weights of the trimmed pathological specimen. t Abbreviations as described in Tables II and III. 70 GEORGE L. COHN AND PATRICK J. MULROW

TABLE V Specific activities of pooled androgens obtained from human adrenal slice incubations

Specific activity Dehydroepiandros- Steroid * Method Solvent system teronet cpm /pmole Mono-2,4-DNPH derivative4 20,81811 Column chromatography§ 3.9% Hydration Benzene:chloroform (4:1) 2e,523 5.0% Hydration Benzene:chloroform (4:1) 20,469 Paper chromatography Upper third 20,364 Heptane: propylene glycol Middle third 20,148 Lowver third 20,287 1 1,8-OH-4- Free compounds androstenedione Androstenedione 8,0481! 10,70811 Paper chromatography 96 % Methanol: benzene (1: 1) 7,748 10,255 Heptane:propylene glycol 7,508 10,340 Ligroin: benzene: methanol: water (667:333:800:200) 7,547 10,028 Bis-2,4-DNPH derivative¶ 7,540 10,001

* The chemical amount of the steroid 2,4-DNPH was measured at 368 mju (mono) and 378 m~i (bis) in a spectrophotometer. Radioactive measurements were performed with a windowed (C14) and windowless (Hs) gas-flow counter. The efficiency for C14 was 24 to 28 per cent. t The specific activity is calculated on the basis of the free compound. t 2,4-DNPH, 2,4-dinitrophenylhydrazone. § The steroid 2,4-DNPH was chromatographed on 1,000 mg of neutral aluminum oxide (Woelm) 12.5 XO.3 cm column. ¶ Specific activity of free androgen. 11 Specific activity of pooled androgens prior to identification procedures. is marked in the left hyperplastic and adenomatous glands. The total androgen production of the tissues of L.H. and carcinomatous slices from Pa- "hypertensive" glands (E.W.) is comparable to tient E.S. (Table III). "Cushing tissue." The amounts of androstenedione are decreased The elevated total urinary 17-ketosteroid levels rather than increased in most other incubations of Patients L.H., M.C., L.J., and T.N. (Cushing's with ACTH. The total androgen production per syndrome) correlated well with the increased total gland by all tissues is shown in: Table IV. These adrenal androgen production in vitro. A similar results are based on the assumption that the in- correlation is present in Patients C.S. and E.S. cubated tissue is representative of the entire gland. with adrenal carcinoma and virilism. The func- The total production of 1 13-hydroxyandrostene- tional in vitro capacity of the contralateral atrophic dione in the absence of ACTH by the hyperplastic, glands of C.S. and E.S. is not known. adenomatous, and carcinomatous tissues (except Identification of adrenal androgens without ad- Patients C.S., B.H., and P.P.) is consistently dition of radioactive precursors. After chemical greater than that of the "normal" and atrophic and radioactive measurements were performed,

TABLE VI Incubation of adrenal slices with progesterone-16-H'

Substrate Progesterone- 1 6-H3 Products Amount Weight of Patient Pathology SA added tissue Androstenedione 1 I#-Hydroxyandrostenedione ,c/Mmole ;ic g pg/g/ AcI %o conv.* Ratio zg/g/ ycl/ % conv. Ratio 3 hrs pmole SAt 3 hrs ;&mole SA J.K. "Normal" 50.3 27 2,734 2.5 1.34 0.12 0.027 1.8 3.29 0.20 0.065 L.H. Right hyperplasia 72.6 4.6 510 2.4 0.32 0.03 0.0044 8.5 1.96 0.061 0.027 Adenoma 72.6 9.12 1,200 2.8 2.46 0.32 0.034 2.1 4.34 0.40 0.059 L.J. Adenoma 40.8 5.3 1,010 1.2 0.21 0.016 0.0051 2.5 5.05 0.80 0.124

* Per cent conversion of radioactive substrate to androstenedione and 1 1#-hydroxyandrostenedione measured after paper chromatography in the heptane:96 per cent methanol system. t Ratio SA is the ratio of the specific activity of the product to the specific activity of the substrate. HUMAN ADULT ADRENAL ANDROGENS 71

0 co W._ - O O NS 0 O o C r.0 0 =0C C 0 '0 U 0 e r. N 00 e 00 S 0 e 04 0 - 0 la - a, 0 W t r. Cd -I~ cj.1'1o o It x 0 0 _ r:tE.tC N

XL - t40 - Oc tCn-66 CU U . "CZi. 0 Z _ U) 0 _ C 0 0._ U. 00 -- tC -e 0 CC) 0 00 .9 r C 10 C 0 0 r. *0 --0!~ 11 00 Cl) Nq :S 0 ~ 6 "6 6 034 r C) I.. 0O 0 0O r. Y)) cO C4) U) CO 0 0 '4

0 0 0 Ct -o~ 0 O 0 0 Cd 0 - ~CU. CUC/2 6 6 U.1 _U) ;C. U) o. o.)~ o C S .C U) > ¢ U) C oC. 0 _ U) O U) U CC 0

._~~~~~~~~~~~~~~~~~~~~~~00 U) 10 mU) oo 0 e 0 0 U) o E 5 -6. oC U)e~o CC C C-Cx o ¢ %. 0 0 OU)Uo .0 cn) C C °% i In a, Cd CS x C 10 ZL| C._ C. . > Q C0U 0 0 0 wI O 0 x) co _U _ o 0 o0 0 U 0 C 'C U)a OoI ._~U U) 0 o t- _ 0 0 .C 0 - U4 CZ * -I X X a V, :L UCU -' '0 I._ 0~ 0C ..C CC U) U U.U~ C.C cd Ct 0, C)d

3*0 - N 0^ 0 0' 0' e C 0 ° w > ~C0=z3 0 0 010 U ='i Cd CU CU cd 0° CU CZ 'C r- ._ OC.¢ CU o :-I .2 C ad .9 a .0 8d 04co '!C X, >) C8-CU(t )C, U 72 GEORGE L. COHN AND PATRICK J. MULROW

TABLE IX The conversion of dehydroepiandrosterone-H3 to androstenedione-H3 and 11,8-hydroxyandrostenedione-H3 by hyperplastic adrenal slices

Products Dehydroepiandros- 1 1O-Hydroxyandro- terone-H' Androstenedione stenedione Wt. of Amt. % % Patient Pathology tissue SA added SA Cons. SA Cons.

g c/I.umole pc pg/g/ c / pg/g/ PCc/ 3 hr pmole 3 hr jmole G.H. Right hyperplasia 2.997 31.53 6.43 23.2 2.55 9.7 38.5 2.76 16.5 the remainder of the eluates (from the 96 per Since these experiments suggested that 17a-hy- cent methanol: heptane system) were pooled and droxypregnenolone alone appeared to be a better specific activities determined. Radiochemical ho- substrate than progesterone for adult adrenal an- mogeneity of the isolated androgens is demon- drogen biosynthesis, an incubation of hyperplastic strated by the data in Table V. adrenal slices was carried out with a mixture of tritium-labeled 17a-hydroxypregnenolone and CA4- B. Radioactive precursor conversion studies labeled progesterone. Ten units of ACTH per In Table VI are listed the data from the incu- gram of slices was added to alternate flasks. In bation of "normal" and pathological adrenal slices Table VIII the results of this study are listed. with progesterone-16-H3. Radioactive andro- The percentage conversion of tritium-labeled 17a- hydroxypregnenolone to stenedione and 11,8f-hydroxyandrostenedione were androstenedione and 11,8- isolated from the media. The percentage conver- hydroxyandrostenedione greatly exceeded that of sion of progesterone to 11,8-hydroxyandrostenedi- C14-labeled progesterone. The percentage conver- one was higher than the percentage conversion of sion of the 17a-hydroxypregnenolone to dehydro- this precursor into androstenedione. The data epiandrosterone was about the same (2.1) as the values in demonstrate that less than one per cent (uncor- observed previous incubations with 17a- rected for manipulative losses) of progesterone- hydroxypregnenolone, 1.6 and 1.4 per cent (Table 16-H3 was converted to the C-19 steroids measured. VII). The "normal," hyperplastic, and adenomatous The low percentage conversion of 1 7a-hydroxy- tissues released approximately the same amount pregnenolone to dehydroepiandrosterone com- of androgen in the presence of progesterone compared to androstenedione and 11,8-hydroxyandro- pared to similar incubations without this substrate stenedione (Tables VII and VIII) prompted a (Tables II, III). In contrast, incubation of 17a- study of the 3,8-ol dehydrogenase system (and hydroxypregnenolone-7a-H3 with adenomatous and double-bond shifting from A5 to A4) in fresh hy- hyperplastic slices resulted in conversion of 6 and perplastic adrenal slices. The A5-3,8-ol substrate 7 per cent of the substrate to androstenedione and dehydroepiandrosterone was readily converted to 11,8-hydroxyandrostenedione, respectively (Table the A4-3-ketone androgens androstenedione and VII). In addition, 1.6 and 1.4 per cent of this sub- 11,/3-hydroxyandrostenedione as shown in Table strate was converted to tritium-labeled dehydro- IX. epiandrosterone. There was a slight decrease in Identification of adrenal androgens from incuba- the specific activity (SA) between dehydroepian- tions with radioactive substrates. After chemical drosterone, androstenedione, and 11,1/-hydroxyan- determinations were performed, the C-19 steroids drostenedione with the adenomatous slices of M.C. derived from a common precursor were pooled On the other hand, the SA of the androgens syn- separately and carrier androstenedione, 11,3-hy- thesized by the hyperplastic slices of P.P. are of droxyandrostenedione, and dehydroepiandroste- the same order of magnitude, 14.3, 12.5, and 13.0 rone were added. (Carrier dehydroepiandrosterone ,uc per Mmole (Table VII). was not added to paper eluates from the incuba- HUMAN ADULT ADRENAL ANDROGENS 73

0C -- r. 600-) O C c .1, 00 ca4) e 4 C t 00 0 C O t00 - I 0 CN ON00 (=O:N 6 & t I -~~ C- 00t~~~0r)c- 6- 6 660Oo 6 ooNo oo 00S90Nooc- 0d -e 6 * -I * -- 006e 00 0o - 00 \o0 4 4t 44 4 CZ5 0 \0000 ,004 -v ooi X le WE't :zI- --t1-4l 06 t- \ !~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dN _ c-c .0 6 r. 0 0 Co U)U, --I so t- * Z 4 0 'I" r,f (' M U)e 00oo c- c)- -0 6\0 0t. 6 s .0 a. C) c.40 C Om 0 - *-a.--- - o *oc o- 00 C ON00O U)~~~~~~~44~~~~ & 4 U)4 UU U) 0 004 -0 1 U) s C-c) 0 CN 00 c-a v:M 0 0 4lt C~ -a \6 ,- -. C, :L d C o-co 10 C- c-c .-r 6 C

c-c a)

a) 33- 0)

0 _) - 0)_ 0 CZ u bwo> t-jCb 0 0

0 N O:^ :o N~CO:

.C 0 *O e *~ .X C:Q 0+z*_. * a-,,

; W;QU

m -0 lb 0 0L) 0 0 0

0) W 4-j0 cr. Ir b1 C-L 0) 0L) Q.) :0 0 0l. )- 0 >- =L n1 ._-L0 x U Q) 0 0 0) G) O 0) 3,- C.) 0L4z ra OL - PL- 74 GEORGE L. COHN AND PATRICK J. MULROW tion experiments with progesterone-H3) .1 Iden- per 24 hours) determined by Vande Wiele and tification was verified by identical observed and Lieberman (18).12 theoretical distributions of the free androgens in On the other hand, the adrenal tissue from the countercurrent systems and characteristic mo- three patients with Cushing's syndrome, L.H., bilities in two different paper chromatographic L.J., and M.C. (Table IV) released 11/,-hydroxy- systems. The reported column (dehydroepian- androstenedione, only in the presence of ACTH, drosterone only) and paper chromatographic mo- commensurate with normal secretory rates of 2.0 bilities and spectrophotometric spectra of the ster- mg per 24 hours reported by Bradlow and Gal- oid 2,4-dinitrophenylhydrazone derivatives were lagher (19). The comparison between the two observed, further substantiating the identification of methods is made on the assumption that the 3-hour the isolated androgens. The SA of the pooled an- in vitro release rate projected to 24 hours is con- drogens (and carriers) from individual substrate stant over that interval. The data of Cooper and experiments (Tables VI-IX) along with the ex- associates (3) substantiate the validity of this perimental conditions used, for identification are speculation for C-21 steroids and 11f-hydroxyan- listed in Table X. Constant SA were maintained, drostenedione. indicating radiochemical purity of the measured The stimulatory effect of ACTH on 11f-hy- compounds. droxyandrostenedione production in vitro (Table III) is in agreement with the data of Dyrenfurth, DISCUSSION Lucis, Beck, and Venning (6); Touchstone, Androgen release by adult human adrenal gland Glazer, Cooper, and Roberts (46); and Cooper slices. The use of an in vitro incubation technique and associates (3), despite the differences of with human adrenal slices affords a direct ap- glandular weights and pathological findings of the proach to an understanding of adrenal androgen adrenal tissues. The most striking increase of metabolism. The data from the present incuba- 11,18-hydroxyandrostenedione production occurred tions (Tables II-IV) indicate that the slices 1) with ACTH in the atrophic glands of L.J. and release detectable amounts of free androgens, de- M.C. (Table II), resulting in 8- and 34-fold in- hydroepiandrosterone, androstenedione, and 11,1- creases, respectively. The functional capacity of hydroxyandrostenedione into the media and 2) these two tissues is demonstrated also by the ef- respond to adrenocorticotropin (ACTH) in vitro. fect of ACTH on dehydroepiandrosterone and In the present study the amounts of dehydro- androstenedione production. There is a slight rise epiandrosterone and androstenedione released per of the androstenedione level in both tissues. How- gland per three hours in vitro are less than the ever, the elevated (control) dehydroepiandros- quantities found by Hirschmann, deCourcy, Levy, terone in J.C. fell with ACTH, 14.3 to 5.3 ug per and Miller (17) in the adrenal vein blood of two g per 3 hours, and rose in L.J. from a control value patients with breast carcinoma and one patient of 0 to 4.4 4g per g per 3 hours. The effect of with essential hypertension. In contrast, the ACTH on dehydroepiandrosterone release is adrenal venous blood levels of 11,8-hydroxyandro- ambiguous with other glands from patients with stenedione are of the same order of magnitude Cushing's syndrome. The failure to detect tes- found by the in vitro tissue slice method (Table tosterone in the incubation media implies that this II). There is also a marked discrepancy between steroid is not usually one of the adrenal androgens the low dehydroepiandrosterone levels determined produced. by the in vitro technique (and analysis of adrenal Adrenal androgen biosynthesis. The studies vein blood) (17) compared to the normal in vivo demonstrate the conversion of radioactive steroid secretory rates of this C-19 steroid (15 to 25 mg 12 Recent observations (47) by the authors indicate that 11 In the 17a-hydroxypregnenolone-H' studies, 100 ,g human adrenal tissues form dehydroepiandrosterone sul- of carrier testosterone was added to a portion of the fate from 17a-hydroxypregnenolone and dehydroepian- overflow from the cyclohexane: dioxane: methanol: water drosterone. It is certain that our extraction procedures system. After acetylation, the extract was chromato- would not remove steroid sulfates from the media, which graphed in the heptane: propylene glycol system. The may possibly explain the low observed unconjugated de- isolated testosterone acetate was devoid of radioactivity. hydroepiandrosterone values. HUMAN ADULT ADRENAL ANDROGENS 75

CH3 GH3 C=O CZO ---OH 0

HO HO HO4 a -PREGNENOLONE 170< OH-PREGNENOLONE DEHYDROEPIANDROSTERONE I I CH3C (DHEA)

i PROGESTERON E/ I CH3 I 1 _ 00

11/3-OH A KInnnIQrC7KlC ninki c 170< OH-PROGESTERONE ", vnvb rl N t Ual FIG. 2. PATHWAYS OF ADULT ADRENAL ANDROGEN BIOSYNTHESIS. The solid line indicates the major path-way in the present study. substrates to C-19 steroids with "normal," hyper- droepiandrosterone not involving C-21 intermedi- plastic, and adenomatous tissue slices. The data ates, pregnenolone and 1 7a-hydroxypregnenolone, from Tables VI to IX suggest the probable bio- has been suggested by Dorfman (50). synthetic sequence of androgen production shown The formation of dehydroepiandrosterone from in Figure 2. pregnenolone and 17a-hydroxypregnenolone is The biosynthesis of dehydroepiandrosterone closely associated with the hypothesis that andro- from 17a-hydroxypregnenolone with adenomatous stenedione may arise from these steroids via an al- and hyperplastic adrenal slices (Table VII) con- ternative pathway without the intermediate pro- firms previous observations of Lipsett and Hok- gesterone (Figure 2). Since a 3,L8-ol dehydroge- felt (24). Their results, however, seem to indi- nase system and double-bond shift are required cate that 17a-hydroxypregnenolone is not the im- for the conversion of the /'\-3,8-ol steroids to the portant "intra-adrenal" precursor of dehydroepi- corresponding A'-3-one derivatives (51, 52), stud- androsterone in the "normal" adrenal gland slice. ies were designed to investigate this hypothesis. The in vitro data in Table VII and VIII show The data in Tables VI to VIII indicate that adult that adrenal tissues from three patients with Cush- adrenal tissues convert a higher percentage of the ing's syndrome are capable of this conversion, sub- A'-3,8-ol substrate, 17a-hydroxypregnenolone, than stantiating in vivo observations of this pathway in the A4-3-one precursor, progesterone, to the prod- a patient with metastatic adrenal carcinoma (48).13 uct androstenedione. A common intermediate Although this pathway is now well documented, to both these reactions may be 17a-hydroxyan alternative pathway from cholesterol to dehy- progesterone as seen in Figure 2. Also, 17a- 13 Recently, Roberts, Vande Wiele, and Lieberman (49) hydroxypregnenolone can be converted to an- demonstrated that only 0.5 per cent of 17a-hydroxypreg- drostenedione via dehydroepiandrosterone. Vil- nenolone was converted to dehydroepiandrosterone in a lee, Engel, Loring, and Villee observed that patient with a virilizing adrenal adenoma, strongly sug- gesting the adrenal rather than peripheral origin of this fresh human fetal adrenal tissues formed andro- androgen. stenedione from progesterone 4-C14 early in gesta- 76 GEORGE L. COHN AND PATRICK J. MULROW tion but not after 12 to 15 weeks gestational age The results of substrate conversions with (53). These investigators suggested the 21-hy- adrenal slice incubations must be interpreted with droxylation appeared to reach a maximum at 12 reservation. Tissue membrane permeability, to 15 weeks gestation with concomitant diminished availability of intracellular co-factors, endogenous 17-hydroxylation and disappearance of andro- pools of steroid intermediates and products, and stenedione. These findings are opposite to those enzymatic activity are limiting factors not con- of Solomon, Lanman, Lind, and Lieberman, who trolled by the tissue slice method. Further stud- described 17a-hydroxyprogesterone and andro- ies are thus necessary with purified human adrenal stenedione conversion from progesterone-4-C14 enzyme preparations to elucidate the conditions with frozen fetal adrenal tissues of 10 to 22 weeks which regulate the pathways of pregnenolone me- gestation (25). Differences in co-factor additions tabolism to the corticosteroids, or androgens, or (Solomon added diphosphopyridine nucleotide both. and adenosine triphosphate and Villee, triphospho- pyridine nucleotide and glucose 6-phosphate) and SUMMARY the status of the tissues may account for the differ- These studies demonstrate that fresh human ences of the results. Although 17a-hydroxypro- adult "normal," atrophic, "hypertensive," hyper- gesterone is present in normal and hyperplastic plastic, adenomatous, and carcinomatous tissue adult adrenal tissues (7, 28), the factors regulating slices release dehydroepiandrosterone, androstene- subsequent 21-hydroxylation of 17a-hydroxy- dione, and 11,8-hydroxyandrostenedione into the progesterone to 11-deoxycortisol (Compound S) media after three hours of incubation in Krebs- compared to side-chain cleavage to androstenedione Ringer bicarbonate buffer, fortified with glucose. may determine the predominant direction of the The addition of ACTH to the system increases pathway. When a mixture of tritium-labeled 17a- 11,8-hydroxyandrostenedione release in all tissues. hydroxypregnenolone and C14-labeled progesterone ACTH did not stimulate dehydroepiandrosterone are incubated with hyperplastic adrenal slices of and androstenedione release. A.D. (Cushing's syndrome, Table VIII), the In addition, data are presented for the pathwaxs H3: C14 ratios of androstenedione (15.8: 1) com- of adrenal androgen synthesis in "normal," hyper- pared to cortisol (1.3: 1) (28) are observed. The plastic, and adenomatous tissues. The data sug- high H3: C14 in androstenedione, despite the fact gest that adrenal androgens are converted pre- that much less androstenedione was produced com- dominantly from 17a-hydroxypregnenolone -> de- pared to cortisol, suggests that the major propor- hydroepiandrosterone -- androstenedione -* 11,3- tion of this androgen and 11,8-hydroxyandro- hydroxyandrostenedione rather than from proges- stenedione are probably derived via a pathway terone -* androstenedione -* 11,8-hydroxyandro- which does not utilize progesterone, presumably stenedione. by way of 17a-hydroxypregnenolone > DHEA * androstenedione -* 11,8-hydroxyandrostenedione. ACKNOWLEDGM ENT However, further studies are necessary to deter- The authors wish to express their appreciation to Mr. mine the significance of 17a-hydroxyprogesterone Albert Kuljian, Mr. Walter Hirniak, Miss Eleanor as an intermediate for adrenal androgen biosyn- Laime, and Miss Vera Dunne for technical assistance. thesis. In addition, the high H3: C14 ratio of 11/3-hy- REFERENCES droxyandrostenedione (22: 1) compared to cor- 1. Berliner, M. L., Berliner, D. L., and Dougherty, tisol (28) implies that "intra-adrenal" side-chain T. F. Metabolism of progesterone by adrenal cleavage of cortisol is an unlikely pathway for tissue from patients with Cushing's syndrome and 11,8- mammary carcinoma. J. clin. Endocr. 1958, 18, hydroxyandrostenedione formation. The data 109. indicate that 11-hydroxylation of androstenedione 2. Davignon, J., Koiw, E., Nowaaczynski, W., Trem- is the more plausible explanation for the biosyn- blay, G., and Genest, J. Studies on the in vitro production of corticosteroids by adrenal glands from thesis of 11,1/-hydroxyandrostenedione in the tis- normotensive individuals and hypertensive patients. sues studied. Canad. J. Biochem. 1962, 40, 285. HUMAN ADULT ADRENAL ANDROGENS 77

3. Cooper, D. Y., Touchstone, J. C., Roberts, J. M., 18. Vande Wiele, R., and Lieberman, S. The metabolism Blakemore, W. S., and Rosenthal, 0. Steroid of dehydroepiandrosterone in Biological Activities formation by adrenal tissue from hypertensives. of Steroids in Relation to Cancer, G. Pincus and J. clin. Invest. 1958, 37, 1524. E. P. Vollmer, Eds. New York, Academic Press, 4. Bongiovanni, A. M. In vitro hydroxylation of ster- 1960, p. 93. oids by whole adrenal homogenates of beef, nor- 19. Bradlow, H. L., and Gallagher, T. F. Metabolism mal man, and patients with adrenogenital syn- of llp-hydroxy-A4-androstene-3,17-dione in man. drome. J. clin. Invest. 1958, 37, 1342. J. biol. Chem. 1957, 229, 505. 5. Bailey, E. E., Slade, C. I., Lieberman, A. H., and 20. Migeon, C. J. Androgens in human plasma in Luetscher, J. A., Jr. Steroid production by hu- Hormones in Human Plasma, H. N. Antoniades, man adrenal adenomata and nontumorous adrenal Ed. Boston, Little, Brown, 1960, p. 297. tissue in vitro. J. clin. Endocr. 1960, 20, 457. 21. Engel, L. L., and Langer, L. J. Biochemistry of 6. Dyrenfurth, I., Lucis, 0. J., Beck, J. C., and Ven- steroid hormones in Annual Review of Biochem- ning, E. H. Studies in patients with adrenocor- istry, J. M. Luck, Ed. Palo Alto, Annual Re- tical hyperfunction. III. In vitro secretion of ster- views, 1961, vol. 30, p. 499. oids by human adrenal glands. J. clin. Endocr. 22. Bloch, E., Dorfman, R. I., and Pincus, G. The con- 1960, 20, 765. version of acetate to C,9 steroids by human adrenal 7. Lombardo, M. E., and Hudson, P. B. The bio- gland slices. J. biol. Chem. 1957, 224, 737. synthesis of adrenocortical hormones by the human adrenal gland in vitro. Endocrinology 1959, 23. Goldstein, N., Gut, M., Dorfman, R. I. Conversion 65, 417. of pregnenolone to dehydroepiandrosterone. Bio- 8. Grant, J. K., Symington, T., Duguid, W. P. Effect chim. biophys. Acta (Amst.) 1960, 38, 190. of adrenocorticotrophic therapy on the in vitro 24. Lipsett, M. B., and H6kfelt, B. Conversion of 17a- 1lp-hydroxylation of desoxycorticosterone by hu- hydroxypregnenolone to cortisol. Experientia man adrenal homogenates. J. clin. Endocr. 1957, (Basel) 1961, 17, 449. 17, 933. 25. Solomon, S., Lanman, J. T., Lind, J., and Lieberman, 9. Mulrow, P. J., and Cohn, G. L. Conversion of cor- S. The biosynthesis of A'-androstenedione and 17a- ticosterone-4-C'4 to aldosterone by human adrenal hydroxyprogesterone from progesterone by sur- slices. Proc. Soc. exp. Biol. (N. Y.) 1959, 101, viving human fetal adrenals. J. biol. Chem. 1958, 731. 233, 1084. 10. Ayres, P. J., Eichhorn, J., Hechter, O., Soba, N., 26. Bloch, E., and Benirschke, K. Synthesis in vitro of Tait, J. F., and Tait, S. A. S. Some studies on steroids by human fetal adrenal gland slices. J. the biosynthesis of aldosterone and other adrenal biol. Chem. 1959, 234, 1085. steroids. Acta endocr. (Kbh.) 1960, 33, 27. 27. Mulrow, P. J., Cohn, G. L., and Kuljian, A. Con- 11. Mulrow, P. J., and Cohn, G. L. Corticosteroid re- version of 17-hydroxypregnenolone to cortisol by lease and synthesis in vitro by adrenal slices from normal and hyperplastic human adrenal slices. J. patients with Cushing's syndrome. J. clin. Invest. clin. Invest. 1962, 41, 1584. 1961, 40. 1250. 28. Hayes, M. A., and Goldenberg, I. S. Operative treat- 12. Plantin, L. O., Diczfalusy, E., and Birke, G. Iso- ment of adrenal cortical hyperfunctioning diseases lation of dehydroepiandrosterone from an adrenal (suppl.). Ann. Surg. 1961, 154, 33. cortical tumour. Nature (Lond.) 1957, 179, 421. 29. Norymberski, J. K. Determination of urinary cor- 13. Anliker, R., Rohr, O., and Marti, M. Untersuch- ticosteroids. Nature (Lond.) 1952, 170, 1074. ungen jiber Organextrakte. t~ber den Nachweis 30. Lakshmanan, T. K., and Lieberman, S. An improved von Testosteron in einem virilsierenden Neben- method of gradient elution chromatography and nierenrindentumor. Helv. chim. Acta 1956, 39, its application to the separation of urinary keto- 1100. steroids. Arch. Biochem. 1954, 53, 528. 14. Romanoff, E. B., Hudson, P., and Pincus, G. Iso- 31. Jailer, J. W., Vande Wiele, R., Christy, N. P., and lation of hydrocortisone and corticosterone from Lieberman, S. Studies in Cushing's syndrome. human adrenal vein blood. J. clin. Endocr. 1953, III. Urinary 17-ketosteroids in patients with bi- 13, 1546. lateral adrenal cortical hyperplasia. J. clin. In- 15. Bush, I. E., and Mahesh, V. B. Adrenocortical hy- vest. 1959, 38, 357. perfunction with sudden onset of hirsutism. J. 32. Philip, B. A., Cohn, G. L., and Bondy, P. K. Corti- Endocr. 1959, 18, 1. sol secretion rates and intermediary metabolism 16. Lombardo, M. E., McMorris, C., and Hudson, P. B. in normal volunteers and in patients with adrenal The isolation of steroidal substances from human dysfunction. Trans. Ass. Amer. Phycns 1961, 74, adrenal vein blood. Endocrinology 1959, 65, 426. 163. 17. Hirschmann, H., DeCourcy, C., Levy, R. P., and 33. Bush, I. E. Methods of paper chromatography of Miller, K. L. Adrenal precursors of urinary 17- steroids applicable to the study of steroids in mam- ketosteroids. J. biol. Chem. 1960, 235, PC 48. malian blood and tissues. Biochem. J. 1952, 50, 370. 78 GEORGE L. COHN AND PATRICK J. MULROW 34. Kliman, B., and Peterson, R. E. Double isotope studies of a virilizing tumor in the testis. J. clin. derivative assay of aldosterone in biological ex- Invest. 1960, 39, 534. tracts. J. biol. Chem. 1960, 235, 1639. 45. Lieberman, S., Katzenellenbogen, E. R., Schneider, 35. Carstensen, H., and Oertel, G. N., Eik-Nes, K. B. R., Studer, P. E., and Dobriner, K. Isolation of Secretion of 17a-hydroxy-A'-pregnenolone by the urinary steroids after cortisone and adrenocorti- canine adrenal gland during stimulation with adreno- cotropic hormone. J. biol. Chem. 1953, 205, 87. corticotropin. J. biol. Chem. 1959, 234, 2570. 46. Touchstone, J. C., Glazer, L., Cooper, D. Y., and 36. Cohn, G. L., Bondy, P. K., and Castiglione, C. Roberts, J. M. The isolation of A4-androstene-1llp- Studies on pyrogenic steroids. I. Separation, iden- ol-3,17-dione from human adrenal incubates. J. tification, and measurement of unconjugated dehy- clin. Endocr. 1955, 15, 382. droepiandrosterone, etiocholanolone, and andros- 47. Cohn, G. L., Mulrow, P. J., and Dunne, V. C. The terone in human plasma. J. clin. Invest. 1961, 40, in vitro synthesis of dehydroepiandrosterone-sul- 400. fate by an adrenal adenoma in The Program of 37. Bush, I. E. and Willoughby, M. The excretion of the Forty-Fourth Meeting of the Endocrine So- allotetrahydrocortisol in human urine. Biochem. ciety, June, 1962, p. 48. J. 1957, 67, 689. 48. Solomon, S., Carter, A. C., and Lieberman, S. The 38. Burton, R. B., Zaffaroni, A., and Keutmann, E. H. conversion in vivo of 17-a-hydroxypregnenolone to Paper chromatography of steroids. II. Corticos- dehydroisoandrosterone and other 17-ketosteroids. teroids and related compounds. J. biol. Chem. 1951, J. biol. Chem. 1960, 235, 351. 188, 763. 49. Roberts, K. D., Vande Wiele, R. L., Lieberman, S. 39. Abelson, D., and Bondy, P. K. Fluorometric de- 17a-Hydroxypregnenolone as a precursor of urinary termination of A4-3-ketosteroids. Arch. Biochem. steroids in a patient with a virilizing adenoma of 1955, 57, 208. the adrenal. J. clin. Endocr. 1961, 21, 1522. 40. Bondy, P. K., Abelson, D., Scheuer, J., Tseu, T. K. L., 50. Dorfman, R. I. A system for evaluating the func- and Upton, V. Determination of cortisol in hu- tional status of the adrenal cortex. Metabolism man plasma by quantitative paper chromatography. 1961, 10, 902. J. biol. Chem. 1957, 224, 47. 51. Samuels, L. T. Studies of the enzymes involved in the 41. Rubin, B. L., Dorfman, R. I., and Pincus, G. A synthesis and degradation of the hormones of the method for the quantitative analysis of 17-keto- adrenal cortex in Ciba Foundation Colloquium on steroid mixtures. J. biol. Chem. 1953, 203, 629. Endocr., W. Klyne, G. E. W. Wolstenholme, and 42. Migeon, C. J., and Plager, J. E. Neutral 17-keto- M. P. Cameron, Eds. Boston, Little, Brown, 1953, steroids in human plasma in Recent Progress in vol. 7, p. 176. Hormone Research, G. Pincus, Ed. New York, 52. Rubin, B. L., Leipsner, G., Deane, H. W. A rapid, Academic Press, 1954, vol. 9, p. 235. sensitive assay procedure for adrenal steroid-3p8-ol- 43. Way, E. L., and Bennett, B. M. Rapid estimation dehydrogenase activity. Endocrinology 1961, 69, of theoretical counter-current distribution values. 619. J. biol. Chem. 1951, 192, 335. 53. Villee, D. B., Engel, L. L., Loring, J. M., and Villee, 44. Savard, K., Dorfman, R. I., Baggett, B., Fielding, C. A. Steroid hydroxylation in human fetal L. L., Engel, L. L., McPherson, H. T., Lister, adrenals: formation of 16a-hydroxyprogesterone, L. M., Johnson, D. S., Hamblen, E. C., and Engel, 17-hydroxyprogesterone, and deoxycorticosterone. F. L. Clinical, morphological and biochemical Endocrinology 1961, 69, 354.